--------------------------------------------------------------------------------



                                                                               
                                                                          Final
                       Assessment of Occupational Exposure to Formaldehyde From
                                                        Composite Wood Products
                                                                               

                                       
                                                                               
                                                                               
                                                                               
                                           U.S. Environmental Protection Agency
                                      Office of Pollution Prevention and Toxics
                                                   	Chemical Engineering Branch
                                                   	1200 Pennsylvania Avenue NW
                                                          	Washington, DC 20460
                                                                               
                                                                               
                                                                               
                                                                               
                                                                               
                                                                               
                                                                 6 October 2011
--------------------------------------------------------------------------------

                                                                               

                                       
Executive Summary
Introduction

This assessment of the occupational exposure to formaldehyde from composite wood products was developed to support EPA rulemaking under the Formaldehyde Standards for Composite Wood Products Act (FSCWPA).
In July 2010, President Obama signed into law the FSCWPA.  This law is an amendment to TSCA and directs EPA to regulate the formaldehyde emissions from hardwood plywood (HWPW), medium-density fiberboard (MDF), and particleboard (PB).  FSCWPA sets formaldehyde limits equal to the California Air Resources Board (CARB) Airborne Toxic Control Measure (ATCM) promulgated in 2007.  As part of the development of the rule under FSCWPA, EPA is conducting a cost-benefit analysis of implementing the formaldehyde limits cited by FSCWPA.  In support of this cost-benefit analysis, EPA assessed occupational exposure to formaldehyde during the life cycle of HWPW, PB, and MDF.
Purpose and Scope

The purpose of the assessment is, first, the prediction of baseline formaldehyde inhalation and dermal occupational exposures that are due to all sources of exposure (aggregate exposure) during the life cycle of composite wood products (CWPs), and, second, the determination of the effects on baseline exposures of the analytical options for limits on emission levels for off-gassing from HWPW, PB, and MDF.
The baseline occurs in the year in which EPA will promulgate limits under FSCWPA, which will be 2013.  The analytical options consist of the following limits on off-gassing emissions for HWPW, PB, and MDF:
Option 1: emission limits equal to the CARB ATCM Phase 1 limits;
Option 2: emission limits equal to the CARB ATCM Phase 2 limits (which are equal to the limits specified in FSCWPA); and
Option 3: emission limits equal to the no-added formaldehyde (NAF) limits specified in FSCWPA.
The assessed life cycle stages include the following:
      1. Manufacture of HWPW, PB and MDF within the United States (domestic manufacturing);
      2. Manufacture of products constructed with domestically-manufactured and imported HWPW, PB and MDF (secondary products) such as the manufacture of furniture (also referred to as composite wood product fabrication);
      3. Wholesale and retail of domestically-manufactured and imported HWPW, PB, and MDF and downstream products; and
      4. Office building occupancy and associated use of office furniture fabricated with domestically-manufactured and imported HWPW, PB, and MDF.

The following life cycle stages were not assessed due to data gaps in information needed for their assessment: the commercial use of HWPW, PB, and MDF and their secondary products in construction work, and the recycling and disposal of HWPW, PB, and MDF and their secondary products.
Approach

Inhalation exposure monitoring data were compiled for all of the assessed life cycle stages but were not deemed to be adequate for assessment purposes.  Time-weighted average (TWA) inhalation exposure monitoring data for sites in which the manufacture, fabrication, wholesale and retail of the three composite wood products likely occurs were compiled from the OSHA Integrated Management Information System (IMIS) database.  Inhalation exposure monitoring data for office buildings were compiled from EPA's Building Assessment Survey and Evaluation (BASE) study, which was conducted from 1994 through 1998.  As explained in Section 3.2, these monitoring data were deemed not adequate for the assessment of baseline exposures and the effects of the analytical options on baseline exposures, which were assessed using mathematical modeling where possible.  The compiled monitoring data are presented for perspective only.
Baseline average inhalation exposures in composite wood fabrication sites and in office buildings and the effects of analytical options on these inhalation exposures were assessed using mathematical models for the estimation of indoor air concentrations in combination with estimates of exposure duration (the hours of exposure per day) and frequency (the days of exposure per year).  For composite wood fabrication, exposure duration and frequency were estimated using census data on pre-baseline aggregate production hours.
Inhalation exposures in the manufacturing, wholesale, and retail life cycle stages were assessed qualitatively due to gaps in data needed for mathematically modeling indoor air concentration.
Dermal exposures to formaldehyde in liquid mixtures were quantitatively assessed using EPA dermal exposure models for CWP manufacturing for pre-baseline conditions only due to data gaps.  Dermal exposures to CWPs are not quantifiable using the EPA dermal exposure models for liquids and solids and, therefore, dermal exposure was not assessed for the CWP fabrication, wholesale, retail and office occupancy life cycle stages.
For all life cycle stages with the exception of occupancy of office buildings, the number of exposed workers was estimated based on aggregate census data (Census 2007a).  The number of exposed office workers was not determined due to data gaps.  The extent of use of personal protective equipment (PPE) was noted for manufacture and fabrication based on a review of OSHA inspection reports.  However, a limited number of inspection reports were available for review and a conclusion as to the prevalence of the use of respiratory protection in these life cycle stages could not be made.
Average formaldehyde indoor air concentrations in a CWP fabrication site and an office building were estimated using mathematical models that are based on the principle of mass conservation and the assumption that the work space is a well mixed box.  In these steady-state models the generation rate of formaldehyde due to off-gassing from CWPs is equated to the rate at which airborne formaldehyde is exhausted from the work space through ventilation with outdoor air to satisfy a formaldehyde mass balance on the work space.  The generation rate was expressed as the product of a formaldehyde off-gassing emission rate and an amount of emitting surface area.  The formaldehyde off-gassing emission rate was determined from average emission levels for the three composite wood products that would be domestically produced or imported at the baseline (provided in the EPA report Market Profile for Industries Potentially Affected by the Proposed Formaldehyde Standards for Composite Wood Products Rule (USEPA 2010a)) and under the analytical options (provided in the EPA report Formaldehyde from Pressed Wood Products: Exposure Assessment (USEPA 2009b)).
Method for Estimating Average Indoor Air Concentration in CWP Fabrication Sites
As discussed in Section 4.2.1, there are indications that sources of formaldehyde other than the off-gassing of formaldehyde from CWPs may exist in a CWP fabrication site.  Aggregate indoor air concentrations are the sum of the concentration contributed by off-gassing of the three CWPs, the concentration contributed by other formaldehyde sources in the workplace, and the concentration of formaldehyde in ambient outdoor air.  However, the extent to which sources in a CWP fabrication site other than off-gassing of CWP contribute to indoor air concentration in CWP fabrication sites is unknown.  Attempts to develop methods to estimate concentration contributions from other sources during CWP fabrication were not successful due to data gaps.  Therefore, only the contributions from off-gassing of the three CWPs and ventilation with outdoor ambient air were accounted for in the estimation of aggregate indoor air concentration in CWP fabrication sites.
Average indoor air concentration in a CWP fabrication site was calculated from the following mathematical model parameters:
Formaldehyde emission rate (off-gassing emissive flux) of the composite wood products;
Total emitting surface area of the composite wood products located in a CWP fabrication site;
Effective ventilation rate for an entire CWP fabrication site; and
The concentration of formaldehyde in the outdoor air for industrial land use areas.

The mathematical model is described in detail in Appendix E.  The literature search did not yield a CWP fabrication site population distribution for these mathematical model parameters, nor did it yield representative values for these parameters for CWP fabrication sites in general (with the exception of concentration of formaldehyde in the outdoor air).  Due to these data gaps with respect to the model parameters, what-if values for indoor air concentration were estimated.  This what-if analysis was accomplished by estimating the range of possible values for each model parameter (with the exception of outside air concentration) and calculating a range of what-if indoor air concentration values from the estimated ranges of the model parameters as discussed in detail in Section 3.3.1.1.  A range of what-if indoor air concentration values was calculated assuming a CWP fabrication site contains only a single CWP type.  Each model parameter (except for outside air concentration) was varied between the two bounding values of its range (the low- and high-end values) while keeping the other model parameters constant at an average, a typical, or an intermediate value.  Section 3.3.1.1 describes in detail how an average, a typical, or an intermediate value was determined for each model parameter.  In this assessment report, an intermediate value is defined as a value in-between a low-end and high-end value.  This computational procedure resulted in three different low-end indoor air concentration values and three different high-end indoor air concentration values.  The broadest range of these calculated low- and high-end values constitutes a range of values for calculated indoor air concentration for a single CWP that is intended to be a broad range that accounts for uncertainty in coinciding values for the model parameters.  An indoor air concentration value was also calculated from the average, typical, or intermediate value of each model parameter.  These calculations were repeated for each CWP type: HWPW, PB, and MDF.
The low-end values of calculated indoor air concentrations for all three CWP types (a total of nine low-end values) were combined into a single range of low-end indoor air concentration values.  The same procedure was followed for the high-end values for all three CWP types, thereby yielding a single range of high-end values.  Finally, the intermediate values of indoor air concentration for each CWP type were also combined into a single range of intermediate values.  Therefore, three ranges of indoor air concentration values (one for low-end values, one for intermediate values, and one for high-end values) were obtained that account for any mixture of the different CWP types in a CWP fabrication site.
A final range of results for indoor air concentration was obtained by combining these three ranges into a single encompassing range and a best estimate for indoor air concentration was chosen as the midpoint of this overall range as discussed in Section 4.2.5.3.  The resulting values in the estimated range for average indoor air concentration at a CWP fabrication site are characterized as what-if results because each concentration value in this range corresponds to combination of choices, or what-if values, for the model parameters.  Although the prevalence of the what-if values for each model parameter at actual CWP fabrication sites is unknown, the ranges of the model parameter values should be realistic such that any given choice, or what-if value, for any model parameter represents a possible value at actual CWP fabrication sites.
Method for Estimating Average Indoor Air Concentration in Office Buildings
Indoor air concentration in office buildings was estimated using a mathematical model for estimation of indoor air concentration resulting from off-gassing emissions from office workstations fabricated with the three CWPs.  This mathematical model was applied to two different office workstation types: open plan workstations and private office workstations.  Average indoor air concentration in an office building was calculated for each workstation type from the following mathematical model parameters:
Emitting surface area of panels in the workstation;
Emitting surface area of work surfaces in the workstation;
The outdoor air ventilation rate that affects the volume of the office building containing the workstation;
Emission rate (emissive flux) of the composite wood products used to construct the workstation panels and work surfaces;
The fraction of the panels' surface area that is finished, or overlayed, and the effectiveness of this finish or overlay as a barrier to off-gassing emissions from the CWP substrate;
Fraction of the work surfaces' surface area that is finished, or overlayed, and the effectiveness of this finish or overlay as a barrier to off-gassing emissions from the CWP substrate; and
Concentration of formaldehyde in the outdoor air for commercial land use areas.

The mathematical model is described in detail in Appendix E.  Representative values for the emitting surface area of each workstation component (panel and work surface) and the work space ventilation rate were obtained from the Business and Institutional Furniture Manufacturer's Association (BIFMA) / American National Standards Institute (ANSI) M7.1-2007 standard test method.  The workstation component emission rates were estimated using CWP emission rates in combination with the extent of finishing or overlay on a workstation component and the reduction in the emission rate of the CWP due to the finish or overlay acting as a barrier to off-gassing emissions.
Information was not identified on the relative amounts of the three CWPs used in the fabrication of office workstations, representative values for the extent to which the workstation components are finished or overlayed, and the barrier effectiveness of the finish or overlay.  Due to these data gaps in information on representative values for these model parameters, what-if values for indoor air concentration were estimated.  What-if values for indoor air concentration were estimated by first estimating the range of possible values for the following parameters: extent of finishing or overlay of each workstation component and the barrier effectiveness of the finishing or overlay.  Next, these estimated ranges of parameter values were used in combination with the range of CWP emission rates obtained from the EPA reports mentioned above and the representative values for the remaining model parameters to estimate a range of what-if values for indoor air concentration in an office building due to each workstation type.  A range of what-if indoor air concentration values was calculated assuming a workstation consists of only a single CWP type.  Low and high ends of the range of what-if average formaldehyde indoor air concentrations in the building volume containing a workstation were estimated using the mathematical model and (1) the bounding values of the range of values for the model parameters for which ranges were estimated and (2) the representative values of the remaining model parameters so as to obtain a wide range of indoor air concentration.  Additionally, an intermediate what-if value of indoor air concentration was calculated using average or intermediate values for the model parameters for which ranges were estimated and the representative values of the remaining model parameters.  Section 3.3.1.2 describes in detail how an average or intermediate value was determined for each model parameter.  In this assessment report, an intermediate value is defined as a value in-between a low-end and high-end value.
These calculations were repeated for each CWP type: HWPW, PB, and MDF, thus resulting in nine values of what-if indoor air concentrations for each workstation type (a low-end, high-end, and intermediate value of what-if indoor air concentration for each of the three CWP types).  These values were then aggregated into low-end ranges, intermediate ranges, and high-end ranges to account for various mixtures of the different CWP types in each office workstation type.  The best estimate concentration was calculated as the average of the following two values: the midpoint of the intermediate range for open plan workstations and the midpoint of the intermediate range for private office workstations as discussed in Section 4.5.4.1.
These calculated indoor air concentrations are initial steady-state concentrations resulting from off-gassing of office workstations fabricated from new CWPs.  Temporal decay in emission rates from the CWPs was not accounted for due to data gaps in information necessary for the modeling of the temporal decay of CWP emission rates in office buildings.
Results

The results for each assessed life cycle stage are summarized below.
Results for Manufacturing

At baseline, approximately all domestic production of HWPW, MDF, and PB will be in compliance with the CARB ATCM Phase 2 standard.  Only small fractions of HWPW and PB domestic production will not be in compliance with the CARB ATCM Phase 2 standard at baseline.  Hence, the CARB Phase 1 and FSCWPA/CARB Phase 2 analytical options will result in a negligible effect on baseline exposure levels for manufacturing.  Since approximately all domestic production of CWPs are expected to meet the CARB ATCM Phase 2 standard at baseline, only the NAF analytical option will affect baseline exposures.
NAF resins do not contain any formaldehyde.  Therefore, mills that use NAF resins in compliance with NAF standards should achieve a reduction in formaldehyde exposures due to the elimination of the formaldehyde content of resins.  However, wood naturally emits formaldehyde, and it is possible that workers in mills that use only NAF resins may still experience some formaldehyde exposures due to the natural emission of formaldehyde from wood.  Other sources of formaldehyde can exist for mills that apply value-added surfaces.  However, it is uncertain how many mills apply value-added surfaces.
Dermal exposures to workers during manufacturing can occur during unloading of liquid UF resin.  For the purpose of this assessment, EPA assumes a formaldehyde concentration of 0.59% in the UF resin.  The EPA/OPPT 2-Hand Dermal Contact with Liquid Model estimates a dermal exposure of 590 to 1,800 mg/day for exposure to a chemical at 100% concentration.  For the assumed formaldehyde concentration of 0.59%, the model estimates worker exposure of 3.5 to 10.6 mg/day.  This results in an estimated dose of 0.05 to 0.15 mg/kg/day for an average 70 kg worker.  It is uncertain how the free formaldehyde content of CWP resins will change as manufacturers adopt the CARB ATCM limits.  Therefore, both baseline dermal exposures and the effect of analytical options on baseline dermal exposures are uncertain and cannot be estimated.
Results for CWP Fabrication

The results of the assessment of indoor air concentrations in CWP fabrication sites at the baseline and the effect of the analytical options on baseline indoor air concentrations are presented in Table ES-1 through Table ES-4.  Table ES-2 provides a summary of the estimates of average indoor air concentration in a CWP fabrication site due to off-gassing from the three CWP types as a range of values and as a single best estimated value for the baseline and the analytical options.  Tables ES-3 and ES-4 provide average daily concentration (ADC) and lifetime average daily concentration (LADC) values, respectively, that correspond to the values given in Table ES-2.  The results show that the maximum formaldehyde indoor air concentration decreases as lower (more stringent) emission standards are implemented.  However, the CARB ATCM Phase 1 analytical option would result in little or no reduction in exposure from the baseline year because most composite wood products are expected to be already meeting more stringent standards by the baseline year (2013).  The minimum indoor air concentration is identical for all of the different analytical options because a small portion of the market is expected to be already meeting the NAF standards at the baseline.  The emission levels of NAF-compliant composite wood products were assumed to remain unchanged and not be affected by the analytical options.
Results for Wholesale and Retail

The formaldehyde concentration resulting from off-gassing is expected to decrease with the implementation of more stringent regulations.  However, as discussed previously, CARB ATCM Phase 1 analytical option would result in little or no reduction in exposure from the baseline year.  This effect is because most composite wood products in the market would already be meeting more stringent limits by 2013.
Results for Office Buildings

The results of the assessment of indoor air concentrations in office buildings at the baseline and the effect of the analytical options on baseline indoor air concentrations are presented in Table ES-5 through Table ES-8.  Table ES-5 provides for the baseline and each of the analytical options the three ranges of estimated indoor air concentration in an office building for each of the two types of office furniture workstations.  Tables ES-6 and ES-7 provide average daily concentration (ADC) and lifetime average daily concentration (LADC) values, respectively, that correspond to the values given in Table ES-5.  Table ES-8 presents the best estimated indoor air concentration values from among the results presented in Table ES-5.  Similarly as with the CWP fabrication results, the office building results show that the maximum formaldehyde indoor air concentration decreases as lower (more stringent) emission standards are implemented.  Again, the minimum indoor air concentrations remain constant throughout the different analytical options due to the presence of NAF-compliant products at the baseline.  The office building results in Table ES-8 show the most significant reduction in indoor air concentration from the NAF emission limits.


Table ES-1. Estimated Low-End, Intermediate, and High-End Ranges of Values for the Site-Average Indoor Air Concentration for CWP Fabrication at the Baseline and for the Analytical Options
                                     Case
                              Intermediate (ppm)
                                 Low End (ppm)
                                High End (ppm)
Baseline 
                               0.0095  -  0.0181
                               0.0055  -  0.0092
                               0.0218  -  0.3934
Phase 1
                      0.0077 (-18.8%)  -  0.0176 (-2.7%)
                        0.0054 (-3.2%)  -  0.0092 (0%)
                      0.0111 (-49.3%)  -  0.3787 (-3.7%)
FSCWPA / Phase 2
                      0.0072 (-23.5%)  -  0.0175 (-3.2%)
                        0.0053 (-4.0%)  -  0.0092 (0%)
                      0.0084 (-61.5%)  -  0.3763 (-4.3%)
NAF
                      0.0064 (-32.0%)  -  0.0110 (-39.3%)
                        0.0052 (-5.5%)  -  0.0092 (0%)
                      0.0064 (-70.5%)  -  0.1807 (-54.1%)

Table ES-2. Estimated Overall Range and Best Estimate for the Site-Average Indoor Air Concentration for CWP Fabrication at the Baseline and for the Analytical Options
                                     Case
                              Overall Range (ppm)
                                Best Estimate 
                             Concentration1 (ppm)
                                Reduction from 
                                 Baseline (%)
Baseline 
                               0.0055  -  0.3934
                                    0.1995
                                      --
Phase 1
                               0.0054  -  0.3787
                                    0.1920
                                     -3.7
FSCWPA / Phase 2
                               0.0053  -  0.3763
                                    0.1908
                                     -4.3
NAF
                               0.0052  -  0.1807
                                    0.0930
                                     -53.4
                        [1] The best estimate concentration is the midpoint of the overall range. The overall range is defined as the low-end value from the low-end range and the high-end value from the high-end range. The reduction from baseline, ADC, and LADC values are calculated from the best estimate concentration value.

Table ES-3. Overall Range and Best Estimate ADC for CWP Fabrication at the Baseline and for the Analytical Options
                                     Case
                           Overall Range (ug/m[3])
                         Best Estimate ADC (ug/m[3])
Baseline 
                                1.556  -  110.4
                                     56.0
Phase 1
                                1.506  -  106.3
                                     53.9
FSCWPA / Phase 2
                                1.494  -  105.6
                                     53.5
NAF
                                1.471  -  50.71
                                     26.1

Table ES-4. Overall Range and Best Estimate LADC for CWP Fabrication at the Baseline and for the Analytical Options
                                     Case
                           Overall Range (ug/m[3])
                         Best Estimate LADC (ug/m[3])
Baseline 
                               0.8894  -  63.08
                                     32.0
Phase 1
                               0.8607  -  60.72
                                     30.8
FSCWPA / Phase 2
                               0.8536  -  60.34
                                     30.6
NAF
                               0.8407  -  28.98
                                     14.9

Table ES-5. Summary of Formaldehyde Indoor Air Concentrations for Office Buildings at the Baseline and for the Analytical Options
                            Open Plan Workstations
                                     Case
                              Intermediate (ppm)
                                 Low End (ppm)
                                High End (ppm)
Baseline 
                                0.0265 - 0.0732
                                0.0035 - 0.0053
                                0.2002 - 1.0089
Phase 1
                       0.0167 (-37.0%) - 0.0705 (-3.7%)
                           0.0035 (0%) - 0.0053 (0%)
                       0.0729 (-63.6%) - 0.2833 (-71.9%)
FSCWPA / Phase2
                       0.0143 (-46.0%) - 0.0701 (-4.2%)
                           0.0035 (0%) - 0.0053 (0%)
                       0.0414 (-79.3%) - 0.167 (-83.4%)
NAF
                       0.0099 (-62.6%) - 0.0346 (-52.7%)
                           0.0035 (0%) - 0.0053 (0%)
                       0.0184 (-90.8%) - 0.0728 (-92.8%)
                          Private Office Workstations
                                     Case
                              Intermediate (ppm)
                                 Low End (ppm)
                                High End (ppm)
Baseline 
                                0.013 - 0.0332
                                0.003 - 0.0036
                                0.0719 - 0.3555
Phase 1
                        0.0087 (-33.1%) - 0.032 (-3.6%)
                           0.003 (0%) - 0.0036 (0%)
                       0.0273 (-62.0%) - 0.1011 (-71.6%)
FSCWPA / Phase2
                       0.0077 (-40.8%) - 0.0318 (-4.2%)
                           0.003 (0%) - 0.0036 (0%)
                       0.0163 (-77.3%) - 0.0603 (-83.0%)
NAF
                       0.0058 (-55.4%) - 0.0165 (-50.3%)
                           0.003 (0%) - 0.0036 (0%)
                       0.0082 (-88.6%) - 0.0273 (-92.3%)
                  Note: Values in parentheses are the percent reductions in the indoor air concentration from the baseline value.


Table ES-6. Summary of Formaldehyde Exposure Concentrations for Office Buildings as Average Daily Concentrations at the Baseline and for the Analytical Options
                            Open Plan Workstations
                                     Case
                            Intermediate (ug/m[3])
                              Low End (ug/m[3])
                              High End (ug/m[3])
Baseline 
                                 7.44 - 20.54
                                  0.98 - 1.49
                                 56.17 - 283.1
Phase 1
                                 4.69 - 19.78
                                  0.98 - 1.49
                                 20.46 - 79.49
FSCWPA / Phase 2
                                 4.01 - 19.67
                                  0.98 - 1.49
                                 11.62 - 46.86
NAF
                                  2.78 - 9.71
                                  0.98 - 1.49
                                 5.16 - 20.43
                          Private Office Workstations
                                     Case
                            Intermediate (ug/m[3])
                              Low End (ug/m[3])
                              High End (ug/m[3])
Baseline 
                                  3.65 - 9.32
                                  0.84 - 1.01
                                 20.17 - 99.75
Phase 1
                                  2.44 - 8.98
                                  0.84 - 1.01
                                 7.66 - 28.37
FSCWPA / Phase 2
                                  2.16 - 8.92
                                  0.84 - 1.01
                                 4.57 - 16.92
NAF
                                  1.63 - 4.63
                                  0.84 - 1.01
                                  2.30 - 7.66

Table ES-7. Summary of Formaldehyde Exposure Concentrations for Office Buildings as Lifetime Average Daily Concentrations at the Baseline and for the Analytical Options
                            Open Plan Workstations
                                     Case
                            Intermediate (ug/m[3])
                              Low End (ug/m[3])
                              High End (ug/m[3])
Baseline 
                                 4.25 - 11.74
                                  0.56 - 0.85
                                 32.1 - 161.8
Phase 1
                                  2.68 - 11.3
                                  0.56 - 0.85
                                 11.69 - 45.42
FSCWPA / Phase 2
                                 2.29 - 11.24
                                  0.56 - 0.85
                                 6.64 - 26.78
NAF
                                  1.59 - 5.55
                                  0.56 - 0.85
                                 2.95 - 11.67
                          Private Office Workstations
                                     Case
                            Intermediate (ug/m[3])
                              Low End (ug/m[3])
                              High End (ug/m[3])
Baseline 
                                  2.08 - 5.32
                                  0.48 - 0.58
                                11.53  -  57.00
Phase 1
                                  1.39 - 5.13
                                  0.48 - 0.58
                                 4.38 - 16.21
FSCWPA / Phase 2
                                  1.23 - 5.10
                                  0.48 - 0.58
                                  2.61 - 9.67
NAF
                                  0.93 - 2.65
                                  0.48 - 0.58
                                  1.31 - 4.38


Table ES-8. Summary of Best Estimate Results for Office Buildings at the Baseline and for the Analytical Options
                                     Case
                      Best Estimate Concentration1 (ppm)
                          Reduction from Baseline (%)
                                 ADC (ug/m3)
                                 LADC (ug/m3)
Baseline 
                                    0.0365
                                      --
                                     10.23
                                     5.85
Phase 1
                                    0.0320
                                     -12.3
                                     8.97
                                     5.13
FSCWPA / Phase 2
                                    0.0310
                                     -15.1
                                     8.69
                                     4.97
NAF
                                    0.0167
                                     -54.2
                                     4.69
                                     2.68
                              [1] The best estimate concentration is the average of the following two values: the midpoint of the intermediate range for open plan workstations and the midpoint of the intermediate range for private office workstations. The reduction from baseline, ADC, and LADC values are calculated from the best estimate concentration value.

Uncertainties

There is uncertainty in the values of the parameters of the mathematical model for the estimation of average indoor air concentration in a CWP fabrication site.  In particular, as discussed in Section 3.3.1.1, there is considerable uncertainty in the range of values for emitting surface area, and some uncertainty in the range of values for effective ventilation rate.  Therefore, there is uncertainty in the calculated range of baseline indoor air concentration.  It is not possible to determine the direction of uncertainty (i.e., whether the estimated values for indoor air concentration are an underestimate or an overestimate).  Furthermore, as discussed above, if sources other than off-gassing from CWPs are present in a CWP fabrication site, then the calculated indoor air concentrations would not be representative of aggregate indoor air concentrations.
The composition of composite wood products used in office furniture was not determined and, therefore, panels and work surfaces of office workstations were assumed to be entirely composed of CWPs.  Furthermore, the type and extent of coverage by finishing or overlay materials on the CWPs used to fabricate office workstations were also not determined, and a range of values was assumed for these parameters.  This assumption introduces uncertainty into the calculated indoor air concentrations.  Furthermore, the mathematical model for the estimation of indoor air concentrations in an office building simplifies the complex nature of office building ventilation systems.  Typically in these systems, outdoor air is mixed with a recirculation stream of indoor air prior to being supplied to the various ventilation zones.  The return air from these ventilation zones is mixed and then a stream is split for recirculation and mixing with the outdoor air while the balance is exhausted.  Therefore, the office environment model neglects the effect of one exposure zone on another resulting from the recirculation of indoor air or the movement of air directly between exposure zones without passing through the ventilation system.  The uncertainty in the results due to neglecting the effect of one exposure zone on another is uncertain and cannot be quantified without a detailed, rigorous mathematical model with detailed supporting monitoring data.  It is not possible to determine the direction of uncertainty in the estimated results for indoor air concentration in office buildings.


                               Table of Contents

1.	Introduction	1
2.	Purpose and Scope of Exposure Assessment	1
2.1	Manufacturing	9
2.2	Composite Wood Product Fabrication	9
2.3	Wholesale	10
2.4	Retail	10
2.5	Commercial Use	10
2.6	Disposal and Recycle	10
3.	Exposure Assessment Method	11
3.1	Literature Search	12
3.2	Compilation of Pre-Baseline Exposure Monitoring Data	13
3.3	Exposure Assessment Methods and Assumptions	17
3.3.1	Estimation of Indoor Air Concentrations Related to Baseline and Analytical Options for CWP Fabrication and Office Buildings	18
3.3.1.1	Method for Estimation of Indoor Air Concentration for CWP Fabrication	19
3.3.1.2	Method for Estimation of Indoor Air Concentration for Office Buildings	27
3.3.2	Estimation of Dermal Exposures to Formaldehyde in Liquid Resin	37
4.	Exposure Assessment By Life Cycle Stage	37
4.1	Exposure Assessment for Manufacturing	37
4.1.1	Description of Life Cycle Stage	37
4.1.2	Production Volumes and Number of Establishments, Workers, Production Hours, and Work Days	41
4.1.3	Pre-Baseline Exposure Monitoring Data	43
4.1.4	Respiratory Protection Usage During Manufacturing	47
4.1.5	Baseline Exposure and the Effect of Analytical Options on Baseline Exposure for Manufacturing	48
4.1.6	Dermal Exposures during Manufacturing	49
4.2	Exposure Assessment for Composite Wood Product Fabrication	50
4.2.1	Description of Life Cycle Stage	50
4.2.2	Number of Establishments, Workers, Production Hours, and Work Days		53
4.2.3	Pre-Baseline Exposure Monitoring Data	57
4.2.4	Respiratory Protection Usage During Composite Wood Product Fabrication	57
4.2.5	Baseline Exposure and the Effect of Analytical Options on Baseline Exposure for Composite Wood Product Fabrication	59
4.2.5.1	Results for Indoor Air Concentration Calculation Trials and Sensitivity Analysis	59
4.2.5.2	Estimation of a Range of Values for the What-If Low-End and the What-If High-End Indoor Air Concentrations that Account for All CWP Types	60
4.2.5.3	Final Results for the Estimation of What-If Indoor Air Concentrations	61
4.2.5.4	The Assessment of Exposure Level for the Baseline and the Analytical Options	63
4.2.6	Dermal Exposures during Composite Wood Product Fabrication	69
4.3	Exposure Assessment for Wholesale	69
4.3.1	Description of Life Cycle Stage	69
4.3.2	Use Volumes and Number of Establishments, Employees, Employed Hours, and Work Days	69
4.3.3	Pre-Baseline Exposure Monitoring Data	70
4.3.4	Baseline Exposure and the Effect of Analytical Options on Baseline Exposure for Wholesale	71
4.3.5	Dermal Exposures during Wholesale	71
4.4	Exposure Assessment for Retail	71
4.4.1	Description of Life Cycle Stage	71
4.4.2	Use Volumes and Number of Establishments, Employees, Employed Hours, and Work Days	72
4.4.3	Pre-Baseline Exposure Monitoring Data	73
4.4.4	Baseline Exposure and the Effect of Analytical Options on Baseline Exposure for Retail	75
4.4.5	Dermal Exposures during Retail	75
4.5	Exposure Assessment for Office Workers during Commercial Use	75
4.5.1	Description of Life Cycle Stage	75
4.5.2	Use Volumes and Number of Establishments, Employees, Employed Hours, and Work Days	75
4.5.3	Pre-Baseline Exposure Monitoring Data	76
4.5.4	Baseline Exposure and the Effect of Analytical Options on Baseline Exposure for Office Buildings	77
4.5.4.1	Determination of a Best Estimate for Indoor Air Concentration	77
4.5.5	Dermal Exposures during Commercial Use	81
5.	Uncertainties In Assessment Results	82
5.1	Uncertainties in Assessment Results for Composite Wood Product Manufacturing, Fabrication, Wholesale, Retail, and Office Buildings	82
5.1.1	Uncertainty in the Estimates of Indoor Air Concentration for Composite Wood Product Fabrication	82
5.1.2	Uncertainty in the Estimates of Indoor Air Concentration for Office Buildings	83
5.2	Uncertainties in Pre-Baseline Monitoring Data	84
5.2.1	OSHA IMIS Data	84
5.2.2	BASE Study Monitoring Data	92
5.3	Uncertainty in Number of Production Workers and Employees	93
6.	References	94

Appendix A: Definition of Terms
Appendix B: Data Quality Acceptance Criteria Specifications
Appendix C: Existing Formaldehyde Occupational Exposure Limits and Formaldehyde Off-Gassing Emission Standards for Composite Wood Products
Appendix D: Formaldehyde CWP Off-Gassing Emission Level Data for the Pre-Baseline Period, Baseline, and the Analytical Options
Appendix E: Modeling Methodology for Estimating Indoor Air Concentrations Related to Baseline and Analytical Options
Appendix F: Pre-Baseline Formaldehyde Exposures from OSHA IMIS by Worker Activity and Industry Sector for Composite Wood Product Fabricators
Appendix G: Summary of OSHA Inspection Reports
Appendix H: Summary of Additional Exposure Monitoring Data from Manufacturing
Appendix I: Model Results for Indoor Air Concentrations at CWP Fabrications Sites using Pre-Baseline Emission Rates
                                List of Tables
                                       
Table ES-1. Estimated Low-End, Intermediate, and High-End Ranges of Values for the Site-Average Indoor Air Concentration for CWP Fabrication at the Baseline and for the Analytical Options	viii
Table ES-2. Estimated Overall Range and Best Estimate for the Site-Average Indoor Air Concentration for CWP Fabrication at the Baseline and for the Analytical Options	viii
Table ES-3. Overall Range and Best Estimate ADC for CWP Fabrication at the Baseline and for the Analytical Options	viii
Table ES-4. Overall Range and Best Estimate LADC for CWP Fabrication at the Baseline and for the Analytical Options	ix
Table ES-5. Summary of Formaldehyde Indoor Air Concentrations for Office Buildings at the Baseline and for the Analytical Options	ix
Table ES-6. Summary of Formaldehyde Exposure Concentrations for Office Buildings as Average Daily Concentrations at the Baseline and for the Analytical Options	x
Table ES-7. Summary of Formaldehyde Exposure Concentrations for Office Buildings as Lifetime Average Daily Concentrations at the Baseline and for the Analytical Options	xi
Table ES-8. Summary of Best Estimate Results for Office Buildings at the Baseline and for the Analytical Options	xi
Table 2-1.  NAICS Industry Sectors Identified for the Life Cycle Stages of HWPW, MDF, and PB	6
Table 3-1.  Pre-Baseline Formaldehyde Emission Levels for Domestically-Produced Composite Wood Products	15
Table 3-2. Summary of What-If Method for the Estimation of the What-If Range of Average Indoor Air Concentrations of Formaldehyde at CWP Fabrication Sites	21
Table 3-3. Matrix of Ventilation Rates and Mixing Factors	22
Table 3-4. Estimated Values for Parameters in the Mathematical Model for Average Indoor Air Concentration at a CWP Fabrication Site for the Baseline and the Analytical Options	26
Table 3-5. Data Sources for the Estimation of Parameters of the Mathematical Model for Average Indoor Air Concentration at a CWP Fabrication Site	27
Table 3-6. Summary of What-if Method for the Estimation of the What-if Range of Average Indoor Air Concentrations of Formaldehyde in Workstation Volumes	30
Table 3-7. Comparison of ASTM E1333 Parameter Values with Typical Office Environment Properties	33
Table 3-8. Estimated Values for Parameters in the Mathematical Model for Average Indoor Air Concentration Resulting from Emissions from Private Office and Open Plan Workstations for the Baseline and Analytical Options	34
Table 4-1.  U.S. Hardwood Plywood Production Data and Number of Mills	42
Table 4-2.  U.S. Reconstituted Wood Product Production Data and Number of Mills	43
Table 4-3.  Summary of 2007 Economic Census Data and Exposure Frequency and  Duration Calculated from These Data for U.S. Manufacturing Mills	43
Table 4-4.  Summary of Formaldehyde Inhalation Monitoring Data for Hardwood Plywood Manufacturing Mills from the OSHA IMIS Database	44
Table 4-5.  Summary of Formaldehyde Inhalation Monitoring Data for Reconstituted Wood Product Manufacturing Mills from the OSHA IMIS Database	45
Table 4-6.  Summary of 2007 Economic Census Data and Exposure Frequency and Duration Calculated from These Data for Composite Wood Product Fabricating Sites	54
Table 4-7.  Summary of All IMIS TWA Personal Monitoring Data for Industries Within the CWP Fabricator Life Cycle Stage	57
Table 4-8. Results for the Calculation Trials for the What-If Low- and High-End Baseline Indoor Air Concentrations	64
Table 4-9. Results for the Calculation Trials for the What-If Low- and High-End CARB Phase 1 Indoor Air Concentrations	65
Table 4-10. Results for the Calculation Trials for the What-If Low- and High-End FSCWPA / CARB Phase 2 Indoor Air Concentrations	66
Table 4-11. Results for the Calculation Trials for the What-If Low- and High-End NAF Indoor Air Concentrations	67
Table 4-12.  Estimated Low-End, Intermediate, and High-End Ranges of Values for the Site-Average Indoor Air Concentration at the Baseline and for the Analytical Options	67
Table 4-13. Estimated Overall Range and Best Estimate Site-Average Indoor Air Concentration at the Baseline and for the Analytical Options	68
Table 4-14. Overall Range and Best Estimate ADC at the Baseline and for the Analytical Options	68
Table 4-15. Overall Range and Best Estimate LADC at the Baseline and for the Analytical Options	68
Table 4-16.  Summary of 2007 Economic Census Data for Wholesale Facilities	70
Table 4-17. Summary of Formaldehyde Inhalation Monitoring Data for Wholesalers from the OSHA IMIS Database	71
Table 4-18.  Summary of 2007 Economic Census Data for Retail Facilities	73
Table 4-19.  Summary of Formaldehyde Inhalation Monitoring Data for Retailers from the OSHA IMIS Database	74
Table 4-20.  Overview of General Characteristics of the 100 BASE Buildings	76
Table 4-21.  Summary of Formaldehyde Inhalation Monitoring Data for Office Occupational Settings and Outdoor Measured Concentrations from the BASE Study	76
Table 4-22.  Results for Calculation of What-If Indoor Air Formaldehyde Concentrations for Each Analytical Option	79
Table 4-23.  Combined (for All CWP) Results: Calculated What-If Indoor Air Concentrations at the Baseline and for the Analytical Options	80
Table 4-24.  Combined (for All CWP) Results: Calculated What-If Indoor Air ADCs at the Baseline and for the Analytical Options	80
Table 4-25.  Combined (for All CWP) Results: Calculated What-If Indoor Air LADCs at the Baseline and for the Analytical Options	81
Table 4-26.  Best Estimate of What-If Formaldehyde Exposure Concentration at the Baseline and for the Analytical Options	81
Table 5-1.  Number of OSHA Inspections at CWP Fabrication Facilities Compared to the Maximum Number of Facilities from the U.S. Economic Census	85

                                List of Figures
                                       
Figure 2-1. Life Cycle and Market Distribution of HWPW	3
Figure 2-2.  Life Cycle and Market Distribution of MDF	4
Figure 2-3.  Life Cycle and Market Distribution of PB	5
Figure 4-1. Comparison of Model Results for Ranges of What-If Indoor Air Concentrations using Pre-Baseline Emission Rates with OSHA IMIS Data	62
Figure 5-1.  TWA Exposures from OSHA IMIS for a Single Inspection for Armstrong Cabinets (NAICS 337110)	88
Figure 5-2.  TWA Exposures from OSHA IMIS for a Single Inspection for Mid Continent Cabinetry (NAICS 337110)	88
Figure 5-3.  TWA Exposures from OSHA IMIS for a Single Inspection for Benvenuti and Stein (NAICS 337122)	89
Figure 5-4.  TWA Exposures from OSHA IMIS for a Single Inspection for Ameriwood Industries (NAICS 337122)	89
Figure 5-5.  TWA Exposures from OSHA IMIS for Two Inspections for The Gunlocke Company (NAICS 337211)	90
Figure 5-6.  TWA Exposures from OSHA IMIS for Two Inspections for Silver Street (NAICS 337211)	90
Figure 5-7.  TWA Exposures from OSHA IMIS for a Single Inspection for The Commercial Furniture Group (NAICS 337127)	91

                             List of Abbreviations

ADC				Average daily concentration
C&D				Construction and demolition
CARB P1			California Air Resources Board Formaldehyde Regulation Phase 1
CARB P2			California Air Resources Board Formaldehyde Regulation Phase 2
CFM				Cubic feet per minute
CWP				Composite wood product
FSCWPA			Formaldehyde Standards for Composite Wood Products Act
HWPW			Hardwood plywood
HWPW-CC			Hardwood plywood  -  composite core
HWPW-VC			Hardwood plywood  -  veneer core
LADC				Lifetime average daily concentration
MDF				Medium-density fiberboard
MF				Melamine formaldehyde
MSW				Municipal solid waste
NAF				No-added formaldehyde
OSHA				Occupational Safety and Health Administration
PB				Particleboard
PF				Phenol formaldehyde
pMDI				Poly(methylene diphenyl diisocyanate)
PPE				Personal protective equipment
PVA				Poly(vinyl acetate)
tMDF				Thin medium-density fiberboard
TWA				Time-weighted average
UF				Urea formaldehyde
ULEF				Ultra low emitting formaldehyde



Introduction
In March 2008, the Environmental Protection Agency (EPA) was petitioned by the Sierra Club and other signatories under Section 21 of the Toxic Substances Control Act (TSCA) to adopt nationally an Airborne Toxic Control Measure (ATCM) promulgated by the California Air Resources Board (CARB) in 2008 that establishes maximum formaldehyde emissions from the three types of composite wood products: hardwood plywood (HWPW), medium-density fiberboard (MDF), and particleboard (PB).  As described in EPA's Advance Notice of Proposed Rulemaking (ANPR) published in December 2008 (73 FR 73620) and in response to this petition, EPA investigated whether regulatory action, including possible action under TSCA Section 6, or non-regulatory action might be appropriate to control the levels of formaldehyde emitted from composite wood products.
In July 2010, President Obama signed into law the Formaldehyde Standards for Composite Wood Products Act (FSCWPA).  This law is an amendment to TSCA and directs EPA to regulate the formaldehyde emissions from HWPW, MDF, and PB.  FSCWPA sets formaldehyde limits equal to the CARB ATCM standards.  It also requires EPA to promulgate implementing regulations by January 1, 2013.  As part of the development of a rule under FSCWPA, EPA is conducting a cost-benefit analysis of implementing the formaldehyde limits cited by FSCWPA.  In support of this cost-benefit analysis, EPA assessed occupational inhalation and dermal exposure to formaldehyde during the life cycle of HWPW, PB, and MDF.
Purpose and Scope of Exposure Assessment
The purpose of the assessment is, first, the prediction of baseline formaldehyde inhalation and dermal occupational exposure that is due to all sources of exposure (aggregate exposure) during the life cycle of composite wood products (CWPs), and, second, the determination of the effect on baseline exposure of the analytical options for limits on emission levels for off-gassing from HWPW, PB, and MDF.
The baseline occurs in the year in which EPA will promulgate limits under FSCWPA, which will be 2013.  This document discusses various analytical options for limits on emission levels for off-gassing from HWPW, PB, and MDF (i.e., various emission standards) simply for informational purposes.  The breadth of the discussion here should not imply that EPA necessarily has corresponding flexibility in implementing the statute.  The following specific analytical options were assessed in support of the cost-benefit analysis for the formaldehyde rulemaking activity:
Analytical Option 1: a scenario in which EPA would implement maximum formaldehyde emission limits equal to the CARB ATCM Phase 1 limits;
Analytical Option 2: a scenario in which EPA would implement maximum formaldehyde emission limits equal to the CARB ATCM Phase 2 limits (which are equal to the limits specified in FSCWPA); and
Analytical Option 3: a scenario in which EPA would implement maximum formaldehyde emission limits equal to the no-added formaldehyde (NAF) limits specified in FSCWPA.
Figure 2-1, Figure 2-2, and Figure 2-3 depict the life cycle and market distribution for HWPW, MDF, and PB, respectively.  Manufacture, import, and export volumes are expressed in cubic meters of composite wood products and are annual volumes for the year 2005 (Table 2-5 of USEPA 2010a).  Market distributions are expressed as percentages that represent the amounts of North American produced CWP that are sold to the various CWP fabrication industry sectors and directly to retail (section 2.3.2 of USEPA 2010a).  The market distributions of imports are unknown.  Percentages may not sum to 100 percent due to rounding.  The description of the life cycles, i.e., the markets and market channels, were obtained from the Hardwood Plywood and Veneer Association (HPVA) (for HWPW) and the Composite Panel Association (CPA) (for PB and MDF) (HPVA 2009a; CPA 2008a).
This assessment assesses potential worker exposure to formaldehyde from the following life cycle stages of HWPW, PB and MDF:
Manufacture of HWPW, PB and MDF within the United States (domestic manufacturing);
Downstream manufacture of products constructed from domestically manufactured and imported HWPW, PB and MDF (e.g. manufacture of furniture using HWPW, PB and MDF), referred to in this report as composite wood product fabrication;
Wholesale and retail of domestically manufactured and imported HWPW, PB, MDF, and downstream products fabricated from these CWPs; and
Occupancy by office workers of office buildings containing domestically manufactured and imported HWPW, PB, MDF, and downstream products fabricated from these CWPs.
The life cycle stages not assessed due to major data gaps include the following:
Commercial construction using HWPW, PB, MDF, and products fabricated from these CWPs; and
Recycling and disposal of HWPW, PB, MDF, and products fabricated from these CWPs.
The life cycle stages that are assessed as well as those that were not assessed are further discussed below in Sections 2.1 through 2.6.  Table 2-1 summarizes the applicable NAICS codes for the assessed life cycle stages.


                                       
Figure 2-1. Life Cycle and Market Distribution of HWPW

                                       
Figure 2-2.  Life Cycle and Market Distribution of MDF
                                       
Figure 2-3.  Life Cycle and Market Distribution of PB

Table 2-1.  NAICS Industry Sectors Identified for the Life Cycle Stages of HWPW, MDF, and PB
                               Life Cycle Stage
                             Included NAICS Codes
                   Corresponding NAICS Industry Descriptions
HWPW Manufacturing
                                                                         321211
Hardwood veneer & plywood manufacturing
MDF Manufacturing
                                                                         321219
Reconstituted wood product manufacturing
PB Manufacturing


Composite Wood Product Fabrication
                                                                         321911
Wood window & door manufacturing

                                                                         321918
Other millwork (including flooring)

                                                                         321991
Manufactured home (mobile home) manufacturing

                                                                         321999
All Other Miscellaneous Wood Product Manufacturing

                                                                         336213
Motor Home Manufacturing

                                                                         336214
Travel Trailer & Camper Manufacturing

                                                                         337110
Wood kitchen cabinet & counter top manufacturing

                                                                         337121
Upholstered household furniture manufacturing

                                                                         337122
Non-upholstered wood household furniture manufacturing

                                                                         337124
Metal Household Furniture Manufacturing

                                                                         337127
Institutional furniture manufacturing

                                                                         337129
Wood television, radio, & sewing machine cabinet manufacturing

                                                                         337211
Wood office furniture manufacturing

                                                                         337212
Custom architectural woodwork & millwork manufacturing

                                                                         337214
Office Furniture (except Wood) Manufacturing 

                                                                         337215
Showcase, partition, shelving, & locker manufacturing

                                                                         339950
Sign Manufacturing
Wholesale
                                                                         423210
Furniture merchant wholesaler

                                                                         423220
Home furnishing merchant wholesaler

                                                                         423310
Lumber, plywood, & wood panel merchant wholesaler

                                                                         423320
Brick, stone, & construction material merchant wholesaler

                                                                         423330
Roofing, siding, & insulation material merchant wholesaler

                                                                         423390
Other construction material merchant wholesaler

                                                                         423440
Other commercial equipment merchant wholesalers

                                                                         423450
Medical, dental, & hospital equipment & supplies merchant wholesalers

                                                                         423490
Other professional equipment & supplies merchant wholesaler

                                                                         423510
Metal service centers and other metal merchant wholesalers

                                                                         423610
Elec. equip. and wiring merchant wholesalers

                                                                         423620
Electric appliance, TV & radio merchant wholesalers

                                                                         423710
Hardware merchant wholesalers

                                                                         423720
Plumbing equip. merchant wholesalers

                                                                         423730
HVAC equip. merchant wholesalers

                                                                         423740
Refrigeration equipment and supplies merchant wholesalers

                                                                         423830
Industrial machinery & equipment merchant wholesalers

                                                                         423850
Service establishment equipment & supp merchant wholesalers

                                                                         423910
Sporting & recreational goods & supplies merchant wholesalers

                                                                         424610
Plastics materials and basic forms and shapes merchant wholesaler

                                                                         424950
Paint, varnish, and supplies merchant wholesalers
Retail
                                                                         441210
Recreational vehicle dealers

                                                                         442110
Furniture stores

                                                                         442210
Floor covering stores

                                                                         442291
Window treatment stores

                                                                         442299
All other home furnishings stores

                                                                         443111
Household appliance stores

                                                                         443112
Radio, television, & other electronics stores

                                                                         443120
Computer & software stores

                                                                         444110
Home centers

                                                                         444120
Paint and wallpaper stores

                                                                         444130
Hardware stores

                                                                         444190
Other building material dealers

                                                                         444210
Outdoor power equipment stores

                                                                         444220
Nursery, garden center, & farm supply stores

                                                                         445110
Supermarkets and other grocery (except convenience) stores

                                                                         448110
Men's clothing stores

                                                                         448120
Women's clothing stores

                                                                         448130
Children's and infants' clothing stores

                                                                         448140
Family clothing stores

                                                                         448150
Clothing accessories stores

                                                                         448190
Other clothing stores

                                                                         451120
Hobby, toy, and game stores

                                                                         451130
Sewing, needlework, and piece goods stores

                                                                         452111
Department stores (except discount department stores)

                                                                         452112
Discount department stores

                                                                         452910
Warehouse clubs and supercenters

                                                                         452990
All other general merchandise stores

                                                                         453110
Florists

                                                                         453210
Office supplies and stationery stores

                                                                         453220
Gift, novelty, and souvenir stores

                                                                         453920
Art dealers

                                                                         453930
Manufactured (mobile) home dealers

                                                                         453998
All other miscellaneous store retailers (except tobacco stores)

                                                                         454113
Mail-order houses

                                                                         454390
Other direct selling establishments
   Source: EPA market profile (USEPA 2010a)


Manufacturing
The term "manufacturing" in this report refers specifically to the manufacture of HWPW, MDF, and PB (and is also referred to as "primary manufacturing").  The scope of this assessment includes the domestic manufacture of these products themselves but excludes any part of the upstream supply chain for the manufacture of these products such as resin or veneer manufacturing.  Occupational exposures of formaldehyde in UF resin manufacture is covered under the "Draft Final Occupational Exposure and Environmental Release Assessment of Formaldehyde in Urea-Formaldehyde Resin Manufacture and of Substitute Technologies to Reduce Emissions from Composite Wood Products" (USEPA 2011a).  The manufacture of imports, i.e., the manufacture of HWPW, MDF, and PB in countries outside of the U.S., is not included in the scope of this assessment.
The NAICS code 321219 - Reconstituted Wood Product Manufacturing includes both the manufacture of PB and MDF.  It also includes the manufacture of composite wood products not included in the scope of this assessment, such as hardboard, oriented strandboard, and waferboard (Census 2007a).
The NAICS code 321211 - Hardwood Veneer and Plywood Manufacturing includes the manufacture of hardwood plywood.  It also includes the manufacture of hardwood veneer (Census 2007a).  These veneers are typically not manufactured at the same facilities that manufacture hardwood plywood.  Therefore, hardwood veneer manufacturing facilities would not be subject to any requirements under the proposed FSCWPA rule (USEPA 2010a).  Therefore, hardwood veneer manufacturing facilities are not included in the scope of this assessment.
Composite Wood Product Fabrication
HWPW, MDF, and PB may be converted into articles or other products.  For example, HWPW may be pressed into panels and plywood components (e.g., curved HWPW, seat backs, chair arms, etc.) or used for furniture, cabinets, architectural millwork, paneling for commercial buildings, flooring, store fixtures and doors (OAQPS 2000b).  Similarly, MDF and PB are used in the manufacture of many products including virtually every application where interior panels or strips are required, especially in furniture and cabinetry.  However, MDF is not used as floor underlayment or decking except on special request (Kirk-Othmer 2011a).  In addition to furniture and cabinetry, PB is used as floor decking in manufactured homes, underlayment in conventional homes, shelving, countertops, door cores, stair stepping, door frames, and many other products that require small flat parts as a starting point.  A small portion of PB is also extruded or molded into furniture parts, paper roll plugs, brush bases and toilet seats (Kirk-Othmer 2011a).
Composite wood products also may have "value-added surfaces" added to the panel product.  "Value-added surfaces" include decorative finishes or overlays and coatings, which may be applied at a "sister" facility of the manufacturer or at unaffiliated finishing sites (CPA 2007a; CPA 2009a).  A technical bulletin published by the Composite Panel Association (CPA) further discusses the types of value-added surfaces (CPA 2007a).
Wholesale
Wholesalers buy products from manufacturers in bulk for the purpose of resale to various industries and individuals.  Wholesale products may include CWP fabrication products as well as composite wood products (i.e., panels).  A single wholesaler may sell many different types of composite wood products (Census 2007a).
Retail
Retailers operate from fixed locations and sell products for direct consumption by the purchaser (individuals or businesses).  These retailers may receive HWPW, MDF, or PB directly or as components in CWP fabrication products.  Like the wholesale sector, the retail sector is not distinguished by composite wood product type (Census 2007a).
Commercial Use
The commercial use of composite wood products includes (1) the use of composite wood products in on-site construction, and (2) the occupancy by office workers of workplace buildings (commercial buildings) containing composite wood products.  CWP fabrication products such as furniture and cabinets, in addition to composite wood products, are present in commercial buildings.
Data gaps for on-site construction using HWPW, PB, and MDF include the following:
The volumes of HWPW, PB, and MDF used during on-site construction could not be determined.
The number of sites and workers engaged in on-site construction could not be determined.  Therefore, the size of this life cycle stage and the number of workers exposed during this life cycle stage are unknown.
Formaldehyde monitoring data for on-site construction could not be determined.  Therefore, the range of potential exposures that occur during this life cycle stage is unknown.
Due to these data gaps, this assessment does not include an occupational exposure assessment of formaldehyde during on-site construction.
Disposal and Recycle
At end of life, composite wood is either disposed of as municipal solid wastes (MSW) or construction and demolition (C&D) debris materials, or recycled into new products.
A total of 15.84 million tons of wood MSW was generated (before recycling) in 2009 in the U.S.  Of these 15.84 million tons, 10.08 million tons were from wood pallets and wood crates (USEPA 2010b).  Wood pallets and wood crates are likely not constructed of CWPs and are outside the scope of this assessment.  Of these 15.84 million tons of wood MSW, the remaining 5.76 million tons of wood were from furniture and furnishings (USEPA 2010b).  As discussed in this report, CWPs are used in the fabrication of furniture; therefore, an unknown fraction of the 5.76 million tons of wood likely consists of CWPs.  EPA reports that the only identified recovery of materials from MSW composed of furniture and furnishings was mattress recovery (USEPA 2010b).  Of all MSW discarded in 2009 (after recovery for recycling and composting), 82% was landfilled or otherwise disposed and 18% was incinerated for energy recovery (USEPA 2010b).  Therefore, it is likely that CWPs in furniture that enters the MSW stream are not recovered and are discarded via landfilling or incineration.
C&D materials consist of the debris generated during the construction, renovation, and demolition of buildings, roads, and bridges, which can include wood materials (USEPA 2011c).  The majority of building-related C&D materials come from building demolition and renovation with the remainder from new construction (USEPA 2011c).  A report from 2003 estimated that 20% to 30% of building related C&D debris generated was wood (USEPA 2011c).  However, it is unknown what fraction of wood C&D debris consists of CWPs.
In 1995, the Department of Agriculture estimated that 3.96 million tons of wood products were recovered for recycling into miscellaneous uses (USDA 1996a).  The sources and types of the recovered wood products referred to here were not given, and it is unknown whether the recovered wood material includes CWPs.  The term "miscellaneous uses" was not explicitly defined in the text but referred to examples such as insulation, molded pulp products, hardboard, insulating paperboard, and composite wood panels.  More recently, of the 10.08 million tons of wood MSW from wood pallets and wood crates, 22% were recovered for recycling as mulch and bedding material (USEPA 2011b); however, as stated above, wood pallets and wood crates are likely not constructed of CWPs.  Several existing and developing technologies exist for producing composites using recycled wood-based fiber.  These composites include dry-formed wood composites, wet-formed structural fiber products, wood-plastic composites, and wood-inorganic composites (Hamilton 1996a).
Data gaps for the disposal and recycling of HWPW, PB, and MDF include the following:
The volumes of HWPW, PB, and MDF disposed of as waste and recycled could not be determined.
Formaldehyde monitoring data for the disposal and recycling of composite wood products could not be determined.

Due to these uncertainties, this assessment does not include an occupational exposure assessment of formaldehyde during disposal and recycling.  It is noted that due to the decay of formaldehyde emission rates from CWPs with time (see Section 3.3.1.1), CWPs can have reduced emission rates and, therefore, a diminished contribution to exposure concentrations when disposed or recycled at their end of life.
Exposure Assessment Method
This section describes the data collection efforts and the methods used for assessing occupational exposures, and their underlying assumptions.
Literature Search
This exposure assessment relied on the use of readily available data, including exposure monitoring data and mathematical modeling data.  A comprehensive literature search of readily available government, industry, and academic sources was conducted, which included sources from EPA, OSHA, NIOSH, the California Air Resources Board (CARB), SRI International, and the Organization for Economic Co-operation and Development (OECD), as well as industry websites, academic journals, and other publications.  Additionally, submissions in response to EPA's ANPR (73 FR 73620) were reviewed.  The ANPR review included 123 public submissions to Docket Number EPA-HQ-OPPT-2008-0627 and 30 public submissions to Docket Number EPA-HQ-OPPT-2008-0267.  Also reviewed were regulations, such as the National Emissions Standards for Hazardous Air Pollutants (NESHAP) for plywood and composite wood products and the Housing and Urban Development (HUD) Rule on formaldehyde emissions from certain wood products in manufactured homes (24 CFR § 3280.308).
The following specific databases and studies were searched for formaldehyde air monitoring data:
OSHA's Integrated Management Information System (IMIS);
EPA's Building Assessment Survey and Evaluation (BASE) study;
The Relationship of Indoor, Outdoor, and Personal Air (RIOPA) project; 
The International Agency for Research into Cancer (IARC) Formaldehyde Monograph (Volume 88); and
The Detroit Exposure and Aerosol Research Study (DEARS).
The literature search did not yield any formaldehyde air monitoring data that are directly relevant to this exposure assessment.  Data from OSHA IMIS and EPA's BASE study were the most relevant data obtained and were used in this assessment to provide perspective as discussed in Section 3.2.  Additionally, other exposure monitoring studies of CWP manufacturers from the literature are cited in this assessment, but data from OSHA IMIS are more comprehensive and more relevant to this exposure assessment than these additional studies as OSHA IMIS contains only U.S. data.  A summary of the abstracts of these additional CWP manufacturer studies are provided in Appendix H.  Some of these additional studies are from monitoring studies of CWP manufacturing mills outside the U.S.
The IARC study cited exposure monitoring studies for various types of settings (including offices, homes, and institutional and public buildings) in various countries (including countries in North America, Europe, South America, Asia, and Australia).  Monitoring data for U.S. office buildings from the IARC study were not used because they were not recent data (they were dated prior to 1990).  The RIOPA data were not used as the RIOPA project focused on individuals' exposures through-out their day.  Although limited information was collected to estimate an individual's exposures during their occupations, occupational exposures were not the focus of the project.  The DEARS also focused on personal, residential, and outdoor monitoring.  Occupational exposures were not the focus of DEARS.  In contrast, OSHA IMIS and EPA's BASE study are both focused on occupational exposures.  Recent data on formaldehyde concentrations in Hong Kong office buildings were reported in Wong et al. (Wong 2006a).  The minimum of the range of these concentrations is similar to the lowest value reported in the BASE study (BASE 1998a), while the maximum of the range is over twenty times greater than the maximum reported in the BASE study.  However, these Hong Kong data were not used because they may not be relevant to exposure in U.S. office buildings.  Specifically, it is unknown how determinants of formaldehyde inhalation exposure in Hong Kong office buildings compare to those in U.S. office buildings, such as sources of airborne formaldehyde including the amount of CWP present in office buildings and their formaldehyde emission rates and ventilation rates.  As compared to the above mentioned studies, the BASE study results are the most relevant to this assessment of formaldehyde inhalation exposure in office buildings and are adequate and sufficient for the purpose of providing perspective on the results of this exposure assessment.  Therefore, the BASE study results are presented in this assessment while information from the other studies is not included.
Dermal exposure was not included in data searches.  Instead, this assessment has relied solely upon mathematical modeling for dermal exposure assessments.
Appendix B of this report presents the quality criteria used during the literature search and data compilation.  These quality criteria represent the preference for selecting data.  The collected data used in this assessment meet these quality criteria.
Compilation of Pre-Baseline Exposure Monitoring Data
Baseline formaldehyde occupational exposure monitoring data are not available because the baseline occurs in the future in the year 2013.  As described in Section 3.1, a literature search for pre-baseline exposure monitoring data was conducted as part of the data gathering for this assessment.  Pre-baseline exposure monitoring data from OSHA IMIS and the EPA BASE study were compiled and evaluated for adequacy for use in the assessment of baseline exposure and the effect of analytical options on baseline exposure.
The compiled inhalation exposure monitoring data, which are described below, were not used in the assessment of exposure at the baseline or for the analytical options.  The reasons that these data were not used to assess exposure are discussed below.  However, these data are still considered useful for reasons also discussed below.
Manufacturing, Composite Wood Product Fabrication, Wholesale and Retail
Pre-baseline formaldehyde exposure monitoring data for manufacturing, fabrication, wholesale and retail facilities were obtained from the OSHA IMIS database (OSHA 2009a).  At the time the OSHA IMIS data were obtained (June 2010), the most current data were from 2009.  Of the collected IMIS data, only data from the period of 2002 to 2009 were considered for use in the assessment because some information on the production-weighted average emission levels and the range of emission levels of domestically-produced CWP during the pre-baseline period of 2002 to 2009 is available.  This information is presented in Table 3-1; as seen in this table, the production-weighted average emission levels for CWP domestically-produced during the period of 2002 to 2009 were close to the CARB Phase 1 limits, especially after 2008.
The pre-baseline OSHA IMIS inhalation exposure monitoring data were determined to not be adequate for the assessment of baseline exposure concentrations and the effect of the analytical options on baseline exposure concentrations due to the following three reasons:
   1. The extent to which exposure levels reported in OSHA IMIS will be affected by the limits on CWP formaldehyde emission rates set by FSCWPA is uncertain;
   2. The OSHA IMIS data may not be representative of exposure levels that are associated with formaldehyde off-gassing emissions from the three CWPs with emission rates equal to baseline or analytical option emission rates; and
   3. It is unknown whether the OSHA IMIS exposure monitoring data are representative of the exposure of the entire population of exposed workers during the pre-baseline period of time associated with these data.
The above three reasons are further discussed below.
First, the extent to which the sources of exposure associated with the compiled OSHA IMIS monitoring data will be affected by the FSCWPA is uncertain.  OSHA's occupational exposure limits apply to aggregate exposure and not to exposure due to any specific source.  Accordingly, the OSHA IMIS, which is described in Appendix F, does not contain information on the sources of formaldehyde that contributed to the observed exposure concentrations, which are a measure of aggregate exposure where more than one source of exposure existed.  As discussed in Section 4, sources other than the off-gassing of formaldehyde from the three CWPs may contribute to formaldehyde occupational exposure during the life cycle of CWPs.  Therefore, the compiled OSHA IMIS monitoring data may be associated with sources of exposure other than off-gassing from the three CWPs.  The contributions of any such sources to the observed inhalation exposure concentrations are not expected to be affected by the limits on CWP formaldehyde emission rates set by the FSCWPA.  Hence, EPA is not able to use the OSHA IMIS data to accurately assess the effect of analytical options on baseline exposure because the extent to which the exposure levels reported in OSHA IMIS will be affected by the CWP emission rate limits set by FSCWPA is uncertain.
Second, the pre-baseline OSHA IMIS data may not represent baseline or analytical option exposure levels.  Levels of formaldehyde in U.S. workplaces across multiple industries have decreased from 1979 to 2001 according to a 2008 study of IMIS data (Lavoue 2008a).  This decrease in formaldehyde levels could be due to OSHA's implementation of a formaldehyde PEL of 1 ppm in 1987 and implementation of the current PEL of 0.75 ppm in 1992 (Lavoue 2008a).  Manufacturers reduced formaldehyde emissions from composite wood products by 80 to 90 percent from the levels of the early 1980s (CPSC 1997a).  A decline in emission levels also occurred during the last decade as illustrated in Table 3-1.  As discussed above, this table contains emission level data for domestically-manufactured CWPs during the pre-baseline period of 2002 to 2009.  The emission level data in Table 3-1 indicate that domestic manufacturers produced national production-weighted average emission levels close to the CARB ATCM Phase 1 limits in 2002; however, the manufacturers also produced CWPs with a wide range of emission levels.  Furthermore, the emission levels of domestically-manufactured CWPs were trending towards a greater adoption of the CARB ATCM Phase 1 limit between 2002 to 2009, but a high percentage of CARB ATCM Phase 1 adoption was not expected until 2009.  Therefore, to the extent that the compiled OSHA IMIS pre-baseline inhalation exposure monitoring data are associated with off-gassing of formaldehyde from the three CWPs of interest, these data are associated with pre-baseline emission rates for these CWPs.  Therefore, the OSHA IMIS exposure levels may not be representative of exposure levels resulting from off-gassing of formaldehyde from CWPs with baseline or analytical option emission rates.
Third, it is uncertain whether the OSHA IMIS data are representative of the exposure concentrations for the entire population of exposed workers during the pre-baseline period of time in which exposure monitoring to obtain these data occurred.  For some life cycle stages, very few monitoring data were compiled.  However, even if an adequate amount of data were available, it is uncertain whether the inspected facilities in which the exposure concentrations were observed were selected randomly.  Therefore, it is unknown whether such exposure monitoring data are representative of the exposure of the entire population of exposed workers during the pre-baseline period of time associated with these data.
Due to the three considerations discussed above, the OSHA IMIS monitoring data were not used to assess baseline exposure concentrations and exposure concentrations for the analytical options.  Due to lack of adequate exposure monitoring data, baseline exposure and the effect of analytical options on baseline exposure are assessed using mathematical modeling as described in Section 3.3.  However, the monitoring data are still useful.  The monitoring data can provide a perspective as to observed formaldehyde exposures in industries.  Additionally, monitoring data obtained from OSHA IMIS provide exposure measurements by worker activity and thereby provide perspective as to the different worker activities in each industry that can result in formaldehyde exposure.
Table 3-1.  Pre-Baseline Formaldehyde Emission Levels for Domestically-Produced Composite Wood Products
                                    Timeline
                               Information Source
                                    HWPW [a]
                                     PB [b]
                                     MDF [b]
2002
CARB survey
                          Production-Weighted Average


                                   0.09 ppm
                                   0.18 ppm
                                   0.25 ppm


                     % of Production Represented in Survey


                                      73%
                                      53%
                                      83%


                           Range in Survey Responses


                               0.07 to 0.75 ppm
                               0.13 to 0.24 ppm
                               0.03 to 0.31 ppm
End of 2008 / Beginning of 2009
ANPR Submissions
75% of domestically-produced CWP were CARB Phase 1 compliant
96% of domestically-produced CWP were CARB P1 compliant
96% of domestically-produced CWP were CARB P1 compliant
Later in 2009
ANPR Submissions
90% of domestics expected to be CARB P1 compliant
                                      --
                                      --
 a  -  A submission from HPVA in May 2008 estimated that domestically-produced HWPW will be in compliance with CARB Phase 1. Phase 2 limits require new technologies, which have not been in commercial use and some only recently invented and patented. Additionally, a submission from HPVA in February 2009 estimated that 75% of North American HWPW was CARB certified and soon 90% or more will be CARB certified. "CARB certified" is assumed to mean CARB Phase 1 per the statements made in the May 2008 submission. The term "soon" is uncertain; therefore, the point in time in which 90% of domestics were expected to meet CARB Phase 1 is uncertain.
 b  -  A submission from CPA in May 2008 stated that "almost all" domestic production is expected to be compliant with CARB Phase 1 and 2. The volume related to "almost all" is uncertain. This submission also states that "eventually" this rate of compliance would be reached; the time frame associated with "eventually" is uncertain. CPA did estimate that 96% of domestics were Phase 1 compliant by the end of 2008.

Commercial Use (Office Buildings)
For office buildings, formaldehyde inhalation exposure monitoring data were obtained from the EPA Building Assessment Survey Evaluation (BASE) study.  The BASE study was a five-year study to characterize determinants of indoor air quality (IAQ).  The BASE study collected indoor air quality data from 100 randomly selected public and commercial office buildings in 37 cities in 25 states from 1994 to 1998.  The buildings were all occupied buildings randomly selected from cities with populations greater than 100,000 within ten climatic regions.  The studied buildings were primarily (73%) located in urban surroundings.  Ninety-eight percent of the buildings had mechanical ventilation and 44 percent had operable windows.  Within each building, a test space with a target population greater than 50 occupants, residing on no more than three floors, and served by no more than two air-handling units was randomly selected.  Buildings with highly publicized indoor environmental quality problems were excluded.  The data compiled through the BASE study include general information regarding the building itself, area monitoring of indoor pollutant concentrations, basic occupant statistics, occupant health symptoms and perceptions of indoor air quality, and information about the design, maintenance and operation of building heating, ventilating, and air conditioning (HVAC) systems.  A summary of the results of the BASE study is given in Section 4.5.3.
The BASE study does not contain any information on the sources of the monitored formaldehyde indoor air concentrations.  Even while assuming that CWPs were the main source of monitored formaldehyde, the BASE study inhalation exposure monitoring data are not adequate for assessing exposure concentration in this assessment because the emission levels of such CWPs in the studied buildings are unknown.  As indicated in Table 3-1 and discussed in the subsection above, the formaldehyde emission levels of new CWPs have been decreasing with time and, therefore, the emission levels of new CWPs depend on their date of manufacture.  Furthermore, formaldehyde emission levels of CWPs decay with time and, therefore, the emission level of aged CWPs depend on the emission level when manufactured and the duration of time they have been in use.  Although the BASE study was conducted in 1994 to 1998, the vintage of the office furniture (i.e., the year in which the office furniture were built), which may contain CWPs and are a likely source of formaldehyde emissions as further discussed in this document, is unknown for each studied building.  Therefore, the emission levels of any CWPs that are components of office furniture in the monitored office buildings at the time the monitoring occurred are unknown.
Exposure Assessment Methods and Assumptions
This section describes the methods used for assessing baseline exposures, which are due solely to off-gassing from CWPs and ventilation with ambient air, the effect of the analytical options on these baseline exposures, and the assumptions underlying the assessment methods.
Baseline indoor air concentration and the effect of the analytical options on baseline indoor air concentration were assessed for CWP fabrication and office buildings; the methods for these assessments are discussed below.  Baseline indoor air concentration and the effects of the analytical options on baseline indoor air concentration were not assessed quantitatively for the CWP manufacturing, wholesale and retail life cycle stages due to data gaps, and only qualitative assessments were made.
The goal of this assessment was to assess aggregate formaldehyde indoor air concentration in each assessed life cycle stage.  Aggregate indoor air concentration is the sum of the concentration contributed by off-gassing of the three CWPs, the concentration contributed by other formaldehyde sources in the workplace, and the concentration of formaldehyde in ambient outdoor air.  However, due to data gaps, this assessment only assesses formaldehyde indoor air concentration that is due to off-gassing of the three CWPs and the concentration of formaldehyde in ambient outdoor air.
For CWP fabrication sites, as discussed in Section 4.2.1, there is indication that sources of formaldehyde other than the off-gassing of formaldehyde from CWPs may exist.  However, the extent to which other sources of formaldehyde contribute to worker exposure in CWP fabrication sites is unknown.  Attempts to develop methods to estimate concentration contributions from other sources during CWP fabrication were not successful due to data gaps.
On the other hand, in homes the most significant sources of formaldehyde are likely to be CWPs made from UF resins (USEPA 2011b).  The literature search did not identify information to indicate that sources of formaldehyde in office buildings are significantly different than in homes.  Therefore, it is assumed that CWPs are the most significant source of formaldehyde in office buildings.  It is noted in this regard that the concentration of airborne formaldehyde tends to be lower in office buildings as compared to residential structures.  EPA compiled residential indoor air formaldehyde monitoring data (USEPA 2009b).  A comparison of these data with the monitoring data of the BASE study (BASE 1998a) shows that the central value of the indoor air concentrations in office buildings are for the most part lower than the central values of residences.  This observation is also indicated by a Health Effects Institute report (HEI 2007a).  However, this comparison neither supports nor detracts from the assumption made here about the predominant type of emission source in office buildings because airborne concentration is a function of not only the type of emission source, but also its quantity as well as the ventilation rate.  These last two factors are not expected to be similar for office buildings and residential structures.
Dermal exposures to formaldehyde in liquid mixtures were quantitatively assessed for manufacturing.  Dermal exposures were assessed only qualitatively for the CWP fabrication, wholesale, retail, and commercial use life cycle stages.  Due to data gaps, quantitative dermal exposure estimates for manufacturing were only made for pre-baseline conditions; baseline and analytical option dermal exposures could not be estimated.  The qualitative dermal exposure assessments for CWP fabrication, wholesale, retail, and commercial use were made for baseline and analytical options.
The use of respirators as personal protective equipment (PPE) for CWP manufacture and fabrication was considered based on the review of OSHA inspection reports.  However, a limited number of inspection reports were available for review and a conclusion as to the prevalence of the use of respiratory protection in these life cycle stages could not be determined.  OSHA guidance on safe practices in the woodworking industry was also considered and is summarized in this assessment report.
Exposure duration (the hours of exposure per day) and frequency (the days of exposure per year) for all assessed life cycle stages with the exception of office buildings were assessed based on census data on pre-baseline aggregate production hours.
The number of exposed workers in each life cycle stage with the exception of office buildings was estimated to be equal to the pre-baseline number of production workers according to census data.
Inhalation exposure results for baseline and each analytical option are presented as average daily concentration (ADC) and lifetime average daily concentration (LADC).  The equations for ADC and LADC are presented in Appendix E.  ADC and LADC are used as measures of exposure when considering non-cancer and cancer risks, respectively (USEPA 2009a).  Formaldehyde may result in both non-cancer and cancer health effects (ATSDR 1999a; ATSDR 2010a).
The assessment results are presented in Section 4.
 Estimation of Indoor Air Concentrations Related to Baseline and Analytical Options for CWP Fabrication and Office Buildings
This section discusses the mathematical models used for estimating baseline indoor air formaldehyde concentrations due to off-gassing of CWP in CWP fabrication sites and office buildings, and the effect on baseline indoor air concentrations of the FSCWPA analytical options.  As discussed in Section 3.2, the OSHA IMIS and BASE data are likely not representative of exposure concentrations at baseline or for the analytical options.  Therefore, mathematical modeling was used to estimate indoor air concentration for the baseline and the analytical options.
The mathematical models for the estimation of formaldehyde indoor air concentration in CWP fabrication sites and office buildings are presented in this section, while a detailed description of these models and their underlying assumptions is included in Appendix E.  These models are based on the principle of mass conservation.  As further discussed below, the generation rate of formaldehyde in these models is expressed as the product of a formaldehyde off-gassing emission rate and an amount of emitting surface area.  This generation rate is used in combination with the rate of ventilation of the work space area containing the CWP emitting surface area.  This section also contains a discussion of the determination of the values of these and the other model parameters of these models and the use of the models for estimating what-if values of indoor air concentration at CWP fabrication sites and office buildings.  Mathematical modeling was not applied to other life cycle stages because data were not available to estimate the model parameters for the other life cycle stages.
Method for Estimation of Indoor Air Concentration for CWP Fabrication
At steady-state, a mass balance equates the formaldehyde generation rate due to off-gassing from HWPW, PB, and MDF present at a CWP fabrication site to the rate at which airborne formaldehyde is exhausted from the CWP fabrication site through ventilation with outdoor air.  As further discussed in detail in Appendix E, the mathematical model for indoor air concentration, which is derived from the mass balance, is as follows:
					C=FxAQeffective+Coind			Eqn (1)
where:
      C	=	Average airborne formaldehyde concentration in a CWP fabrication site (ug/m[3]);
      F	=	Emission rate (off-gassing emissive flux) of formaldehyde from the composite wood product (ug/m[2]-h);
      A	=	Total emitting surface area of the composite wood products located in a CWP fabrication site (m[2]);
      Qeffective	=	Effective volumetric ventilation rate for an entire CWP fabrication site (m[3]/h); and
      Co[ind]	=	Concentration of formaldehyde in the outdoor air for industrial land use areas (ug/m[3]).

The literature search did not yield an exposure site population distribution for the input parameters of Equation 1, nor did it yield representative values for these input parameters for CWP fabrication sites.  Due to these data gaps with respect to the model parameters, what-if values for indoor air concentration were estimated.  This what-if analysis was accomplished by estimating the range of possible values for each model parameter (with the exception of outside air concentration as further explained below) and calculating a range of what-if indoor air concentration values from the estimated ranges of the model parameters as further discussed below.  The resulting values in the estimated range for average indoor air concentration at a CWP fabrication site are characterized as what-if results because each concentration value in this range corresponds to combinations of choices, or what-if values, for the model parameters.  Although the prevalence of the model parameter what-if values at actual CWP fabrication sites is unknown, the ranges of the model parameter values should be realistic such that any given choice, or what-if value, for any model parameter represents a possible value at actual CWP fabrication sites.  However, as further discussed below, the ranges of the model parameters were estimated with varying levels of accuracy.
What-If Method for Calculating Indoor Air Concentration
The ranges of possible values for the model parameters were estimated as further discussed below.  These ranges consist of low- and high-end values and average, typical, or intermediate values, which are presented in Table 3-4.  For the purpose of this assessment, "intermediate" is defined as a value in-between the low-end and the high-end estimates.  Low and high ends of the range of what-if average formaldehyde indoor air concentrations in CWP fabrication sites were estimated using Equation 1 from what-if values of the model parameters that are within the estimated ranges for these parameters and that result in a wide but realistic range of indoor air concentration.  Due to data gaps with respect to the relative amounts of the three CWPs used at fabrication sites, this what-if approach for estimating indoor air concentration was applied to three limiting cases for the type of CWP that is present at CWP fabrication sites: only HWPW, only PB, or only MDF.  The results of these three limiting cases are combined into ranges to account for mixtures of different CWP types as discussed in Section 4.2.5.
To estimate the what-if low-end indoor air concentration, three estimates of indoor air concentration are calculated, and the minimum value is used as the estimate for the what-if low-end indoor air concentration.  The three estimates are calculated by using the low-end value for one parameter and the average, typical, or intermediate value for the two other varying parameters and repeating this twice with a different parameter having a low-end value each time.  The estimate of the what-if high-end indoor air concentration was similarly calculated.  The what-if estimation method is summarized in Table 3-2.  An intermediate indoor air concentration was also calculated for each of the three limiting cases for CWP type from the typical, average, or intermediate values of the model parameters.  Due to varying uncertainty of the representativeness of the estimated typical, average, or intermediate values of the model parameters, the extent to which these calculated intermediate indoor air concentrations are representative of all exposure levels is unknown.
The model results for the range of average formaldehyde indoor air concentration at CWP fabrication sites for the three limiting cases are presented in Section 4.2.5.

Table 3-2. Summary of What-If Method for the Estimation of the What-If Range of Average Indoor Air Concentrations of Formaldehyde at CWP Fabrication Sites
                          Limiting Cases for CWP Type
   Method for Calculating What-if Low- and High-End Indoor Air Concentration

Determination of What-if Values of the Indoor Air Concentration Model Parameters that Result in a What-if Low-End Indoor Air Concentration
Determination of What-if Values of the Indoor Air Concentration Model Parameters that Result in a What-if High-End Indoor Air Concentration
A total of three cases: HWPW only, PB only, or MDF only is used at a fabrication site
What-if low-end indoor air concentration is estimated as the minimum value from amongst three low-end estimates of indoor-air concentration that are calculated from the following three combinations of parameters:
   1. High-end ventilation rate, intermediate emitting surface area, weighted-average emission rate, and average value for the ambient air concentration of formaldehyde;
   2. Low-end emitting surface area, typical ventilation rate, weighted-average emission rate, and average value for the ambient air concentration of formaldehyde;
   3. Low-end emission rate, typical ventilation rate, intermediate emitting surface area, and average value for the ambient air concentration of formaldehyde.

What-if high-end indoor air concentration is estimated as the maximum value from amongst three high-end estimates of indoor-air concentration that are calculated from the following three combinations of parameters:
   1. Low-end ventilation rate, intermediate emitting surface area, weighted-average emission rate, and average value for the ambient air concentration of formaldehyde;
   2. High-end emitting surface area, typical ventilation rate, weighted-average emission rate, and average value for the ambient air concentration of formaldehyde;
   3. High-end emission rate, typical ventilation rate, intermediate emitting surface area, and average value for the ambient air concentration of formaldehyde.

The basis for the ranges of values of the indoor air concentration model parameters is discussed below and summarized in Table 3-5.
Range of Values for Effective Ventilation Rate
Industrial ventilation rates for all industries, including those unrelated to CWP, vary from 500 to over 10,000 cubic feet per minute (cfm) with a typical value of 3,000 cfm (USEPA 1991a).  The adopted range for this analysis is 500 cfm to 10,000 cfm.  A mixing factor is applied to account for non-ideal mixing of air; the mixing factor is multiplied by the ventilation rate to estimate an effective ventilation rate.  The value of the mixing factor ranges from 0.1 (worst-case) to 1 (ideal mixing) with a typical value of 0.5.  The various combinations of ventilation rate and mixing factor are given in Table 3-3 and the resulting range of effective ventilation rate is 50 to 10,000 cfm.  The final values of effective ventilation rate selected for use in the mass balance equation are the following three values from Table 3-3: 50 cfm (low-end value), 1,500 cfm (typical value), and 10,000 cfm (high-end value).

Table 3-3. Matrix of Ventilation Rates and Mixing Factors
                            Ventilation Rates (cfm)
           Effective Ventilation Rates for each Mixing Factor (cfm)
                                       
                           Worst-Case Mixing Factor
                             Typical Mixing Factor
                           Ideal (no mixing factor)
                                       
                                      0.1
                                      0.5
                                       1
10,000 (high end)
                                     1,000
                                     5,000
                                    10,000
3,000 (typical)
                                      300
                                     1,500
                                     3,000
500 (worst case)
                                      50
                                      250
                                      500
  Source: USEPA 1991a
  cfm = cubic feet per minute

Range of Values for Emitting Surface Area
The emitting surface area of composite wood products in CWP fabrication sites is the most uncertain of the three parameters of the mathematical model.  In CWP fabrication sites, CWP panels can be stacked, in which case the emitting surface area of the stack would be the top surface of the top panel plus the summation of the surface area of the edges of each panel in the stack.  It is uncertain how many hours per day any given panel is expected to remain within a stack and how a stack of panels may grow or shrink throughout the day.  Additionally, panels are handled by workers as they process the panels, which can include sawing, sanding, routing, and assembling (OSHA 2009a; see Appendix F for a categorization of the worker activities identified in the obtained OSHA IMIS data).  During processing activities, all sides of the panel can be exposed to the air as workers handle, transfer, and process the panels.  Therefore, the emitting surface area of a panel can be equal to all surfaces and edges of the panel for some part of a single day of operation.  Finally, when the panels are assembled into a finished fabricated product, some panel surfaces can be covered due to stacking or overlapping the panels with other materials.  There is uncertainty as to the emitting surface area of panels incorporated into a final fabricated product.  Additionally, panels can be covered with decorative finishes or overlays.  However, the effect of barriers due to decorative finishes or overlays is to reduce the emission rate of the panel but not the emitting surface area (CPA 2003a).  The principle is that a finished or overlayed surface is still available for emission, but the actual emission rate is reduced due to the overlay.
Due to data gaps in information on the range of possible values for emitting surface area and due to the expected high variation in the value for this model parameter based on the above discussion, a range for this model parameter is estimated as follows.  The estimate of the emitting surface area is based on the aggregate average throughput of CWP panels per site.  Throughput is estimated from the estimated domestic consumption of CWP at baseline, the total number of CWP fabrication sites at baseline, and an estimate of the sites' operating days per year.  The baseline U.S. consumed volume of CWP for the three CWP types combined, which accounts for domestic production less exports and plus imports, is 18,045,282 m[3] (USEPA 2010a) with a total surface area of 1,895 million m[2] (assuming each CWP has a thickness of (3/4) inch and that both sides of the panel are exposed to air).  This quantity of material was divided by the total number of assessed CWP fabrication sites, 33,737 sites (USEPA 2010a), and the number of operating days (which was not determined but was assumed to be equal to the assessed frequency of exposure of 250 days per year) to obtain 224.6 m[2] per site per day, which is an estimate of the daily throughput, or amount processed daily, of all three CWP types combined (the number of sites and the frequency of exposure are given in Section 4.2.2).
This intermediate estimate of emitting surface area is predicated on the aggregate average site-specific throughput of CWP panels (assuming a (3/4) inch thickness) and that both sides of the panel throughput are available for emission.  As discussed above, it is recognized that panels processed in a single day can have less than their full two sides available for emission for part of the day.  It is additionally recognized that inventoried panels that are not processed in a given day can also contribute to formaldehyde emissions (such as from a stack of panels).  The effect of these uncertainties on the intermediate estimate cannot be quantified.  Since the estimation method of emitting surface area is highly uncertain, a range of values for emitting surface area was adopted by assuming that the low-end and high-end values are equal to 10% and 1,000%, respectively, of the intermediate estimate of emitting surface area based on throughput.  The 10% and 1,000% values of this intermediate estimate are hereinafter referred to as the low-end and high-end estimates, respectively.  The adopted range for emitting surface area including the intermediate value is given in Table 3-4.
Range of Values for Emission Rate
The emission rate is the only parameter of the mass balance equation that will be affected by the FSCWPA; therefore, values for this parameter were obtained for the baseline and for each of the analytical options.  Emission levels for the baseline and for the analytical options were obtained from the EPA market profile (USEPA 2010a) and the EPA draft exposure assessment (USEPA 2009b), respectively, and are presented in Appendix D.  Uncertainty in the values of these emission levels and their applicability to the estimation of average indoor air concentration in a CWP fabrication site is discussed below.
Representativeness of Emission Levels Reported in the EPA Market Profile
The emission level estimates for domestic and Canadian CWPs at baseline were obtained from an EPA survey of domestic CWP manufacturers.  The survey received the following response rates from the manufacturing facilities (mills) that received the survey: 85% of the 39 HWPW mills surveyed; 100% of the 20 MDF mills surveyed; and 97% of the 31 PB mills surveyed.  The mills surveyed only included mills subject to FSCWPA and did not include mills that only produce products exempted or not otherwise covered by FSCWPA.  Therefore, the survey results for domestic CWPs are likely representative of all U.S. CWP manufacturing mills subject to FSCWPA.  However, the EPA market profile notes that some survey responders only provided information about the compliance standards being achieved but not specific emission levels (USEPA 2010a).  Therefore, the surveyed population of mills is likely representative of U.S. mills subject to FSCWPA.  The emission level estimates for non-Canadian imported CWPs at baseline were estimated from data from a 2009 presentation by Professional Service Industries (PSI), a CARB third-party certifier (TPC) that mainly certifies mills located in Asia.  The mills certified by PSI are major sources of CWPs imported into the U.S. (USEPA 2010a).  The EPA market profile discusses the method in which data were obtained from the presentation and used to estimate imported CWP emission levels at baseline (USEPA 2010a).  The representativeness of the emission level values for non-Canadian imported CWP is uncertain.
Accuracy of Emission Level Values Reported in the EPA Market Profile
As discussed above, some survey responders only stated the compliance standard they would achieve and not the specific emission levels of their products; in such case, the actual emission level may be lower than the compliance standard.  All survey data were assumed to represent actual emission levels of products, and therefore these data may be an overestimate of emission level.  CARB ATCM emission standards are based on the ASTM E1333 chamber test method (ASTM 2002a).  Therefore, it is assumed that emission levels reported in the survey and provided from PSI are based on the ASTM E1333 chamber test method (see Appendix E for a description of ASTM E1333).  The emission levels from the survey and the PSI presentation are converted to emission rates in units of flux for use in the indoor air concentration mathematical model according to the equations described in ASTM E1333 (see Appendix E).  Another consideration with respect to the accuracy of the emission levels reported in the EPA market profile is the accuracy associated with the measurements made to obtain these values.  ASTM E1333 discusses the precision and bias of the large chamber test method in general.  However, ASTM E1333 does not discuss the accuracy and precision of the analytical test method recommended in ASTM E1333 for measuring formaldehyde levels in the chamber air samples: NIOSH 3500 chromotropic acid test procedure.  The accuracy and precision of this procedure is discussed in NIOSH 3500 (NIOSH 1994a).  ASTM E1333 does not discuss the overall accuracy of the test chamber method for determining accurate formaldehyde off-gassing emission rates.  Additionally, it is uncertain if chamber test results from ASTM E1333 accurately represent actual off-gassing rates in real world industrial settings, including CWP fabrication mills.  For ASTM E1333 chamber test results to accurately represent off-gassing rates in actual CWP fabrication mills, the ASTM E1333 parameters should be consistent with the actual conditions in the fabrication sites.  However, adequate data to perform this comparison were not available.  This assessment did not consider how ASTM E1333 uncertainties would impact the overall uncertainty of the assessment results.  However, it is likely that ASTM E1333 uncertainties are of small impact to the assessment results compared to the uncertainties of other mathematical model parameters.  It is also uncertain as to what extent manufacturers will actually obtain their predicted emission levels in 2013.
Applicability of Emission Level Values Reported in the EPA Market Profile to the Estimation of Average Indoor Air Concentration in a CWP Fabrication Site
An additional consideration is the decay with time of CWP formaldehyde emission levels.  Various literature sources have discussed the decay of formaldehyde emission levels with time, such as Groah 2005a, Zinn 1990a, and Gammage 1988a.  A wide range of values for the formaldehyde emission level half life for PB, MDF, and HWPW were reported in these articles.  The studies, which span a period of approximately 20 years, cite CWP emission level half lives ranging from less than three months to less than two years.  Not all tests used the same test method for determining emission level half life.  Additionally, the test method conditions may not be similar to the conditions of storage and transport between CWP manufacture and fabrication.  Therefore, the test results may not be applicable for the emission level decay experienced during storage and transport.  The initial emission level has a moderately strong influence on half life (Groah 2005a).  A higher initial emission level will result in a more rapid decay (Groah 2005a).  Therefore, half lives from older studies may not be applicable for newer or future CWPs with lower initial emission levels, which would have less rapid decays.
The available literature on formaldehyde emission level decay indicate that CWPs at CWP fabrication sites could have emission levels less than their emission levels at the time of their manufacture if the time for distribution from manufacturer to fabricator is considerable and depending on the conditions of storage and transport prior to delivery to the fabricator.  However, the residence time of CWPs at CWP fabrication sites is small and would likely not result in significant further emission level decay while the CWPs are at the fabrication site.  Formaldehyde emission level decay is not accounted for in this assessment due to data gaps in decay rates for CWPs manufactured at the baseline or under any of the analytical options.  In this assessment, formaldehyde concentration modeling at CWP fabrication sites assumes the emission levels from the EPA market profile (USEPA 2010a) are for newly-manufactured CWPs, which may overestimate the actual emission levels at CWP fabrication sites.
The range of emission levels and corresponding emission rates, including weighted-average values for the baseline and analytical options, is given in Table 3-4.  These values account for both domestically manufactured and imported CWPs.  See Appendix D for a complete list of the emission levels for baseline and each analytical option.  It is important to note that CWPs will continue to emit formaldehyde as they are overlayed and after they are overlayed.  However, emissions from overlayed panels at CWP fabrication sites were not accounted for in the estimation of indoor air concentration.  Many overlay materials serve as a barrier and reduce formaldehyde emissions from 50% to as much as 95% or higher (CPA 2003a).  Therefore, emissions from overlayed products have a diminished effect on indoor air concentrations compared to non-overlayed products at CWP fabrication sites.
Value for Ambient Air Formaldehyde Concentration
Limited statistics of ambient air formaldehyde concentrations are available in the EPA draft report Formaldehyde from Pressed Wood Products: Exposure Assessment (USEPA 2009b).  Table 3-1 of the draft exposure assessment provides ambient air formaldehyde concentrations for various land use categories.  The CWP fabrication indoor air concentration mathematical model uses the average value of the ambient formaldehyde concentration for industrial land use areas.  The average value is chosen for use in this mathematical model since the low standard deviation indicates limited variation in this parameter and the average value is representative.  This average value of 6.28 ug/m[3] is kept constant throughout the what-if calculations.
Table 3-4. Estimated Values for Parameters in the Mathematical Model for Average Indoor Air Concentration at a CWP Fabrication Site for the Baseline and the Analytical Options
                                Parameter Name
                                   Scenario
                              (Analytical Option)
                              Value of Parameter
                                       
                                       
                                     Units
                    Intermediate / Typical / Average Value
                                 Low-End Value
                                High-End Value
Emitting Surface area
All
                                     m[2]
                                     224.6
                                     22.5
                                     2,246
Ventilation Rate 
All
                                      cfm
                                   (m[3]/h)
                                     1,500
                                    (2,549)
                                      50
                                     (85)
                                    10,000
                                   (16,990)
HWPW Emission Level
Baseline
                                      ppm
                                       
                                     0.04
                                     0.013
                                     0.163

CARB Phase 1
                                       
                                     0.025
                                     0.013
                                     0.058

FSCWPA / CARB Phase 2
                                       
                                     0.021
                                     0.013
                                     0.032

NAF
                                       
                                     0.013
                                     0.013
                                     0.013
PB  Emission Level
Baseline
                                      ppm
                                       
                                     0.06
                                     0.013
                                     0.384

CARB Phase 1
                                       
                                     0.05
                                     0.013
                                     0.090

FSCWPA / CARB Phase 2
                                       
                                     0.053
                                     0.013
                                     0.057

NAF
                                       
                                     0.013
                                     0.013
                                     0.013
MDF Emission Level
Baseline
                                      ppm
                                       
                                     0.08
                                     0.024
                                     0.502

CARB Phase 1

                                     0.07
                                     0.024
                                     0.140

FSCWPA / CARB Phase 2

                                     0.073
                                     0.024
                                     0.082

NAF

                                     0.035
                                     0.024
                                     0.035
Ambient (outdoor) Air Concentration of Formaldehyde
All
                                   ug/m[3]
                                     6.28
                                Parameter Name
                                   Scenario
                              (Analytical Option)
                              Value of Parameter
                                       
                                       
                                     Units
                    Intermediate / Typical / Average Value
                                 Low-End Value
                                High-End Value
HWPW Emission Rate
Baseline
                                  ug/m[2]-h
                                     60.9
                                     18.6
                                     232.9

CARB Phase 1
                                       
                                     36.0
                                     18.6
                                     82.9

CARB Phase 2
                                       
                                     29.8
                                     18.6
                                     45.7

NAF
                                       
                                     18.6
                                     18.6
                                     18.6
PB  Emission Rate
Baseline
                                  ug/m[2]-h
                                     81.7
                                     18.6
                                     548.8

CARB Phase 1
                                       
                                     77.1
                                     18.6
                                     128.6

CARB Phase 2
                                       
                                     75.8
                                     18.6
                                     81.5

NAF
                                       
                                     18.6
                                     18.6
                                     18.6
MDF Emission Rate
Baseline
                                  ug/m[2]-h
                                     180.5
                                     56.7
                                    1186.5

CARB Phase 1

                                     173.6
                                     56.7
                                     330.9

CARB Phase 2

                                     172.5
                                     56.7
                                     193.8

NAF

                                     81.6
                                     56.7
                                     82.7
 Notes:  Table 3-5 indicates the sources of these data.  Emission rates are shown as both emission levels in units of ppm and as emission rates in units of flux in ug/m[2]-h for convenience of the reader.  Emission levels are assumed to represent the measured concentration from ASTM E1333, while emission rates in units of flux represent the emission rate value calculated from the measured concentration and the ASTM E1333 test parameters.  Appendix E describes ASTM E1333 and the conversion between emission levels in units of concentration and emission rates in units of flux.

Table 3-5. Data Sources for the Estimation of Parameters of the Mathematical Model for Average Indoor Air Concentration at a CWP Fabrication Site
                                Parameter Name
                              Intermediate Value
                                 Low-End Value
                                High-End Value
Emitting Surface Area
(see "Range of Values for Emitting Surface Area" in Section 3.3.1.1)
Aggregate average daily throughput estimated as the baseline U.S. consumed volume of CWP divided by the assessed number of fabrication sites and the assessed exposure frequency
10% of aggregate average daily throughput
1,000% of aggregate average daily throughput
Effective Ventilation Rate
(see "Range of Values for Effective Ventilation Rate" and Table 3-3 in Section 3.3.1.1)
Typical ventilation rate multiplied by typical mixing factor.
Worst-case ventilation rate multiplied by worst-case mixing factor
High-end ventilation rate assuming ideal mixing (no mixing factor applied)
HWPW, PB, or MDF Emission Rates
Baseline  Values
Production-weighted average calculated from data in Table D-3 in Appendix D
Minimum emission rate for data in Table D-3 in Appendix D
Maximum emission rate for data in Table D-3 in Appendix D

Analytical Option 1: CARB Phase 1
Production-weighted average calculated from data in Table D-4 in Appendix D
Minimum emission rate for data in Table D-4 in Appendix D
Maximum emission rate for data in Table D-4 in Appendix D

Analytical Option 2: FSCWPA/ CARB Phase 2 
Production-weighted average calculated from data in Table D-5 in Appendix D
Minimum emission rate for data in Table D-5 in Appendix D
Maximum emission rate for data in Table D-5 in Appendix D

Analytical Option 3: NAF 
Production-weighted average calculated from data in Table D-6 in Appendix D
Minimum emission rate for data in Table D-6 in Appendix D
Maximum emission rate for data in Table D-6 in Appendix D
Ambient (outdoor) Air Concentration of Formaldehyde
Average value for the industrial land use category from Table 3-1 in USEPA 2009b.

Method for Estimation of Indoor Air Concentration for Office Buildings
At steady-state, a mass balance equates the formaldehyde generation rate due to off-gassing from HWPW, PB, and MDF that are components of an office workstation to the rate at which airborne formaldehyde is exhausted from the volume of the office building containing the workstation through ventilation with outdoor air.  Based on the work by Carter and Zhang (Carter 2007a), formaldehyde indoor air concentration is estimated for two types of workstations: open plan workstations and private office workstations.  Furthermore, both workstation types contain the following components that may emit formaldehyde: panels, work surfaces, and storage external areas, but storage external areas are assumed to not have any formaldehyde emissions.
The office building mathematical model is described in greater detail in Appendix E and is displayed in Equation 2.
C=ECWPx1-BExFinpanels+ECWPx1-FinpanelsxApanels+ECWPx1-BExFinwork surfaces+ECWPx1-Finwork surfacesxAwork surfacesQ+Cocom          Eqn (2)

where:
      C	= 	Average airborne formaldehyde concentration in the volume of the office building containing the workstation (ug/m[3]);
      Apanels	= 	Emitting surface area of panels in the office workstation (m[2]);
      Awork surfaces	= 	Emitting surface area of work surfaces in the office workstation (m[2]);
      Q	= 	The rate of ventilation with outdoor air that affects the volume of the office building containing the workstation (m[3]/h);
      ECWP	=	Emissive flux (emission rate) of the composite wood product used to construct the workstation component type (ug/m[2]-h) (CWP = HWPW, PB, or MDF);
      BE	=	Barrier effectiveness of the finishing, or overlay, of the workstation component type, expressed as a unitless fraction;
      Finpanels	=	Fraction of the panels' surface area that is finished, or overlayed, with a barrier effectiveness of BE, expressed as a unitless fraction;
      Finwork surfaces	=	Fraction of the work surfaces' surface area that is finished, or overlayed, with a barrier effectiveness of BE, expressed as a unitless fraction; and
      Co[com]	=	Concentration of formaldehyde in the outdoor air for commercial land use areas (ug/m[3]).

Representative values for some of the model parameters of this mathematical model, including workstation component emitting surface area, ventilation rate, formaldehyde emission rates for each of HWPW, PB, and MDF, and outdoor air formaldehyde concentration, were obtained through a literature search.  On the other hand, information was lacking on the relative amounts of the three CWPs used in the fabrication of office workstations, representative values for the extent to which the workstation components are finished or overlayed, and the barrier effectiveness of the finish or overlay.  Due to these data gaps on representative values for these model parameters, what-if values for indoor air concentration were estimated.  What-if values for indoor air concentration were estimated by first estimating the range of possible values for the following parameters: formaldehyde emission rates for all three CWP types; extent of finishing or overlay of each workstation component; and the barrier effectiveness of the finishing or overlay.  Then, these ranges of parameter values and the representative values for the remaining model parameters were used in Equation 2 to estimate what-if values for indoor air concentration.
What-If Method for Calculating Indoor Air Concentration
The ranges of possible values for three model parameters (CWP emission rate, extent of finishing or overlay, and barrier effectiveness) were estimated as further discussed below.  These ranges consist of low- and high-end values and average or intermediate values, which are presented in Table 3-8, which also contains the representative values for the remaining model parameters.  For the purpose of this assessment, "intermediate" is defined as a value in-between the low-end and the high-end estimates.  Low and high ends of the range of what-if average formaldehyde indoor air concentrations in the workstation volumes were estimated using Equation 2 from the bounding values of the range of values for the above mentioned three model parameters and from the representative values of the remaining model parameters so as to obtain a wide range of indoor air concentration.  Due to data gaps with respect to the relative amounts of the three CWP types used in the construction of office furniture, this approach for estimating what-if indoor air concentration was applied to three limiting cases for the type of CWP of which the office furniture is constructed: only HWPW, only PB, or only MDF.  The results of the three limiting cases are combined into ranges to account for mixtures of different CWP types as discussed in Section 4.5.4.  What-if values for low-end and high-end indoor air concentration were estimated as summarized in Table 3-6.  An intermediate indoor air concentration was also calculated for each of the three limiting cases for the CWP type from the intermediate values of the model parameters.  The representativeness of the range of values for intermediate indoor air concentration (calculated from the three limiting cases for CWP type) is limited by the representativeness of the intermediate values for the extent of finishing and barrier effectiveness, which are unknown.  The intermediate values for the remaining model parameters are expected to be representative of real world values.
Table 3-6. Summary of What-if Method for the Estimation of the What-if Range of Average Indoor Air Concentrations of Formaldehyde in Workstation Volumes
                          Limiting Cases for CWP Type
   Method for Calculating What-if Low- and High-End Indoor Air Concentration

What-if Values for Emission Rate, Extent of Finish or Overlay, and Barrier Effectiveness that are used to Calculate a What-if Low-End Indoor Air Concentration [1]
What-if Values for Emission Rate, Extent of Finish or Overlay, and Barrier Effectiveness that are used to Calculate a What-if High-End Indoor Air Concentration [1]
Workstation panels and work surfaces are constructed of:
* Only HWPW;
* Only PB; or 
* Only MDF 
* The emission rate of the (unfinished) CWP is equal to the low-end value in a range of values for the emission rates for the unfinished CWP.
* Both sides of workstation panels and workstation work surface are finished or overlayed, and the barrier effectiveness of the finishing or overlay is equal to a highest value in a range of values for barrier effectiveness.
* The emission rate of the (unfinished) CWP is equal to the high-end value in a range of emission rates for the unfinished CWP. 
* Both sides of workstation panels and the lower side of the workstation work surface are unfinished (their barrier effectiveness is 0%), and the upper side of the workstation work surface is finished or overlayed and has a barrier effectiveness equal to the lowest value in a range of values for the barrier effectiveness.
[1] The representative values for the remaining model parameters (ventilation rate, workstation component emitting surface area, and outdoor air formaldehyde concentration) were used in the calculation.

The values of all of the model parameters and their bases are given in Table 3-8 below.
Values for Emitting Surface Area and Ventilation Rate
The emitting surface area of each workstation component and the ventilation rate of the work space volume containing the workstation with outdoor air were obtained from the BIFMA / ANSI M7.1-2007 standard test method (BIFMA 2007a), which is based on the work by Carter and Zhang (Carter 2007a) and their analysis of actual office building floor plans and actual office workstations.  The BIFMA / ANSI M7.1-2007 standard test method uses representative values of workstation component emitting surface areas and ventilation rates for the purpose of modeling the impact on office indoor air volatile organic compound (VOC) concentrations due to office furniture emissions.  Therefore, the values for these parameters obtained from BIFMA M7.1-2007 are treated as representative values for the purpose of this assessment.  These values are kept constant and are not varied.
Range of Values for Workstation Component Emission Rate
The BIFMA standard M7.1-2007 and the work by Carter and Zhang do not contain any information on observed formaldehyde emission rates of actual office workstations.  Carter and Zhang provide example workstation emission rates of formaldehyde but do not discuss the statistical significance of these values such as being representative, worst case, etc.  Therefore, workstation component emission rates (emission rates for finished CWPs used to construct the workstation) were estimated using information on CWP emission rates, extent of finishing, and barrier effectiveness.  These three model parameters are discussed in this subsection.
The CWP emission rate is the only parameter of the mass balance equation that will be affected by the FSCWPA; therefore, values for this parameter were obtained for the baseline and for each of the analytical options.  Emission levels for the baseline and for the analytical options were obtained from the EPA market profile (USEPA 2010a) and the EPA draft exposure assessment (USEPA 2009b), respectively, and are presented in Appendix D.  The low-end and high-end emission rates for each CWP type are chosen from the low-end and high-end values from Appendix D.  An average emission rate for each CWP type is calculated as a volume-weighted average of the domestic and imported CWPs that are consumed domestically as was done in the CWP fabrication calculation method.  Table 3-8 summarizes the emission rates used for each calculation step of the what-if method.
As summarized in Table 3-8, the what-if values of the extent of finishing or overlaying of panels used for calculating what-if low-end and high-end indoor air concentrations are zero percent and 100 percent of their emitting surface area.  These values are logical bounding values, as panels may have both surfaces finished for aesthetic reasons or no finishing at all as a conservative estimate.  Similarly, the what-if values of the extent of finishing or overlaying of work surfaces are 50 percent and 100 percent of their emitting surface area.  These values are again logical bounding values, as work surfaces are expected to at least be finished on their upper surface for functional purposes.  In both cases, the surface area of edges was not accounted for because its effect is negligible.
The what-if values for barrier effectiveness used for calculating what-if low- and high-end indoor air concentrations are given in Table 3-8.  The low-end and high-end values for barrier effectiveness were the low-end and high-end values identified in the source CPA 2003a.  Although this source listed example barrier effectiveness values, it did not describe the extent to which any given value represented actual office workstations.  The intermediate barrier effectiveness was calculated using BIFMA standard M7.1-2007 and BIFMA standard X7.1-2007 (BIFMA 2007b).  BIFMA standard X7.1-2007, which was also adopted as an ANSI standard, is a voluntary standard for meeting VOC emission limits from office furniture.  This standard specifies a formaldehyde limit of 50 ppb as measured using the BIFMA M7.1-2007 chamber test for all workstation configuration types.  Fabricators of office furniture who choose to fabricate workstations that meet the requirement of this standard must construct workstations such that they result in a measured chamber concentration using BIFMA M7.1-2007 that does not exceed 50 ppb.  The intermediate value for barrier effectiveness was calculated using Equation 2 subject to the following conditions:
The workstation is tested according to M7.1-2007 and results in a test chamber formaldehyde concentration of 50 ppb according to X7.1-2007;
The panels have 100 percent of their surfaces finished and the work surfaces have 50 percent of their surfaces finished; and
The workstation panel and work surface components are constructed of CWPs such that the CWP emission rate is equal to an average pre-baseline emission rate.

Pre-baseline emission rates were used because these would be the prevalent emission rates at the time the standard was established.  Therefore, the calculated intermediate barrier effectiveness is for the pre-baseline period and is assumed to be relevant to the baseline.  Appendix E describes the method for calculating an intermediate barrier effectiveness in greater detail.
The emission rates used in this analysis are for new composite wood products; therefore, the estimated indoor air concentrations are for new office furniture.  The applicability of the utilized emission level values to the estimation of emissions from new office furniture is uncertain because decay in formaldehyde emissions from newly-manufactured CWPs, which is discussed in Section 3.3.1.1, can occur in the intervening time between manufacture of the CWPs and their use as components of office furniture in an office building.  Hence, the actual emission levels from the unfinished CWPs would be lower than the values used here resulting in an overestimate of the initial steady-state indoor air concentration associated with new office furniture.  Emission level decay can continue after the initial introduction of office furniture into an office.  Such continued decay of emissions with time and consequent decline of indoor air concentration from the initial steady-state level associated with new furniture was not accounted for because the literature search did not yield the necessary information for the estimation of this effect.
In addition to the uncertainty of the temporal decay of emission rates, the uncertainties of emission rates discussed for the CWP fabrication method also apply for the office buildings method.  Some survey responders only stated the compliance standard they would achieve and not the specific emission levels of their products; therefore, all survey data are assumed to represent actual emission levels of products.  CARB ATCM emission standards are based on the ASTM E1333 chamber test method (ASTM 2002a).  Therefore, the emission levels reported in the EPA survey and provided from PSI described in Section 3.3.1.1 are assumed to be based on the ASTM E1333 chamber test method (see Appendix E for a description of ASTM E1333).  The emission levels from the survey and the PSI presentation are converted to emission rates in units of flux for use in the indoor air concentration mathematical model according to the equations described in ASTM E1333 (see Appendix E).  ASTM E1333 discusses the precision and bias of the large chamber test method in general.  However, ASTM E1333 does not discuss the accuracy and precision of the analytical test method recommended in ASTM E1333 for measuring formaldehyde levels in the chamber air samples: NIOSH 3500 chromotropic acid test procedure.  The accuracy and precision of this procedure is discussed in NIOSH 3500 (NIOSH 1994a).  ASTM E1333 does not discuss the overall accuracy of the test chamber method for determining accurate formaldehyde off-gassing emission rates.
To determine if chamber test results from ASTM E1333 accurately represent actual off-gassing rates in office buildings, the ASTM E1333 test parameters should be compared with typical office building environment properties.  Using the typical values for ventilation rate and component emitting surface area from Table 3-8 and the typical open plan and private office volumes determined by Carter and Zhang (Carter 2007a) and BIFMA M7.1-2007, the CWP loading ratio and air change rate can be calculated and compared to those values specified in ASTM E1333.  See Appendix E for the associated equations, a description of the ASTM E1333 test parameters, and the typical office environment dimensions.  For this comparison, only the workstation panels and work surfaces were used in the calculation of loading ratio, since this assessment assumes storage surfaces are not constructed of CWPs.  Table 3-7 provides a comparison of the ASTM E1333 parameter values for CWP loading ratio and air change rate with those in typical office environments.  The ASTM E1333 parameter values for the private office environment result in an air change rate consistent with that in a typical private office.  The ASTM E1333 loading ratio for MDF is similar to the total CWP loading ratio in typical private offices, but the loading ratios for HWPW and PB are approximately twice the value for typical private offices.  The ASTM E1333 parameter values for air change rate and all three loading ratios are significantly different from those specified in typical open plan environments.  This assessment did not consider how ASTM E1333 uncertainties would impact the overall uncertainty of the assessment results.  However, it is likely that ASTM E1333 uncertainties are of small impact to the assessment results compared to the uncertainties of other mathematical model parameters.  It is also uncertain as to what extent manufacturers will actually obtain their predicted emission levels in 2013.
Table 3-7. Comparison of ASTM E1333 Parameter Values with Typical Office Environment Properties
                                   Parameter
                               ASTM E1333 Values
                  Typical Office Environment Property Values
                                       
                                       
                             Open Plan Environment
                          Private Office Environment
Air Change Rate (h[-1])
                                  0.5 +-0.05
                                     0.92
                                     0.53
CWP Loading Ratio (m[2]/m[3]):
HWPW
                                     0.43
                                      1.1
                                     0.22
PB
                                     0.43
                                       
                                       
MDF
                                     0.26
                                       
                                       
         Sources: ASTM 2002a; BIFMA 2007a, and Carter 2007a

This office environment model treats a workstation (either private office or open plan) as a single exposure zone.  It also only models the mitigation of airborne concentrations due to the ventilation rate of outdoor air.  The ventilation rate of recirculated indoor air, which can vary from building to building, is not included in the model.  The premise behind this assumption is that recirculated indoor air would contain the same concentration of formaldehyde as is in the exposure zone.  This model assumes that each exposure zone is of the same type (open plan or private office), results in the same formaldehyde concentration, and, therefore, the exposure zones do not affect one another.  Additionally, this model does not account for materials present in office buildings that may act as sinks of formaldehyde, such as carpets and wallboards (Koontz 1996a).
Ambient (Outdoor) Air Formaldehyde Concentration
An average value for the ambient air concentration of formaldehyde is available from the EPA draft report Formaldehyde from Pressed Wood Products: Exposure Assessment (USEPA 2009b).  Table 3-1 of the EPA draft exposure assessment provides ambient air formaldehyde concentrations for various land use categories.  The average value for the commercial land use category, 3.26 ug/m[3], is used as an estimate of the formaldehyde concentration in the outdoor air used in office building ventilation.  This value is kept constant throughout the what-if analysis.


Table 3-8. Estimated Values for Parameters in the Mathematical Model for Average Indoor Air Concentration Resulting from Emissions from Private Office and Open Plan Workstations for the Baseline and Analytical Options
                                Parameter Name
                               Workstation Type
                                   Scenario
                     Value of Parameter Used to Determine:
                                Source of Data
                                       
                                       
                                       
                                     Units
                          Intermediate Concentration
                             Low-End Concentration
                            High-End Concentration
                                       
Component Emitting Surface Area
Private office workstation
All scenarios
m[2]
Panels: 7.63
Work surfaces: 6.73
External storage surfaces: 10.55
Indoor air concentration was not assessed for a range of values of emitting surface area because representative values for this parameter were obtained from the BIFMA / ANSI M7.1-2007 test method and used for all calculations. Storage surfaces are assumed not to be constructed from CWPs and do not factor into the final equation.
BIFMA /ANSI M7.1-2007 test method

Open Plan workstation
All scenarios

Panels: 11.08
Work surfaces: 6.10
External storage surfaces: 4.57


Ventilation Rate 
Private office work station
All Scenarios
m[3]/h
34.7
Indoor air concentration was not assessed for a range of values of ventilation rate because representative values for this parameter were obtained from the BIFMA / ANSI M7.1-2007 test method and used for all calculations.
BIFMA /ANSI M7.1-2007 test method

Open Plan workstation


15.0


HWPW Emission Rate
Applicable to both private office and open plan workstations.
Baseline
ppm (ug/m[2]-h)

0.043 (60.9)
0.013 (18.6)
0.163 (232.9)
USEPA 2010a


Analytical Option 1: CARB Phase 1

0.025 (36.0)
0.013 (18.6)
0.058 (82.9)
Appendix D: Table D-4 through Table D-6


Analytical Option 2: CARB Phase 2

0.021 (29.8)
0.013 (18.6)
0.032 (45.7)



Analytical Option 3: NAF

0.013 (18.6)
0.013 (18.6) 
0.013 (18.6)

PB  Emission Rate
Applicable to both private office and open plan workstations
Baseline
ppm (ug/m[2]-h)
0.057 (81.7)
0.013 (18.6)
0.384 (548.8)
USEPA 2010a


Analytical Option 1: CARB Phase 1

0.054 (77.1)
0.013 (18.6)
0.09 (128.6)
Appendix D: Table D-4 through Table D-6


Analytical Option 2: CARB Phase 2

0.053 (75.8)
0.013 (18.6)
0.057 (81.5)



Analytical Option 3: NAF

0.013 (18.6)
0.013 (18.6)
0.013 (18.6)

MDF Emission Rate
Applicable to both private office and open plan workstations
Baseline
ppm (ug/m[2]-h)
0.076 (180.5)
0.024 (56.7)
0.502 (1,186.5)
USEPA 2010a


Analytical Option 1: CARB Phase 1

0.073 (173.6)
0.024 (56.7)
0.140 (330.9)
Appendix D: Table D-4 through Table D-6


Analytical Option 2: CARB Phase 2

0.073 (172.5)
0.024 (56.7)
0.082 (193.8)



Analytical Option 3: NAF

0.035 (81.6)
0.024 (56.7)
0.035 (82.7)

Extent to Which CWP is Finished 
Applicable to both private office and open plan workstations
All Scenarios
Percent of surface area that is finished
100% of panel; 50% of work surface
100% of panel; 100% of work surface
0% of panel; 50% of work surface
Assumed by EPA
Barrier Effectiveness of CWP Finish

Private office workstation
All Scenarios
Percent reduction in emission factor of unfinished CWP
65 [a]
95 [b]
50 [c]
CPA 2007a

Open Plan workstation


71 [a]



Ambient (outdoor) Air Concentration of Formaldehyde
Applicable to both private office and open plan workstations
All Scenarios
ug/m[3]
                                     3.26
Indoor air concentration was not assessed for a range of values of ambient air concentration of formaldehyde because an average value for this parameter was obtained from the EPA draft exposure assessment and used for all calculations.
USEPA 2009b
    a - The barrier effectiveness used to estimate the intermediate concentration was estimated by using the average pre-baseline emission rate of all CWPs (306.7 ug/m[2]-h) and calculating the barrier effectiveness needed to match the BIFMA/ANSI X7.1-2007 emission standard of 50 ppb assuming that 100% of panels and 50% of work surfaces are finished. The calculated barrier effectiveness for private office and open plan workstations were 65% and 71%, respectively.
    b - The highest barrier effectiveness for various barrier types provided in CPA 2003a was 95%.  The highest barrier effectiveness yields the lowest concentration for all else being equal.
    c - The lowest barrier effectiveness for various barrier types provided in CPA 2003a was 50%.  The lowest barrier effectiveness yields the highest concentration for all else being equal.


Estimation of Dermal Exposures to Formaldehyde in Liquid Resin
EPA has developed a series of standard models for quantitatively estimating worker dermal exposures to liquid and solid chemicals during various types of activities.  To estimate dermal exposure, all of these dermal exposure models assume a specific surface area of the skin that is contacted by a material containing the chemical of interest, as well as a specific surface density of the material on the skin.  These exposures can be further normalized by accounting for the body weight of the worker.  The specific methods used to estimate dermal exposures are discussed in detail in Appendix E.
Dermal exposures to formaldehyde contained in liquid urea-formaldehyde (UF) resins are assessed quantitatively during manufacturing according to EPA's standard default model for estimating dermal exposures due to two-hand dermal contact with liquids.  Dermal exposures to formaldehyde due to pressed CWPs are not quantifiable using the EPA dermal exposure models for liquids and solids.  Due to data gaps, quantitative dermal exposure estimates for manufacturing were only made for pre-baseline conditions; baseline and analytical option dermal exposures could not be estimated.  The qualitative dermal exposure assessments for CWP fabrication, wholesale, retail, and commercial use were made for baseline and analytical options.
Exposure Assessment By Life Cycle Stage
Occupational exposures resulting from the manufacture and use of baseline CWPs and effects of the analytical options on the baseline exposures were estimated for CWP fabrication and office building occupancy using the methods described in Section 3.  Exposures were not assessed quantitatively for CWP manufacture, wholesale, and retail due to data gaps.  Instead, qualitative assessments were made for these life cycle stages.  Exposures were not assessed for on-site construction using CWPs and for recycling and disposal of CWPs due to data gaps as discussed in Section 2.
Exposure Assessment for Manufacturing
The assessment of formaldehyde exposures resulting from manufacturing of composite wood products is presented in this section.
Description of Life Cycle Stage
The HWPW, PB, and MDF manufacturing processes include the following general steps: the wood components are prepared, assembled, and bonded together using resin, heat, and pressure.  Urea-formaldehyde (UF) has historically been used in the manufacture of these CWP (USEPA 2010a), and has recently been used in the production of most of the volumes of these CWPs (OAQPS 1999a; OAQPS 2000a).  Hot pressing cures this thermosetting resin which bonds the wood components into panel products (OAQPS 2000b).  UF resin is synthesized from the reaction of urea with formaldehyde; it contains un-reacted formaldehyde and formaldehyde that is generated during its synthesis reaction.  Also, formaldehyde may be formed as the resin cures during hot pressing in the CWP manufacturing process.  Emissions from the CWP result from hydrolytic degradation of the cured resin (which is the reverse of the synthesis reaction), or from the un-reacted (free) formaldehyde (Tohmura 2000a); these emissions are a function of humidity and temperature (USEPA 1996a).  Currently other resins in addition to UF resin are used in the manufacture of the three CWPs.  Poly(vinyl acetate) (PVA) and phenol-formaldehyde (PF) resins are used in the manufacture of HWPW (USEPA 2010a).  Melamine-urea-formaldehyde (MUF) and poly(methylene diphenyl diisocyanate) (pMDI) resins are used in the manufacture of PB and MDF and PF is used in the manufacture of PB as well (OAQPS 2000a).
The HWPW, PB, and MDF manufacturing processes are further described below with emphasis on aspects related to resin processing and potential sources of formaldehyde exposure.  The market profile (USEPA 2010a) provides broad descriptions of the manufacturing processes that complement the descriptions given here.  Most of the references used to obtain the process descriptions below are dated prior to 2006.  In 2006, EPA implemented the National Emission Standards for Hazardous Air Pollutants (NESHAP) for plywood and composite wood products manufacturing industries.  It is unknown whether any process changes were implemented in response to this NESHAP.
Description of HWPW Manufacturing Process
HWPW is generally made by assembling a hardwood veneer to the face and back of a core of veneer, lumber, particleboard, MDF, a combination of these materials, or other special cores (USEPA 2010a).
The overall HWPW manufacturing process consists of the following steps that are presented here as two groups of processing steps (OAQPS 2002b):
Log selection, debarking, bucking (cutting), heating and then further cutting into veneers and drying of veneers; and
Adhesive mixing, veneer assembly and adhesive application, hot pressing, cutting, and finishing.
The two groups combined constitute the manufacturing process in log-based mills; in these mills, the process starts with the conversion of logs into veneer and ends with the production of hardwood plywood.  The processing steps in the first group constitute the manufacturing process in a hardwood veneer plant, which are outside the scope of work because they will not be impacted by the FSCWPA.  The processing steps in the second group constitute the manufacturing process in purchased veneer-based mills and 3-ply mills for which the raw materials are wood veneers or other platforms (USEPA 2010a).
Resin constitutes between 4.5 and eight percent of HWPW by weight (HPVA 2009a).  It is mixed with fillers and extenders prior to application.  The resin is used during the veneer assembly step of HWPW manufacture.  In general, veneer assembly involves coating resin onto various plies of hardwood veneer or other materials and stacking the plies into a glue-bonded assembly.  Most hardwood plywood mills manually lay-up their assemblies.  In the majority of hardwood plywood mills (regardless of size), roll coaters are used to spread resin (USEPA 2010a), and the panels are fed manually into these coaters.  Generally, resin is applied to one side of a back or face ply and then a core ply is laid onto the resin-coated surface of the back or face ply.  This process continues with a subsequent resin application to the core ply followed by the lay-up of another ply.  This process is repeated until the last ply (a back or face ply) is in place (OAQPS 2000b).  An alternative to roll coaters is curtain coaters (pressure head or gravity feed), spray coaters (the glue is either mixed with atomized air or directly atomized using high pressure), and extrusion (glue is turned into foam at 5 times its original volume and the foam is laid on the veneer surface in beads that are spread during pressing) (USEPA 2010a).  Spray coating occurs in spray booths.  The alternative glue application systems may be used in automated lay-up and may allow for reduction in waste glue resulting from the operation of roll coaters.  Other means of reduced adhesive consumption include improved process control (for example, avoidance of over-dried or unevenly dried veneer) and production line cleanliness (to allow quick identification of resin spill sources).  Panel design is also a factor in reduced adhesive consumption (Baldwin 1995a).  After application of the resin, the bundle then moves to a prepress or hot press pre-loader.  Once pressed, the hardwood plywood then is taken to a finishing process where edges are trimmed and the face and back may or may not be sanded smooth.
The primary source of formaldehyde emissions in the HWPW manufacturing process is the veneer dryer (OAQPS 2002b).  Emissions from dryer exhausts are may be controlled with an air pollution control device (OAQPS 2002b).  Even when that is not so, it is likely that a dryer exhaust is vented through a stack.  Hence, it is unknown whether this emission source results in worker exposure.  The press is another source of formaldehyde emissions; emissions from press vents are typically not controlled (OAQPS 2002b).  In addition to vent emissions, formaldehyde may be emitted from the hot press as it opens during the pressing operation (Tohmura 2000a).  Exposure to formaldehyde may also occur while handling the HWPW or performing activities in proximity to the press (OAQPS 1997a; OAQPS 1998a, and OAQPS 1999b).
Description of PB Manufacturing Process
The raw materials for PB manufacturing, or furnish, are wood particles (wood chips, sawdust, planer shavings and other recycled materials).  The manufacturing process consists of wood particle generation or procurement, milling, drying, and blending of the furnish with resin and other additives, forming the resinated material into a mat, hot pressing, and finishing (USEPA 2010a).
In order to ensure that the resin bonds correctly with the particles, which affects both the board's structural characteristics and final HCHO emissions, it is important that the furnish attain a specific moisture content (USEPA 2010a).  The moisture content of the particles entering the dryer may be as high as 50 percent on a wet basis (OAQPS 2000b).  The furnish is dried to a very low moisture content to allow for moisture gained by adding resins and other additives (USEPA 1996a).  Dryer temperature is routinely adjusted based on furnish moisture measurements (OAQPS 2000b).  After drying, face material, or the furnish that will be present on the face of a panel, may be screened for a second time to remove fines, which often absorb too much resin (OAQPS 2002b).  The dried furnish is sent to holding bins, which provide surge capacity (USEPA 2010a).
From the holding bins the furnish is conveyed to blenders and mixed with resin and other additives.  Resin is usually received on-site as a solution of 65% solids in water, stored in tanks, and pumped to the blenders.  Additives may include wax (to impart water resistance to the manufactured product and reduce the tendency for equipment plugging), catalyst (to accelerate resin cure and reduce press time), and scavengers for reduction of emissions from the process and the manufactured product (USEPA 1996a; OAQPS 2000b; USEPA 2010a, and Maloney 1993a).  Furnishes are generally no warmer than 100 F when blended to avoid pre-curing and drying out of the resin (USEPA 1996a).  The furnish may be sampled before and after blending to ensure proper moisture control (OAQPS 1998c).
The dominant type of blender consists of a small horizontal drum with high-speed, high shear impellers (USEPA 1996a; Maloney 1993a).  Resin and its additives are sprayed through air or airless atomizing nozzles or applied by centrifugal atomization.  In the air spray system, the supplied air must be exhausted from the blender but in the airless systems the elimination of atomizing air eliminates the associated air pollution problem.  Additionally, because of the high speed and number of its impellers, the blender, especially when not fully loaded with furnish, can act as an axial-type fan and discharge air causing dusting problems; this may be limited by restricting its air intake (Maloney 1993a).  Most blenders vent through baghouses (OAQPS 2000b).  Blenders generally are designed to discharge the resinated particles into a plenum over a belt conveyor that feeds the blended material to a forming machine (OAQPS 2000b).  In some installations, this plenum is vented back to the blender for recirculation of fines while in other installations it is vented to the suction side of the forming machine (Maloney 1993a).  Some evaporation of moisture (and hence formaldehyde as well) from the furnish may occur from air exposure when conveyed from the blenders to the forming station (USEPA 1996a).
After the blending process, the resinated material is fed to the forming machine that deposits the blended material into a continuous mat. To produce multilayer particleboard, several forming heads can be used in series, or air currents can produce a gradation of particle sizes from face to core (OAQPS 2002b).  Most formers vent to a baghouse (OAQPS 2000b).  The formed mat is pre-pressed (optional), cut into desired lengths, and conveyed to the hot press.  Few presses are vented to an air pollution control device and some of these presses are fully enclosed (OAQPS 2000b).  After pressing, the particleboard panels are cooled, sanded, trimmed to desired final dimensions, and may undergo other finishing operations.  Emissions from most coolers are not controlled and some coolers whose emissions are controlled are fully enclosed.
The primary sources of formaldehyde emissions are the dryers, hot presses, board coolers and blenders (OAQPS 2002b).
Description of MDF Manufacturing Process
The raw materials for MDF manufacturing consist of wood chips, sawdust, planer shavings and other recycled materials which are typically delivered from off-site locations.  The manufacturing process consists of pulping the raw material to fibers in a refiner, drying, blending the fibers with resin and additives, forming the resinated material in a mat, pressing, and finishing (OAQPS 2002b; USEPA 2010a).
Raw material is processed through a steam pressurized refiner which discharges through a blow valve into a blowpipe that is connected to a tube dryer (USEPA 1996a; OAQPS 2000b, and USEPA 2010a).  The pressurized refiner may also vent into the tube dryer.  The tube dryers are directly fired or indirectly heated.  Dryer inlet temperature are on average as high as 270°F.  The wood fiber is pneumatically drawn through the tube dryers and is separated from the dryer exhaust gas in a cyclonic collector.  The dried wood fibers are discharged from the cyclonic collectors through air locks into dry storage bins which provide surge capacity.  The dryer gas exhaust may be vented to an air pollution control device, including incineration based controls or scrubbers (OAQPS 2000b; USEPA 2010a).  Most mills blend resin, wax and other additives by injection into the blowline where they readily mix with the fibers under the turbulent flow conditions present in the bowline.  This occurs prior to the drying of the fiber in the tube dryer but some mills use mechanical blending that applies the additives to the dried fiber (USEPA 2010a).  The blowline system is more frequently used because it is fully enclosed, thereby reducing worker exposure and has superior blending performance.  However, the resin may lose some reactivity as a result of the hot, wet environment in the blowline tube and therefore a higher resin-to-fiber ratio is required as compared to the conventional blender system.  Based on a 1998 site visit report to the Medite Division of Sierra Pine MDF plant, the plant personnel stated that methylene diphenyl diisocyanate (MDI) resins are always added using the blowline system due to potential worker exposure concerns (OAQPS 1998b).
The resinated fibers are conveyed from the dry storage bin, where they are deposited as a mat on a continuously moving screen system (OAQPS 2002b).  Most formers vent to a baghouse (USEPA 1996a).  The deposited mat is pre-pressed and sometimes pre-trimmed prior to transferring to the hot press.  Hot pressing activates the thermosetting resin and bonds the resinated material into a solid panel.  The press may be vented to an air pollution control device, including incineration based controls, scrubbers or a baghouse (OAQPS 2000b).  Additionally, the press area may be partially or fully enclosed and may be vented with fans (OAQPS 2002a).  Based on non-CBI responses from the 1998 information collection request, seven of the 21 presses were fully enclosed at MDF manufacturing facilities (OAQPS 2000a).  After hot pressing, the panels are cooled and finished. The board cooler may be vented to incineration based controls and additionally the cooler may be fully enclosed in a vented enclosure (OAQPS 2000b).  Finishing operations include sanding, trimming, and sawing.
The primary sources of hazardous air pollutant (HAP) emissions (including formaldehyde emissions) at MDF plants include emissions from pressurized refiners, wood dryers, blenders, formers, presses, and board coolers; the finishing operations may also be a source of HAP emissions (OAQPS 2000b).  The largest emission sources are wood dryers and press vents. Formaldehyde emissions are significantly higher after resin application.  For example, dryer emissions from a blowline system are much higher than the dryer emissions from a conventional blender system due to the presence of formaldehyde-based resin in the wood material in the former dryer system.  While fugitive emissions from the blender, former, and cooler may occur, these emissions are much lower and are typically not controlled (OAQPS 2002b).
 Production Volumes and Number of Establishments, Workers, Production Hours, and Work Days
The domestic production volumes and number of sites, or mills, for the manufacture of the three composite wood products are given in Table 4-1 and Table 4-2.  These data are provided for reference purposes only and are not used in the assessment.  These tables present data on the number of mills from the Composite Panel Association (CPA) (AWFS 2009a) and the U.S. International Trade Commission (USITC) (USITC 2008a).  The number of mills reported by the 2007 Economic Census (Census 2007a) is presented in Table 4-3.  The U.S. Census reported a greater number of mills for both NAICS codes than the CPA and USITC sources.  Note that the U.S. Census classifies PB and MDF manufacture under a single NAICS code: 321219  -  Reconstituted Wood Product Manufacturing.  This NAICS code also includes fiberboard, hardboard (HB), oriented-strand board (OSB), waferboard, flakeboard, and chipboard, which are outside of the scope of the assessment.  Similarly, NAICS code 321211 - Hardwood Veneer and Plywood Manufacturing includes the manufacture of hardwood veneers, which are outside the scope of the assessment (see Section 2).  Therefore, the number of mills data for both manufacturing NAICS codes represent mills that manufacture products outside the scope of this assessment.
Estimates for the total number of workers and aggregate production hours worked for the manufacturing of the three composite wood products were obtained from the EPA market profile (EPA 2010a) and the 2007 Economic Census (Census 2007a) and are presented in Table 4-3.  The 2007 Economic Census data are likely an overestimate of the number of workers for the following reasons:
Trade associations reported that mills have closed down; and
NAICS codes 321219  -  Reconstituted Wood Product Manufacturing and 321211 - Hardwood Veneer and Plywood Manufacturing both include products outside the scope of this assessment.
However, it is assumed that the census data on number of workers and aggregate production hours worked are proportionally overestimated such that their ratio, used to calculate production hours per day and days per year, is representative of mills within the scope of the assessment.
Using these data, the average production days per year per worker (assuming eight hours per day), and average production hours per day per worker (assuming 250 days per year) were calculated, and the results are given in Table 4-3.  These calculated values are compared to the OPPT default values.  As default, OPPT assumes people work five days per week and 50 weeks per year (with two weeks of vacation), which results in 250 work days per year.  OPPT also assumes a typical work day is eight hours per day.  As seen in Table 4-3, the calculated production days per year (assuming 8 hours per day) do not deviate from the OPPT default value for frequency of exposure of 250 days per year by more than 10 percent.  Based on this comparison, the frequency of exposure for manufacturing is assessed as 250 days per year.  Similarly, all of the calculated values for production hours per day (assuming 250 days per year) do not deviate from the OPPT default value for duration of exposure of 8 hours per day by more than 10 percent.  Based on this comparison, the duration of exposure for manufacturing is assessed as 8 hours per day.
Table 4-1.  U.S. Hardwood Plywood Production Data and Number of Mills
Hardwood Veneer and Plywood Mill
                          Composite Wood Product Type
             2008 U.S. Production Volume of Composite Wood Product
                            (thousands of ft[2])[1]
                            Number of U.S. Mills[2]
Hardwood Plywood Mills
Hardwood Plywood
                                    748,400
                                      24
  [1] U.S. production volume data obtained from personal communication with HPVA (HPVA 2009a). The plywood thickness basis was not provided.
  [2] Number of U.S. mills obtained from the U.S. International Trade Commission report on wood flooring and hardwood plywood (USITC 2008a).

Table 4-2.  U.S. Reconstituted Wood Product Production Data and Number of Mills
                     Reconstituted Wood Product Mill Type
                          Composite Wood Product Type
2007 U.S. Production Volume of Composite Wood Product (thousands of ft[2] at 3/4" thickness basis)[1]
                            Number of U.S. Mills[2]
Particleboard Mills
Particleboard
                                   3,542,666
                                      34
Medium-Density Fiberboard Mills
Medium-Density Fiberboard
                                   1,884,264
                                      27
  [1] U.S. production volume data obtained from the 2007 North American Downstream Market Report (CPA 2008a).
  [2] Number of U.S. mills obtained from "Update on California's Composite Wood Products Regulation" presented at AWFS Show in Las Vegas, NV on July 18, 2009 provided by CARB to EPA (AWFS 2009a). The presentation was co-presented by CARB and CPA. The number of MDF mills is a high-end estimate that may include some hardboard mills: the presentation did not differentiate between non-CPA-member MDF and HB mills.

Table 4-3.  Summary of 2007 Economic Census Data and Exposure Frequency and 
Duration Calculated from These Data for U.S. Manufacturing Mills
                           2007 Economic Census Data
    Parameters Calculated from the Corresponding 2007 Economic Census Data
                                  NAICS Code
                          NAICS Industry Description
                           Number of Establishments
                     Maximum Number of Production Workers
                   Aggregate Production Hours Worked (hours)
            Average Production Days per Year (assuming 8 hours/day)
            Average Production Hours per Day (assuming 250 days/yr)
321219
Reconstituted wood product manufacturing
                                      262
                                    16,076
                                  34,615,000
                                      269
                                     8.61
321211
Hardwood veneer & plywood manufacturing
                                      304
                                    15,504
                                  32,256,000
                                      260
                                     8.32
Source: EPA market profile (USEPA 2010a) and 2007 Economic Census (Census 2007a).

Pre-Baseline Exposure Monitoring Data
This section presents a summary of the inhalation exposure monitoring data collected from OSHA IMIS for the period 2002 to 2009 on the manufacture of HWPW, MDF, and PB as well as a summary of the other compiled inhalation exposure monitoring data on CWP manufacturing.
Table 4-4 provides the maximum, median, and minimum 8-hour TWA personal inhalation monitoring formaldehyde concentrations by worker activity category for the manufacture of HWPW from OSHA IMIS.  Table 4-5 provides the maximum, median, and minimum 8-hour TWA personal inhalation monitoring formaldehyde concentrations by worker activity category for the manufacture of reconstituted wood products, including PB and MDF from OSHA IMIS.


Table 4-4.  Summary of Formaldehyde Inhalation Monitoring Data for Hardwood Plywood Manufacturing Mills from the OSHA IMIS Database
                          Worker Activity Description
                Total Number of Sites where Monitoring Occurred
                          Total Number of Data Points
                    Monitored Formaldehyde Concentration[1]
             Year Monitoring Occurred for the Summary Data Points
                                       
                                       
                                       
                                    Maximum
                                    Median
                                    Minimum
                             Maximum Concentration
                            Median Concentration[2]
                             Minimum Concentration
                                       
                                       
                                       
                                     (ppm)
                                  (ug/m[3])
                                     (ppm)
                                  (ug/m[3])
                                     (ppm)
                                  (ug/m[3])
                                       
                                       
                                       
Press Operation
                                       2
                                       4
                                     1.56
                                    1,917.2
                                     0.695
                                     854.2
                                     0.213
                                     261.8
                                     2003
                                  2003 / 2003
                                     2006
Glue Spreading
                                       2
                                       2
                                     0.32
                                     393.3
                                     0.28
                                     344.1
                                     0.24
                                     295.0
                                     2003
                                  2003 / 2006
                                     2006
Mixing Adhesive
                                       1
                                       1
                                     0.119
                                     146.3
                                     0.119
                                     146.3
                                     0.119
                                     146.3
                                     2006
                                     2006
                                     2006
Lathe Operation
                                       1
                                       1
                                     0.022
                                     27.0
                                     0.022
                                     27.0
                                     0.022
                                     27.0
                                     2004
                                     2004
                                     2004
Production Laborer or Other Operator
                                       2
                                       4
                                     0.142
                                     174.5
                                    0.0695
                                     85.4
                                     0.021
                                     25.8
                                     2007
                                  2004 / 2007
                                     2004
Overall Statistics for All Data Points
                                       4
                                      12
                                     1.56
                                    1,917.2
                                    0.1775
                                     218.1
                                     0.021
                                     25.8
                                     2003
                                  2006 / 2007
                                     2004
    [1] The formaldehyde concentration in ug/m[3] was calculated from the provided concentration in ppm by multiplying the concentration in ppm by 1,229 ug/m[3]/ppm (see Appendix E).
    [2] For worker activities with an even number of data points, the median was calculated as the average of the two middle values.  The years are provided for the two middle values.




Table 4-5.  Summary of Formaldehyde Inhalation Monitoring Data for Reconstituted Wood Product Manufacturing Mills from the OSHA IMIS Database
                                    Worker
                                   Activity
                                  Description
                Total Number of Sites where Monitoring Occurred
                          Total Number of Data Points
                    Monitored Formaldehyde Concentration[1]
             Year Monitoring Occurred for the Summary Data Points
                                       
                                       
                                       
                                    Maximum
                                    Median
                                    Minimum
                             Maximum Concentration
                                    Median 
                               Concentration[2]
                                   Minimum 
                                 Concentration
                                       
                                       
                                       
                                     (ppm)
                                  (ug/m[3])
                                     (ppm)
                                  (ug/m[3])
                                     (ppm)
                                  (ug/m[3])
                                                                               
                                                                               
                                                                               
Press Operation
                                       1
                                       3
                                     0.12
                                      147
                                     0.089
                                      109
                                     0.009
                                      11
                                     2005
                                     2005
                                     2005
Production Laborer or Other Operator and CSHO
                                       4
                                       6
                                     0.243
                                      299
                                     0.043
                                      53
                                     0.02
                                      25
                                     2005
                                  2006 / 2005
                                     2006
Overall Statistics for All Data Points
                                       4
                                       9
                                     0.243
                                      299
                                     0.051
                                     62.7
                                     0.009
                                      11
                                     2005
                                     2005
                                     2005
   [1] The formaldehyde concentration in ug/m[3] was calculated from the provided concentration in ppm by multiplying the concentration in ppm by 1,229 ug/m[3]/ppm (see Appendix E).
   [2] For worker activities with an even number of data points, the median was calculated as the average of the two middle values.  The years are provided for the two middle values.


Appendix H presents additional historical exposure monitoring data from a variety of studies and reports.  These data are briefly summarized here for reference purposes.  Note that these data include studies from countries other than the United States and their relevance to U.S. mills has not been established.
Plywood Manufacturing Mills Monitoring Data
Sampling data collected in British Columbia from 1992 to 2003 indicate that formaldehyde levels in the plywood manufacturing industry are predominantly below 0.3 ppm (WorksafeBC 2009a).  Other data indicate an exposure range of 0.08 ppm to 0.60 ppm (Fransman 2003a) in the plywood manufacturing sector.
Short-term exposure levels and instantaneous ceiling limit values for plywood manufacturing available from one discussion paper (WorksafeBC 2009a) ranged from 0.14 to 0.29 ppm and 0.33 to 0.44 ppm, respectively, in a 2002 survey.
MDF Manufacturing Mills Monitoring Data
For MDF manufacturing, the highest estimated geometric mean for ambient formaldehyde levels was reported as 0.43 ppm in the main production area of manufacturing facilities in Quebec (Lavoue 2005a).  In one study (Chung et al. 2000a), free formaldehyde levels were determined to be less than 0.14 ppm during machining of the Class B (higher formaldehyde potential) MDF.  According to one Swedish report, formaldehyde levels were at 0.2 ppm (arithmetic mean) between 1980 and 1989 in facilities manufacturing MDF (IARC 1995a).  
Particleboard and Fiberboard Manufacturing Mills Monitoring Data
Time-weighted average (8-hour) levels of formaldehyde have been noted to be generally higher than 0.3 ppm in sampling data collected over a period of 15 years (1990 to 2005) for particleboard and fiberboard manufacturing in British Columbia, Ontario, Quebec, and the Netherlands (WorksafeBC 2009a).  Exposures in fiberboard manufacturing ranged from 0.07 to 5.0 ppm for samples taken between 1996 and 2000 (WorksafeBC 2009a). Exposure levels from between 0.1 and 0.4 ppm (Quebec) (Lavoue 2005a) to 1.15 ppm (Scandinavia) (Niemelä and Vainio 1981a) have been noted in other studies of the particleboard and MDF manufacturing industry.
Eight-hour TWA levels of formaldehyde ranging from 0.07 to 0.20 ppm and instantaneous ceiling values that were as high as 7.47 ppm were reported in particleboard manufacturing plants in 2002 (WorksafeBC 2009a).  A median of 0.62 ppm was observed for particleboard workers in measurements taken in 1987/1988 (Horvath et al. 1988a) and a value of 0.3 ppm was reported for particleboard manufacture between 1980 and 1989 (IARC 1995a).
For particleboard manufacturing, formaldehyde exposure levels in the range of 0.08 to 0.9 ppm, with peaks reaching 4.1 ppm, were observed from measurements taken between 1975 to 1983 (Edling 1988a).  Mean levels between 0.4 and 2.3 ppm have been reported for particleboard manufacture by other researchers between 1975 and 1984 (Kauppinen and Niemelä 1985a).  A mean level less than 1 ppm and a maximum level of 1.4 ppm during particleboard manufacture were also reported in 1982 (Preuss et al. 1985a).
Occupational hygienists from WorkSafeBC sampled different locations in a fiberboard plant in British Columbia from January to June 2006 and found formaldehyde levels ranging from 0.03 to 1.19 ppm.  The highest exposure was in the press outfeed area, ranging from 0.32 to 1.19 ppm.  
Respiratory Protection Usage During Manufacturing
Since OSHA IMIS data do not provide information on personal protective equipment use, OSHA inspection reports associated with the IMIS data were examined to get an idea of the use of respiratory protection by workers when engaged in manufacturing activities.  Appendix G presents a summary of each reviewed inspection report.
The nine TWA, personal monitoring IMIS data points for PB and MDF manufacturing in Table 4-5 discussed in Section 4.1.3 are associated with four inspections.  Of these four inspections, a total of two inspection reports for PB and MDF manufacturing facilities were requested from OSHA.  One of the two inspection reports (inspection number 307061689; see Appendix G), which accounts for four of the nine TWA, personal monitoring IMIS data points, has been received and reviewed, but the other inspection report was no longer available due to its age.  Additionally, two other inspection reports were requested from inspections prior to 2002 and hence are not associated with the nine data points discussed above.  One of these two inspection reports (inspection number 303864029; see Appendix G) has been received and reviewed, but the other inspection report was no longer available due to its age.  Finally, an additional inspection report (inspection number 312479835; see Appendix G) associated with the CWP fabrication IMIS data (see Section 4.2.3) was requested, received, and reviewed.  Although the facility was identified in IMIS by a fabricator NAICS code, the inspection report states that the facility manufactures MDF and then fabricates flooring from the MDF.  As the monitored workers were associated with the manufacture of MDF, this inspection report is discussed in this section and Appendix G with the manufacturer inspection reports.
Of the three reviewed inspection reports, only one report for a facility that manufactures MDF primarily for flooring indicated the use of respirators by production workers (press assistants) (inspection number 312479835; see Section 4.2.4 and Appendix G).  The workers were observed wearing 3M 6300 half masks with formaldehyde cartridges and had the option to use full-facepiece respirators.  The two TWA measurements taken for press assistants were both below the PEL of 0.75 ppm, although one exceeded OSHA's TWA action level (AL) of 0.5 ppm for formaldehyde.  The two facilities associated with the remaining two inspection reports had TWA measurements below the action level.
The 12 TWA, personal monitoring IMIS data points for HWPW manufacturing in Table 4-4 discussed in Section 4.1.3 are associated with four inspections.  None of these four inspection reports for HWPW manufacturing facilities were requested from OSHA.  However, two inspection reports for HWPW manufacturing facilities were requested from inspections prior to 2002 and hence are not associated with the 12 data points discussed above.  Both of these inspection reports have been received and reviewed.  Neither report indicated the use of respiratory protection.  The first facility had four TWA measurements and three STEL measurements (inspection number 303713325; see Appendix G).  All four TWA measurements were below the action level and all three STEL measurements were below the OSHA STEL of 2 ppm.  The second facility had ten TWA measurements and six STEL measurements (inspection number 123445595); see Appendix G).  All STEL measurements were below the OSHA STEL of 2 ppm.  Two of the TWA measurements were above the OSHA AL but below the PEL; the remainder were below the action level.  This facility was cited by OSHA for formaldehyde overexposure.  The inspection report noted that workers were not observed wearing respiratory protection during the inspection.
Although one inspection report (inspection number 123445595; see Appendix G) for HWPW manufacturing noted that workers were not wearing respirators in spite of exposure levels that exceeded the AL (but were below the PEL), respirator use has been noted in another case in MDF manufacturing (inspection number 312479835; see Appendix G) when exposure levels exceeded the AL (but were below the PEL).  The need to protect against chemicals and particulates other than formaldehyde might also dictate respirator use.  Therefore, the reported use of respirators in the inspection report for the MDF manufacturing facility could have been for protection against chemicals other than formaldehyde or it could be a precautionary practice.  Overall, the limited number of inspection reports that were available and examined does not provide conclusive evidence that respirators are necessarily worn by workers when their exposure meets or exceeds the formaldehyde AL or PEL.
Baseline Exposure and the Effect of Analytical Options on Baseline Exposure for Manufacturing
Baseline exposures and exposures for the analytical options were not assessed quantitatively due to data gaps.  The emission rate data for CWPs presented in Appendix D apply to the CWPs after their manufacture.  Hence, these emission rates are only relevant to the manufacturing process after the pressing operation and are not applicable to the remainder of the process, which includes unit operations in which uncured resins are present and the press.  These sources of formaldehyde cannot be assumed negligible in manufacturing mills.  As indicated in the pre-baseline OSHA IMIS monitoring data presented in Section 4.1.3, formaldehyde exposures have been observed associated with mixing and applying adhesives and pressing operations.  The literature search did not yield information on modeling indoor air concentrations from these sources.  Therefore, exposures during baseline and analytical options are assessed qualitatively.
At baseline, approximately all domestic production of HWPW, MDF, and PB will be in compliance with the CARB ATCM Phase 2 standard, as shown in Table D-3 in Appendix D.  Only small fractions of HWPW and PB domestic production will not be in compliance with the CARB ATCM Phase 2 standard at baseline.
Approximately 0.28% of domestic HWPW production will not be in compliance with any CARB ATCM standard.  However, the emission level of this fraction of HWPW production is estimated to be 0.058 ppm, which is close to the CARB ATCM Phase 2 standard of 0.05 ppm and less than the CARB ATCM Phase 1 standard of 0.08 ppm for HWPW.
Approximately 2% of domestic PB production will not be in compliance with the CARB ATCM Phase 2 standard.[7]  Of this 2% of domestic PB production, approximately 0.2% of domestic PB production will not meet any CARB ATCM emission standard.  However, the emission level of this 0.2% of domestic PB production is estimated to be 0.058 ppm, which is less than the CARB ATCM Phase 2 standard of 0.09 ppm for PB.  Additionally, approximately 1.8% of domestic PB production will be in compliance with the CARB ATCM Phase 1 standard with an estimated emission level of 0.09 ppm.  Hence, the CARB Phase 1 and FSCWPA/CARB Phase 2 analytical options will result in a negligible effect on baseline exposure levels for manufacturing.  Since approximately all domestic production of CWPs are expected to meet the CARB ATCM Phase 2 standard at baseline, only the NAF analytical option will affect baseline exposures.
NAF resins do not contain any formaldehyde.  Therefore, mills that use NAF resins in compliance with NAF standards should achieve a reduction in formaldehyde exposures due to the elimination of the formaldehyde content of resins.  However, wood naturally emits formaldehyde, and it is possible that workers in mills that use only NAF resins may still experience some formaldehyde exposures due to the natural emission of formaldehyde from wood.  Other sources of formaldehyde can exist for mills that apply value-added surfaces.  However, it is uncertain how many mills apply value-added surfaces.
Dermal Exposures during Manufacturing
Dermal exposures to workers during manufacturing can occur during unloading of liquid UF resin.  The free formaldehyde content immediately after manufacture is approximately 0.59% but reduces to 0.15% after 24 hours (Fink 2005a).  For the purpose of this assessment, EPA assumes a formaldehyde concentration of 0.59% in the UF resin.
The EPA/OPPT 2-Hand Dermal Contact with Liquid Model estimates a dermal exposure of 590 to 1,800 mg/day for exposure to a chemical at 100% concentration.  For the assumed formaldehyde concentration of 0.59%, the model estimates worker exposure of 3.5 to 10.6 mg/day.  This results in an estimated dose of 0.05 to 0.15 mg/kg/day for an average 70 kg worker.
The source (Fink 2005) does not specify whether the provided free formaldehyde contents in UF resins are specific to UF resins for CWPs or other applications.  Additionally, Fink (2005) does not specify the statistical significance of the formaldehyde contents, such as being typical, maximum, minimum, etc.  This assessment assumes these formaldehyde contents are typical for UF resins specific for CWPs and representative of pre-baseline conditions.  Additionally, it is uncertain how the free formaldehyde content of CWP resins will change as manufacturers adopt the CARB ATCM limits.  The source (Fink 2005) is from 2005 and may not represent resins at baseline or analytical options.  Therefore, both baseline dermal exposures and the effect of analytical options on baseline dermal exposures are uncertain and cannot be estimated.
Exposure Assessment for Composite Wood Product Fabrication
The assessment of formaldehyde exposures resulting from the use of the three composite wood products (CWPs) in composite wood product fabrication is presented in this section.
Description of Life Cycle Stage
A general description of the manufacturing process is presented below.  The literature search did not yield detailed information on the manufacture of specific composite wood fabrication products (downstream products).  The Office of Enforcement and Compliance Assurance (OECA) sector notebook Profile of the Wood Furniture and Fixtures Industry (OECA 1995a) provides a general process description for wood furniture and fixtures constructed from raw lumber, which does not include composite wood products.  Activities involved in fabrication of wood furniture from raw lumber include drying raw wood; sawing into appropriate sizes, planing and bending (i.e., machining); assembling the wood material into parts; sanding; prefinishing (e.g., watering, sanding, derosination, bleaching); coating by spraying, brushing, or dipping; and finishing (OECA 1995a).  As described below, similar activities are used to produce furniture and fixtures from the composite wood products that are the focus of this assessment.
The following activities (unit operations) are typically used to manufacture wood products, based on a survey of Georgia wood products manufacturers (GDNR 1996a).  Although the reference does not describe if the surveyed manufacturers used composite wood products or raw lumber, it explains that wood product manufacturers may use similar processes to produce different products using different raw materials.  Therefore, the cited activities are likely used to manufacture downstream products using composite wood products as raw materials.  The cited activities include the following:
Application of overlays or decorative surfaces (other than wood or woody-grass veneers);
Assembling;
Gluing;
Spray application;
Painting and coating;
Water-based coating;
Solvent-based coating;
Staining;
Sealing;
Water-based cleaning; and
Solvent-based coating.
These process activities apply to the following CWP fabrication sectors: millwork, cabinet makers, veneer plywood, trusses, prefabricated construction, particleboard and reconstituted wood, wood furniture, miscellaneous furniture, exhibits and showrooms (GDNR 1996a).  Workers are potentially exposed to formaldehyde during the above-listed activities from unfinished and finished (e.g., overlayed, coated, or painted) composite wood products.  However, emission rates from the finished products are generally lower than those from the unfinished products (Battelle 1996a).
Additionally, the following materials, which are used in CWP fabrication, may also be a potential source of formaldehyde exposure during CWP fabrication:
Adhesives used in gluing and overlaying barrier surfaces;
Overlay materials; and
Coatings and varnishes used in finishing operations.
Formaldehyde-based adhesives may be used in upholstered furniture manufacturing based on the contents of an OSHA inspection report associated with the compiled IMIS data (OSHA 2003a).  At the inspected site, formaldehyde-based adhesive is mixed and applied by hand to glue wood sections together.  The inspection report noted that proper PPE was not used during this activity and formaldehyde sampling results were below the action level (OSHA 2003a; see Appendix G).  Workers also use a formaldehyde-based adhesive with a pneumatic gun to assemble wooden chairs.  Workers wear a half-face respirator while assembling the wooden chairs, although the inspection report stated past air sampling indicated formaldehyde levels were below the PEL (OSHA 2003a; see Appendix G).
During the application of overlay materials, adhesive is applied to the decorative overlay, and the overlay and the substrate (composite wood product) are subsequently bonded through a pressing operation.  Based on information associated with similar activities that occur during manufacturing of composite wood products, it is inferred that formaldehyde vapor would be emitted during glue application and hot pressing, resulting in potential for inhalation exposure.  Some overlay materials do not require additional adhesive to be bonded with the wood substrate as they are saturated with reactive resins and partially cured as supplied (CPA 2007a).  The overlays may be saturated with formaldehyde-based resins and thereby serve as another source of formaldehyde.  A CPA technical bulletin indicates that several types of adhesives may be used for the application of overlays to PB and MDF, such as hot melt adhesives, PVA, acrylic-based formulations, UF, PF, and MF resins (CPA 2007a).
The finishing operation for wood furniture and cabinets includes the application of several clear and/or color coatings.  The coating process may be manual or automatic.  The most common manual techniques are wiping, dipping and spraying via spray guns.  Automatic application techniques include robotic spraying, roll coating and curtain coating.  While spraying and coating processes typically occur in ventilated booths equipped with dry filters to control particulates (USEPA 2000a), workers inside the booth can be significantly exposed if they are located in the direction of flow of the coating mist (IRSST undated).  An OSHA inspection report indicates the use of formaldehyde-based coating in the finishing of wood cabinets.  The employer provided respirators to reduce worker exposure to formaldehyde from the coating (OSHA 2008a; see Appendix G).  OSHA offers guidance on safe practices in the woodworking industry through its Woodworking eTool (OSHA 2002a).  OSHA recommends using ventilated, automated systems for applying coatings and adhesives when feasible.  OSHA also recommends that manual spray operations be performed in a spray booth or separate, ventilated area.  Dip coating should be ventilated with an enclosure or capture hood.  The prevalence of the engineering controls described in the OSHA Woodworking eTool is unknown.  In addition to the Woodworking eTool, OSHA has a standard for spray finishing operations (29 CFR 1910.107), which provides detailed requirements for the design and construction of spray booths and rooms (OSHA 2002a).
Some information is available on the amount of natural wood products and composite wood products outside the scope of this assessment in fabricator sites.  According to the EPA market profile (USEPA 2010a), natural wood products and composite wood products outside the scope of this assessment (e.g., hardboard) may be used in the composite wood fabrication sectors.  While the three composite wood products within the scope of work have the highest formaldehyde emission rates, the relative contribution to overall exposure of these other wood products may be considerable if they are present at a site in large quantities relative to the three composite wood products.  Tables 6-4 through 6-20 in the EPA market profile (USEPA 2010a) provide the value of raw materials in each of the sectors.  For the following sectors, the other composite wood products (such as softwood plywood and hardboard) comprise 10% or greater of the total composite wood product material costs, and therefore, could be present in significant quantities relative to HWPW, PB and MDF:
Wood window and door manufacturing (NAICS code 321911);
Other millwork (including flooring) (NAICS code 321918);
Manufactured home (mobile home) manufacturing (NAICS code 321991);
Upholstered household furniture manufacturing (NAICS code 337121);
Wood television, radio, and sewing machine cabinet manufacturing (NAICS code 337129); and
Showcase, partition, shelving, and locker manufacturing (NAICS code 337215).
Therefore, for these sectors, other composite wood products may contribute to overall exposure.  However, this analysis is based only on the costs of the materials and not on the actual volumes of the materials.  Data on actual volumes consumed per sector and data to convert material costs to material volumes were not available.
The information above indicates the possibility of the existence of potential formaldehyde exposure sources other than the composite wood products of interest.  The relative significance of these other sources is not known.  For example, the extent to which formaldehyde-containing coatings are used in CWP fabrication is not known and where such materials are used, the extent to which engineering controls such as ventilated spray booths are used to control exposure is also not known.
Number of Establishments, Workers, Production Hours, and Work Days
Estimates for the number of establishments, total number of workers, and aggregate production hours worked for each of the industry sectors that comprise the composite wood product fabrication life cycle stage were obtained from the 2007 U.S. Economic Census (Census 2007a), and are presented in Table 4-6.  It is uncertain whether all sites within each identified NAICS sector use the composite wood products of interest.  Therefore, the number of sites and workers identified in Table 4-6 may represent facilities that do not use composite wood products and are considered maximum values.  Using these data, the average number of workers per site, average production days per year per worker (assuming eight hours per day), and average production hours per day per worker (assuming 250 days per year) were calculated, and the results are given in Table 4-6.  As seen in Table 4-6, all of the calculated values for production days per year (assuming 8 hours per day) do not deviate from the OPPT default value for frequency of exposure of 250 days per year by more than 10 percent.  Based on this observation, the frequency of exposure for composite wood fabrication in general is assessed to be 250 days per year.  Similarly, all of the calculated values for production hours per day (assuming 250 days per year) do not deviate from the OPPT default value for duration of exposure of 8 hours days by more than 10 percent.  Based on this observation, the average duration of exposure for composite wood fabrication in general is assessed to be 8 hours per day.
The result for duration of exposure of 8 hours per day derived from aggregate census data is supported by duration data given in IMIS.  For all composite wood product fabrication industries identified in Table 4-6 and all exposure types (e.g., TWA, STEL), 182 exposure data points (sampled in the year 2002 through 2009) were collected.  These 182 data points have the following exposure duration data:
Exposure durations of blank, "daily", or other frequency estimate without a duration estimate: 88 data points (48%);
Exposure durations of eight hours per day: 80 data points (44%);
Exposure durations of less than eight hours per day: 10 data points (5%); and
Exposure durations of greater than eight hours per day: four data points (2%).

Table 4-6.  Summary of 2007 Economic Census Data and Exposure Frequency and Duration Calculated from These Data for Composite Wood Product Fabricating Sites
                           2007 Economic Census Data
    Parameters Calculated from the Corresponding 2007 Economic Census Data
                                  NAICS Code
                          NAICS Industry Description
                       Maximum Number of Establishments
                     Maximum Number of Production Workers
                  Maximum Aggregate Production Hours Worked 
                             (thousands of hours)
                     Average Production Workers per Site 
      Average Production Days per Year Per Worker (assuming 8 hours/day)
      Average Production Hours per Day Per Worker (assuming 250 days/yr)
321911
Wood Window and Door Manufacturing
                                     1,494
                                    62,089
                                    116,209
                                      42
                                      234
                                     7.49
321918
Other Millwork (including flooring) Manufacturing
                                     2,101
                                    34,394
                                    68,475
                                      16
                                      249
                                     7.96
321991
Manufactured Home (mobile home) Manufacturing
                                      376
                                    32,010
                                    60,725
                                      85
                                      237
                                     7.59
321999
All Other Miscellaneous Wood Product Manufacturing
                                     1,964
                                    27,952
                                    54,290
                                      14
                                      243
                                     7.77
337110
Wood Kitchen Cabinets and Countertop Manufacturing
                                     9,683
                                    107,206
                                    213,487
                                      11
                                      249
                                     7.97
337121
Upholstered Household Furniture Manufacturing
                                     1,633
                                    60,721
                                    118,966
                                      37
                                      245
                                     7.84
337122
Nonupholstered Household Furniture Manufacturing
                                     3,428
                                    49,764
                                    95,685
                                      15
                                      240
                                     7.69
337124
Metal Household Furniture Manufacturing
                                      330
                                     8,827
                                    18,527
                                      27
                                      262
                                     8.40
337127
Institutional Furniture Manufacturing
                                      782
                                    25,848
                                    52,006
                                      33
                                      251
                                     8.05
337129
Wood Television, Radio, Phonograph, and Sewing Machine Cabinet Manufacturing
                                      270
                                     1,541
                                     3,094
                                       6
                                      251
                                     8.03
337211
Wood Office Furniture Manufacturing
                                      457
                                    15,167
                                    31,614
                                      33
                                      261
                                     8.34
337212
Custom Architectural Woodwork and Millwork Manufacturing
                                     2,228
                                    32,184
                                    64,362
                                      14
                                      250
                                     8.00
337214
Office Furniture (except wood) Manufacturing
                                      289
                                    19,307
                                    40,614
                                      67
                                      263
                                     8.41
337215
Showcase, Partition, Shelving, and Locker Manufacturing
                                     1,377
                                    34,481
                                    67,716
                                      25
                                      245
                                     7.86
339950
Sign Manufacturing
                                     6,383
                                    52,755
                                    103,305
                                       8
                                      245
                                     7.83
336213
Motor Home Manufacturing
                                      80
                                    14,152
                                    27,550
                                      177
                                      243
                                     7.79
336214
Travel Trailer & Camper Manufacturing
                                      862
                                    39,282
                                    75,327
                                      46
                                      240
                                     7.67
Totals
                                    33,737
                                    617,680
                                   1,211,952
                                      18
                                      245
                                     7.85
     Sources: 2007 U.S. Economic Census (Census 2007a) and EPA market profile (USEPA 2010a)
Pre-Baseline Exposure Monitoring Data
The exposure monitoring data collected from IMIS for the period 2002 to 2009 on composite wood product fabrication are presented in this section.
Table 4-7 provides a summary of all IMIS 8-hour TWA, personal inhalation monitoring data for industry sectors within the composite wood product fabrication life cycle stage.  Summary data for the individual sectors is provided in Appendix F.  Appendix F also provides the maximum, median, and minimum values of the compiled formaldehyde monitoring data for worker activity categories in each sector.  As discussed in Section 4.2.1, sources of potential exposure in CWP fabrication sites include off-gassing of composite wood products and possibly other sources such as paints, adhesives and coatings, etc.  However, the specific sources of exposure associated with the compiled IMIS data are unknown.
Table 4-7.  Summary of All IMIS TWA Personal Monitoring Data for Industries Within the CWP Fabricator Life Cycle Stage
                                   Statistic
                               Concentration [1]

                                      ppm
                                    mg/m[3]
Maximum
                                       1
                                     1.229
Median
                                     0.091
                                     0.112
Minimum
                                    0.00001
                                   0.000012
Average
                                     0.170
                                     0.209
Number of Data Points
                                      120
          Source: OSHA 2009a
          [1] The formaldehyde concentration in mg/m[3] was calculated from the provided concentration in ppm by multiplying the concentration in ppm by 1.229 mg/m[3]/ppm (see Appendix E).

Respiratory Protection Usage During Composite Wood Product Fabrication
Since OSHA IMIS data do not provide information on personal protective equipment use, OSHA guidance on safe practices in the woodworking industry was reviewed to get an idea of OSHA's recommendations for respiratory protection use during composite wood product fabrication.  Inspection reports associated with the IMIS data were also examined to get an idea of the observed use of respiratory protection by workers when engaged in composite wood product fabrication activities.  Appendix G contains a summary of each reviewed inspection report.
The OSHA Woodworking eTool (OSHA 2002a) offers guidance on safe practices in the woodworking industry.  The eTool requires provision of respirators and gloves to workers performing manual spray activities.  Workers downstream of the spraying operation within the spray booth must be provided with an air-supplying respirator (OSHA 2002a).  The prevalence of the practices required or recommended in the OSHA Woodworking eTool is unknown for the composite wood product fabrication industry.
The 120 TWA, personal monitoring data points discussed in Section 4.2.3 are associated with 61 inspections.  Of these 61 inspections, a total of 13 inspection reports for composite wood product fabrication facilities were obtained from OSHA and reviewed.  These 13 inspection reports account for 39 of the 120 TWA, personal monitoring data points.  Six other inspection reports were requested but not reviewed either because they are no longer available due to their age or because they were not obtained in time to be considered for this assessment.  As previously discussed in Section 4.1.4, one of the 13 reviewed inspection reports (inspection number 312479835; see Appendix G) was for a facility identified by NAICS code as a fabricator.  However, the inspection report stated that the facility manufactures MDF and then fabricates flooring from the MDF.  As the monitored workers were associated with the manufacture of MDF, this inspection report is discussed in Section 4.1.4 and Appendix G with the manufacturer inspection reports.  Although this inspection report is associated with the 120 data points discussed in Section 4.2.3, it is not included in the discussion of this section below.
In addition to the 13 reviewed inspection reports, two inspection reports (inspection numbers 303424014 and 304064835) were obtained and reviewed for inspections that occurred prior to 2002 and hence are not associated with the 120 data points discussed above.  Four other inspection reports for inspections dating prior to 2002 were requested but not reviewed either because they are no longer available due to their age or because they were not obtained in time to be considered for this assessment.  Additionally, a single wholesale facility inspection report was requested, received, and reviewed (inspection number 306629171; see Appendix G).  This inspection report is associated with the single wholesale facility IMIS data discussed in Section 4.3.3.  Although IMIS identifies the facility with a wholesaler NAICS code, worker activities described in the IMIS data and inspection report indicate the presence of fabrication activities.  Therefore, this inspection report is discussed in this section and Appendix G with the CWP fabrication inspection reports.  In conclusion, a total of 16 inspection reports were reviewed, and 15 of the reviewed inspection reports are discussed in this section.
In the 15 reviewed reports discussed in this section, all measured formaldehyde concentrations were below the OSHA PEL of 0.75 ppm for formaldehyde, with levels exceeding the action level (AL) of 0.5 ppm in only one facility (inspection number 306021239; see Appendix G).  This single facility is a manufacturer of bath and kitchen wood cabinets and had three TWA measurements exceed the action level.  Workers were not observed to be wearing respirators during the inspection at this site.
Five of the fifteen reports (33%) reviewed indicated the use of a respirator by workers as follows:
One of the five reports identified the use of a half-face air purifying respirator in the mixing area where a formaldehyde-based glue is mixed and applied to wooden frame chairs via pneumatic guns.
Two other reports identified the use of cartridge respirators during spray application of formaldehyde-based coatings.
Another report identified a facility that did not require respirator use in the spray booths, although some workers were observed to be wearing organic vapor cartridge respirators during spraying activities.
The fifth report was for the inspection of a facility in which the only identified inhalation PPE in use was the N95 disposal particulate respirator.
An additional report indicated voluntary use of respirators but respirators were not worn during the inspection.  The remaining nine inspection reports did not note the use or availability of respirators.
As in MDF manufacturing, respirator use by workers in composite wood product fabrication facilities has been noted when formaldehyde levels are below its AL.  This may be a precautionary practice because spraying activities can increase the potential for exposure to formaldehyde or it might result from the need to protect against chemicals and particulates other than formaldehyde.  Only a single instance of the lack of respirator use has been noted (inspection number 306021239; see Appendix G) when exposure levels exceeded the AL (but were below the PEL).
The limited number of inspection reports that were available and examined show worker exposures to be below the formaldehyde PEL but do not provide enough information to conclude that respirators would necessarily be worn by workers when their exposure meets or exceeds the formaldehyde action level or PEL.
Baseline Exposure and the Effect of Analytical Options on Baseline Exposure for Composite Wood Product Fabrication
Monitoring data are a preferred approach for assessing exposure.  However, as previously discussed in Section 3.2, the monitoring data are not adequate for assessing exposure at the baseline and for the analytical options.  Therefore, this exposure assessment uses mathematical modeling to estimate indoor air concentration for the baseline and the analytical options.  The mathematical modeling for indoor air concentration is presented in Section 3.3.1.1 and described in detail in Appendix E.
The model parameters include total emitting surface areas for the three CWPs in a CWP fabrication site, CWP formaldehyde off-gassing emission rates, total effective ventilation rate for a CWP fabrication site, and concentration of formaldehyde in the outside air.  As discussed in Section 3.3.1.1, due to data gaps in information on the values of these four parameters at representative CWP fabrication sites, a range of what-if values for indoor air concentration was calculated from bounding values for the ranges of three of the above four parameters.  The what-if method for estimating indoor air concentration was applied to three limiting cases for the CWP type present in the CWP fabrication sites: only HWPW, or only PB, or only MDF.  The calculation of concentration followed the algorithm outlined in Table 3-2.  For each iteration of the algorithm, one parameter is varied between its minimum and maximum values while the other two varying parameters remain at their intermediate values.  The fourth parameter, ambient air concentration of formaldehyde, is kept at a constant value.  These steps were repeated for each CWP type.  The what-if method used values for each parameter according to Table 3-4.
Results for Indoor Air Concentration Calculation Trials and Sensitivity Analysis
Results for the calculation trials for indoor air concentration that summarized in Table 3-2 are presented below.  A sensitivity analysis based on these results was conducted and is discussed below.  Results are presented in Table 4-8 for each of the three calculation trials for the baseline low-end air concentration for each of the three limiting cases for CWP type and similarly for the high-end indoor air concentration.  This table also contains the results of the calculation of a baseline intermediate indoor air concentration for each of the three CWP types.  Similar results are presented in Table 4-9 through Table 4-11 for the CARB ATCM Phase 1, the FSCWPA / CARB ATCM Phase 2, and the NAF analytical options.
The results in Table 4-8 through Table 4-11 demonstrate the sensitivity of the calculated indoor air concentration to variation in the value of each model parameter.  The purpose of observing the sensitivity of the mass balance equation is to understand the impact of the estimated range of values for each parameter of the equation on the calculated concentration.  As the value of one parameter is varied while the other parameters are kept constant, the changes in the calculated concentration resulting from the variation of a single parameter can be compared to the changes in concentration resulting from the variation of other parameters.  For example, the row labeled "Vary Emitting Surface Area" shows the changes in indoor air concentration from varying the emitting surface area from its minimum to its maximum values while keeping effective ventilation rate and emission rate constant at their intermediate values.  This trend is shown for each CWP type.  The results indicate that, for each CWP type, the indoor air concentration is least sensitive to emission rate.  The row "Vary Emission Rate" shows the smallest relative changes in indoor air concentration from low-end to high-end value (relative change is defined as the difference between the high-end and low-end values divided by the low-end value).  Note that for each "vary scenario" the MDF scenario estimates higher exposures than the HWPW and PB scenarios due to the higher emission rates of MDF.
The rows "Vary Emitting Surface Area" and "Vary Effective Ventilation Rate" show relative changes in indoor air concentration of a similar order of magnitude.  At baseline, the relative changes in indoor air concentration due to varying emitting surface area and effective ventilation rate are approximately two to nine times higher than the relative changes due to varying emission rate.  The relative changes in indoor air concentration due to varying emitting surface area are higher than the relative changes due to varying effective ventilation rate.  This effect is due to the wide range of values of emitting surface area chosen (10% and 1,000% of the intermediate value).  This is especially notable given that mathematically the mass balance equation would show greater sensitivity to the effective ventilation rate in general since indoor air concentration is inversely proportional to the effective ventilation rate.  Therefore, particularly at lower values of effective ventilation rate, the indoor air concentration will change more rapidly as effective ventilation rate is changed.  On the other hand, indoor air concentration is directly proportional to emitting surface area and emission rate.  Therefore, indoor air concentration changes linearly with both emitting surface area and emission rate.
Estimation of a Range of Values for the What-If Low-End and the What-If High-End Indoor Air Concentrations that Account for All CWP Types
As summarized in Table 3-2, the what-if low-end and high-end indoor air concentrations for each CWP type were estimated by choosing the minimum and maximum, respectively, of the results of the three calculation trials made for each of these two concentration values.  As discussed in Section 3.3.1.1, the concentration values that were chosen as the estimate of the what-if low-end indoor air concentration for each CWP type were combined into a single range of what-if low-end indoor air concentrations that encompasses all three estimates to account for the possibility that a mixture of the three CWP types is used in a CWP fabrication site.  This was also done for the chosen high-end indoor air concentration for each CWP type.  These ranges for the values of the what-if low-end indoor air concentration and what-if high-end indoor air concentration are given Table 4-12 in for the baseline and the analytical options.
Final Results for the Estimation of What-If Indoor Air Concentrations
The estimated ranges for low- and high-end what-if indoor air concentration given in Table 4-12 were combined into an overall range for the baseline and each of the analytical options.  The overall range is defined as the range bound by the low-end concentration from the low-end range and the high-end concentration from the high-end range.  These overall ranges are presented in Table 4-13.  The best estimate concentration values summarized in Table 4-13 are calculated as the midpoint of the overall range of estimated concentration values for the baseline and each analytical option.  The midpoint of the overall range of estimated concentrations is chosen to represent the best estimate due to the varying uncertainties of the model parameter values, which are discussed in Section 3.3.1.1.  Table 4-13 also provides the percentage reductions in indoor air concentration from the baseline value for the analytical options.
The rationale for this best estimate comes from considering pre-baseline IMIS data for fabrication and uncertainties in model parameters.  Modeling results obtained using pre-baseline emission levels were compared to the pre-baseline IMIS data for fabrication.  This comparison is presented in Figure 4-1 and shows that the midpoint of the lowest and highest model results (the midpoint of the overall range of model results) compares well to the median value of the IMIS data.  The pre-baseline emission levels used in the calculations are given in Appendix D and the corresponding modeling results are given in Appendix I.  However, comparing IMIS data, which are aggregate exposures that may have one or more sources of formaldehyde of which CWPs may or may not be included, to model results for off-gassing alone, is highly uncertain.  Likewise, the representativeness of the model parameter values are of varying uncertainty, as discussed in Section 3.3.1.1.  Therefore, it is most logical to select the midpoint of the lowest and highest model results as the best estimate of formaldehyde concentration.

                                       
Figure 4-1. Comparison of Model Results for Ranges of What-If Indoor Air Concentrations using Pre-Baseline Emission Rates with OSHA IMIS Data
The range of what-if indoor air concentrations given in Table 4-13 is the estimate of the range of values for the average concentration of airborne formaldehyde in a CWP fabrication site that is due to off-gassing from HWPW, PB, and MDF.  As presented in Section 3.3.1.1 and described in detail in Appendix E, these concentration values were estimated based on a steady-state mass balance that equates the formaldehyde generation rate due to off-gassing from HWPW, PB, and MDF present at a CWP fabrication site to the rate at which airborne formaldehyde is exhausted from the CWP fabrication site through ventilation with outdoor air.  In this model, which is discussed in detail in Section 3.3.1.1, the off-gassing emissions from all of the exposed CWP surfaces are diluted with outside air, which contains the ambient concentration of formaldehyde, and the fabrication site is assumed to be a well-mixed box resulting in a uniform indoor air concentration throughout the site.  The assumption of a well-mixed box model is based on the assumption that the CWP as well as the ventilation sources are for the most part uniformly distributed in a CWP fabrication site.
As discussed in Section 3.3.1.1, due to data gaps in information on the values of the four model parameters at representative CWP fabrication sites, a range of what-if values for indoor air concentration was calculated from bounding values for the ranges of three of the four parameters.  The estimated range for average indoor air concentration at a CWP fabrication site is characterized as a range of what-if results because each concentration value in this range corresponds to combinations of choices, or what-if values, for the model parameters.  Although the prevalence of the model parameter what-if values at actual CWP fabrication sites is unknown, the ranges of the model parameter values should be realistic such that any given choice, or what-if value, for any model parameter represents a possible value at actual CWP fabrication sites.  However, as discussed in Section 3.3.1.1, the ranges of the model parameters were estimated with varying levels of accuracy.  It was not possible to validate the modeling results for indoor air concentration as further discussed in Section 5.1.1.
The Assessment of Exposure Level for the Baseline and the Analytical Options
The duration of exposure was assessed to be eight hours per day as presented in Section 4.2.2; therefore, the estimated what-if steady-state average indoor air concentrations in a CWP fabrication site presented in Table 4-13 are assessed as the 8-hour TWA exposure concentration.  As discussed in Section 4.2.4, there is insufficient information to determine whether respiratory PPE is utilized by workers and therefore the estimated what-if indoor air concentration was assessed as the exposure concentration.
This exposure concentration is due only to off-gassing from the CWP products.  As discussed in Section 4.2.1, there is some indication of potential sources of formaldehyde inhalation exposure in a CWP fabrication site other than off-gassing of formaldehyde from the three CWPs (such as adhesives and the application of overlays).  The relative significance of the contribution of these other potential sources to aggregate exposure is unknown.  Different approaches were considered for their assessment but they were not assessed due to data gaps.
Average daily concentration (ADC) and lifetime average daily concentration (LADC) were calculated for each indoor air concentration value reported in Table 4-13.  As described in Appendix E, ADC and LADC were calculated assuming exposure duration of 8 hours per day and exposure frequency of 250 days per year over 40 working years per lifetime for ADC and over a lifetime of 70 years for LADC.   The ADC and LADC results are presented in Table 4-14 and Table 4-15.
Results in Table 4-13 show that the maximum formaldehyde exposure concentration decreases as lower (more stringent) emission standards are implemented.  However, the CARB ATCM Phase 1 limits result in little or no reduction in exposure from the baseline year.  This effect is because many composite wood products are expected to be already meeting more stringent standards by 2013.  Note the minimum exposure concentrations remain constant throughout the different analytical option periods.  This effect is because a small portion of the market is expected to be already meeting the NAF standards by the baseline period.  The emission levels of NAF-compliant composite wood products are assumed to remain unchanged and no effect of analytical options is expected for these products.


Table 4-8. Results for the Calculation Trials for the What-If Low- and High-End Baseline Indoor Air Concentrations
                               Varied Parameter
                  Calculated Concentration for Baseline, ppm
                                       
                               Intermediate [4]
                                  Low End [5]
                                 High End [6]
                                       
                                     HWPW
                                      PB
                                      MDF
                                     HWPW
                                      PB
                                      MDF
                                     HWPW
                                      PB
                                      MDF
Vary Emitting Surface Area [1]
                                    0.0095
                                       
                                    0.0110
                                       
                                    0.0181
                                       
                                    0.0055
                                    0.0057
                                    0.0064
                                    0.0488
                                    0.0637
                                    0.1345
Vary Effective Ventilation Rate [2]
                                       
                                       
                                       
                                    0.0058
                                    0.0060
                                    0.0071
                                    0.1362
                                    0.1809
                                    0.3934
Vary Emission Rate [3]
                                       
                                       
                                       
                                    0.0064
                                    0.0064
                                    0.0092
                                    0.0218
                                    0.0445
                                    0.0902
  [1] For area, the intermediate value is based on the aggregate average per-site throughput calculated using domestic composite wood product consumption data from the EPA market profile. The high-end and low-end values are calculated assuming 1,000% and 10% of the aggregate average throughput, respectively.
  [2] Effective ventilation is varied based on values shown in Table 3-3. The low-end, typical, and high-end values are 50, 1,500 and 10,000 cfm, respectively.
  3 Emission rate is varied based on values shown in Table 3-4.
  [4] Intermediate concentration is calculated using intermediate value for surface area, typical value for effective ventilation rate, and weighted-average baseline emission rate for each CWP type.
  [5] Low-end values calculated according to Table 3-2.
  [6] High-end values calculated according to Table 3-2.


Table 4-9. Results for the Calculation Trials for the What-If Low- and High-End CARB Phase 1 Indoor Air Concentrations
                               Varied Parameter
Calculated Concentration for Phase 1, ppm, and % Change from Corresponding Baseline Value
                                       
                               Intermediate [4]
                                  Low End [5]
                                 High End [6]
                                       
                                     HWPW
                                      PB
                                      MDF
                                     HWPW
                                      PB
                                      MDF
                                     HWPW
                                      PB
                                      MDF
Vary Emitting Surface Area [1]
                                    0.0077
                                   (-18.8%)
                                       
                                     0.011
                                    (-3.0%)
                                       
                                    0.0176
                                    (-2.7%)
                                       
                                    0.0054
                                    (-3.2%)
                                    0.0057
                                    (-0.6%)
                                    0.0064
                                    (-0.8%)
                                    0.0309
                                   (-36.6%)
                                    0.0604
                                    (-5.2%)
                                    0.1296
                                    (-3.6%)
Vary Effective Ventilation Rate [2]
                                       
                                       
                                       
                                    0.0055
                                    (-4.6%)
                                    0.0059
                                    (-0.8%)
                                    0.0070
                                    (-1.0%)
                                    0.0826
                                   (-39.3%)
                                    0.1710
                                    (-5.5%)
                                    0.3787
                                    (-3.7%)
Vary Emission Rate [3]
                                       
                                       
                                       
                                    0.0064
                                     (0%)
                                    0.0064
                                     (0%)
                                    0.0092
                                     (0%)
                                    0.0111
                                   (-49.3%)
                                    0.0143
                                   (-67.8%)
                                    0.0288
                                   (-68.0%)
  [1] For area, the intermediate value is based on the aggregate average per-site throughput calculated using domestic composite wood product consumption data from the EPA market profile. The high-end and low-end values are calculated assuming 1,000% and 10% of the aggregate average throughput, respectively.
  [2] Effective ventilation is varied based on values shown in Table 3-3. The low-end, typical, and high-end values are 50, 1,500 and 10,000 cfm, respectively.
  3 Emission rate is varied based on values shown in Table 3-4.
  [4] Intermediate concentration is calculated using intermediate value for surface area, typical value for effective ventilation rate, and weighted-average Phase 1 emission rate for each CWP type.
  [5] Low-end values calculated according to Table 3-2.
  [6] High-end values calculated according to Table 3-2.

Table 4-10. Results for the Calculation Trials for the What-If Low- and High-End FSCWPA / CARB Phase 2 Indoor Air Concentrations
                               Varied Parameter
Calculated Concentration for FSCWPA / Phase 2, ppm, and % Change from Corresponding Baseline Value
                                       
                               Intermediate [4]
                                  Low End [5]
                                 High End [6]
                                       
                                     HWPW
                                      PB
                                      MDF
                                     HWPW
                                      PB
                                      MDF
                                     HWPW
                                      PB
                                      MDF
Vary Emitting Surface Area [1]
                                    0.0072
                                   (-23.5%)
                                       
                                     0.011
                                    (-3.9%)
                                       
                                    0.0175
                                    (-3.2%)
                                       
                                    0.0053
                                    (-4.0%)
                                    0.0057
                                    (-0.7%)
                                    0.0063
                                    (-0.9%)
                                    0.0265
                                   (-45.7%)
                                    0.0595
                                    (-6.7%)
                                    0.1288
                                    (-4.2%)
Vary Effective Ventilation Rate [2]
                                       
                                       
                                       
                                    0.0054
                                    (-5.8%)
                                    0.0059
                                    (-1.1%)
                                    0.0070
                                    (-1.2%)
                                    0.0693
                                   (-49.1%)
                                    0.1681
                                    (-7.1%)
                                    0.3763
                                    (-4.3%)
Vary Emission Rate [3]
                                       
                                       
                                       
                                    0.0064
                                     (0%)
                                    0.0064
                                     (0%)
                                    0.0092
                                     (0%)
                                    0.0084
                                   (-61.5%)
                                    0.0110
                                   (-75.4%)
                                    0.0190
                                   (-78.9%)
  [1] For area, the intermediate value is based on the aggregate average per-site throughput calculated using domestic composite wood product consumption data from the EPA market profile. The high-end and low-end values are calculated assuming 1,000% and 10% of the aggregate average throughput, respectively.
  [2] Effective ventilation is varied based on values shown in Table 3-3. The low-end, typical, and high-end values are 50, 1,500 and 10,000 cfm, respectively.
  3 Emission rate is varied based on values shown in Table 3-4.
  [4] Intermediate concentration is calculated using intermediate value for surface area, typical value for effective ventilation rate, and weighted-average Phase 2 emission rate for each CWP type.
  [5] Low-end values calculated according to Table 3-2.
  [6] High-end values calculated according to Table 3-2.


Table 4-11. Results for the Calculation Trials for the What-If Low- and High-End NAF Indoor Air Concentrations
                               Varied Parameter
Calculated Concentration for Phase NAF, ppm, and % Change from Corresponding Baseline Value
                                       
                               Intermediate [4]
                                  Low End [5]
                                 High End [6]
                                       
                                     HWPW
                                      PB
                                      MDF
                                     HWPW
                                      PB
                                      MDF
                                     HWPW
                                      PB
                                      MDF
Vary Emitting Surface Area [1]
                                    0.0064
                                   (-32.0%)
                                       
                                    0.0064
                                   (-41.3%)
                                       
                                    0.0110
                                   (-39.3%)
                                       
                                    0.0052
                                    (-5.5%)
                                    0.0052
                                    (-8.0%)
                                    0.0057
                                   (-11.1%)
                                    0.0184
                                   (-62.2%)
                                    0.0184
                                   (-71.1%)
                                    0.0637
                                   (-52.7%)
Vary Effective Ventilation Rate [2]
                                       
                                       
                                       
                                    0.0053
                                    (-7.9%)
                                    0.0053
                                   (-11.3%)
                                    0.0060
                                   (-15.1%)
                                    0.0451
                                   (-66.9%)
                                    0.0451
                                   (-75.1%)
                                    0.1807
                                   (-54.1%)
Vary Emission Rate [3]
                                       
                                       
                                       
                                    0.0064
                                     (0%)
                                    0.0064
                                     (0%)
                                    0.0092
                                     (0%)
                                    0.0064
                                   (-70.5%)
                                    0.0064
                                   (-85.5%)
                                    0.0110
                                   (-87.8%)
    [1] For area, the intermediate value is based on the aggregate average per-site throughput calculated using domestic composite wood product consumption data from the EPA market profile. The high-end and low-end values are calculated assuming 1,000% and 10% of the aggregate average throughput, respectively.
    [2] Effective ventilation is varied based on values shown in Table 3-3. The low-end, typical, and high-end values are 50, 1,500 and 10,000 cfm, respectively.
    3 Emission rate is varied based on values shown in Table 3-4.
    [4] Intermediate concentration is calculated using intermediate value for surface area, typical value for effective ventilation rate, and weighted-average NAF emission rate for each CWP type.
    [5] Low-end values calculated according to Table 3-2.
    [6] High-end values calculated according to Table 3-2.

Table 4-12.  Estimated Low-End, Intermediate, and High-End Ranges of Values for the Site-Average Indoor Air Concentration at the Baseline and for the Analytical Options
                                     Case
                              Intermediate (ppm)
                                 Low End (ppm)
                                High End (ppm)
Baseline 
                               0.0095  -  0.0181
                               0.0055  -  0.0092
                               0.0218  -  0.3934
Phase 1
                      0.0077 (-18.8%)  -  0.0176 (-2.7%)
                        0.0054 (-3.2%)  -  0.0092 (0%)
                      0.0111 (-49.3%)  -  0.3787 (-3.7%)
FSCWPA / Phase 2
                      0.0072 (-23.5%)  -  0.0175 (-3.2%)
                        0.0053 (-4.0%)  -  0.0092 (0%)
                      0.0084 (-61.5%)  -  0.3763 (-4.3%)
NAF
                      0.0064 (-32.0%)  -  0.0110 (-39.3%)
                        0.0052 (-5.5%)  -  0.0092 (0%)
                      0.0064 (-70.5%)  -  0.1807 (-54.1%)


Table 4-13. Estimated Overall Range and Best Estimate Site-Average Indoor Air Concentration at the Baseline and for the Analytical Options
                                     Case
                              Overall Range (ppm)
                      Best Estimate Concentration1 (ppm)
                          Reduction from Baseline (%)
Baseline 
                               0.0055  -  0.3934
                                    0.1995
                                      --
Phase 1
                               0.0054  -  0.3787
                                    0.1920
                                     -3.7
FSCWPA / Phase 2
                               0.0053  -  0.3763
                                    0.1908
                                     -4.3
NAF
                               0.0052  -  0.1807
                                    0.0930
                                     -53.4
                               [1] The best estimate concentration is the midpoint of the overall range. The overall range is defined as the low-end value from the low-end range and the high-end value from the high-end range. The reduction from baseline, ADC, and LADC values are calculated from the best estimate concentration value.

Table 4-14. Overall Range and Best Estimate ADC at the Baseline and for the Analytical Options
                                     Case
                           Overall Range (ug/m[3])
                         Best Estimate ADC (ug/m[3])
Baseline 
                                1.556  -  110.4
                                     56.0
Phase 1
                                1.506  -  106.3
                                     53.9
FSCWPA / Phase 2
                                1.494  -  105.6
                                     53.5
NAF
                                1.471  -  50.71
                                     26.1

Table 4-15. Overall Range and Best Estimate LADC at the Baseline and for the Analytical Options
                                     Case
                            Intermediate (ug/m[3])
                         Best Estimate LADC (ug/m[3])
Baseline 
                               0.8894  -  63.08
                                     32.0
Phase 1
                               0.8607  -  60.72
                                     30.8
FSCWPA / Phase 2
                               0.8536  -  60.34
                                     30.6
NAF
                               0.8407  -  28.98
                                     14.9

Dermal Exposures during Composite Wood Product Fabrication
Dermal exposures to CWPs are not quantifiable using the EPA dermal exposure models for liquids and solids (these models are only applicable to dermal exposure to uncured resin, which may occur upstream of the press in the CWP manufacturing process).
Exposure Assessment for Wholesale
This section presents and discusses the assessment of formaldehyde exposures from the wholesale of composite wood products.
Description of Life Cycle Stage
Wholesalers buy products from manufacturers in bulk for the purpose of resale to various industries and individuals.  Wholesale products may include composite wood products and CWP fabrication products (downstream products).  A single wholesaler may sell many different types of composite wood products.  The U.S. Census Bureau defines wholesale facilities as those who "sell or arrange the purchase or sale of capital or durable goods to other businesses" (Census 2007a).  However, IMIS data, which are presented in Section 4.3.3, indicate that some wholesale facilities may also process composite wood products on site (the reported worker activities at wholesale facilities include "sawing").
Use Volumes and Number of Establishments, Employees, Employed Hours, and Work Days
Information related to the number of sites, number of employees, and facility-specific use rates (including storage and wholesale) of composite wood products and CWP fabrication articles made of composite wood products were researched.  Data for the number of sites and employees from the 2007 U.S. Economic Census are provided in the EPA market profile (USEPA 2010a).  However, actual loading data of the total emitting surface area of composite wood products at wholesale settings were not found.  Table 4-16 summarizes the census data identified for wholesale facilities that may sell composite wood products or articles fabricated from composite wood products as obtained from the EPA market profile.  It is uncertain whether all facilities in the identified NAICS codes sell composite wood products or downstream articles made of composite wood products.  Therefore, the number of sites and employees identified in Table 4-16 may represent wholesalers that do not use composite wood products and are considered maximum values.

Table 4-16.  Summary of 2007 Economic Census Data for Wholesale Facilities
                                  NAICS Code
                          NAICS Industry Description
                       Maximum Number of Establishments
                       Maximum Number of Paid Employees
                                                                         423210
Furniture merchant wholesalers
                                                                          6,640
                                                                         69,103
                                                                         423220
Home furnishing merchant wholesalers
                                                                          7,540
                                                                         93,060
                                                                         423310
Lumber, plywood, millwork, and wood panel merchant wholesalers
                                                                          8,428
                                                                        138,739
                                                                         423320
Brick, stone, and related construction material merchant wholesale
                                                                          3,383
                                                                         35,005
                                                                         423330
Roofing, siding, and insulation material merchant wholesalers
                                                                          2,587
                                                                         32,662
                                                                         423390
Other construction material merchant wholesalers
                                                                          3,383
                                                                         37,920
                                                                         423440
Other commercial equipment merchant wholesalers
                                                                          4,352
                                                                         48,976
                                                                         423450
Medical equipment merchant wholesalers
                                                                          8,282
                                                                        141,282
                                                                         423490
Other professional equipment and supplies merchant wholesalers
                                                                          2,297
                                                                              D
                                                                         423510
Metal service centers and other metal merchant wholesalers
                                                                         10,148
                                                                        144,954
                                                                         423610
Elec. equip. and wiring merchant wholesalers
                                                                         13,277
                                                                        169,355
                                                                         423620
Electric appliance merchant wholesalers
                                                                          3,110
                                                                         44,819
                                                                         423710
Hardware merchant wholesalers
                                                                          6,828
                                                                         83,542
                                                                         423720
Plumbing equip. merchant wholesalers
                                                                          6,939
                                                                         86,092
                                                                         423730
HVAC equip. merchant wholesalers
                                                                          5,071
                                                                         50,495
                                                                         423740
Refrigeration equipment and supplies merchant wholesalers
                                                                          1,265
                                                                         11,024
                                                                         423830
Industrial machinery and equipment merchant wholesalers
                                                                         29,838
                                                                        319,557
                                                                         423850
Service establishment equipment and supplies merchant wholesalers
                                                                          4,576
                                                                          D[10]
                                                                         423910
Sporting and recreational goods and supplies merchant wholesalers
                                                                          6,192
                                                                         55,842
                                                                         424610
Plastics materials and basic forms and shapes merchant wholesaler
                                                                          3,248
                                                                         32,552
                                                                         424950
Paint, varnish, and supplies merchant wholesalers
                                                                          1,649
                                                                         14,631
   Source: EPA market profile (USEPA 2010a)

Pre-Baseline Exposure Monitoring Data
Formaldehyde monitoring data for various worker activities at various wholesale settings were obtained from the OSHA IMIS database (OSHA 2009a).  Only two data points from a single wholesale site classified under one of the NAICS codes identified in Table 2-1 were available for the pre-baseline period (2002 to 2009).  Table 4-17 presents these data by worker activity.
Table 4-17. Summary of Formaldehyde Inhalation Monitoring Data for Wholesalers from the OSHA IMIS Database
                                   Industry
                          Worker Activity Description
                           Number of Sites Monitored
                  Monitored Formaldehyde Concentration (ppm)
              Monitored Formaldehyde Concentration (ug/m[3])[1]
                                Monitoring Type
                                     Year
Furniture Merchant Wholesalers
Band Saw Operator
                                       1
                                     0.002
                                     2.46
                                   Personal
                                     2003



                                 Not Detected
                                 Not Detected
                                   Personal
                                     2003

Panel Saw Operator

                                     0.002
                                     2.46
                                   Personal
                                     2003

CNC Operator

                                 Not Valid[2]
                                   Not Valid
                                   Personal
                                     2003
 [1] The formaldehyde concentration in ug/m[3] was calculated from the provided concentration in ppm by multiplying the concentration in ppm by 1,229 ug/m[3]/ppm (see Appendix E).
 [2] Data point reported as "not valid" in the IMIS database; likely an invalid measurement. 

Baseline Exposure and the Effect of Analytical Options on Baseline Exposure for Wholesale
Exposures in wholesale settings are not assessed quantitatively due to data gaps.  There is a lack of information on the ventilation rate at wholesale settings to allow for a what-if analysis similar to what was completed for the fabrication life cycle stage.  Therefore, it is not possible to accurately quantify the exposure concentrations during the baseline period and the reduction of exposure concentrations resulting from the analytical options.
The formaldehyde concentration resulting from off-gassing is expected to decrease with the implementation of more stringent regulations.  However, as discussed previously, the implementation of CARB ATCM Phase 1 limits is expected to result in little or no reduction in exposure from the baseline year.  This is because most composite wood products in the market would already be meeting more stringent limits by 2013.
Dermal Exposures during Wholesale
Dermal exposures to CWPs are not quantifiable using the EPA dermal exposure models for liquids and solids.
Exposure Assessment for Retail
This section presents and discusses the assessment of formaldehyde exposures from retail due to composite wood products.
Description of Life Cycle Stage
Retailers operate from fixed locations and sell products for direct consumption by the purchaser (individuals or businesses).  These retailers may receive HWPW, MDF, or PB directly as products or included in articles fabricated using composite wood products.  Like wholesale sectors, retail sectors are not distinguished by composite wood product type (Census 2007a).
Use Volumes and Number of Establishments, Employees, Employed Hours, and Work Days
Information related to the number of sites, number of employees, and facility-specific use rates (including storage and retail) of composite wood products and CWP fabrication products were researched.  The number of sites and employees data from the 2007 U.S. Economic Census (Census 2007a) was identified.  However, actual loading data of the total emitting surface area of composite wood products at wholesale settings were not found.  Table 4-18 summarizes the census data identified for retail facilities that may sell composite wood products or articles fabricated from composite wood products.  It is uncertain whether all facilities in the identified NAICS codes sell composite wood products or downstream articles made of composite wood products.  Therefore, the number of sites and employees identified in Table 4-18 may represent retailers that do not use composite wood products and are considered maximum values.
The duration and frequency of exposure were assumed to be 8 hours per day and 250 days per year, respectively.  This is supported to some extent by IMIS data on duration.  Some exposure data points in IMIS do not include an exposure duration and other data points include an exposure duration of "daily" or other frequency without comment on the number of hours of exposure per day.  For retail, 17 exposure data points (sampled in 2002 through 2009) were collected.  For these 17 data points, the following exposure duration data were included in IMIS:
Exposure durations of blank or "daily": seven data points (41%);
Exposure durations of eight hours per day: three data points (18%);
Exposure duration of three hours per week: one data point (6%); 
Exposure duration of 40 hours per week: three data points (18%); and
Exposure duration of "monthly", "routinely", or "varies": three data points (18%).

Table 4-18.  Summary of 2007 Economic Census Data for Retail Facilities
                                    NAICS 
                                     Code
                          NAICS Industry Description
                                Maximum Number
                               of Establishments
                                Maximum Number 
                               of Paid Employees
                                    441210
Recreational vehicle dealers
                                                                          3,100
                                                                         42,669
                                    442110
Furniture stores
                                                                         28,834
                                                                        266,374
                                    442210
Floor covering stores
                                                                         14,534
                                                                         89,420
                                    442291
Window treatment stores
                                                                          2,652
                                                                          9,626
                                    442299
All other home furnishings stores
                                                                         19,124
                                                                        192,057
                                    443111
Household appliance stores
                                                                          9,336
                                                                         70,045
                                    443112
Radio, television, and other electronics stores
                                                                         29,051
                                                                        309,374
                                    443120
Computer and software stores
                                                                         10,428
                                                                         92,777
                                    444110
Home centers
                                                                          6,953
                                                                        579,442
                                    444120
Paint and wallpaper stores
                                                                          7,860
                                                                         35,491
                                    444130
Hardware stores
                                                                         16,454
                                                                        143,738
                                    444190
Other building material dealers
                                                                         39,551
                                                                        405,755
                                    444210
Outdoor power equipment stores
                                                                          4,357
                                                                         25,882
                                    444220
Nursery, garden center, and farm supply stores
                                                                         15,895
                                                                        141,014
                                    445110
Supermarkets and other grocery (except convenience) stores
                                                                         64,881
                                                                      2,432,425
                                    448110
Men's clothing stores
                                                                          8,321
                                                                         57,340
                                    448120
Women's clothing stores
                                                                         36,220
                                                                        339,798
                                    448130
Children's and infants' clothing stores
                                                                          7,102
                                                                         98,729
                                    448140
Family clothing stores
                                                                         28,254
                                                                        642,117
                                    448150
Clothing accessories stores
                                                                          7,494
                                                                         37,575
                                    448190
Other clothing stores
                                                                         12,034
                                                                         97,581
                                    451120
Hobby, toy, and game stores
                                                                          9,357
                                                                        135,688
                                    451130
Sewing, needlework, and piece goods stores
                                                                          5,663
                                                                         49,450
                                    452111
Department stores (except discount department stores)
                                                                          3,533
                                                                        534,189
                                    452112
Discount department stores
                                                                          5,001
                                                                        690,265
                                    452910
Warehouse clubs and supercenters
                                                                          4,260
                                                                      1,244,237
                                    452990
All other general merchandise stores
                                                                         33,061
                                                                        294,783
                                    453110
Florists
                                                                         19,822
                                                                         93,967
                                    453210
Office supplies and stationery stores
                                                                          9,558
                                                                        112,734
                                    453220
Gift, novelty, and souvenir stores
                                                                         31,391
                                                                        193,317
                                    453920
Art dealers
                                                                          6,677
                                                                         23,126
                                    453930
Manufactured (mobile) home dealers
                                                                          3,878
                                                                         21,068
                                    453998
All other miscellaneous store retailers (except tobacco stores)
                                                                         17,270
                                                                         88,296
                                    454113
Mail-order houses
                                                                          7,324
                                                                        220,516
                                    454390
Other direct selling establishments
                                                                         22,077
                                                                        131,013
Source: EPA market profile (USEPA 2010a)

Pre-Baseline Exposure Monitoring Data
Formaldehyde exposure monitoring data for various worker activities at various retail settings obtained from the OSHA IMIS database (OSHA 2009a) for the period of 2002 to 2009 are presented in Table 4-19.  Pre-baseline data are available for seven retail sectors only, and for most only a single monitoring data point is available.  A total of 17 data points are available for the entire sector during the pre-baseline period.

Table 4-19.  Summary of Formaldehyde Inhalation Monitoring Data for Retailers from the OSHA IMIS Database
                                   Industry
                          Worker Activity Description
                           Number of Sites Monitored
                             Number of Data points
                  Monitored Formaldehyde Concentration (ppm)
              Monitored Formaldehyde Concentration (ug/m[3])[1]
                                Monitoring Type
                                     Year
Recreational Vehicle Dealers
Sawing
                                       1
                                       1
0.002
2.46
Personal
                                     2006
Furniture Stores
Cabinet Maker
                                       1
                                       1
0.069
84.8
Personal
                                     2002

Front Right of Store
                                       1
                                       5
Not Detected [2]
Not Detected [2]
Area
                                     2008

Front Left of Store
                                       
                                       
Not Detected [2]
Not Detected [2]
Area
                                     2008

Back Left of Store
                                       
                                       
Not Detected [2]
Not Detected [2]
Area
                                     2008

Back Right of Store
                                       
                                       
Not Detected [2]
Not Detected [2]
Area
                                     2008

Middle of Store
                                       
                                       
Not Detected [2]
Not Detected [2]
Area
                                     2008
All Other Home Furnishing Stores
Sale
                                       1
                                       1
0.017
20.9
Area
2006
Radio, Television & Other Electronic Stores
Warehouse Handler
                                       1
                                       1
0.14
172.1
Personal
2004
Home Centers
Cabinet Maker/Painter
                                       1
                                       1
0.00001 [2]
0.012 [2]
Personal
2007
Other Building Material Dealers
Laminator
                                       1
                                       1
0.055
67.6
Personal
2003

Painting
                                       1
                                       1
0.1
122.9
Personal
2008

Operator
                                       
                                       1
0.19
233.5
Personal
2008

Driver
                                       1
                                       1
Not Detected [2]
Not Detected [2]
Area
2004

Sales Rep
                                       1
                                       1
Not Detected [2]
Not Detected [2]
Area
2002
Department Stores (except Discount Department Stores)
Merchandise Coordinator
                                       1
                                       1
Not Detected [2]
Not Detected [2]
Area
2003

Unpack Clothing
                                       1
                                       1
Not Detected [2]
Not Detected [2]
Area
2006
 [1] The formaldehyde concentration in ug/m[3] was calculated from the provided concentration in ppm by multiplying the concentration in ppm by 1,229 ug/m[3]/ppm (see Appendix E).
 [2] The minimum reliable quantitation limit (RQL) identified for OSHA sampling and analysis methods is 0.00058 ppm.  Non-detected values and values below this RQL are set equal to the RQL for the exposure calculations.

Baseline Exposure and the Effect of Analytical Options on Baseline Exposure for Retail
Exposures in retail settings are not assessed quantitatively due to data gaps.  There is a lack of information on the ventilation rate at retail settings to allow for a what-if analysis similar to what was completed for the fabrication life cycle stage.  Therefore, it is not possible to accurately quantify the exposure concentrations during the baseline period and the reduction of exposure concentrations resulting from the analytical options.
The formaldehyde concentration resulting from off-gassing is expected to decrease with the implementation of more stringent regulations.  However, as discussed previously, the implementation of CARB ATCM Phase 1 limits is expected to result in little or no reduction in exposure from the baseline year.  This is because most composite wood products in the market would already be meeting more stringent limits by 2013.
Although the U.S. Census Bureau defines the retail trade sector as "establishments engaged in retailing merchandise, generally without transformation, and rendering services incidental to the sale of merchandise," the limited IMIS data indicate that sawing, assembling, building, painting, and other laboring activities are performed at some retail facilities.  A further search of one facility reported in the IMIS suggests that some fabrication sites could be involved in direct retail.  For example, one company makes custom cabinets and sells these cabinets to builders, dealers, and consumers.  This company is classified under NAICS 444190: "Other Building Material Dealers."  The exposures at fabrication sites have been assessed in Section 4.2.
Dermal Exposures during Retail
Dermal exposures to CWPs are not quantifiable using the EPA dermal exposure models for liquids and solids.
Exposure Assessment for Office Workers during Commercial Use
This section presents and discusses the assessment of formaldehyde exposures resulting from the occupancy of office buildings (commercial buildings) containing composite wood products by office workers.
Description of Life Cycle Stage
The commercial use of composite wood products includes the occupancy by office workers of office buildings (commercial buildings) containing composite wood products.  These office workers may be exposed to airborne formaldehyde emitted from CWPs or CWP fabrication products that are present in the commercial building.
Use Volumes and Number of Establishments, Employees, Employed Hours, and Work Days
Data that provide inventories of each composite wood product within commercial office buildings were not found.  Additionally, census data that provide the number of commercial use sites, the total number of employees, and employee hours worked per day and days worked per year in office settings were not found.  These data are difficult to obtain as commercial use settings, such as office buildings, occur across multiple industries.  For example, the industries described in the upstream life cycle stages in this report (manufacturing, wholesale, retail) all may involve office settings in addition to the settings described in their respective sections above.
Pre-Baseline Exposure Monitoring Data
This section provides pre-baseline indoor air concentration monitoring data obtained from the BASE study from 1994 to 1998.  These data are presented to compare to the model results presented in the next subsection.  Table 4-20 presents some general characteristics of the 100 commercial and public office buildings collected from the BASE study, including the construction date, floor area, number of floors and occupants.
Table 4-20.  Overview of General Characteristics of the 100 BASE Buildings
                             Characteristic (unit)
                                     Range
                                    Average
                                    Median
Construction Date (year)
                                  1850 - 1996
                                     1961
                                     1972
Gross Floor Area (m[2])
                                1,665 - 134,195
                                    24,756
                                    13,964
Occupied Floor Area (m[2])
                                 629 - 98,474
                                    16,380
                                     8,477
Number of Floors Above Grade
                                   1  -  53
                                       9
                                       5
Number of Building
Occupants (including visitors)
                                 87  -  6,500
                                     1,020
                                      705
  Source: BASE 1998a

Formaldehyde was detected in a quantifiable amount in all 100 buildings.  Table 4-21 presents a summary of the monitored formaldehyde concentrations in office buildings that were obtained from the BASE study (BASE 1998a).  The study only presented the overall formaldehyde concentrations within the buildings and did not distinguish exposures according to worker activities.  Table 4-21 also provides the summary statistics of the monitored outdoor formaldehyde concentrations from the BASE study, which were measurements made outside of the office buildings.
Table 4-21.  Summary of Formaldehyde Inhalation Monitoring Data for Office Occupational Settings and Outdoor Measured Concentrations from the BASE Study
                                  Percentile
                     Formaldehyde Concentration (ug/m[3])
                       Formaldehyde Concentration (ppm)

                                    Indoors
                                   Outdoors
                                    Indoors
                                   Outdoors
100[th] (Maximum)
                                      51
                                      13
                                     0.041
                                     0.011
95[th]
                                      32
                                      10
                                     0.026
                                     0.008
75[th]
                                      21
                                      5.7
                                     0.017
                                     0.005
50[th] (Median)
                                      15
                                      3.0
                                     0.012
                                     0.002
25[th]
                                      9.2
                                      1.4
                                     0.007
                                     0.001
5[th]
                                      4.4
                                     0.40
                                     0.004
                                     0.000
Arithmetic Mean
                                      16
                                      3.9
                                     0.013
                                     0.003
Source: BASE 1998a

Baseline Exposure and the Effect of Analytical Options on Baseline Exposure for Office Buildings
What-if values for indoor air concentrations in an office building related to baseline and the analytical options were estimated using mathematical modeling.  The mathematical modeling for indoor air concentration is presented in Section 3.3.1.2 and described in detail in Appendix E.
The estimation results are given in Table 4-22.  The results in Table 4-22 are for the three limiting cases of the office workstation being fabricated from 100% HWPW, 100% PB, and 100% MDF.  These results were combined into a single range, which is given in Table 4-23.
The range of what-if indoor air concentrations given in Table 4-23 is the estimate of the range of values for the steady-state indoor air formaldehyde concentration in a work space resulting from the off-gassing emissions from new HWPW, PB, and MDF components of the workstation occupying the work space and due to the ventilation with outside air containing ambient formaldehyde.  The emitting surface area of each workstation component and the ventilation rate of the work space containing the workstation are values given in BIFMA M7.1-2007  -  "Standard Test Method for Determining VOC Emissions from Office Furniture Systems, Components and Seating".
In homes, the most significant sources of formaldehyde are likely to be CWPs made from UF resins (USEPA 2011b).  The literature search did not identify information to indicate that sources of formaldehyde in office buildings are significantly different than in homes.  Therefore, it is assumed that CWPs are the most significant source of formaldehyde in office buildings.
Office building ventilation systems likely contain a recirculation component, which results in the mixing of air from various sections of the office building with outside air and the subsequent recirculation of the mixed air throughout the office building.  Thus, the formaldehyde concentration in any given work space can be affected by the formaldehyde concentration in other sections of the office building.  The mathematical modeling does not account for this effect.
Determination of a Best Estimate for Indoor Air Concentration
The best estimate exposure concentrations, ADC, and LADC results are presented in Table 4-26.  The best estimate exposure concentrations for the baseline and each analytical option were determined by averaging the following values: the midpoint of the intermediate range of concentrations for open plan workstations and the midpoint of the intermediate range of concentrations for private office workstations.  The midpoints of the intermediate ranges of concentration estimates were used for the following reasons.  First, all of the high-end concentration estimates at the baseline are on the order of or greater than 0.1 ppm for both workstation types.  Indoor air in buildings with a formaldehyde concentration of 0.1 ppm is considered highly polluted air.  It is not likely that typical office buildings will have workstations with highly polluted air at the baseline even if the buildings do use new furniture.  Second, the low-end concentration estimates at the baseline and for each analytical option are on the order of 0.003 to 0.005 ppm for both workstation types.  The ambient air formaldehyde concentration for commercial land use areas used for this assessment is 3.26 ug/m[3] or about 0.00265 ppm.  The low-end concentration estimates are the same order of magnitude as the ambient air formaldehyde concentration.  It is likely that formaldehyde concentrations in office buildings at the baseline would be greater than the ambient air formaldehyde concentration.  Therefore, the intermediate range of concentration estimates is more reasonable in comparison with the low- and high-end ranges and was used to derive the best estimate concentrations.  The midpoints of the intermediate ranges for both workstation types are averaged together to provide the best estimate concentrations.  The best estimate reduction from baseline, ADC, and LADC values are calculated from the best estimate concentration values.  It is assumed that office workers spend most of their time at workstations.  Therefore, this indoor air concentration is the exposure concentration for office workers.  As described in Appendix E, ADC and LADC were calculated assuming exposure duration of 8 hours per day and exposure frequency of 250 days per year over 40 working years per lifetime for ADC and over a lifetime of 70 years for LADC.
Similarly as with the CWP fabrication results, the office building results show that the maximum formaldehyde exposure concentration decreases as lower (more stringent) emission standards are implemented.  Again, the minimum exposure concentrations in Table 4-22 and Table 4-23 remain constant throughout the different analytical options due to the presence of NAF-compliant products at the baseline.  The office building results in Table 4-26 show the most significant reduction in exposure concentration from the NAF emission limits.

Table 4-22.  Results for Calculation of What-If Indoor Air Formaldehyde Concentrations for Each Analytical Option
                                     Case
                                   CWP Type
                               Low-End of Range
                              Intermediate Value
                               High-End of Range
                        Open Plan Workstations (ppm)[1]
Baseline 
HWPW
                                    0.0035
                                    0.0265
                                    0.2002

PB
                                    0.0035
                                    0.0346
                                    0.4680

MDF
                                    0.0053
                                    0.0732
                                    1.0089
Analytical Option (1): CARB Phase 1 
HWPW
                                  0.0035 (0%)
                                0.0167 (-36.8%)
                                0.0729 (-63.6%)

PB
                                  0.0035 (0%)
                                0.0328 (-5.2%)
                                0.1117 (-76.1%)

MDF
                                  0.0053 (0%)
                                0.0705 (-3.7%)
                                0.2833 (-71.9%)
Analytical Option (2): FSCWPA / CARB Phase 2 
HWPW
                                  0.0035 (0%)
                                0.0143 (-45.9%)
                                0.0414 (-79.3%)

PB
                                  0.0035 (0%)
                                0.0323 (-6.7%)
                                0.0717 (-84.7%)

MDF
                                  0.0053 (0%)
                                0.0701 (-4.2%)
                                0.1670 (-83.4%)
Analytical Option (3): NAF 
HWPW
                                  0.0035 (0%)
                                0.0099 (-62.5%)
                                0.0184 (-90.8%)

PB
                                  0.0035 (0%)
                                0.0099 (-71.3%)
                                0.0184 (-96.1%)

MDF
                                  0.0053 (0%)
                                0.0346 (-52.8%)
                                0.0728 (-92.8%)
                     Private Office Workstations (ppm)[1]
Baseline 
HWPW
                                    0.0030
                                    0.0130
                                    0.0719

PB
                                    0.0030
                                    0.0165
                                    0.1659

MDF
                                    0.0036
                                    0.0332
                                    0.3555
Analytical Option (1): CARB Phase 1 
HWPW
                                  0.0030 (0%)
                                0.0087 (-32.5%)
                                0.0273 (-62.0%)

PB
                                  0.0030 (0%)
                                0.0157 (-4.7%)
                                0.0409 (-75.3%)

MDF
                                  0.0036 (0%)
                                0.0320 (-3.5%)
                                0.1011 (-71.6%)
Analytical Option (2): FSCWPA / CARB Phase 2 
HWPW
                                  0.0030 (0%)
                                0.0077 (-40.6%)
                                0.0163 (-77.4%)

PB
                                  0.0030 (0%)
                                0.0155 (-6.1%)
                                0.0269 (-83.8%)

MDF
                                  0.0036 (0%)
                                0.0318 (-4.0%)
                                0.0603 (-83.0%)
Analytical Option (3): NAF 
HWPW
                                  0.0030 (0%)
                                0.0058 (-55.3%)
                                0.0082 (-88.6%)

PB
                                  0.0030 (0%)
                                0.0058 (-64.8%)
                                0.0082 (-95.1%)

MDF
                                  0.0036 (0%)
                                0.0165 (-50.4%)
                                0.0273 (-92.3%)
Note: For the purpose of this assessment, "intermediate" is defined as a value in-between the low-end and the high-end (see Section 3.3.1.2).
[1] For the analytical options, the percent decrease from the corresponding baseline value is given in parentheses.  For example, for Analytical Option (1) for open plan workstations, the high-end value for HWPW is 0.0729 ppm - a 63.6 percent decrease from the high-end baseline value.

Table 4-23.  Combined (for All CWP) Results: Calculated What-If Indoor Air Concentrations at the Baseline and for the Analytical Options
                           Open Plan Workstations[1]
                                     Case
                              Intermediate (ppm)
                                 Low End (ppm)
                                High End (ppm)
Baseline 
                                0.0265 - 0.0732
                                0.0035 - 0.0053
                                0.2002 - 1.0089
Phase 1
                       0.0167 (-37.0%) - 0.0705 (-3.7%)
                           0.0035 (0%) - 0.0053 (0%)
                       0.0729 (-63.6%) - 0.2833 (-71.9%)
FSCWPA / Phase 2
                       0.0143 (-46.0%) - 0.0701 (-4.2%)
                           0.0035 (0%) - 0.0053 (0%)
                       0.0414 (-79.3%) - 0.167 (-83.4%)
NAF
                       0.0099 (-62.6%) - 0.0346 (-52.7%)
                           0.0035 (0%) - 0.0053 (0%)
                       0.0184 (-90.8%) - 0.0728 (-92.8%)
                        Private Office Workstations[1]
                                     Case
                              Intermediate (ppm)
                                 Low End (ppm)
                                High End (ppm)
Baseline 
                                0.013 - 0.0332
                                0.003 - 0.0036
                                0.0719 - 0.3555
Phase 1
                        0.0087 (-33.1%) - 0.032 (-3.6%)
                           0.003 (0%) - 0.0036 (0%)
                       0.0273 (-62.0%) - 0.1011 (-71.6%)
FSCWPA / Phase 2
                       0.0077 (-40.8%) - 0.0318 (-4.2%)
                           0.003 (0%) - 0.0036 (0%)
                       0.0163 (-77.3%) - 0.0603 (-83.0%)
NAF
                       0.0058 (-55.4%) - 0.0165 (-50.3%)
                           0.003 (0%) - 0.0036 (0%)
                       0.0082 (-88.6%) - 0.0273 (-92.3%)
[1] For the analytical options, the percent decrease from the corresponding baseline value is given in parentheses.  For example, for Analytical Option (1) for open plan workstations, the minimum of the high-end values is 0.0729 ppm - a 63.6 percent decrease from the high-end baseline value.

Table 4-24.  Combined (for All CWP) Results: Calculated What-If Indoor Air ADCs at the Baseline and for the Analytical Options
                            Open Plan Workstations
                                     Case
                            Intermediate (ug/m[3])
                              Low End (ug/m[3])
                              High End (ug/m[3])
Baseline 
                                 7.44 - 20.54
                                  0.98 - 1.49
                                 56.17 - 283.1
Phase 1
                                 4.69 - 19.78
                                  0.98 - 1.49
                                 20.46 - 79.49
FSCWPA / Phase 2
                                 4.01 - 19.67
                                  0.98 - 1.49
                                 11.62 - 46.86
NAF
                                  2.78 - 9.71
                                  0.98 - 1.49
                                 5.16 - 20.43
                          Private Office Workstations
                                     Case
                            Intermediate (ug/m[3])
                              Low End (ug/m[3])
                              High End (ug/m[3])
Baseline 
                                  3.65 - 9.32
                                  0.84 - 1.01
                                 20.17 - 99.75
Phase 1
                                  2.44 - 8.98
                                  0.84 - 1.01
                                 7.66 - 28.37
FSCWPA / Phase 2
                                  2.16 - 8.92
                                  0.84 - 1.01
                                 4.57 - 16.92
NAF
                                  1.63 - 4.63
                                  0.84 - 1.01
                                  2.3 - 7.66

Table 4-25.  Combined (for All CWP) Results: Calculated What-If Indoor Air LADCs at the Baseline and for the Analytical Options
                            Open Plan Workstations
                                     Case
                            Intermediate (ug/m[3])
                              Low End (ug/m[3])
                              High End (ug/m[3])
Baseline 
                                 4.25 - 11.74
                                  0.56 - 0.85
                                 32.1 - 161.8
Phase 1
                                  2.68 - 11.3
                                  0.56 - 0.85
                                 11.69 - 45.42
FSCWPA / Phase 2
                                 2.29 - 11.24
                                  0.56 - 0.85
                                 6.64 - 26.78
NAF
                                  1.59 - 5.55
                                  0.56 - 0.85
                                 2.95 - 11.67
                          Private Office Workstations
                                     Case
                            Intermediate (ug/m[3])
                              Low End (ug/m[3])
                              High End (ug/m[3])
Baseline 
                                  2.08 - 5.32
                                  0.48 - 0.58
                                 11.53 - 57.00
Phase 1
                                  1.39 - 5.13
                                  0.48 - 0.58
                                 4.38 - 16.21
FSCWPA / Phase 2
                                  1.23 - 5.1
                                  0.48 - 0.58
                                  2.61 - 9.67
NAF
                                  0.93 - 2.65
                                  0.48 - 0.58
                                  1.31 - 4.38

Table 4-26.  Best Estimate of What-If Formaldehyde Exposure Concentration at the Baseline and for the Analytical Options
                                     Case
                      Best Estimate Concentration1 (ppm)
                          Reduction from Baseline (%)
                                 ADC (ug/m3)
                                 LADC (ug/m3)
Baseline 
                                    0.0365
                                      --
                                     10.23
                                     5.85
Phase 1
                                    0.0320
                                     -12.3
                                     8.97
                                     5.13
FSCWPA / Phase 2
                                    0.0310
                                     -15.1
                                     8.69
                                     4.97
NAF
                                    0.0167
                                     -54.2
                                     4.69
                                     2.68
            [1] The best estimate concentration is the average of the following two values: the midpoint of the intermediate range for open plan workstations and the midpoint of the intermediate range for private office workstations. The reduction from baseline, ADC, and LADC values are calculated from the best estimate concentration value.

Dermal Exposures during Commercial Use
Dermal exposures to CWPs are not quantifiable using the EPA dermal exposure models for liquids and solids.

Uncertainties In Assessment Results
Uncertainties in Assessment Results for Composite Wood Product Manufacturing, Fabrication, Wholesale, Retail, and Office Buildings
This section discusses the uncertainties in the results of the assessment of occupational formaldehyde exposure at composite wood product manufacturing, fabrication, wholesale, and retail sites, and office buildings.  Baseline indoor air concentration and the effect of the analytical options on it were assessed quantitatively for CWP fabrication and office buildings only, and therefore the discussion of uncertainty in the assessment of indoor air concentration is limited to these life cycle stages and is presented in Section 5.1.1 and Section 5.1.2, respectively.  Uncertainty in the assessed number of production workers for the manufacturing and fabrication life cycle stages and in the number of employees in the wholesale and retail life cycle stages is discussed in Section 5.3.
Dermal exposures were assessed quantitatively for manufacturing.  Due to data gaps in information on the free (un-reacted) formaldehyde content in resins that will be used at baseline and for each analytical option, only pre-baseline dermal exposures were estimated for manufacturing.  The dermal exposure level at the baseline and for the analytical options was assumed to be equal to the estimated pre-baseline dermal exposure level.  This assumption would overestimate the dermal exposures at baseline and the analytical options if the free formaldehyde content of resins is lower at the baseline and for the analytical options.
Uncertainty in the Estimates of Indoor Air Concentration for Composite Wood Product Fabrication
The assessment of site-average formaldehyde indoor air concentrations for composite wood product fabrication relies on a mass balance approach that equates the emission rates due to off-gassing to the rate at which airborne formaldehyde is exhausted from the work space through ventilation.  As discussed in Section 3.3.1.1 and Appendix E, indoor air concentration was calculated from the four parameters in the mass balance equation, which are emitting surface area, emission rate due to off-gassing, effective ventilation rate, and ambient air formaldehyde concentration, as well as the assumption of a well-mixed box for the fabrication site.  As discussed in Section 3.3.1.1, due to data gaps on the values of these four parameters at representative CWP fabrication sites, a range of what-if values for indoor air concentration was calculated from bounding values for the ranges of three of the four parameters.  CWP fabrication sites were assumed to contain only HWPW, or only PB, or only MDF in this what-if approach.  As discussed in Section 3.3.1.1, there is considerable uncertainty in the range of values for emitting surface area, and some uncertainty in the ranges of values for effective ventilation rate and for emission rates.  Therefore, there is uncertainty in the calculated ranges of baseline indoor air concentration due to off-gassing and the effect of the analytical options on the baseline concentrations.  It is not possible to determine the direction of uncertainty (i.e., whether the estimated values for indoor air concentration are an underestimate or an overestimate).
Although IMIS data and emission rates for the pre-baseline period are available, they cannot be used to perform a general validation of the calculation method nor can this information be used to estimate values for effective ventilation rate and emitting surface area for the following reasons:
The calculated assessment results are only an estimate of airborne concentrations due to off-gassing.  IMIS monitoring data are intended to be measurements of aggregate airborne concentrations and do not contain information on the source or sources of exposure and the contributions to the aggregate concentrations where different sources of exposure existed when the monitoring occurred.
It is uncertain as to whether the IMIS monitoring data are representative of the exposures experienced by workers in the CWP fabrication industries during the pre-baseline period (see Section 5.2.1).

The modeling method used to estimate formaldehyde concentrations in this assessment only accounts for the off-gassing of formaldehyde from the three composite wood products of interest and the formaldehyde concentration in ambient air.  As discussed in Section 4.2.1, there is some indication that sources other than the three CWPs of interest may contribute to formaldehyde airborne concentrations.  The number of sites in which these other sources may be present, the number of workers that would be potentially exposed to emissions from these other sources, and the magnitude of such potential exposure are not known.  The assessment results for the baseline and analytical options are not an estimate of aggregate indoor air concentration where other sources of formaldehyde exist.
Uncertainty in the Estimates of Indoor Air Concentration for Office Buildings 
Data on the composition of composite wood products in office furniture was not found.  Therefore, the composition of each of the three types of composite wood products in office workstations is unknown.  As discussed in Section 3.3.1.2, panels and work surfaces were assumed to be wholly made of composite wood products.  Information on the type and extent of overlays on the composite wood products was also not found.  The effect of uncertainty about the composition of CWPs in office workstations and the overlays on these CWPs on the assessment of baseline exposure and the effect of analytical options on baseline exposure are undetermined.
The office building method uses emission rates for new composite wood products and do not account for the decay of emissions with time.  Therefore, this method to estimate formaldehyde concentrations in office buildings only accounts for new office furniture and does not account for the decay of emission rates with time.  Data to describe the decay of emission rates with time in office buildings were not found through the literature search.  Therefore, the estimated airborne concentrations presented in this assessment are only applicable to offices with new office furniture.
The method used simplifies the complex nature of office building ventilation systems.  Typically in these systems, outdoor air is mixed with a recirculation stream of indoor air prior to being supplied to the various ventilation zones.  The return air from these ventilation zones is mixed and then a stream is split for recirculation and mixing with the outdoor air while the balance is exhausted.  Therefore, the office environment model neglects the effect of one exposure zone on another resulting from the recirculation of indoor air or the movement of air directly between exposure zones without passing through the ventilation system.  The uncertainty in the results due to neglecting the effect of one exposure zone on another is uncertain and cannot be quantified without a detailed, rigorous mathematical model with detailed supporting monitoring data.
Although monitoring data from the BASE study and emission rates for the 2002 to 2009 pre-baseline period are available, they cannot be used to perform a general validation of the calculation method or a validation of the assumed values for the surface area covered by overlay materials and their associated barrier effectiveness.  The ages of the office furniture workstations present in the buildings studied in the BASE study are unknown.  It is likely that much of the office furniture that contributed to the exposure concentrations observed in the BASE study were of various ages and not representative of new office furniture.  The emission rates for CWP that are components of this aged office furniture at the time of monitoring are not known because their initial value when these CWP were new is not known and the extent of their decay with time until the time of monitoring is not known.  Furthermore, even where monitoring occurred for new office furniture (a situation that cannot be determined from the BASE study data), the emission rates of CWPs used in this furniture is unknown.  Hence, model validation is not possible using the BASE study data because the emission rates of CWPs present in office furniture in the monitored buildings is unknown.
Uncertainties in Pre-Baseline Monitoring Data
OSHA IMIS Data
This section discusses uncertainties associated with OSHA IMIS monitoring data.  The uncertainties discussed below apply to the IMIS data collected for each life cycle stage (CWP manufacturing, fabrication, wholesale, and retail).  The major conclusion of this discussion is that it is uncertain whether the collected IMIS data are representative of the formaldehyde exposure concentrations that occurred during the pre-baseline period of 2002 to 2009 in facilities of the industries within each life cycle stage assessed in this report.
Representativeness of IMIS Data
The representativeness of the measured exposure concentrations in IMIS for the entire population of exposed workers is examined below.  The discussion below presents statistics of the IMIS data obtained for the CWP fabrication life cycle stage only since a larger amount of data were obtained for this life cycle stage.  Very few data were obtained for the CWP manufacturing, wholesale, and retail life cycle stages.
Table 5-1 presents data on the total number of unique OSHA inspections conducted within each NAICS industry code of the CWP fabrication life cycle stage.  A total of 78 inspections were conducted at 77 facilities between 2002 and 2009, indicating that most inspections occurred at separate facilities.  The table compares the number of inspected facilities to the total number of facilities for that particular industry sector based on estimates from the 2007 Economic Census.  The sample size (the number of inspected facilities) ranges from zero to 19 over all of the industries.  Sampling theory guarantees an accurate degree of confidence in the estimate of a population mean when the sample size is at least 30 samples.  When at least 30 samples are obtained, it is not required for the population to have a normal distribution to guarantee an accurate degree of confidence.  However, accuracy does require that the samples are random (Walpole 2002a).
Table 5-1.  Number of OSHA Inspections at CWP Fabrication Facilities Compared to the Maximum Number of Facilities from the U.S. Economic Census
                                     NAICS
                                  Description
                        2007 U.S. Economic Census Data
                         OSHA IMIS Data (2002 to 2009)


                       Maximum Number of Establishments
                         Number of Unique Inspections
                          Number of Unique Facilities
321911
Wood window & door Manufacturing
                                     1,494
                                       4
                                       4
321918
Other millwork (including flooring)
                                     2,101
                                       9
                                       9
321991
Manufactured home (mobile home) Manufacturing
                                      376
                                       0
                                       0
321999
All other misc. wood product Manufacturing
                                     1,964
                                       8
                                       8
337110
Wood kitchen cabinet & countertop Manufacturing
                                     9,683
                                      20
                                      19
337121
Upholstered household furniture Manufacturing
                                     1,633
                                       4
                                       4
337122
Nonupholstered wood household furniture Manufacturing
                                     3,428
                                      12
                                      12
337124
Metal household furniture Manufacturing
                                      330
                                       0
                                       0
337127
Institutional furniture Manufacturing
                                      782
                                       5
                                       5
337129
Wood television, radio, & sewing machine cabinet Manufacturing
                                      270
                                       0
                                       0
337211
Wood office furniture Manufacturing
                                      457
                                       6
                                       6
337212
Custom architectural woodwork & millwork Manufacturing
                                     2,228
                                       2
                                       2
337214
Office furniture (except wood) Manufacturing
                                      289
                                       2
                                       2
337215
Showcase, partition, shelving, & locker Manufacturing
                                     1,377
                                       5
                                       5
339950
Sign manufacturing
                                     6,383
                                       0
                                       0
336213
Motor home Manufacturing
                                      80
                                       1
                                       1
336214
Travel trailer & camper Manufacturing
                                      862
                                       0
                                       0
                                                                          Total
                                    33,737
                                      78
                                      77


For each fabrication industry, less than 30 samples were obtained between 2002 and 2009.  Therefore, an accurate degree of confidence in the estimate of the population mean cannot be guaranteed without knowledge of whether the population is represented by a normal distribution.  While all of the fabrication industries in Table 5-1 can be aggregated together to yield a sample size of 78 (the total number of inspections), such aggregation would not resolve the second criterion for an accurate degree of confidence: random sampling.  Examination of the randomness of the sampling is made using knowledge of how OSHA conducts inspections.
Lavoue et al. (Lavoue 2008a) analyzed formaldehyde exposures from U.S. industries from the OSHA IMIS database using data from 1979 to 2001.  The authors note that exposures in IMIS have long been suspected to not be representative of exposures in the U.S. population.  The authors found the IMIS exposures to mainly result from complaints, referrals, or programs targeted at specific industrial sectors.  However, the authors also note that, "within industries, available evidence suggests that bias relating to selecting companies with high exposure levels during complaint visits should be of small amplitude".  The pre-baseline exposure monitoring data presented in the assessment were examined to further evaluate any selection bias.
OSHA may initiate an inspection due to a referral by a federal or state government agency or a worker complaint regarding workplace hazards.  Alternatively, OSHA may inspect establishments that were selected under the agency's programmed inspection plan.  The "inspection types" for the compiled IMIS data are: (1) referral (15%); (2) complaint (58%); (3) program-planning (25%); and (4) follow-up (2%) (see Appendix A, Definition of Terms for exposure and inspection types).  These inspection types are further discussed below.
The Site-Specific Targeting program is OSHA's main programmed inspection plan for non-construction workplaces that have 40 or more workers.  The selection criteria involve several parameters such as the number of injury and illness cases, the number of days away from work, and the number of workers who receive job transfers or work restrictions due to injury or illness.  In addition to the Site-Specific Targeting program, OSHA also implements both national and local emphasis inspection programs to target high-risk hazards and industries.  Some areas that OSHA focuses on are amputations, lead, crystalline silica, shipbreaking, trenching and excavations, process safety management in petroleum refineries, hexavalent chromium, diacetyl, recordkeeping and combustible dust (OSHA 2010a).  Programmed and planned inspections may or may not be considered random.  If a particular industry were chosen for a programmed or planned inspection, the facilities inspected within the selected industry would have been chosen randomly.  However, it is possible a facility was specifically chosen as part of a program or plan if the facility had a history of problems or hazards, in which case the choice of facility would not have been random.
Complaint and referral inspection types would not be considered random, since OSHA inspected the given facility based on evidence that a problem existed from an employee or other agency with knowledge of the facility.  It is difficult to determine whether follow-up inspections would be considered random without knowing the inspection type for the original inspection.
Although OSHA typically inspects establishments with potential workplace hazards, these hazards may or may not be related to formaldehyde overexposure.  For example, OSHA may inspect an establishment that manufactures PB after an employee files a complaint concerning dust exposure.  Since the establishment processes formaldehyde-containing materials, personal monitoring for formaldehyde was also conducted during the inspection although there may or may not be a formaldehyde overexposure.  Therefore, it cannot be assumed that a facility inspected by OSHA through one of the programs described above is necessarily an outlying or atypical facility with high formaldehyde exposures.
It is uncertain whether the programmed and planned inspections included in the IMIS data presented in this assessment randomly selected the inspected facilities.  Therefore, it cannot be concluded whether any of the OSHA IMIS data are the result of random sampling.  In conclusion, it is uncertain whether OSHA IMIS monitoring data are representative of the exposure concentrations of formaldehyde in the industries from which the data were obtained.
Intra-Site Variability in Monitored Indoor Air Concentration
This subsection discusses the variability in the measured indoor air concentrations during a single inspection at a single facility.  This discussion is limited to CWP fabrication sites since only this life cycle stage had an adequate amount of data to allow this analysis.
The variation in indoor air concentration of fabricator sites was considered by examination of IMIS monitoring data.  Figure 5-1 through Figure 5-8 display formaldehyde indoor air concentrations from OSHA IMIS for composite wood product fabricators in which multiple worker activities were monitored for a single site.  The graphs display formaldehyde concentration as a function of monitored worker activity.  CSHO stands for Compliance Safety and Health Officer and represents the exposures for the inspector.  Worker activities with only a single exposure level are shown with only a median exposure concentration measurement.
These figures illustrate that many of these site inspections resulted in similar exposure concentrations for all monitored worker activities (exposure level for the worker activity with the highest exposure level is typically greater than the exposure level of the worker activity with the lowest exposure level by up to approximately 100%).  However, for some sites, there are worker activities with greater variation than this, including the frame operator at Armstrong Cabinets, the painter U cell at The Gunlocke Company, and the August 25th auto spray line at Silver Street.  Most of these have higher values than the other measurements.  Most of the pre-baseline monitoring data presented in the assessment consist of multiple measurements for a single worker activity at a given site.  There is considerable variability in the monitored worker activities from site to site.
In conclusion, intra-site variability in the monitored exposure level ranges from small to as much as 100% or greater based on examination of IMIS data for which multiple worker activities at a single site were monitored.

                                       
Figure 5-1.  TWA Exposures from OSHA IMIS for a Single Inspection for Armstrong Cabinets (NAICS 337110)
                                       
Figure 5-2.  TWA Exposures from OSHA IMIS for a Single Inspection for Mid Continent Cabinetry (NAICS 337110)
                                       
Figure 5-3.  TWA Exposures from OSHA IMIS for a Single Inspection for Benvenuti and Stein (NAICS 337122)
                                       
Figure 5-4.  TWA Exposures from OSHA IMIS for a Single Inspection for Ameriwood Industries (NAICS 337122)
                                       
Figure 5-5.  TWA Exposures from OSHA IMIS for Two Inspections for The Gunlocke Company (NAICS 337211)

                                       
                                       
Figure 5-6.  TWA Exposures from OSHA IMIS for Two Inspections for Silver Street (NAICS 337211)

                                       
                                       
Figure 5-7.  TWA Exposures from OSHA IMIS for a Single Inspection for The Commercial Furniture Group (NAICS 337127)

                                       
                                       
Figure 5-8.  TWA Exposures from OSHA IMIS for a Single Inspection for Parenti & Raffaelli (NAICS 321918)


BASE Study Monitoring Data
A total of 100 commercial and public office buildings were monitored in the BASE study between 1994 and 1998.  To determine the representativeness of the BASE sample of office buildings, the BASE study authors compared this sample to larger samples of U.S. office buildings using data collected by the Energy Information Administration of the U.S. Department of Energy through the 1995 Commercial Building Energy Consumption Survey (CBECS).  The BASE study authors were able to perform two comparisons: 1) a comparison of the BASE buildings to a larger sample of large urban office buildings; and 2) a comparison of the BASE buildings to a larger sample of all U.S. office buildings.  The BASE study authors determined that the BASE buildings are most similar to the CBECS sample of large, urban office buildings; specifically, buildings with more than 50 occupants in Metropolitan Statistical Areas (MSAs) (BASE 2001a).  The BASE study authors note that the BASE buildings were selected from cities with populations greater than 100,000.  However, MSAs are urbanized areas with populations greater than 50,000.  Therefore, there is a slight discrepancy in the comparison of the surrounding populations of the BASE study buildings and the large, urban office buildings from CBECS.  Furthermore, the BASE buildings were selected from downtown areas.  The BASE study authors found the BASE buildings to be three to four times larger than the CBECS sample of large, urban office buildings.  Consistent with this finding, a separate survey conducted by the Building Owners and Managers Association (BOMA) found that downtown office buildings tend to be larger than other metropolitan area office buildings.  Although these differences in surrounding population and building size are noted, the BASE study authors found that BASE study buildings most likely represent large, urban office buildings (BASE 2001a).
Based on the CBECS sample of all U.S. office buildings, the BASE buildings do not represent the entire population of all U.S. office buildings well.  However, the BASE study authors note that, although the large, urban office buildings account for only 11% (or about 80,000) of all U.S. office buildings (about 705,000), this building type represents 73% (or about 19,428,229) of all U.S. office workers (about 26,563,566) according to the CBECS data (BASE 2001a).  Therefore, the BASE study office buildings are most similar to the office building type (large, urban office buildings) that house approximately 73% of all U.S. office workers.  Note that these statistics are based on survey data from 1995.  It is unknown how these statistics have changed since 1995.  The second criterion, random sample selection, is analyzed next.
The building selection criteria described in the BASE study documents indicate that the BASE study did randomly select office buildings for study.  Buildings were selected randomly in randomly selected cities.  The study authors selected buildings in ten climatic regions of the U.S. (BASE 2001a; NIST 2008a).
Whether the measured formaldehyde concentrations are representative of year-round concentrations in these buildings is also analyzed.  The BASE study focused its building measurements during the winter and summer months.  Building ventilation systems are more likely to operate at minimum outdoor air ventilation rates during the winter and summer to minimize the amount of air heating or cooling required.  NIST conducted an analysis of the ambient temperatures recorded during the BASE study and concluded that approximately 62% of the data could possibly correspond to buildings operating with minimum outdoor air ventilation rates (NIST 2008a).  However, the building operations during winter and summer were not compared against their operations during spring or autumn.  Therefore, it cannot be concluded for certain whether buildings were operating at lower outdoor air ventilation rates during the BASE study.  NIST additionally analyzed the BASE study data on the ventilation system designs and the measured ventilation rates.  NIST concluded that the ratio of the minimum outdoor air ventilation rate to total supply air ventilation based on the ventilation system designs had an average value of 0.19 and a median of 0.12.  The observed ratio of actual outdoor air ventilation rate to total supply air ventilation had an average value of 0.38 and a median of 0.22 (NIST 2008a).  This observation illustrates that, in general, many of the studied buildings operated at outdoor air ventilation rates greater than the system design minimum.
In conclusion, it has been established that the studied buildings are representative of large, urban commercial and public office buildings from the time frame of the study.  However, it is uncertain whether the measured formaldehyde concentrations are representative of concentrations that occurred in these buildings during an entire year or if they are only representative of winter and summer concentrations.
Uncertainty in Number of Production Workers and Employees
U.S. Census data on the number of production workers (for CWP manufacturing and fabrication) and employees (for wholesale and retail) are presented for the NAICS industry sectors associated with each life cycle stage.  While these data were not used for the quantitative assessment, they were presented to indicate the total number of production workers and employees potentially exposed to formaldehyde emissions from composite wood products.  These data likely represent an overestimation for the following reasons.
Census data represent the overall statistics for the entire industry sector.  Additionally, Census data do not contain specific information on the type of composite wood products associated with these statistics.  Some sites within these industry sectors may not manufacture, use, or sell the composite wood products of interest.
For any given site, it is uncertain whether the breathing zone for all production workers or employees would be affected by the off-gassing of formaldehyde from the composite wood products of interest.  Therefore, there is some uncertainty in assessing whether the number of production workers and employees presented in this assessment is an overestimate.  However, for the reasons listed above, the number of production workers and employees from U.S. Census data is likely an overestimate of the actual exposed population who would be affected by the regulation.  More specific information on the actual exposed population is not available.


References
Alexandersson 1982a. Alexandersson, R.; Hedenstierna, G.; Kolmodin-Hedman, B. Exposure to formaldehyde: effects on pulmonary function. Arch. Environ. Health 1982, 37 (5), 279-284.
ASTM 2002a. Standard Test Method for Determining Formaldehyde Concentrations in Air and Emission Rates from Wood Products Using a Large Chamber: ASTM E 1333-96 (Reapproved 2002); ASTM International: West Conshohocken, PA, 2002.
ATSDR 1999a. Toxicological Profile for Formaldehyde; Agency for Toxic Substances and Disease Registry (ATSDR): July 1999.
ATSDR 2010a. Addendum to the Toxicological Profile for Formaldehyde. Agency for Toxic Substances and Disease Registry (ATSDR): Atlanta, GA, October 2010.
AWFS 2009a. California Air Resources Board; Masco Retail Cabinet Group; Composite Panel Association. Update on California's Composite Wood Products Regulations, Association of Woodworking and Furnishing Suppliers (AWFS) Show, Las Vegas, Nevada, July 18, 2009.
Baldwin 1995a. Baldwin, Richard F. Plywood and Veneer-Based Products: Manufacturing Practices; Miller Freeman Books: 1995.
Ballarin 1992a. Ballarin, Cinzia; Sarto, Franco; Giacomelli, Luciano; Bartolucci, Giovanni Battista; Clonfero, Erminio. Micronucleated cells in nasal mucosa of formaldehyde-exposed workers. Mutation Research 1992, 280 (1), 1-7.
BASE 1998a. U.S. Environmental Protection Agency (USEPA). Building Assessment Survey and Evaluation (BASE) Study: Volatile Organic Compounds Master List. http://www.epa.gov/iaq/base/voc_master_list.html (accessed June 16, 2009).
BASE 2001a. Environmental Health & Engineering, Inc. Summary and Analysis Report of the BASE Study Building Selection Process; 11995; Newton, MA, 2001.
Battelle 1996a. Kelly, Thomas J. Determination of Formaldehyde and Toluene Diisocyanate Emissions from Indoor Residential Sources; Battelle: Columbus, OH, November 1996.
BIFMA 2007a. Standard Test Method for Determining VOC Emissions from Office Furniture Systems, Components and Seating; ANSI/BIFMA M7.1-2007; Business and Institutional Furniture Manufacturer's Association (BIFMA) International: Grand Rapids, MI, September 2007.
BIFMA 2007b. Standard for Formaldehyde and TVOC Emissions of Low-emitting Office Furniture Systems and Seating; ANSI/BIFMA X7.1-2007; Business and Institutional Furniture Manufacturer's Association (BIFMA) International: Grand Rapids, MI, August 2007.
CARB 2007a. Proposed Airborne Toxic Control Measure to Reduce Formaldehyde Emissions from Composite Wood Products: Initial Statement of Reasons for Proposed Rulemaking (ISOR); California Air Resources Board (CARB): Sacramento, CA, March 2007.
Carter 2007a. Carter, Randal D.; Zhang, Jianshun S. Definition of Standard Office Environments for Evaluating the Impact of Office Furniture Emissions on Indoor VOC Concentrations. ASHRAE Transactions. 2007, 113, 466-477.
Census 2007a. 2007 Economic Census [Online]; U.S. Census Bureau, 2007. http://www.census.gov/econ/census07/.
Chimar Hellas undated. Update on the formaldehyde release from wood-based panels; Chimar Hellas S.A.: Undated. http://www.chimarhellas.com/wp-content/uploads/2008/07/formaldehyde_2008.pdf (accessed February 12, 2009).
Chung 2000a. Chung, K.Y.; Cuthbert, R.J.; Revell, G.S.; Wassel, S.G.; Summers, N. A study on dust emission, particle size distribution and formaldehyde concentration during machining of medium density fibreboard. Ann. Occup. Hyg. 2000, 44, 455-466.
CPA 2003a. VOC Emission Barrier Effects of Laminates, Overlays and Coatings for Particleboard, Medium Density Fiberboard (MDF) and Hardboard: Technical Bulletin; Composite Panel Association (CPA): Gaithersburg, MD, 2003.
CPA 2007a. Laminating Composite Panels: Technical Bulletin; Composite Panel Association (CPA): Gaithersburg, MD, 2007.
CPA 2008a. 2007 North American Downstream Market Report; Composite Panel Association (CPA): Leesburg, VA, 2008.
CPA 2009a. Meeting Summary  -  Discussion with CPA in Response to Comments Submitted to Formaldehyde ANPR Docket; Eastern Research Group, Inc.: Chantilly, VA, June 18, 2009.
CPSC 1997a. An Update on Formaldehyde: 1997 Revision; CPSC Document # 725; Consumer Product Safety Commission (CPSC): Washington, DC, 1997. http://www.cpsc.gov/CPSCPUB/PUBS/725.html (accessed January 26, 2009).
Edling 1988a. Edling, C.; Hellquist, H.; Odkvist, L. Occupational exposure to formaldehyde and histopathological changes in the nasal mucosa. Br. J. Ind. Med. 1988, 45, 761-765.
Fink 2005a. Fink, Johannes Karl. Reactive Polymers Fundamentals and Applications: A Concise Guide to Industrial Polymers; William Andrew, Inc.: Norwich, NY, 2005: pp 288-289.
Fransman 2003a. Fransman, W.; McLean, D.; Douwes, J.; Demers, P.A.; Leung, V.; Pearce, N. Respiratory symptoms and occupational exposures in New Zealand plywood mill workers. Ann. Occup. Hyg. 2003, 47, 287-295.
GDNR 1996a. Crumpler, Paul. An Analysis of Pollution Prevention Opportunities and Impediments in the Wood Products Manufacturing Sector in Georgia; Georgia Department of Natural Resources (GDNR), Sustainability Division. April 1, 1996 (accessed June 2, 2009).
Gammage 1988a. Gammage, R.; Matthews, T. Volatile organic compounds in indoor air: types, sources, and characteristics. Environmental Progress 1988, 7, 279-283.
Groah 2005a. Groah, William J. Decay or the Decrease in Formaldehyde Concentrations or Emissions over Time and UF-bonded Wood Panel Products [Online]; December 21, 2005. http://www.ecobind.com/research/Decay_or_Decrease_in_Emissions.pdf.
Hamilton 1996a. Hamilton, Thomas E.; Alig, Joanne T.; Falk, Robert H. Products from Recycled Fibers  -  A U.S. Perspective, Proceedings of the International Union of Forest Research Organizations (IUFRO) XX World Congress, Tampere, Finland, August 6-12, 1995. November 21, 1996. http://www.metla.fi/iufro/iufro95abs/rsp21.htm.
HEI 2007a. HEI Air Toxics Review Panel. Formaldehyde. In Mobile-Source Air Toxics: A Critical Review of the Literature on Exposure and Health Effects; HEI Special Report 16; Health Effects Institute: Boston, MA, 2007; pp 87-105.
Herbert 1994a. Herbert, F.A.; Hessel, P.A.; Melenka, L.S.; Yoshida, K.; Nakaza, M. Respiratory consequences of exposure to wood dust and formaldehyde of workers manufacturing oriented strand board. Arch. Environ. Health 1994, 49 (6), 465-470.
Hodgson 2007a. Hodgson, Alfred T.; Tannous, Raja S. Meeting the Requirements of the California Composite Wood ATCM Using Chambers of Different Sizes; Berkeley Analytical Associates, LLC: Richmond, CA, September 7, 2007.
Horvath 1988a. Horvath, E.P. Jr.; Anderson, H. Jr.; Pierce, W.E.; Hanrahan, L.; Wendlick, J.D. Effects of formaldehyde on the mucous membranes and lungs. A study of an industrial population. JAMA. 1988, 259, 701-707.
HPVA 2009a. Howlett, Kip. Hardwood Plywood and Veneer Association (HPVA), Reston, VA. Personal communication, June 11, 2009.
IARC 1995a. IARC Monographs on the Evaluation of Carcinogenic Risks to Humans; Volume 62: Wood Dust and Formaldehyde, IARC Working Group on the Evaluation of Carcinogenic Risks to Humans, Lyon, France, October 11-18, 1994; World Health Organization: Lyon, 1995.
IARC 2006a. IARC Monographs; Volume 88: Formaldehyde, 2006.
IRSST undated. Exposure to Formaldehyde in the Workplace: Wood Furniture Manufacturing; Institut de recherche Robert-Sauvé en santé et en sécurité du travail (IRSST): Montréal, Québec, undated.
Kauppinen and Niemelä 1985a. Kauppinen, T.P.; Niemelä, R.I. Occupational exposure to chemical agents in the particleboard industry. Scand. J. Work Environ. Health. 1985, 11, 357-363.
Kim 2007a. Kim, Sumin; Kim, Jin-A; Kim, Hyun-Joong. Application of Field and Laboratory Emission Cell (FLEC) to Determine Formaldehyde and VOCs Emissions from Wood-Based Composites. Mokchae Konghak (Journal of the Korean Wood Science and Technology). 2007, 35, 24-37.
Kirk-Othmer 2011a. Lehmann, W.F. Wood-Based Composites and Laminates. Kirk-Othmer Encyclopedia of Chemical Technology; John Wiley & Sons, Inc.: New York, 2011; pp 1-47.
Koontz 1996a. Koontz, Michael D.; Rector, Harry E.; Cade, Donald R.; Wilkes, Charles R.; Niang, Laura L. Residential Indoor Air Formaldehyde Testing Program: Pilot Study; GEOMET Technologies, Inc.: Germantown, MD, March 21, 1996.
Lavoue 2005a. Lavoué, Jérôme; Beaudry, Charles; Goyer, Nicole; Perrault, Guy; Gérin, Michel. Investigation of Determinants of Past and Current Exposures to Formaldehyde in the Reconstituted Wood Panel Industry in Quebec. Ann. Occup. Hyg. 2005, 49, 587-602.
Lavoue 2008a. Lavoué, Jérôme; Vincent, Raymond; Gérin, Michel. Formaldehyde Exposure in U.S. Industries from OSHA Air Sampling Data. Journal of Occupational and Environmental Hygiene 2008, 5, 575-587.
Malaka 1990a. Malaka, T.; Kodama, A.M. Respiratory health of plywood workers occupationally exposed to formaldehyde. Arch. Environ. Health 1990, 45 (5), 288-294.
Maloney 1993a. Maloney, T.M. Modern Particleboard & Dry-Process Fiberboard Manufacturing; Miller Freeman Publications, Inc.: San Francisco, CA, 1993.
Niemelä and Vainio 1981a. Niemelä, R.; Vainio, H. Formaldehyde exposure in work and the general environment: Occurrence and possibilities for prevention. Scand. J. Work Environ. Health 1981, 7, 95-100.
NIOSH 1994a. NIOSH Method 3500: Formaldehyde by Vis (1994). In NIOSH Manual of Analytical Methods, 4th ed. [Online]; Schlecht, P.C.; O'Connor, P.F., Eds.; 2003; 3rd Supplement. http://www.cdc.gov/niosh/docs/2003-154/.
NIST 2008a. Persily, Andrew; Gorfain, Josh. Analysis of Ventilation Data from the U.S. Environmental Protection Agency Building Assessment Survey and Evaluation (BASE) Study; NISTIR 7145-Revised; National Institute of Standards and Technology (NIST): November 2008. http://fire.nist.gov/bfrlpubs/build04/art043.html.
OAQPS 1997a. Hanks, Katie; Bullock, David. Site Visit  -  Champion International Corporation Camden Plywood and Stud Operations, Camden, Texas; Memorandum from Midwest Research Institute to U.S. Environmental Protection Agency (USEPA), Office of Air Quality and Planning Standards (OAQPS): December 1, 1997.
OAQPS 1998a. Hanks, Katie; Nicholson, Becky. Site Visit  -  Timber Products Company Particleboard and Hardwood Plywood Plant in Medford, Oregon; Memorandum from Midwest Research Institute to U.S. Environmental Protection Agency (USEPA), Office of Air Quality and Planning Standards (OAQPS): April 16, 1998.
OAQPS 1998b. Hanks, Katie; Nicholson, Becky. Site Visit  -  Medite Division of Sierra Pine Medium Density Fiberboard Plant in Medford, Oregon; Memorandum from Midwest Research Institute to U.S. Environmental Protection Agency (USEPA), Office of Air Quality and Planning Standards (OAQPS): May 27, 1998.
OAQPS 1998c. Hanks, Katie; Bullock, David. Site Visit  -  Triwood, Incorporated Particleboard Plant in Bassett, Virginia; Memorandum from Midwest Research Institute to U.S. Environmental Protection Agency (USEPA), Office of Air Quality and Planning Standards (OAQPS): August 18, 1998.
OAQPS 1999a. Hanks, Katie; Threatt, Brigit; Nicholson, Becky. Summary of Responses to the 1998 EPA Information Collection Request (MACT Survey)  -  Hardwood Plywood and Veneer; Development of a NESHAP for Plywood and Composite Wood Products Manufacturing Industry; Memorandum from Midwest Research Institute to U.S. Environmental Protection Agency (USEPA), Office of Air Quality and Planning Standards (OAQPS): May 19, 1999.
OAQPS 1999b. Bullock, David; Nicholson, Becky; Threatt, Brigit. Site Visit  -  Willamette Industries, Incorporated, Moncure, North Carolina; Memorandum from Midwest Research Institute to U.S. Environmental Protection Agency (USEPA), Office of Air Quality and Planning Standards (OAQPS): July 27, 1999.
OAQPS 2000a. Hanks, Katie; Bullock, David; Nicholson, Becky. Summary of Responses to the 1998 EPA Information Collection Request (MACT Survey)  -  General Survey; Development of a NESHAP for Plywood and Composite Wood Products Manufacturing Industry; Memorandum from Midwest Research Institute to U.S. Environmental Protection Agency (USEPA), Office of Air Quality and Planning Standards (OAQPS): April 28, 2000.
OAQPS 2000b. Background Information Document for Proposed Plywood and Composite Wood Products NESHAP; EPA-453/R-01-004; U.S. Environmental Protection Agency (USEPA), Office of Air Quality and Planning Standards (OAQPS): Research Triangle Park, NC, September 2000.
OAQPS 2002a. Economic Impact Analysis of the Plywood and Composite Wood Products NESHAP: Final Report; U.S. Environmental Protection Agency (USEPA), Office of Air Quality and Planning Standards (OAQPS): Research Triangle Park, NC, November 2002.
OAQPS 2002b. Compilation of Air Pollutants Emission Factors, Volume I: Stationary Point and Area Sources (AP-42), Chapter 10, Wood Products Industry; U.S. Environmental Protection Agency (USEPA), Office of Air Quality and Planning Standards (OAQPS): 2002. http://www.epa.gov/ttnchie1/ap42/ch10/.
OECA 1995a. EPA Office of Compliance Sector Notebook Project: Profile of the Wood Furniture and Fixtures Industry; EPA/310-R-95-003; U.S. Environmental Protection Agency (USEPA), Office of Enforcement and Compliance Assurance (OECA), Office of Compliance: Washington, DC, September 1995.
OSHA 1999a. Inspection Report for McKnight Plywood, West Helena, AR; Inspection number 123445595; U.S. Department of Labor, Occupational Safety and Health Administration (OSHA): Little Rock, AR, August 30, 1999.
OSHA 2000a. Inspection Report for Carolina Curves, Inc., Conover, NC; Inspection number 303713325; North Carolina Department of Labor, Division of Occupational Safety and Health: Charlotte, NC, August 15, 2000.
OSHA 2000b. Inspection Report for Jimson Manufacturing Company, Inc., Haleyville, AL; Inspection number 303424014; U.S. Department of Labor, Occupational Safety and Health Administration (OSHA): Birmingham, AL, September 20, 2000.
OSHA 2001a. Inspection Report for Marshfield Door Systems, Inc., Marshfield, WI; Inspection number 303864029; U.S. Department of Labor, Occupational Safety and Health Administration (OSHA): Madison, WI, November 9, 2001.
OSHA 2001b. Inspection Report for Newood Display Fixtures Manufacturing Company, Eugene, OR; Inspection number 304064835; Oregon Department of Consumer and Business Services, Oregon Occupational Safety and Health Division (OR-OSHA): Salem, OR, February 15, 2001.
OSHA 2002a. Woodworking eTool [Online]; Occupational Safety and Health Administration (OSHA): April 2002. http://www.osha.gov/SLTC/etools/woodworking/index.html (accessed September 7, 2011).
OSHA 2002b. Inspection Report for Parenti & Rafaelli, Ltd, Mount Prospect, IL; Inspection number 305228652; U.S. Department of Labor, Occupational Safety and Health Administration (OSHA): Des Plaines, IL, August 1, 2002.
OSHA 2003a. Inspection Report for Bushline, Inc., New Tazewell, TN; Inspection number 306132275; State of Tennessee, Department of Labor and Workforce Development, Division of Occupational Safety and Health: Nashville, TN, June 19, 2003.
OSHA 2003b. Inspection Report for Tru-Wood Cabinets, Inc., Ashland, AL; Inspection number 307013250; U.S. Department of Labor, Occupational Safety and Health Administration (OSHA): Birmingham, AL, December 5, 2003.
OSHA 2003c. Inspection Report for Simple Designs Manufacturing, Inc., Tualatin, OR; Inspection number 306629171; Oregon Department of Consumer and Business Services, Oregon Occupational Safety and Health Division (OR-OSHA): Salem, OR, October 21, 2003.
OSHA 2004a. Inspection Report for Armstrong Cabinets, Auburn, NE; Inspection number 306021239; U.S. Department of Labor, Occupational Safety and Health Administration (OSHA): Omaha, NE, September 30, 2004.
OSHA 2004b. Inspection Report for England, Inc., New Tazewell, TN; Inspection number 307674051; State of Tennessee, Department of Labor and Workforce Development, Division of Occupational Safety and Health: Nashville, TN, June 11, 2004.
OSHA 2004c. Inspection Report for Harris-Tarkett, Inc., Johnson City, TN; Inspection number 307206987; State of Tennessee, Department of Labor and Workforce Development, Division of Occupational Safety and Health: Nashville, TN, March 19, 2004.
OSHA 2005a. Inspection Report for Fiberesin Industries, Inc., Oconomowoc, WI; Inspection number 307061689; U.S. Department of Labor, Occupational Safety and Health Administration (OSHA): Milwaukee, WI, September 21, 2005.
OSHA 2005b. Inspection Report for Plymart, Inc., Boring, OR; Inspection number 308245372; Oregon Department of Consumer and Business Services, Oregon Occupational Safety and Health Division (OR-OSHA): Portland, OR, January 10, 2005.
OSHA 2005c. Inspection Report for The Gunlocke Company, LLC, Wayland, NY; Inspection number 307690578; U.S. Department of Labor, Occupational Safety and Health Administration (OSHA): North Syracuse, NY, September 21, 2005.
OSHA 2006a. Inspection Report for Wood Goods Industries, Inc., Luck, WI; Inspection number 307041707; U.S. Department of Labor, Occupational Safety and Health Administration (OSHA): Eau Claire, WI, December 18, 2006.
OSHA 2007a. Inspection Report for The Commercial Furniture Group, Morristown, TN; Inspection number 311144059; State of Tennessee, Department of Labor and Workforce Development, Division of Occupational Safety and Health: Nashville, TN, July 9, 2007.
OSHA 2008a. Inspection Report for Benvenuti and Stein, Inc., Evanston, IL; Inspection number 312188972; U.S. Department of Labor, Occupational Safety and Health Administration (OSHA): Des Plaines, IL, September 26, 2008.
OSHA 2008b. Occupational Safety and Health Standards: Toxic and Hazardous Substances: Formaldehyde; 29 C.F.R. 1910.1048 (December 12, 2008). http://www.osha.gov/pls/oshaweb/owadisp.show_document?p_table=STANDARDS&p_id=10075 (accessed September 7, 2011).
OSHA 2008c. Inspection Report for Unilin Flooring NC, LLC, Mt. Gilead, NC; Inspection number 312479835; North Carolina Department of Labor, Division of Occupational Safety and Health: Charlotte, NC, July 28, 2008.
OSHA 2009a. Occupational Safety and Health Administration (OSHA) Integrated Management Information System (IMIS) database for formaldehyde (substance codes 1290, 1291, 1293) in all NAICS codes, encompassing the dates 1979 to 2009 (accessed June 2009).
OSHA 2009b. Inspection Report for Ameriwood Industries, Inc., Tiffin, OH; Inspection number 311606883; U.S. Department of Labor, Occupational Safety and Health Administration (OSHA): Toledo, OH, March 23, 2009.
OSHA 2010a. OSHA targets high hazard worksites for inspection. OSHA QuickTakes [Online] 2010, 9 (19). http://www.osha.gov/as/opa/quicktakes/qt10012010.html (accessed September 29, 2011).
OSHA 2010b. Chemical Sampling Information: Formaldehyde [Online]; Occupational Safety and Health Administration (OSHA): 2010. http://www.osha.gov/dts/chemicalsampling/data/CH_242600.html.
Preuss et al. 1985a. Preuss, P.W.; Dailey, R.L.; Lehman, E.S. Exposure to formaldehyde. In Formaldehyde. Analytical Chemistry and Toxicology; Turoski, V., Ed.; Advances in Chemistry Series, Vol. 210; American Chemical Society: Washington, DC, 1985; pp 247-259.
Reinke 2009a. Reinke, Patricia H.; Keil, Charles B. Well-Mixed Box Model. In Mathematical Models for Estimating Occupational Exposure to Chemicals, 2nd ed.; Keil, Charles B.; Simmons, Catherine E.; Anthony, T. Renée, Eds.; American Industrial Hygiene Association: Fairfax, VA, 2009; pp 23-31.
RIOPA 2005a. Weisel, Clifford P.; Zhang, Junfeng (Jim); Turpin, Barbara J.; Morandi, Maria T.; Colome, Steven; Stock, Thomas H.; Spektor, Dalia M.; and Others. Relationships of Indoor, Outdoor, and Personal Air (RIOPA): Part I. Collective Methods and Descriptive Analyses; Number 130 Part I; Research Report for the Health Effects Institute (HEI): November 2005.

Tohmura 2000a. Tohmura, Shin-ichiro; Hse, Chung-Yun; Higuchi, Mitsuo. Formaldehyde emission and high-temperature stability of cured urea-formaldehyde resins. J. Wood Sci. 2000, 46, 303-309.
USDA 1996a. Ince, Peter J. Recycling of Wood and Paper Products in the United States; FPL-GTR-89; U.S. Department of Agriculture (USDA), Forest Service, Forest Products Laboratory: Madison, WI, 1996.
USEPA 1991a. Chemical Engineering Branch Manual for the Preparation of Engineering Assessments, Volume 1; U.S. Environmental Protection Agency (USEPA), Office of Pollution Prevention and Toxics, Chemical Engineering Branch: Washington, DC, February 1991.
USEPA 1996a. Turner, Sonji; Martin, Cybele; Hetes, Robert; Northeim, Coleen. Sources and Factors Affecting Indoor Emissions from Engineered Wood Products: Summary and Evaluation of Current Literature; EPA-600-R-96-067; U.S. Environmental Protection Agency (USEPA), National Risk Management Research Laboratory: Research Triangle Park, NC, June 1996.
USEPA 1997a. Exposures Factors Handbook; EPA/600/P-95/002; U.S. Environmental Protection Agency (USEPA), Office of Research and Development, National Center for Environmental Assessment: Washington, DC, August 1997.
USEPA 2000a. Case Studies: Low-VOC / HAP Wood Furniture Coatings; EPA-600/R-00-043; U.S. Environmental Protection Agency (USEPA), National Risk Management Research Laboratory: Research Triangle Park, NC, May 2000.
USEPA 2000b. Options for Revising CEB's Method for Screening-Level Estimates of Dermal Exposure  -  Final Report; U.S. Environmental Protection Agency (USEPA), Office of Pollution Prevention and Toxics, Chemical Engineering Branch: Washington, DC, June 2000.
USEPA 2009a. Risk Assessment Guidance for Superfund Volume I: Human Health Evaluation Manual (Part F, Supplemental Guidance for Inhalation Risk Assessment); U.S. Environmental Protection Agency (USEPA): January 2009.
USEPA 2009b. Formaldehyde from Pressed Wood Products: Exposure Assessment  -  Draft Final Report; U.S. Environmental Protection Agency (USEPA), Office of Pollution Prevention and Toxics, Exposure Assessment Branch: Washington, DC, December 2009.
USEPA 2010a. Market Profile for Industries Potentially Affected by the Proposed Formaldehyde Standards for Composite Wood Products Rule  -  Internal Deliberative; U.S. Environmental Protection Agency (USEPA), Office of Pollution Prevention and Toxics, Economic and Policy Analysis Branch: Washington, DC, December 2010.
USEPA 2010b. Municipal Solid Waste in The United States: 2009 Facts and Figures; EPA530-R-10-012; U.S. Environmental Protection Agency (USEPA), Office of Solid Waste: Washington, DC, December 2010.
USEPA 2011a. Draft Final Occupational and Consumer Exposure and Environmental Release Assessment of Formaldehyde in Urea-Formaldehyde Resin Manufacture and of Certain Chemicals for Substitute Technologies to Reduce Emissions from Composite Wood Products; U.S. Environmental Protection Agency (USEPA), Office of Pollution Prevention and Toxics, Chemical Engineering Branch and Exposure Assessment Branch: Washington, DC, October 2011.
USEPA 2011b. An Introduction to Indoor Air Quality (IAQ): Formaldehyde [Online]; U.S. Environmental Protection Agency (USEPA), Updated February 15, 2011. http://www.epa.gov/iaq/formalde.html.
USEPA 2011c. Construction and Demolition Materials; U.S. Environmental Protection Agency (USEPA), Updated July 27, 2011. http://www.epa.gov/wastes/nonhaz/industrial/cd/index.htm.
USITC 2008a. Wood Flooring and Hardwood Plywood: Competitive Conditions Affecting the U.S. Industries; USITC Publication 4032; U.S. International Trade Commission (USITC), Washington, DC, August 2008.

Walpole 2002a. Walpole, Ronald E.; Myers, Raymond H.; Myers, Sharon L.; Ye, Keying. Probability and Statistics for Engineers and Scientists, 7th ed.; Prentice Hall: Upper Saddle River, NJ, 2002; pp 234-236.
Wong 2006. Wong, L.T.; Mui, K.W.; Hui, P.S.A statistical model for characterizing common air pollutants in air-conditioned offices. Atmospheric Environment 2006, 40, 4246-4257.
WorksafeBC 2009a. Occupational Exposure Limits for Formaldehyde; Formaldehyde OEL Study, Policy and Research Division, WorksafeBC: Vancouver, BC, July 2009.

Zinn 1990a. Zinn, Terry W.; Cline, Dennis; Lehmann, William F. Long-term study of formaldehyde emission decay from particleboard. Forest Prod. J. 1990, 40 (6), 15-18.

Appendix A
DEFINITION OF TERMS

Definition of Terms

American Conference of Governmental Industrial Hygienists (ACGIH)
ACGIH is a member-based organization of industrial hygienists whose goal is to advance occupational and environmental health.  ACGIH performs individual lab tests and develops with their own exposure limits that are sometimes used as suggested standards.
Action Level
Action level means a concentration designated in 29 CFR 1910 for a specific substance, calculated as an 8-hour time-weighted average (TWA), which initiates certain required activities such as exposure monitoring and medical surveillance.
CARB Compliant Products
In 2008, the California Air Resources Board (CARB) issued an airborne toxic control measure (ATCM) for three composite wood products.  The emission standards were issued in two phases.  Products that are manufactured in accordance with emissions standards issued under Phase 1 and Phase 2 of the CARB rule are referred to as CARB Phase 1 compliant and CARB Phase 2 compliant products, respectively.  Products that do not meet the standards set by CARB are herein referred to as non-CARB compliant products.  Table C-2 in Appendix C summarizes the CARB emission standards.
Ceiling Limit (C)
The air concentration for a substance that should not be exceeded during any part of the occupational exposure.
Composite Wood Products (CWPs)
The manufacture of most composite wood products involves an operation where the wood material is bonded with a resin. In this report, composite wood products refer specifically to hardwood plywood, medium-density fiberboard, and particleboard.
Composite Wood Product Fabrication
The manufacture of commercial or consumer products, such as cabinets and furniture, from HWPW, MDF, and PB.
Emission Level
A term used to indicate concentration of formaldehyde in a test chamber for a CWP.  In this report, emission levels are assumed to be measured using the ASTM E1333 large chamber test method.
Emission Rate
A term that is used in this document to describe the emissions of formaldehyde as a flux, meaning a quantity (such as mass) of formaldehyde emitted per unit emitting area and per unit time.  Concentrations measured using ASTM E1333 may be used to calculate emission rates (in units of flux) according to the method described in ASTM E1333 (see Appendix E).  Emission rates are alternatively referred to as emission factors (see the BIFMA M7.1-2007 method).
Exposure
Contact with a substance by inhalation, dermal contact, or ingestion.
The magnitude of exposure is a measure of the concentration of a substance contacted over a period of time.  
Exposure Assessment
The process of determining how people come into contact with a hazardous substance, how often and for how long they are in contact with the substance, and how much of the substance they are in contact with.
Exposure Pathway
The route a substance takes from its source (where it began) to its end point (where it ends), and how people can come into contact with (or get exposed to) it. An exposure pathway has five parts: a source of contamination (such as a composite wood product); an environmental media and transport mechanism (such as the off-gassing of formaldehyde through air); a point of exposure (such as an individual's breathing zone); a route of exposure (inhalation), and a receptor population (people potentially or actually exposed). When all five parts are present, the exposure pathway is termed a completed exposure pathway.
Hardwood Plywood (HWPW)
According to FSCWPA, a hardwood or decorative panel that is:
	i. intended for interior use; and
	ii. composed of (as determined under the
	standard numbered ANSI/HPVA HP - 1 - 2009) an
	assembly of layers or plies of veneer, joined by an
	adhesive with:
		I. lumber core;
		II. particleboard core;
		III. medium-density fiberboard core;	
		IV. hardboard core; or	
		V. any other special core or special back 			material.

This does not include:
	i. military-specified plywood;
	ii. curved plywood; or
	iii. any other product specified in -- 
		I. the standard entitled `Voluntary Product
		Standard -- Structural Plywood' and 				numbered PS1 - 07; or
		II. the standard entitled `Voluntary Product
		Standard -- Performance Standard for Wood-			Based Structural-Use Panels' and numbered 			PS 2 - 04.
Note that the definition of hardwood plywood includes engineered veneer and any laminated product unless EPA exempts them by rule-making.
Laminated Product
According to FSCWPA, a product:

  (I) in which a wood veneer is affixed to:
   (aa) a particleboard platform;
   (bb) a medium-density fiberboard platform; or
   (cc) a veneer-core platform; and
  (II) that is:
   (aa) a component part;
   (bb) used in the construction or assembly of a finished good; and
   (cc) produced by the manufacturer or fabricator of the finished good in which the product is incorporated.

EPA has the discretion to change this definition.
Life Cycle
All stages of a product's development, from manufacture and processing to distribution, use, and disposal and recycling.
Manufacturing
The manufacture of the composite wood products: HWPW, MDF, and PB.  May also be referred to as "primary manufacturing".
Medium-Density Fiberboard (MDF)
According to FSCWPA, a panel composed of cellulosic fibers made by dry forming and pressing a resinated fiber mat (as determined under the standard numbered ANSI A208.2 - 2009).
No Added Formaldehyde (NAF) Resins
According to FSCWPA: 
(A) IN GENERAL:
  (i) a  resin formulated with no added formaldehyde as part of the resin cross-linking structure in a composite wood product that meets the emission standards in subparagraph (C) as measured by: one test conducted pursuant to test method ASTM E - 1333 - 96 (2002) or, subject to clause (ii), ASTM D - 6007 - 02; and 3 months of routine quality control tests pursuant to ASTM D - 6007 - 02 or ASTM D - 5582 or such other routine quality control test methods as may be established by the Administrator through rulemaking.
  (ii) Test results obtained under clause (i)(I) or (II) by any test method other than ASTM E - 1333 - 96 (2002) must include a showing of equivalence by means established by the Administrator through rulemaking.

(B) INCLUSIONS: The term `no-added formaldehyde based resin' may include any resin made from:
  (i) soy;
  (ii) polyvinyl acetate; or
  (iii) methylene diisocyanate.

(C) EMISSION STANDARDS: The following are the emission standards for composite wood products made with no-added formaldehyde-based resins under this paragraph:
  (i) No higher than 0.04 parts per million of formaldehyde for 90 percent of the 3 months of routine quality control testing data required under subparagraph (A)(ii).
  (ii) No test result higher than 0.05 parts per million of formaldehyde for hardwood plywood and 0.06 parts per million for particleboard, medium-density fiberboard, and thin medium-density fiberboard.
National Institute for Occupational Safety and Health (NIOSH)
As part of the Centers for Disease Control, NIOSH is responsible for conducting research and making recommendations for the prevention of work-related illnesses and injuries.  
Occupational Exposure
Contact with a substance as a result of one's work environment.  
Occupational Safety and Health Administration (OSHA)
OSHA is the main federal agency charged with the enforcement of safety and health legislation. OSHA has the regulatory authority to set exposure limits for specific chemicals or groups of chemicals.
OSHA Integrated Management Information System (IMIS)
IMIS is an on-line data entry and information retrieval system designed to collect, process, retrieve, and communicate penalty assessment, arbitration, and collection information regarding OSHA's inspections.
OSHA Inspection Type: 
FAT-CAT
Complaint

OSHA inspections following a fatality or catastrophe at a facility. 
OSHA inspections resulting from worker complaints of unsafe or unhealthful working conditions. 
Follow-Up
A follow-up inspection determines if the employer has corrected previously cited violations. If the employer fails to abate a former violation, he or she is subject to additional penalties until the violation is corrected.  
Planned



Programmed


Related
The inspection was planned based on a particular reason, which can vary.  An example reason may include OSHA determining that a certain industry has a particular hazard more than others, so a random sample of facilities within that industry is selected for inspection. 
Planned and/or programmed inspections aimed at specific high-hazard industries, workplaces, occupations, health substances, or other industries identified in OSHA's current inspection procedures. 
The inspection was conducted due to concern over the facility regarding a different category, and upon investigation, OSHA determined that a related hazard should be monitored. 

Referral


Unprogrammed
OSHA inspections resulting from referrals. Referrals may come from any source including but not limited to: state EPA, health department, fire department, building code inspectors, newspaper, and related OSHA inspections. 
The facility was not inspected based on any planned or programmed list of industries or facilities.  These are the most random of OSHA inspections. 
Particleboard (PB)
According to FSCWPA, a  panel composed of cellulosic material in the form of discrete particles (as distinguished from fibers, flakes, or strands) that are pressed together with resin (as determined under the standard numbered ANSI A208.1 - 2009)

This does not include any product specified in the standard entitled `Voluntary Product Standard -- Performance Standard for Wood-Based Structural-Use Panels' and numbered PS 2 - 04.
Peak Concentration
The maximum air concentration of a chemical that is observed for a given sampling event.  
Permissible Exposure Limit (PEL)
PELs are regulatory limits issued by OSHA for the amount or concentration of a substance in the air. They may also contain a skin designation. OSHA PELs are based on an 8-hour time weighted average (TWA) exposure.
Potential Dose
The amount of a substance that is taken in by the body through ingestion, inhalation or dermal contact.  In this report, the potential dose, also referred to as "dose" applies to the amount of the substance that is inhaled.
Recommended Exposure Limit (REL)
Unless noted otherwise, RELs are time-weighted average (TWA) concentrations for up to a 10-hour workday during a 40-hour workweek. Unlike PELs, which are regulatory limits, RELs are recommended limits issued by NIOSH.
Short-Term Exposure Limit (STEL)
A STEL is defined as a 15-minute TWA exposure that should not be exceeded at any time during the work day even if the 8-hr TWA is within established limits. An averaging period other than 15 minutes may be used when this is warranted by observed biological effects.
The American Conference of Governmental Industrial Hygienists (ACGIH) recommends that exposures beyond the time-weighted average limit (whether regulatory or otherwise) up to the STEL should not occur more than 4 times per day with at least 60 minutes between successive exposures in this range.
Threshold Limit Value (TLV)
TLVs are recommended exposure limits that are issued by the ACGIH. TLVs are airborne concentrations of substances to which it is believed that nearly all workers may be repeatedly exposed day after day without adverse effect. There are three categories of TLVs: TLV-TWA; TLV-STEL; and TLV-C.
Time-Weighted Average (TWA) Concentration
Average air concentration of a substance based on a normal 8-hour work day and a 40-hour work week. OSHA PELs are based on an 8-hour time-weighted average (TWA) exposure.
Ultra Low-Emitting Formaldehyde (ULEF) Resin
According to FSCWPA:
(A) IN GENERAL:
  (i) The term `ultra low-emitting formaldehyde resin' means a resin in a composite wood product that meets the emission standards in subparagraph (C) as measured by:
   (I) 2 quarterly tests conducted pursuant to test method ASTM E - 1333 - 96 (2002) or, subject to clause (ii), ASTM D - 6007 - 02; and
   (II) 6 months of routine quality control tests pursuant to ASTM D - 6007 - 02 or ASTM D - 5582 or such other routine quality control test methods as may be established by the Administrator through rulemaking.
  (ii) Test results obtained under clause (i)(I) or (II) by any test method other than ASTM E - 1333 - 96 (2002) must include a showing of equivalence by means established by the Administrator through rulemaking.

(B) INCLUSIONS: The term `ultra low-emitting formaldehyde resin' may include:
  (i) melamine-urea-formaldehyde resin;
  (ii) phenol formaldehyde resin; and
  (iii) resorcinol formaldehyde resin.

(C) EMISSION STANDARDS:
  (i) The Administrator may, pursuant to regulations
issued under subsection (d), reduce the testing requirements for a manufacturer only if its product made with ultra low-emitting formaldehyde resin meets the following emission standards:
   (I) For hardwood plywood, no higher than 0.05 parts per million of formaldehyde.
   (II) For medium-density fiberboard:
     (aa) no higher than 0.06 parts per million of formaldehyde for 90 percent of 6 months of routine quality control testing data required under subparagraph (A)(ii); and
     (bb) no test result higher than 0.09 parts per million of formaldehyde.	


   (III) For particleboard:
     (aa) no higher than 0.05 parts per million of formaldehyde for 90 percent of 6 months of routine quality control testing data required under subparagraph (A)(ii); and
     (bb) no test result higher than 0.08 parts per million of formaldehyde.
   (IV) For thin medium-density fiberboard:
     (aa) no higher than 0.08 parts per million of formaldehyde for 90 percent of 6 months of routine quality control testing data required under subparagraph (A)(ii); and
     (bb) no test result higher than 0.11 parts per million of formaldehyde.
  (ii) The Administrator may not, pursuant to regulations issued under subsection (d), exempt a manufacturer from third party certification requirements unless its product made with ultra low-emitting formaldehyde resin meets the following emission standards:
   (I) No higher than 0.04 parts per million of formaldehyde for 90 percent of 6 months of routine quality control testing data required under subparagraph (A)(ii).
   (II) No test result higher than 0.05 parts per million of formaldehyde for hardwood plywood and 0.06 parts per million for particleboard, medium-density fiberboard, and thin medium-density fiberboard.


Appendix B
DATA QUALITY ACCEPTANCE CRITERIA SPECIFICATIONS

                Data Quality Acceptance Criteria Specifications
                                       
                       Specifications for Data Elements
                            Acceptance Criterion: 
                            Description/Definition
                                 Specification
                                       
Currency
Is the information up to date?  Does it reflect current conditions?
Process Description: Year of source is 1990 or after

Number of Sites and Production Data: Year of source is 2000 or after and correlates with number of sites

Exposure Data: 
Priority - Year of source is 2004 or after[1] Consider -  Year of source is 1999 or after

Geographic Scope
Does the information reported reflect an area relevant to the assessment?
All Data:  
Priority  -  sources describing U.S. sites

When U.S. data are not available, consider sources from countries with geographic proximity that may adopt similar processes or from other developed countries that may have similar regulations and standards.  

Process Description:
Consider  -  sources describing Canadian, Mexican, or European processes.  

Number of Sites and Production Data: 
Consider  -  North American sites  (and interpolate to estimate U.S. sites and production)

Exposure Data: 
Consider  -  Canadian or European data.  
 
Accuracy/Reliability
Is the information or data from a peer-reviewed, government, or industry-specific source? Is the source published? Is the author engaged in a relevant field such that competent knowledge is expected (i.e., the author writes for an industry trade association publication versus a general newspaper)?  
All Data: 
U.S. or other government publication

For an academic researcher:
   * Publication in peer-reviewed journal 
   * Presentation at a technical conference
   * Source has documented qualifications/ credentials to discuss particular topic 

For an industry expert or trade group: 
   * Presentation at a technical conference where the information is subject to review by other industry experts
   * Source has documented qualifications/ credentials to discuss particular topic 
   * Source represents a large portion of the industry of interest
Unbiased[2]
Is the information objective towards a particular product or outcome?
All Data:
   * Objective of the information is clear.
   * Methodology is designed to answer a specific question.

Comparability
Are the data comparable to other sources that have been identified or are they significantly different than other sources?
Production Data, Exposure Data, and Number of Sites:  Values are within the same order of magnitude of each other.  

Process Description: Matches or is similar to other information for the same industry and will be evaluated on a case-by-case basis.

Representativeness
Do the data reflect typical industry practices? 
All Data: 
Information is relevant to the industry and from a reputable source with qualifications in the industry of interest.

Data are based on a large industry survey or study, as opposed to a case study or sample from a limited number of sites.  
Applicability
For surrogate data, are the data expected to be similar for the industry of interest?  
Surrogate data  used to fill data gaps must be judged to be similar to the data of interest: 
   * Industry is similar in size 
   * Same types of chemicals are used
   * Similar processes and activities are performed
   * Similar levels of engineering controls and personal protective equipment are used

1 EPA promulgated in 2004 the Plywood and Composite Wood Products MACT Rule for hazardous air pollutants (HAPs), including formaldehyde, for regulated units.   
[2] Information from advocacy groups should be collected and judged for acceptability.  While such information may be biased, it may provide references to other useful sources.
Appendix C
EXISTING FORMALDEHYDE OCCUPATIONAL EXPOSURE LIMITS AND FORMALDEHYDE OFF-GASSING EMISSION STANDARDS FOR COMPOSITE WOOD PRODUCTS

Formaldehyde Occupational Exposure Limits

The Occupational Safety and Health Administration (OSHA) sets permissible exposure limits (PELs) to regulate worker exposure to formaldehyde in occupational settings.  OSHA has established two types of exposure limits for airborne exposures to formaldehyde: a 0.75 parts per million (ppm) time-weighted average (TWA) PEL over an 8-hour period and a 2 ppm short-term exposure limit (STEL) over a 15-minute period.
Table C-1 summarizes these OSHA exposure limits as well as other formaldehyde exposure limits for the U.S. and for other countries.
Table C-1. Existing Occupational Exposure Limits for Formaldehyde
                                    Country
                              Concentration (ppm)
                                     Type
USA-ACGIH
                                      0.3
                                    Ceiling
USA-NIOSH
                                     0.016
                                      TWA
USA-OSHA[b]
                                     0.75
                                      TWA
USA-OSHA
                                      2.0
                                     STEL
Canada-Alberta
                                      2.0
                                    Ceiling
Canada-Ontario
                                      0.3
                                    Ceiling
Canada-Quebec
                                      2.0
                                    Ceiling
Sweden
                                      0.5
                                      TWA
New Zealand
                                      1.0
                                    Ceiling
Australia [a]
                                      1.0
                                      TWA
Japan
                                      0.5
                                      TWA
United Kingdom
                                      2.0
                                      TWA
a  -  Australia proposed a lower occupational exposure limit of 0.3 ppm in 2008 (Chimar Hellas undated).
b  -  The OSHA action level for formaldehyde is 0.5 ppm TWA (29 CFR § 1910.1048).
Sources: IARC 2006a, Chimar Hellas undated, and OSHA 2010b

Formaldehyde Emission Standards for Composite Wood Products

Emission standards have been established in the U.S. for HWPW, MDF, and PB.  The American National Standards Institute (ANSI) established a voluntary standard for MDF and PB, while the Department of Housing and Urban Development (HUD) established national standards for plywood and particleboard (CARB 2007a).  In 2008, CARB issued an air toxics control measure (ATCM) for the state of California, which provided stricter formaldehyde emission standards than the existing national standards (see Table C-2).  The ANSI standards have been revised to reflect the CARB ATCM emission standards.
The CARB regulation focuses only on three types of composite wood products: HWPW, PB, and MDF.  CARB set emission standards for each of these products based on best achievable control technologies (BACTs).  The BACTs were selected based on CARB survey results, information in the open literature, and expert judgment on low- and no-added formaldehyde resin technologies.  The selected BACTs are such that the three regulated products have different emission standards (CARB 2007a).
CARB reported the resin types used by the manufacturers who responded to the CARB 2003 survey (CARB 2007a).
Of the HWPW manufacturer responses, which accounted for 73% of domestic HWPW production:
               o All companies reported using ammonia-UF resin.  The lowest emission rates were achieved with the use of catalysts, such as ammonium chloride or hexamethylenetetramine.
         * Of the PB manufacturer responses, which accounted for 53% of domestic PB production:
               o 51% of the reported production volume used a straight polymer UF resin;
               o 48% of the reported production volume used a methanol-UF resin; and
               o 1% of the reported production volume used a blended PF-UF resin.
         * Of the MDF manufacturer responses, which represented 83% of domestic MDF production:
               o 40% of the reported production volume used a straight polymer UF resin;
               o 20% of the reported production volume used a methanol-UF resin;
               o 39% of the reported production volume used a melamine-UF resin with sodium chloride catalyst; and
               o 1% of the reported production volume used MDI resin.

The majority of the reported resins used were UF resins with many using additives, such as ammonia or methanol, and catalysts.

Table C-2. California Phase 1 and Phase 2 Formaldehyde Emission Standards
                                       
                 ASTM E1333 Standard Concentrations (ppm) [a]
                    Phase 1 Formaldehyde Emission Standards
                                Effective Date
                                  HWPW-VC [b]
                                  HWPW-CC [c]
                                      PB
                                      MDF
                                   tMDF [d]
                                   Jan 2009
                                     0.08
                                       -- 
                                     0.18
                                     0.21
                                     0.21
                                   July 2009
                                       -- 
                                     0.08
                                       -- 
                                       -- 
                                       -- 
                    Phase 2 Formaldehyde Emission Standards
                                Effective Date
                                  HWPW-VC [b]
                                  HWPW-CC [c]
                                      PB
                                      MDF
                                   tMDF [d]
                                   Jan 2011
                                     0.05
                                       -- 
                                     0.09
                                     0.11
                                       -- 
                                   Jan 2012
                                       -- 
                                       -- 
                                       -- 
                                       -- 
                                     0.13
                                   July 2012
                                       -- 
                                     0.05
                                       -- 
                                       -- 
                                       -- 
a  -  ppm = parts per million.  In this case, parts of formaldehyde per million parts of air.
b  -  HWPW-VC = hardwood plywood  -  veneer core.
c  -  HWPW-CC = hardwood plywood  -  composite core. 
d  -  tMDF = thin medium-density fiberboard (8-mm or thinner).
Source: CARB 2007a


Table C-3.  Formaldehyde Emission Limits and Corresponding Emission Rates as Measured using ASTM E1333 under HUD, CARB, and FSCWPA
 
           HUD, CARB, and FSCWPA Limits as Measured from ASTM E1333
                          ASTM E1333 Test Parameters
                      Calculated Values (see Appendix E)

                              Concentration (ppm)
                            Concentration (mg/m[3])
                           Loading Ratio (m[2]/m[3])
                           Air Change Rate (hr[-1])
                    Formaldehyde Emission Rate (mg/m[2]-hr)
                   Area-Specific Airflow Rate (m[3]/m[2]-hr)
                                     HWPW
HUD
                                      0.2
                                     0.246
                                     0.43
                                      0.5
                                    0.2858
                                     1.16
CARB Phase 1
                                     0.08
                                     0.098
                                     0.43
                                      0.5
                                    0.1143
                                     1.16
FSCWPA / CARB Phase 2
                                     0.05
                                     0.061
                                     0.43
                                      0.5
                                    0.0715
                                     1.16
ULEF
                                     0.045
                                    0.0553
                                     0.43
                                      0.5
                                    0.0643
                                     1.16
NAF
                                     0.04
                                    0.0492
                                     0.43
                                      0.5
                                    0.0572
                                     1.16
                                      PB
HUD
                                      0.3
                                     0.369
                                     0.43
                                      0.5
                                    0.4287
                                     1.16
CARB Phase 1
                                     0.18
                                     0.221
                                     0.43
                                      0.5
                                    0.2572
                                     1.16
FSCWPA / CARB Phase 2
                                     0.09
                                     0.111
                                     0.43
                                      0.5
                                    0.1286
                                     1.16
ULEF
                                     0.045
                                     0.055
                                     0.43
                                      0.5
                                    0.0643
                                     1.16
NAF
                                     0.04
                                     0.049
                                     0.43
                                      0.5
                                    0.0572
                                     1.16
                                      MDF
HUD
                                     None
                                      --
                                      --
                                      --
                                      --
                                      --
CARB Phase 1
                                     0.21
                                     0.258
                                     0.26
                                      0.5
                                    0.4963
                                     1.92
FSCWPA / CARB Phase 2
                                     0.11
                                     0.135
                                     0.26
                                      0.5
                                    0.2600
                                     1.92
ULEF
                                     0.05
                                     0.061
                                     0.26
                                      0.5
                                    0.1182
                                     1.92
NAF
                                     0.04
                                     0.049
                                     0.26
                                      0.5
                                    0.0945
                                     1.92


Appendix D
FORMALDEHYDE CWP OFF-GASSING EMISSION LEVEL DATA FOR THE PRE-BASELINE PERIOD, BASELINE, AND THE ANALYTICAL OPTIONS

Use of Emission Levels in the Assessment of Formaldehyde Occupational Exposure

Emission levels for formaldehyde off-gassing from CWP, which have units of concentration, are converted to emission rates, which have units of flux, and are used in the mass balance equations used to calculate indoor air concentrations due to off-gassing of CWPs in fabrication sites and in office buildings for the baseline and the three analytical options.  Pre-baseline emission levels are used to calculate pre-baseline indoor air concentrations in CWP fabrication sites for the purpose of comparing model results to pre-baseline OSHA IMIS data as described in Section 4.2.5.3.  Pre-baseline emission levels are also used to estimate the barrier effectiveness of the finishing (coating, overlays, etc) of the CWP from which office building workstation components are fabricated.  This barrier effectiveness is used in the calculation of formaldehyde indoor air concentration in office buildings for the baseline and the three analytical options (see Section 3.3.1.2 and Appendix E).
Emission Levels of Pre-Baseline CWPs

This section discusses the available pre-baseline emission level data and the estimation of pre-baseline emission levels.  The pre-baseline emission levels used in this assessment are summarized in Table D-1 below.  The data used to estimate these emission levels are presented in Table D-2.
Considerable information is available on the emission levels of domestically-produced pre-baseline products as shown in Table D-2.  As seen in this table, at the beginning of the pre-baseline period of 2002 to 2009, the production-weighted average emission levels of domestically-produced CWP were close to the CARB ATCM Phase 1 limits while the range of emission levels for these products had a lower bound that was less than the CARB ATCM Phase 1 standard emission level, and an upper bound that was considerably greater.  In ANPR submissions, it is noted that 75% of HWPW and 96% of PB and MDF met the CARB ATCM Phase 1 limit by 2009.  Therefore, the CARB ATCM Phase 1 limits were assumed to be representative of average pre-baseline emission levels for CWP produced domestically during the pre-baseline period of 2002 to 2009, and the average emission levels for domestically-produced HWPW, PB, and MDF were equated to the CARB ATCM Phase 1 limits (0.08 ppm, 0.18 ppm, and 0.21 ppm, respectively).
Emission levels of imported pre-baseline CWPs are not available and were estimated taking into account temporal trends in the emission levels of domestically-produced CWP and the reported emission levels of imported CWP.  The emission levels of baseline products (domestically produced and imported) and their share of U.S. consumption is provided in the market profile (USEPA 2010a) in Table 12-5.  A comparison of Table D-2 and Table 12-5 in the market profile (reproduced as Table D-3 in this appendix) shows that a considerable decrease in the emission levels of domestically-produced CWP will occur in going from the pre-baseline period to the baseline.  This temporal trend was assumed to apply to imported CWP such that there would be a decline in the emission levels of imported CWP from the pre-baseline period to the baseline.  The pre-baseline emission levels for Canadian imported CWP and non-Canadian imported CWP were estimated differently as discussed below.
According to Table 12-5 in the market profile (USEPA 2010a), baseline CWP imported from Canada have emission levels that are identical to the emission levels of domestically-produced baseline CWP.  This similarity was assumed to extend to the pre-baseline period, and therefore CWP imported from Canada during the pre-baseline period were assumed to have emission levels that are identical to the values assumed for domestically produced pre-baseline CWP.
Most of the non-Canadian imported CWP will adhere to the CARB Phase 2, ULEF, or NAF standards at baseline, but will have emission level values that are different from those of U.S. and Canadian CWP.  The baseline emission levels of non-Canadian imports are likely not representative of pre-baseline emission levels of these imports assuming that there will be a decline in emission levels of imported CWP from the pre-baseline to baseline as discussed above.  The fraction of imported baseline CWP that will not adhere to any of the CARB emission standards will be approximately 26%, 37%, and 10% for HWPW, PB, and MDF, respectively.  The average values of the emission levels for these volumes will be 0.163 ppm, 0.384 ppm, and 0.502 ppm for HWPW, PB, and MDF, respectively, as shown in Table D-2.  It is assumed that these emission levels are representative of the average pre-baseline emission levels for each CWP type for imported CWP.
In conclusion, the average emission levels for domestic and Canadian pre-baseline HWPW, PB, and MDF are assumed to be equal to the CARB ATCM Phase 1 limits (0.08 ppm, 0.18 ppm, and 0.21 ppm, respectively).  The average emission levels for imported pre-baseline HWPW, PB, and MDF are assumed to be equal to 0.163 ppm, 0.384 ppm, and 0.502 ppm, respectively.
An average emission level for all three CWP was used in the estimation of barrier effectiveness for use in the calculation of office building indoor air concentration.  Therefore, these emission levels are averaged together.  First, a volume-weighted average of the emission levels of domestically-produced and imported CWP is calculated for each individual CWP type.  The EPA market profile provides the 2005 volume-based shares of domestic consumption of domestic/Canadian and imported CWP (Table 12-4 in USEPA 2010a).  The volume-based shares of domestic consumption of non-Canadian imported CWPs are 64%, 3.5%, and 7.6% for HWPW, PB, and MDF, respectively.  The shares of domestic/Canadian CWPs are the difference between 100% and the import shares.  A volume-weighted average emission level is calculated for each CWP type as follows.  The emission rates (in units of flux) are similarly calculated.
For HWPW:
0.64x0.163 ppm+0.36x0.08 ppm=0.13 ppm

For PB:
0.035x0.384 ppm+0.965x0.18 ppm=0.19 ppm

For MDF:
0.076x0.502 ppm+0.924x0.21 ppm=0.23 ppm

Next, a single volume-weighted average emission level for all three CWP types is calculated from the three CWP weighted averages calculated above.  Table 12-4 in the EPA market profile (USEPA 2010a) also provides annual consumption volumes of the three CWP for the year 2005.  Based on these data, the consumption volumes of HWPW, PB, and MDF as a fraction of total the total consumption volume for these three CWP was 31.1%, 46.4%, and 22.5%, respectively.  The single volume-weighted average emission level for all three CWPs is calculated as follows.  The emission rates (in units of flux) are similarly calculated.
0.311x0.13 ppm+0.464x0.19 ppm+0.225x0.23 ppm=0.18 ppm

Table D-1. Pre-Baseline Formaldehyde Emission Levels for CWPs used in this Assessment
                                   CWP Type
                              Domestic or Import
                                     Units
                                 Average Value
                            Volume-Weighted Average
                                     Basis
HWPW
Domestics
ppm
                                     0.08
                                     0.13
CARB ATCM Phase 1 Limit

Imports
ppm
                                     0.163

Baseline imports that do not meet any emission standard
PB
Domestics
ppm
                                     0.18
                                     0.19
CARB ATCM Phase 1 Limit

Imports
ppm
                                     0.384

Baseline imports that do not meet any emission standard
MDF
Domestics
ppm
                                     0.21
                                     0.23
CARB ATCM Phase 1 Limit

Imports
ppm
                                     0.502
                                       
Baseline imports that do not meet any emission standard
Volume-Weighted Average of the Volume-Weighted Averages for the Three CWPs (ppm)
                                     0.18

HWPW
Domestics
ug/m[2]-h
                                     114.3
                                     190.3
Calculated from emission level based on ASTM E1333 (see Appendix E)

Imports
ug/m[2]-h
                                     232.9
                                       

PB
Domestics
ug/m[2]-h
                                     257.2
                                     267.3


Imports
ug/m[2]-h
                                     548.8
                                       

MDF
Domestics
ug/m[2]-h
                                     496.3
                                     548.9


Imports
ug/m[2]-h
                                    1,186.5


Volume-Weighted Average of the Volume-Weighted Averages for the Three CWPs (ug/m[2]-h)
                                     306.7



Table D-2.  Data Used to Determine Pre-Baseline Emission Levels for Composite Wood Products
                                    Timeline
                               Information Source
                                    HWPW [a]
                                     PB [b]
                                     MDF [b]
                                 Domestic CWPs
2002
CARB survey
                          Production-Weighted Average


                                   0.09 ppm
                                   0.18 ppm
                                   0.25 ppm


                  % of U.S. Production Represented in Survey


                                      73%
                                      53%
                                      83%


                           Range in Survey Responses


                               0.07 to 0.75 ppm
                               0.13 to 0.24 ppm
                               0.03 to 0.31 ppm
End of 2008 / Beginning of 2009
ANPR Submissions
75% of domestically-produced CWP were CARB Phase 1 compliant
96% of domestically-produced CWP were CARB P1 compliant
96% of domestically-produced CWP were CARB P1 compliant
Later in 2009
ANPR Submissions
90% of domestics expected to be CARB P1 compliant
                                      --
                                      --
Average Emission Levels for Imported Baseline CWPs That Do Not Meet Any of the CARB Emission Standards (USEPA 2010a) [c]
                                   Baseline
EPA market profile
                                   0.163 ppm
                                   0.384 ppm
                                   0.502 ppm
 a  -  A submission from HPVA in May 2008 estimated that domestically-produced HWPW will be in compliance with CARB Phase 1. Phase 2 limits require new technologies, which have not been in commercial use and some only recently invented and patented. Additionally, a submission from HPVA in February 2009 estimated that 75% of North American HWPW was CARB certified and soon 90% or more will be CARB certified. "CARB certified" is assumed to mean CARB Phase 1 per the statements made in the May 2008 submission. The term "soon" is uncertain; therefore, the point in time in which 90% of domestics were expected to meet CARB Phase 1 is uncertain.
 b  -  A submission from CPA in May 2008 stated that "almost all" domestic production is expected to be compliant with CARB Phase 1 and 2. The volume related to "almost all" is uncertain. This submission also states that "eventually" this rate of compliance would be reached; the time frame associated with "eventually" is uncertain. CPA did estimate that 96% of domestics were Phase 1 compliant by the end of 2008.
 c  -  Average emission levels of imported CWPs at baseline that do not meet any of the CARB emission standards are used as estimates of the average emission levels of all imported CWPs during pre-baseline.

Emission Levels of Baseline CWP and CWP that Would be Produced Under the Analytical Options

The EPA market profile provides emission levels for CWPs at baseline.  These emission levels and their corresponding volume-based shares of U.S. consumption are summarized in Table D-3 below.  Emission levels and their corresponding volume-based shares of U.S. consumption for the analytical options were determined based on the baseline values and are given in Table D-4 through Table D-6 (USEPA 2009b).
Table D-3.  Baseline Emission Levels and Corresponding U.S. Consumption (Copy of Table 12-5 in Market Profile)
                               Emission Standard
                               Hardwood Plywood
                                      MDF
                                 Particleboard
                                       
                                   U.S./Can.
                                 Non-U.S./Can.
                                   U.S./Can.
                                 Non-U.S./Can.
                                   U.S./Can.
                                 Non-U.S./Can.
                            Average Emissions (ppm)
None
                                     0.058
                                     0.163
                                       -
                                     0.502
                                     0.058
                                     0.384
CARB P1
                                       -
                                       -
                                       -
                                       -
                                     0.090
                                       -
CARB P2
                                     0.032
                                     0.015
                                     0.082
                                     0.047
                                     0.057
                                     0.036
ULEF
                                       -
                                       -
                                     0.024
                                       -
                                     0.042
                                       -
NAF
                                     0.013
                                       -
                                     0.035
                                       -
                                     0.013
                                       -
                           Share of U.S. Consumption
None
                                     0.1%
                                     16.6%
                                       -
                                     0.8%
                                     0.2%
                                     1.1%
CARB P1
                                       -
                                       -
                                       -
                                       -
                                     1.7%
                                       -
CARB P2
                                     19.7%
                                     47.4%
                                     79.2%
                                     7.2%
                                     85.0%
                                     1.9%
ULEF
                                       -
                                       -
                                     4.2%
                                       -
                                     3.0%
                                       -
NAF
                                     16.2%
                                       -
                                     8.6%
                                       -
                                     7.1%
                                       -
                    Weighted Average Emission Levels (ppm)
National Average
                                     0.043
                                     0.076
                                     0.057



Table D-4.  Emission Levels and Corresponding U.S. Share of Consumption for Analytical Option 1 (CARB Phase 1)
                               Emission Standard
                               Hardwood Plywood
                                      MDF
                                 Particleboard
                                       
                                   U.S./Can.
                                 Non-U.S./Can.
                                   U.S./Can.
                                 Non-U.S./Can.
                                   U.S./Can.
                                 Non-U.S./Can.
                            Average Emissions (ppm)
None
                                       -
                                       -
                                       -
                                       -
                                       -
                                       -
CARB P1 (previously NOT CARB P1)
                                   0.058 [a]
                                     0.058
                                       -
                                   0.14 [b]
                                   0.058 [a]
                                     0.090
Already at CARB P1
                                       -
                                       -
                                       -
                                       -
                                     0.090
                                       -
CARB P2
                                     0.032
                                     0.015
                                     0.082
                                     0.047
                                     0.057
                                     0.036
ULEF
                                       -
                                       -
                                     0.024
                                       -
                                     0.042
                                       -
NAF
                                     0.013
                                       -
                                     0.035
                                       -
                                     0.013
                                       -
                           Share of U.S. Consumption
None
                                       -
                                       -
                                       -
                                       -
                                       -
                                       -
CARB P1 (previously NOT CARB P1)
                                     0.1%
                                     16.6%
                                       -
                                     0.8%
                                     0.2%
                                     1.1%
Already at CARB P1
                                       -
                                       -
                                       -
                                       -
                                     1.7%
                                       -
CARB P2
                                     19.7%
                                     47.4%
                                     79.2%
                                     7.2%
                                     85.0%
                                     1.9%
ULEF
                                       -
                                       -
                                     4.2%
                                       -
                                     3.0%
                                       -
NAF
                                     16.2%
                                       -
                                     8.6%
                                       -
                                     7.1%
                                       -
                    Weighted Average Emission Levels (ppm)
National Average
                                     0.025
                                     0.073
                                     0.054
Source: USEPA 2009b
a  -  This share of consumption was previously given for products not meeting any standard at baseline, but the baseline emission level complies with CARB Phase 1 and thus does not change.
b  -  This emission level is 67% of the CARB Phase 1 limit.


Table D-5.  Emission Levels and Corresponding U.S. Share of Consumption for Analytical Option 2 (FSCWPA/CARB Phase 2)
                               Emission Standard
                               Hardwood Plywood
                                      MDF
                                 Particleboard
                                       
                                   U.S./Can.
                                 Non-U.S./Can.
                                   U.S./Can.
                                 Non-U.S./Can.
                                   U.S./Can.
                                 Non-U.S./Can.
                            Average Emissions (ppm)
None
                                       -
                                       -
                                       -
                                       -
                                       -
                                       -
CARB P1
                                       -
                                       -
                                       -
                                       -
                                       -
                                       -
CARB P2 (previously not CARB P2)
                                     0.032
                                     0.032
                                       -
                                     0.082
                                     0.057
                                     0.057
Already at CARB P2
                                     0.032
                                     0.015
                                     0.082
                                     0.047
                                     0.057
                                     0.036
ULEF
                                       -
                                       -
                                     0.024
                                       -
                                     0.042
                                       -
NAF
                                     0.013
                                       -
                                     0.035
                                       -
                                     0.013
                                       -
                           Share of U.S. Consumption
None
                                       -
                                       -
                                       -
                                       -
                                       -
                                       -
CARB P1
                                       -
                                       -
                                       -
                                       -
                                       -
                                       -
CARB P2 (previously not CARB P2)
                                     0.1%
                                     16.6%
                                       -
                                     0.8%
                                     1.9%
                                     1.1%
Already at CARB P2
                                     19.7%
                                     47.4%
                                     79.2%
                                     7.2%
                                     85.0%
                                     1.9%
ULEF
                                       -
                                       -
                                     4.2%
                                       -
                                     3.0%
                                       -
NAF
                                     16.2%
                                       -
                                     8.6%
                                       -
                                     7.1%
                                       -
                    Weighted Average Emission Levels (ppm)
National Average
                                     0.021
                                     0.073
                                     0.053
Source: USEPA 2009b

Table D-6.  Emission Levels and Corresponding U.S. Share of Consumption for Analytical Option 3 (NAF)
                               Emission Standard
                               Hardwood Plywood
                                      MDF
                                 Particleboard
                                       
                                   U.S./Can.
                                 Non-U.S./Can.
                                   U.S./Can.
                                 Non-U.S./Can.
                                   U.S./Can.
                                 Non-U.S./Can.
                            Average Emissions (ppm)
None
                                       -
                                       -
                                       -
                                       -
                                       -
                                       -
CARB P1
                                       -
                                       -
                                       -
                                       -
                                       -
                                       -
CARB P2
                                       -
                                       -
                                       -
                                       -
                                       -
                                       -
NAF (previously not NAF with non-NAF emissions)[a]
                                     0.013
                                     0.013
                                     0.035
                                     0.035
                                     0.013
                                     0.013
NAF (previously not NAF with NAF emissions)[b]
                                       
                                       
                                     0.024
                                       -
                                       -
                                       
Already at NAF
                                     0.013
                                       -
                                     0.035
                                       -
                                     0.013
                                       -
                           Share of U.S. Consumption
None
                                       -
                                       -
                                       -
                                       -
                                       -
                                       -
CARB P1
                                       -
                                       -
                                       -
                                       -
                                       -
                                       -
CARB P2
                                       -
                                       -
                                       -
                                       -
                                       -
                                       -
NAF (previously not NAF with non-NAF emissions)[a]
                                     19.8%
                                     64.0%
                                     79.2%
                                     8.0%
                                     89.9%
                                      3%
NAF (previously not NAF with NAF emissions)[b]
                                       
                                       
                                     4.2%
                                       -
                                       -
                                       
Already at NAF
                                     16.2%
                                       -
                                     8.6%
                                       -
                                     7.1%
                                       -
                    Weighted Average Emission Levels (ppm)
National Average
                                     0.013
                                     0.035
                                     0.013
Source: USEPA 2009b
a  -  These shares of consumption were previously given for products not meeting the NAF standard at baseline, and the emission levels were changed to match the domestic baseline NAF emission levels.
b - These shares of consumption were previously given for products not meeting the NAF standard at baseline, but the baseline emission level complies with NAF and thus does not change.






Appendix E
MODELING METHODOLOGY FOR ESTIMATING INDOOR AIR CONCENTRATIONS RELATED TO BASELINE AND ANALYTICAL OPTIONS

1.0	Indoor Air Concentration Assessment Methods

This section of this appendix describes the modeling methodology for estimating indoor air concentrations related to baseline and analytical option periods for the three composite wood products of interest: HWPW, PB, and MDF.  Both inhalation and dermal exposures are assessed.
1.1	CWP Fabrication Indoor Air Concentration Assessment Method

An airborne concentration of a chemical contaminant in a control volume may be determined from a mass balance on the chemical as presented in Equation 1 (USEPA 1991a).
					C=SQ						Eqn (1)

where:
      C	=	Airborne concentration of the chemical in the work space (ug/m[3]);
      S	=	Generation rate of the chemical as a vapor or gas from the source of interest in the work space (ug/h); and
      Q	=	Volumetric ventilation rate of the work space (m[3]/h).

The following assumptions apply to this model:
Extinction of the chemical (adsorption, absorption, or chemical transformation) resulting from deposition on walls and equipment, condensation of hot vapors, and photodegradation of chemicals is negligible;
The incoming air is contaminant-free;
The concentration of contaminant at initial time is negligible (there is no background concentration or other generation sources of the contaminant);
The generation and ventilation rates are constant over time (steady-state assumption); and
Room air and ventilation air mix ideally.

Equation 1 is modified to account for the presence of contaminant in the incoming (outdoor) air.  A new mass balance is derived to arrive at Equation 2.
					C=SQ+Coind				Eqn (2)

where:
      Co[ind]	=	Concentration of contaminant in the outdoor air (ug/m[3]).

To account for non-ideal mixing of the ventilation air and the contaminated room air, a dimensionless mixing factor, k, is used.  One interpretation of the mixing factor is a representation of the fraction, k, of the ventilation air that completely mixes with an unknown fraction of the room volume and the contaminant emitted from the source at the generation rate, S.  The remaining fraction, 1-k, of the ventilation air only mixes with the remaining (and unknown) fraction of the room volume, which does not mix with the emitted contaminant at generation rate, S (Reinke 2009a).
The mixing factor is multiplied by the ventilation rate to approximate an effective ventilation rate, which accounts for non-ideal mixing.  The effective ventilation rate, Qxk, is then substituted into Equation 2.
					Qeffective=Qxk				Eqn (3)

					C=SQeffective+Coind			Eqn (4)

Equation 4 is applied to the calculation of the formaldehyde indoor air concentration at a CWP fabrication site as follows:
      * The entire site constitutes the work space.
      * The ideal ventilation rate, Q, is equal to the total ventilation rate of the entire site.
      * The mixing factor, k, represents a fraction of the work space volume that effectively mixes with a fraction of the ventilation air and the formaldehyde emitted at generation rate, S, to result in the average indoor air concentration, C.
      * The contaminant concentration in outdoor air is equal to the mean formaldehyde concentration in ambient air for industrial land use areas according to the EPA exposure assessment (USEPA 2009b), which is equal to 6.28 ug/m[3].
      * The source generation rate, S, is the rate of formaldehyde off-gassing from the three CWP that are present at a CWP fabrication site (which is the only source considered in this assessment).  This input parameter is expressed as the product of the off-gassing emissive flux of formaldehyde (in terms of mass per time per composite wood product emitting surface area) and the emitting surface area.

					C=FxAQeffective+Coind			Eqn (5)

where:
	F	=	Emission rate (off-gassing emissive flux) of formaldehyde from the composite wood product (ug/m[2]-h); and
	A	=	Emitting surface area of the composite wood product (m[2]).

Formaldehyde off-gassing emissions from CWP decline with time, but this phenomenon is not relevant to the estimation of indoor air concentration in a CWP fabrication site because the residence time of the CWP at the site is expected to be small (days to weeks) as compared to the time required for the emission rate, F, to decline to an extent that would significantly affect the resulting indoor air concentration (weeks to months).  The concentration, C, becomes the average steady-state concentration of airborne formaldehyde due to formaldehyde off-gassing from CWP in the entire CWP fabrication site.  Section 3.3.1.1 discusses the input parameters in Equation 5 and their use in the mass balance equation, and the values of these parameters used are shown in Table 3-4 of that section.  The emissive flux values in Table 3-4 are derived from data in Appendix D.
1.2	Office Buildings Indoor Air Concentration Assessment Method

A mass balance approach similar to the one used for estimating formaldehyde indoor air concentration in CWP fabrication sites is used for the estimation of indoor air concentration in office buildings.  The rate of formaldehyde off-gassing from CWP components of office furniture that is expected to be present in a typical office work space is equated to the rate of exhaustion of airborne formaldehyde from this typical office work space through ventilation.  The rate of formaldehyde off-gassing from CWPs is estimated using a formaldehyde source model in which the generation rate of formaldehyde is equal to an emission rate multiplied by an emitting surface area of the office furniture expected to be present in a typical office work space.  The rate of exhaustion of indoor air formaldehyde is based on an ASHRAE outdoor air ventilation standard for this typical office work space.  Several parameters for the model, specifically emitting surface area and ventilation rate, are based on the modeling work presented in Carter and Zhang, 2007 and the BIFMA M7.1-2005 standard test method.  The remaining parameter, the emission rate, is determined using data from the EPA market profile (USEPA 2010a), similarly as was done for the CWP fabrication method, and using other assumptions and calculations described in subsection 1.2.3 and 1.2.4 below.
Carter and Zhang attempted to define a standard, representative "worst-case" office environment, in support of the BIFMA M7.1-2005 standard test method.  Carter and Zhang discussed using the BIFMA office environment model to calculate the impact of office furniture on indoor volatile organic compound (VOC) concentrations, such as formaldehyde.  Steady-state indoor airborne formaldehyde concentrations from office furniture in open work space areas and private offices are calculated from the office environment model with the following basic components:
Inputs:
Outdoor, clean air ventilation flow rate,
Surface area of composite wood products, and
Formaldehyde emission rates of composite wood products as a constant value;
Output:
Steady-state concentration at a specific time (C(t)).

The work by Carter and Zhang assumes the outdoor, clean air does not contain the VOC of interest (in this case, formaldehyde).  This exposure assessment accounts for observed concentrations of formaldehyde in ambient air.  This exposure assessment assumes the outdoor air used for office building ventilation contains formaldehyde at a concentration equal to the mean formaldehyde concentration in ambient air in commercial areas according to the EPA exposure assessment (USEPA 2009b).
1.2.1	Calculation Method

The steady-state concentration in a typical office environment due to emissions from an office workstation system or individual components may be calculated using Equation 6 (Carter and Zhang, 2007), which is based on the same mass balance principles as the mass balance model used for the CWP fabrication method:
				QC=AxE+QCocom				Eqn (6)


				C=AxEQ+Cocom					Eqn (7)

where:
      C	= 	concentration in the defined office environment at the particular time (ug/m[3]) due to a specific component;
      A 	= 	source amount in the office environment (m[2]);
      E	= 	estimated emissive flux at a particular time (ug/m[2]-h);
      Q	= 	outdoor air ventilation rate in the office environment (m[3]/h); and
      Co[com]	=	concentration of formaldehyde in the outdoor air for commercial land use areas (ug/m[3]).

The total overall concentration is calculated by summing the contribution from each component.  This summation results in the following:
				C=AixEiQ+Cocom				Eqn (8)

Here, the summation is performed over all workstation components i, where i represents each component type of the office workstation.
This model is applied to two office workstation types: open plan workstations and private office workstations.  The following subsections present the values used for the model parameters for both office workstation types.  The model parameters are:
The source amount of each workstation component, Ai, where the components are:
Panels,
Work surfaces,
Storage external surfaces;
The emissive flux of each workstation component, Ei; and
The outdoor air ventilation rate in the office workstation environment, Q.

1.2.2	BIFMA 2005 Workstation Model Parameters Ai and Q

As used in Carter and Zhang 2007, the BIFMA standard test method M7.1-2005 specifies two sets of workstation parameters based on typical open plan and private office environments.  These typical workstation parameters are for the purpose of estimating the impact of office furniture on the VOC concentrations in office spaces.  For a "typical open plan office environment of a single workstation system," BIFMA specifies:
Total workstation area dimensions: floor area of 5.94 m[2] (64 ft[2]) by 2.74 m high (9 ft high) (576 ft[3] or 16.3 m[3]);
Accounts for standard 1.83 m x 1.83 m (6 ft x 6 ft) open plan workstation system, traffic area and support space (copiers, files, storage, etc.); and
Outdoor air ventilation rate (Q) of 4.17 L/s (8.84 cfm) (minimum required ventilation rate per ASHRAE Standard 62.1-2004).

For a "typical private office environment of a single workstation system," BIFMA specifies:
Total office dimensions: floor area of 23.78 m[2] (256 ft[2]) by 2.74 m high (9 ft high) (2,304 ft[3] or 65.2 m[3]);
Accounts for standard 13.38 m[2] (144 ft[2]) private office workstation system, traffic area and support space (copiers, files, storage, etc.); and
Outdoor air ventilation rate (Q) of 9.63 L/s (20.4 cfm) (minimum required ventilation rate per ASHRAE Standard 62.1-2004).

Individual open plan and private office workstation component surface areas and outdoor air ventilation rates are tabulated in Table E-1.
Table E-1. BIFMA M7.1-2007 Standard Workstation Model Parameters
                            Workstation System Type
                 Workstation Component Type Surface Area (Ai)
                       Outdoor Air Ventilation Rate (Q)
                                       
                                  Panel Area
                               Work Surface Area
                          Storage Total External Area
                                       
                                   Open Plan
                                  11.08 m[2]
                                 (119.3 ft[2])
                                  6.103 m[2]
                                 (65.69 ft[2])
                                  4.569 m[2]
                                 (49.18 ft[2])
                                  15.0 m[3]/h
                                  (8.84 cfm)
                                Private Office
                                  7.633 m[2]
                                 (82.16 ft[2])
                                  6.734 m[2]
                                 (72.48 ft[2])
                                  10.55 m[2]
                                 (113.6 ft[2])
                                  34.7 m[3]/h
                                  (20.4 cfm)

To determine the workstation surface areas listed above in Table E-1, Carter and Zhang first determined the footprint size for all workstations present on the office building floor plans (4,594 open plan workstations and 430 private office workstations).  The 50[th] percentile workstation footprint size was taken as representative of the standard workstation size, resulting in areas of 3.34 m[2] (36 ft[2]) for an open plan workstation and 13.47 m[2] (145 ft[2]) for a private office.
Because total workstation surface areas vary within workstations of identical floor area, the 90[th] percentile conditions for total furniture surface area within the 50[th] percentile workstation footprint size were chosen to provide a representative office environment.  This resulted in 1,982 open plan workstations and 142 private offices.  Within this data set, the panel, work surface and storage surface areas were determined as a percentage of total workstation surface area and averaged, resulting in the values presented in Table E-1.
The parameter values in Table E-1 are also presented in Table 3-8 in the main body of the report.  The ventilation rate values originally presented in Carter and Zhang, 2007 were in units of L/s and cfm.  The ventilation rate values were converted to units of m[3]/h for use in the assessment method equation.
1.2.3	Formaldehyde Emissive Flux for Workstation Component Types (Ei)

The BIFMA standard and the work by Carter and Zhang do not contain any information on observed formaldehyde emission rates of actual office workstations.  Carter and Zhang provide example workstation emission rates of formaldehyde but do not discuss the statistical significance of these values such as being representative, worst case, etc.
The formaldehyde emission rate of a workstation component is a function of the type of CWP that the component is constructed of and the type of finishing, if any, for the CWP.  The finishing of CWP is the overlay of the CWP with other material that retard emissions from the surface on to which they are applied. This effect is described as barrier effectiveness, or a percent reduction of the emission rate of the CWP substrate (CPA 2003a).  The literature search did not yield any information on how the three CWPs are finished and the extent to which they are used together and in combination with other material to fabricate the various components of the workstations.  This information is needed for the estimation of representative values of the emissions rates.  Due to these data gaps, indoor air formaldehyde concentration was estimated for a range of emission rates that is derived from a what-if analysis.  For the what-if analysis, the panel and work surface components were assumed to be made entirely of a single CWP type, with the CWP type changing for each iteration of the what-if analysis.  Storage external surfaces were assumed to be constructed of materials other than CWPs with no emission of formaldehyde.  The what-if analysis follows the algorithm described in Section 3.3.1.2 and in Table 3-6 of the main body of the report.
For each iteration of the what-if analysis, the emissive flux of the workstation component type is calculated from the emissive flux (emission rate in units of flux) of the CWP type, extent of finishing of the component type, and barrier effectiveness chosen for the given iteration as follows:
			Ei=ECWPx1-BExFini+ECWPx1-Fini	Eqn (9)

where:
      Ei	= 	emissive flux of the workstation component type (ug/m[2]-h) (i = panels, work surfaces, or storage external surfaces);
      ECWP	= 	emissive flux of the composite wood product used to construct the workstation component type (ug/m[2]-h) (CWP = HWPW, PB, or MDF);
      BE	=	barrier effectiveness of the finishing, or overlay, of the workstation component type, expressed as a unitless fraction; and
      Fini	=	fraction of the workstation component surface area that is finished or overlayed with a barrier effectiveness of BE, expressed as a unitless fraction (i = panels, work surfaces, or storage external surfaces).

The values of ECWP used in Equation 9 are provided in Table 3-8 in the main body of the report, which are derived from the tables in Appendix D.  The values of BE are also provided in Table 3-8 in the main body of the report.  The barrier effectiveness values used to estimate the low-end and high-end concentrations were obtained from CPA (CPA 2007a).  The derivation of the intermediate value of barrier effectiveness is described in the following subsection (subsection 1.2.5) of this appendix.
1.2.4	Ambient Air Formaldehyde Concentration in Commercial Land Use Areas (Co[com])

This exposure assessment assumes the outdoor air used for office building ventilation contains formaldehyde at a concentration equal to the mean formaldehyde concentration in ambient air in commercial land use areas according to the EPA exposure assessment (USEPA 2009b).  Table 3-8 in Section 3.3.1.2 of the main body of the report provides the value for the concentration of formaldehyde in the outdoor air for commercial land use areas, which is 3.26 ug/m[3].
The overall equation used to calculate the formaldehyde concentration is presented in Equation 10.  Equation 10 is calculated for each iteration of the algorithm described in Section 3.3.1.2 of the main body of the report using the parameter values given in Table 3-8.  Recall that it is assumed storage external surfaces are not constructed of CWPs and do not emit formaldehyde.  Therefore, storage external surfaces are not included in Equation 10.
C=ECWPx1-BExFinpanels+ECWPx1-FinpanelsxApanels+ECWPx1-BExFinwork surfaces+ECWPx1-Finwork surfacesxAwork surfacesQ+Cocom         Eqn (10)

Note that the emission rates used in this analysis are for new composite wood products and do not account for the decay of emissions with time.  Therefore, this method to estimate formaldehyde exposures in office buildings only accounts for new office furniture and does not account for the decay of emission rates with time.  Data to describe the decay of emission rates with time were not identified.
1.2.5	Determination of Intermediate Value for Barrier Effectiveness

The ANSI / BIFMA M7.1-2007 standard test method specifies a full-scale chamber test method for determining VOC emissions of complete workstation systems, workstation components, and seating.  The determined emissions can then be used to estimate the impact of the workstation on the VOC concentration in a typical office environment.  The standard test method specifies a chamber ventilation rate of 12 to 20 cfm (22 to 36 m[3]/h) per workstation system during emissions testing.  BIFMA standard X7.1-2007 is a voluntary standard for meeting VOC emission limits from office furniture.  This standard specifies a formaldehyde limit of 50 ppb as measured using BIFMA M7.1-2007 for all workstation systems types.  BIFMA test method M7.1-2007 states that the background concentration of the compound of interest (such as formaldehyde) in the chamber should be measured prior to conducting the chamber test.  This background concentration is then subtracted from the measured concentration in the test chamber exhaust or return air.  Therefore, the resulting concentration, which is equated to the emission rate from the workstation divided by the chamber ventilation rate, is a net concentration due only to the workstation.  The 50 ppb formaldehyde limit is therefore the net concentration determined from the chamber test.
To define an intermediate case, EPA assumed a typical office workstation (both open plan and private office) meeting the BIFMA X7.1-2007 standard for formaldehyde.  EPA estimated that 100 percent of the workstation panel and 50 percent of the work surface area were coated with a barrier of unknown effectiveness.  These assumed values are logical bounding values.  Work surfaces are expected to be finished on their upper surface, and panels may be finished on both surfaces for aesthetic purposes.  EPA averaged the pre-baseline emission rates to estimate an average CWP emission rate for the workstation panel and work surface areas (see Appendix D for derivation of the average pre-baseline emission rate).  The parameters used are shown in Table E-2.
Since chamber testing requires that the net formaldehyde concentration be used to determine the emissions from the workstation, the ambient air concentration is subtracted from Equation 10 presented in the previous subsection 1.2.4.  The net formaldehyde concentration, C[net], is then used instead in the left-hand side of Equation 10.  This new equation is presented below and is solved for the intermediate barrier effectiveness using the values for concentration and each parameter given in Table E-2.
Cnet=ECWPx1-BExFinpanels+ECWPx1-FinpanelsxApanels+ECWPx1-BExFinwork surfaces+ECWPx1-Finwork surfacesxAwork surfacesQ 	Eqn (11)

Table E-2. Parameters Used to Determine Intermediate Barrier Effectiveness
                                   Parameter
                                     Value
                                    Source
Net Formaldehyde Concentration in Chamber
61.5 ug/m[3]
BIFMA X7.1-2007 standard for formaldehyde (unit converted from 50 ppb)
Panel Emission Rate
306.7 ug/m[2]-h
Weighted average of domestic and import pre-baseline emission rates, averaged over all three CWP types (see Section 3 and Appendix D)
Work Surface Emission Rate
306.7 ug/m[2]-h

Storage Panel Emission Rate 
0 ug/m[2]-h
Carter 2007a
Panel Area
11.08 m[2] (open plan)
7.633 m[2] (private office)
Carter 2007a
Percent Panel Area with Barrier
100%
EPA Assumption
Work Surface Area
6.1 m[2] (open plan)
6.734 m[2] (private office)
Carter 2007a
Percent Work Surface Area with Barrier
50%
EPA Assumption
Storage Panel Surface Area
4.57 m[2] (open plan)
10.55 m[2] (private office)
Carter 2007a
Storage Panel Surface Area with Barrier
N/A
EPA Assumption
Ventilation Rate
36 m[3]/h
The high-end ventilation rate for chamber testing specified in BIFMA M7.1-2007
N/A  -  Not applicable

The intermediate barrier effectiveness calculated from the parameters for the open plan and private office workstations were 71 percent and 65 percent, respectively.  These intermediate barrier effectiveness values were used for all intermediate indoor air concentration calculations for open plan and private office workstations following the algorithm outlined in Section 3.3.1.2 and Table 3-6 in the main body of the report.
1.3	Calculation of ADC and LADC from Model Results for CWP Fabrication and Office Buildings

Model results from Equation 5 and Equation 10 above are inputs to estimate ADC and LADC.  The average daily concentration (ADC) and lifetime-average daily concentration (LADC) are calculated from the concentration for each modeled scenario outlined in Section 3 of the main body of the report for each baseline and analytical-option scenario.  ADC and LADC are calculated from Equation 12 (EPA 2009a).
					EC=CxEDxEFxWYAT				Eqn (12)

where:
      EC	=	Exposure concentration (μg/m[3]);
      C	=	Contaminant concentration in air (μg/m[3]);
      ED	=	Exposure duration (hr/day; default: 8);
      EF	=	Exposure frequency (days/yr; default: 250);
      WY	=	Working years per lifetime (years; ADC and LADC: 40 years); and
      AT	=	Averaging time (lifetime in years x 365 days/yr x 24 hours/day; ADC: 40 years, LADC: 70 years).

1.4	Dermal Exposure Assessment Method for All Life Cycle Stages

EPA has two general categories of assessing dermal exposure when measurements are not available: (1) quantitative dermal exposure models and (2) qualitative dermal exposure assessments.  Both categories of dermal exposure assessments are discussed below.  EPA has developed a series of standard models for quantitatively estimating worker dermal exposures to liquid and solid chemicals during various types of activities.  To estimate dermal exposure, all of these dermal exposure models assume a specific surface area of the skin that is contacted by a material containing the chemical of interest, as well as a specific surface density of the material on the skin.
EPA has developed a series of standard models for quantitatively estimating worker dermal exposures to liquid and solid chemicals during various types of activities.  To estimate dermal exposure, all of these dermal exposure models assume a specific surface area of the skin that is contacted by a material containing the chemical of interest, as well as a specific surface density of the material on the skin.  The models also assume no use of controls or gloves to reduce the exposure.  These assumptions and default parameters are defined based on the nature of the exposure (e.g., one hand or two hand, immersion in material, contact with surfaces).  The standard EPA dermal model equation is shown in Equation 13 (USEPA 1991a; USEPA 2000b).
		EXPdermal=AREAsurfacexQremain_skinxFchemxNevent	Eqn (13)

where:
      EXPdermal	=	Dermal exposure to the liquid or solid chemical per day (mg chemical/worker-day);
      AREAsurface	=	Surface area of the skin that is in contact with liquid or solid material containing the chemical (840 cm[2]; EPA/OPPT 2-Hand Dermal Contact with Liquid Model);
      Qremain_skin	=	Quantity of the liquid or solid material containing the chemical that remains on the skin after contact (0.7-2.1 mg/cm[2]-event; EPA/OPPT 2-Hand Dermal Contact with Liquid Model);
      Fchem	=	Weight fraction of the chemical of interest in the material being handled in the activity; and
      Nevent	=	Frequency of events for the activity (EPA default = 1 event/worker-day).

Dermal doses in this report are estimated in units further normalized by worker body weight, as indicated in Equation 14:
			EXPdermal_bw=EXPdermalBWworker			Eqn (14)

where:
      EXPdermal_bw	=	Dermal exposure to the liquid or solid chemical per day normalized by worker body weight (mg chemical/kg-day);
      EXPdermal	=	Dermal exposure to the liquid or solid chemical per day (mg chemical/worker-day); and
      BWworker	=	Body weight of an average worker (kg; EPA default value = 70 kg (USEPA 1997a).

For several categories of exposure, EPA uses qualitative assessments to estimate dermal exposure.  Table E-4 summarizes these categories and the resulting qualitative dermal exposure assessments.  Where chemicals were monomers that had been reacted or additives that would not likely to be released from a resin matrix, dermal exposures are characterized as not quantifiable.



Table E-3. Standard EPA Default Values for Use in the Worker Dermal Exposure Models
                                 Default Model
                              Example Activities
                                AREAsurface[a]
                                    (cm[2])
                                Qremain_skin[b]
                               (mg/cm[2]-event)
                               Resulting Contact
                          AREAsurface x Qremain_skin
                                  (mg/event)
Physical Form: Liquids
  EPA/OPPT 1-Hand Dermal Contact with Liquid Model
Liquid sampling activities
Ladling liquid/bench-scale liquid transfer
                                      420
                                 (1 hand mean)
                                   Low: 0.7
                                   High: 2.1
                                   Low: 290
                                   High: 880
  EPA/OPPT 2-Hand Dermal Contact with Liquid Model
Maintenance
Manual cleaning of equipment and containers
Filling drum with liquid
Connecting transfer line
                                      840
                                 (2 hand mean)
                                   Low: 0.7
                                   High: 2.1
                                   Low: 590
                                  High: 1,800
  EPA/OPPT 2-Hand Dermal Immersion in Liquid Model
Handling wet surfaces
Spray painting
                                      840
                                 (2 hand mean)
                                   Low: 1.3
                                  High: 10.3
                                  Low: 1,100
                                  High: 8,650
Physical Form: Solids
  EPA/OPPT 2-Hand Dermal Contact with Container Surfaces Model
Handling bags of solid materials (closed or empty)
                                  No defaults
                                  No defaults
                                  < 1,100c
  EPA/OPPT 2-Hand Dermal Contact with Solids Model
Solid sampling activities
Filling/dumping containers of powders, flakes, granules
Weighing powder/scooping/mixing (i.e., dye weighing)
Cleaning solid residues from process equipment
Handling wet or dried material in a filtration and drying process
                                  No defaults
                                  No defaults
                                 < 3,100[c]
   a - These default values were adopted in the 2000 EPA report on screening-level dermal exposure estimates (USEPA 2000b) and are the mean values for men taken from the EPA Exposure Factors Handbook, 1997 (USEPA 1997a).
   b - These default values were adopted in the 2000 EPA report on screening-level dermal exposure estimates (USEPA 2000b). The report derived the selected ranges of values for liquid handling activities from: U.S. EPA. A Laboratory Method to Determine the Retention of Liquids on the Surface of Hands. U.S. Environmental Protection Agency, Office of Pollution Prevention and Toxics, Exposure Evaluation Division. EPA 747-R-92-003. September 1992.
   c - These default values were adopted in the 2000 EPA report on screening-level dermal exposure estimates (USEPA 2000b). The report derived values for dermal contact for solids handling activities from: Lansink, C.J.M., M.S.C. Breelen, J. Marquart, and J.J. van Hemmen: Skin Exposure to Calcium Carbonate in the Paint Industry. Preliminary Modeling of Skin Exposure Levels to Powders Based on Field Data (TNO Report V 96.064). Rijswijk, The Netherlands: TNO Nutrition and Food Research Institute, 1996.
   

Table E-4. EPA Default Qualitative Assessments for Screening-Level Estimates of Dermal Exposure
                                   Category
                               Dermal Assessment
Corrosive substances (pH>12, pH<2)
Negligible
Materials at temperatures >140F (60C)
Negligible
Cast Solids (e.g., molded plastic parts, extruded pellets)
Non-Quantifiable (Some surface contact may occur if manually transferred)
"Dry" surface coatings (e.g., fiber spin finishes, dried paint)
Non-Quantifiable (If manual handling is necessary and there is an indication that the material may abrade from the surface, quantify contact with fingers/palms as appropriate)
Gases/Vapors
Non-Quantifiable (Some contact may occur in the absence of protective clothing)
  Source: USEPA 1991a

3.0	Converting Units of Concentration

This section presents the methodology for converting between airborne concentrations measured in volume- or mole-based parts-per million (ppm) and airborne concentrations measured in micrograms per cubic meter (ug/m[3]).  This assessment conducts exposure and dose calculations using concentrations in mg/m[3] or ug/m[3] depending on the magnitude of the values.  This section presents concentration units in ug/m[3].
To convert the units of concentration between a volume- or mole-based ppm to ug/m[3] at ambient room conditions, it is assumed that the ideal gas law applies to a mixture of formaldehyde and air at ambient conditions.  The mass-based concentration of formaldehyde in air from the ideal gas law is solved for as follows:
	C=mV=yPMRTx1,000,000μggx1,000molkmol	Eqn (15)

where:
      C	=	formaldehyde concentration (ug/m[3]);
      m	=	total mass of formaldehyde (ug);
      V	=	total volume of gas (m[3]);
      y	=	mole fraction of formaldehyde (mol/mol);
      P	=	total pressure (atm);
      M	=	molecular weight of formaldehyde (g/mol);
      R	=	universal gas constant (m[3]-atm/kmol-K); and
      T	=	temperature (K).

Here, the mole fraction of formaldehyde, y, is equal to the formaldehyde concentration in parts-per million (ppm) divided by one million.  At ambient conditions (1 atm and 298 K), with a formaldehyde molecular weight of 30.03 g/mol and a gas constant of 0.082 m[3]-atm/kmol-K, the unit conversion is 1,229 ug/m[3] per ppm of formaldehyde.
4.0	ASTM E1333 Large Chamber Test Method

In this assessment, an equation is needed to calculate emission rates from the methods used to measure them.  In this assessment, the ASTM E1333 large chamber test method is used to convert between chamber-test measured formaldehyde concentrations (emission levels) and emission rates.  The equation used is Equation 15 below.  This subsection presents some general background information on chamber test methods and then describes the ASTM E1333 large chamber test method.
The standard method for measuring emissions from composite wood products is to use a test chamber (Kim 2007a).  The chamber method is based on a mass balance approach that measures the mass flow rates of formaldehyde entering and leaving the chamber system at steady-state (Hodgson 2007a).  Standard chamber methods are either referred to as the large or small chamber method based on the specified chamber size.  In general, common large chamber methods for estimating formaldehyde emissions require minimum chamber sizes between 10 to 22 cubic meters.  In addition to chamber size, large chamber methods specify several test parameters, including the loading ratio, air exchange rate, chamber temperature and humidity.  This section discusses the ASTM E1333 large chamber test method.
The ASTM E1333 has been used by the composite wood product industry for establishing compliance of their products with the U.S. Department of Housing and Urban Development (HUD) and ANSI standards for formaldehyde emissions (Hodgson 2007a).  It is also the reference test method specified in CARB's regulation.  The purpose of ASTM E1333 is to measure the formaldehyde concentration in air and the emission rate from a number of large test samples (composite wood products) containing formaldehyde under conditions designed to simulate product use.  ASTM E1333 is primarily used for testing newly manufactured panel products (ASTM 2002a).  Prior to testing, this method requires test samples to be conditioned for seven days.  During conditioning, the samples must be placed in a manner such that there is a minimum of six-inch spacing in-between the panels.  The conditioning is followed by a 16- to 20- hour exposure period in a large-scale chamber.  The chamber is operated at specific product loading ratios and an air exchange rate that are typical of the indoor environment.  Air samples are taken from the chamber at steady-state conditions and the formaldehyde concentration is analyzed.  Table E-5 lists the product-specific loading ratios for the composite wood products of interest.  Table E-6 lists the test parameters for ASTM E1333.
Table E-5.  ASTM E1333 Loading Ratios for Hardwood Plywood, Medium-Density Fiberboard, and Particleboard
                            Composite Wood Product
                               Loading Ratio (L)

                                  ft[2]/ft[3]
                                   m[2]/m[3]
Hardwood plywood
                                     0.13
                                     0.43
Medium-density fiberboard
                                     0.08
                                     0.26
Particleboard
                                     0.13
                                     0.43

Table E-6.  Specified Parameters for ASTM E1333
                                Test Parameter
                               ASTM E1333 Value
Loading Ratio (m[2]/m[3]) (L)
                                 See Table E-5
Minimum Chamber Size (m[3])
                                     22.7
Temperature (°C)
                                    25 +-1
Relative Humidity (%)
                                    50 +-4
Air Exchange Rate (h[-1]) (n)
                                  0.5 +-0.05
Edge Sealing
                                     None
Conditioning
                                7 days +-3 hrs
Test Duration
                                 16 to 20 hrs

The emission rate of formaldehyde is dependent on several parameters.  At steady-state conditions, the emission rate (ER) or area-specific emission rate of formaldehyde in ug/m[2]-h is calculated as:
				ER=VAxnxCnet				Eqn (16)

where:
      V	=	chamber volume (m[3]);
      n	=	chamber air exchange rate (h[-1]);
      C[net]	=	net formaldehyde chamber concentration (ug/m[3]); and
      A	=	total surface area of sample (m[2]).

The emission rate represents the flux of formaldehyde or the mass of formaldehyde emitted per unit area of wood material per time (ug/m[2]-h).  As seen in Equation 16, the emission rate is a function of the chamber volume, V, and test sample surface area, A.  The ratio of these parameters is often specified in a large chamber method in terms of a "loading ratio."  Loading ratios (L) are calculated as the emitting surface area of the wood sample divided by the chamber volume (m[2]/m[3]):
					L=AV,   or    A=LxV			Eqn (17)

Substituting Equation 17 into Equation 16 yields Equation 18, which is used to calculate emission rates from the measured net formaldehyde chamber concentration (emission level) using the parameters given in Table E-5 and Table E-6.
					ER=nxCnetL					Eqn (18)

In chamber methods, air is continuously circulated inside the test environment.  The area-specific air flow rate (m3/m2-h or m/h) is the ratio between the air flow rate Q (m3/h) entering the chamber and the total surface area, A, of the test sample.  The area-specific air flow rate is also calculated as the product of the chamber volume, V, and the air exchange rate, n, divided by the test sample surface area as follows.
					QA=VxnA					Eqn (19)

Substituting Equation 17 into 19, the area-specific air flow rate may be expressed as a quotient of n/L.
					QA=VxnLxV=nL	 			Eqn (20)

After steady-state conditions are achieved, a minimum of two air samples are taken from the chamber at a rate of 1+-0.05 L/min for 60 minutes +-5 seconds.  The quantity of formaldehyde in the sample is determined by an adaptation of the National Institute for Occupational Safety and Health (NIOSH) 3500 chromotropic acid test procedure (ASTM 2002a), where air is bubbled through a 1-percent sodium bisulfite solution and analyzed using a UV-Vis spectrophotometer.



Appendix F
PRE-BASELINE FORMALDEHYDE EXPOSURES FROM OSHA IMIS BY WORKER ACTIVITY AND INDUSTRY SECTOR FOR COMPOSITE WOOD PRODUCT FABRICATORS
                            BACKGROUND ON IMIS DATA

IMIS is designed to collect, process, retrieve, and communicate penalty assessment, arbitration, and collection information regarding OSHA inspections.  For each data point, the database contained information on the following data elements:
Inspection date;
Identity of inspected facility;
Inspection type (e.g., inspection in response to a complaint, referral, or as part of a program);
Worker job description;
Exposure concentration;
Exposure concentration measurement type (personal or area monitoring);
Exposure type (i.e., time weighted average (TWA) or peak exposure);
Frequency or duration of exposure;
Number of employees at the inspected facility;
Number of employees exposed; and
Applicable OSHA PEL.
OSHA IMIS data for most of the NAICS codes given in Table 2-1 were obtained from OSHA in June 2010.  The OSHA occupational exposure limits for formaldehyde are given in Appendix C.
The "exposure types" for the obtained data include TWA, STEL, ceiling, peak exposures or not detected or not found.  Only TWA data were used to develop exposure estimates (in terms of average daily concentrations (ADCs) and lifetime-average daily concentrations (LADCs)).  STEL, ceiling, and peak measurements are not representative of daily average concentration levels and were therefore excluded (refer to Appendix A for a definition of these terms).  Also, measurements that resulted in a non-detected formaldehyde concentration do not include their measurement type in the OSHA IMIS database (i.e., TWA or STEL measurement type).  Therefore, it is unknown whether a non-detected value was the result of a measurement over an extended duration of time for the purpose of determining TWA exposure or measurement over a fifteen-minute duration.  For this reason, non-detected and not-found results were excluded except as discussed below.
Additionally, only personal monitoring data were included in the assessment except in the case of the retail life cycle stage.  IMIS contains few personal monitoring data from 2002 and onward for this life cycle stage, and the inclusion of area monitoring data approximately doubled the number of included data points.  Most of these area monitoring data have non-detect values.  Area monitoring data in a strictly retail setting are considered representative of potential worker exposure and are expected to be TWA data.  The non-detected value was set equal to the lowest reliable quantitation limit (RQL) identified among OSHA's sampling and analysis methods for formaldehyde.  The RQL was identified as 0.00058 ppm (OSHA 2010b).  Also, any detected values for all life cycle stages less than this RQL were set equal to the RQL.

Grouping Data by Worker Activity

This appendix presents a summary of OSHA IMIS data for CWP fabrication industry sectors grouped by their worker activity.  This grouping was performed as part of this exposure assessment to provide perspective as to the different worker activities monitored by OSHA and to present statistics of the monitoring data according to worker activity.  There is some uncertainty about the worker activities recorded in IMIS.  The worker activities were not recorded or identified in a consistent manner in IMIS.  For example, for CWP fabrication, gluing activities are identified as "apply wood glue" and "glue machine operator".  Therefore, this appendix presents OSHA IMIS data grouped such that similar worker activities are grouped into a single worker activity category.  Additionally, some worker activities are similar or can be expected to result in similar exposures, such as "sawing", "routing", and "cutting", and are compiled into a single category.  Other worker activities are compiled into a single category due to uncertainty in the exact job description, such as "wood worker", "laborer", or "operator".  Table F-1 presents a summary of the TWA, personal monitoring data from OSHA IMIS for CWP fabrication organized by worker activity description categories.



Table F-1.  Pre-Baseline Formaldehyde Exposures by Organized by Worker Activity Category for Composite Wood Product Fabricators from OSHA IMIS
                                   Industry
                          Worker Activity Description
                           Number of Sites Monitored
                             Number of Data Points
                  Monitored Formaldehyde Concentration (ppm)
                Monitored Formaldehyde Concentration (mg/m[3])
                                     Year
                      Description of Concentration Level
Wood Kitchen Cabinets and Countertop Manufacturing
Painting / Laminating / Finishing / Spraying
                                      14
                                      22
                                     0.69
                                     0.85
                                     2003
Maximum




                                     0.15
                                     0.18
                                  2004, 2007
Median




                                    0.0006
                                     0.001
                                     2004
Minimum

Fabricating / Assembling / Building
                                       1
                                       1
                                      0.2
                                     0.25
                                     2004
1 Data Point

Other Operations / Maintenance
                                       1
                                       2
                                     0.52
                                     0.639
                                     2004
Maximum




                                     0.28
                                     0.343
                                  2004, 2004
Median




                                     0.038
                                     0.047
                                     2004
Minimum

Sawing / Cutting / Routing
                                       1
                                       2
                                     0.070
                                     0.086
                                     2005
Maximum




                                     0.054
                                     0.066
                                  2005, 2005
Median




                                     0.037
                                     0.046
                                     2005
Minimum

Sanding
                                       1
                                       1
                                     0.035
                                     0.043
                                     2003
1 Data Point

Compliance, Safety, and Health Officer
                                       1
                                       1
                                     0.025
                                     0.031
                                     2006
1 Data Point
Upholstered Household Furniture Manufacturing
Painting / Laminating / Finishing / Spraying
                                       1
                                       2
                                     0.092
                                     0.11
                                     2004
Maximum




                                     0.06
                                     0.074
                                  2004, 2004
Median




                                     0.028
                                     0.034
                                     2004
Minimum

Other Operations / Maintenance
                                       2
                                       3
                                     0.11
                                     0.137
                                     2003
Maximum




                                     0.04
                                     0.049
                                     2004
Median




                                     0.04
                                     0.049
                                     2004
Minimum

Gluing
                                       1
                                       1
                                     0.009
                                     0.011
                                     2003
1 Data Point
Nonupholstered Household Furniture Manufacturing
Painting / Laminating / Finishing / Spraying
                                       8
                                      10
                                     0.59
                                     0.73
                                     2004
Maximum




                                     0.330
                                     0.405
                                  2008, 2007
Median




                                     0.031
                                     0.038
                                     2008
Minimum

Fabricating / Assembling / Building
                                       1
                                       1
                                     0.22
                                     0.27
                                     2007
1 Data Point

Other Operations / Maintenance
                                       1
                                       5
                                    0.1129
                                     0.139
                                     2009
Maximum




                                    0.0625
                                     0.077
                                     2009
Median




                                    0.0278
                                     0.034
                                     2009
Minimum

Compliance, Safety, and Health Officer
                                       3
                                       3
                                     0.14
                                     0.172
                                     2008
Maximum




                                     0.089
                                     0.109
                                     2009
Median




                                    0.0712
                                     0.088
                                     2008
Minimum
Wood Office Furniture Manufacturing
Painting / Laminating / Finishing / Spraying
                                       4
                                       8
                                     0.58
                                     0.713
                                     2004
Maximum




                                     0.055
                                     0.068
                                  2004, 2005
Median




                                     0.009
                                     0.011
                                     2004
Minimum

Fabricating / Assembling / Building / Laboring
                                       1
                                       1
                                     0.02
                                     0.025
                                     2004
1 Data Point

Other Operations / Maintenance
                                       1
                                       2
                                     0.04
                                     0.049
                                     2005
Maximum




                                    0.03135
                                     0.039
                                  2005, 2005
Median




                                    0.0227
                                     0.028
                                     2005
Minimum

Sawing / Cutting / Routing
                                       2
                                       4
                                     0.067
                                     0.082
                                     2004
Maximum




                                     0.049
                                     0.060
                                  2004, 2004
Median




                                    0.0048
                                     0.006
                                     2004
Minimum

Sanding
                                       1
                                       3
                                     0.028
                                     0.034
                                     2005
Maximum




                                     0.027
                                     0.033
                                     2005
Median




                                    0.0219
                                     0.027
                                     2005
Minimum
Institutional Furniture Manufacturing
Painting / Laminating / Finishing / Spraying
                                       3
                                       3
                                     0.21
                                     0.258
                                     2005
Maximum




                                     0.06
                                     0.074
                                     2007
Median




                                    0.0189
                                     0.023
                                     2002
Minimum

Other Operations / Maintenance
                                       2
                                       3
                                     0.10
                                     0.123
                                     2007
Maximum




                                     0.06
                                     0.074
                                     2007
Median




                                     0.01
                                     0.012
                                     2002
Minimum
Showcase, Partition, Shelving, and Locker Manufacturing
Painting / Laminating / Finishing / Spraying
                                       2
                                       3
                                     0.228
                                     0.280
                                     2004
Maximum




                                     0.216
                                     0.265
                                     2004
Median




                                     0.13
                                     0.160
                                     2005
Minimum

Press Operating
                                       1
                                       2
                                     0.187
                                     0.230
                                     2004
Maximum




                                     0.138
                                     0.169
                                  2004, 2004
Median




                                     0.088
                                     0.108
                                     2004
Minimum

Compliance, Safety, and Health Officer
                                       1
                                       1
                                     0.184
                                     0.226
                                     2004
1 Data Point
Custom Architectural Woodwork and Millwork Manufacturing
Other Operations / Maintenance
                                       1
                                       3
                                       1
                                     1.229
                                     2008
Maximum




                                     0.44
                                     0.541
                                     2008
Median




                                     0.04
                                     0.049
                                     2008
Minimum

Sawing / Cutting / Routing
                                       1
                                       1
                                     0.068
                                     0.084
                                     2004
1 Data Point
Other Millwork (including flooring) Manufacturing
Painting / Laminating / Finishing / Spraying
                                       1
                                       1
                                     0.13
                                     0.160
                                     2003
1 Data Point

Fabricating / Assembling / Building / Laboring
                                       1
                                       2
                                     0.075
                                     0.092
                                     2002
Maximum




                                    0.03805
                                     0.047
                                  2002, 2002
Median




                                    0.0011
                                     0.001
                                     2002
Minimum

Other Operations / Maintenance
                                       2
                                       3
                                     0.243
                                     0.299
                                     2006
Maximum




                                     0.184
                                     0.226
                                     2006
Median




                                    0.0992
                                     0.122
                                     2002
Minimum

Sawing / Cutting / Routing
                                       1
                                       1
                                     0.05
                                     0.061
                                     2004
1 Data Point

Press Operating
                                       2
                                       3
                                     0.55
                                     0.676
                                     2008
Maximum




                                     0.317
                                     0.390
                                     2008
Median




                                     0.06
                                     0.074
                                     2004
Minimum
Wood Window and Door Manufacturing
Painting / Laminating / Finishing / Spraying
                                       1
                                       1
                                    0.00438
                                     0.005
                                     2007
1 Data Point

Fabricating / Assembling / Building / Laboring
                                       2
                                       2
                                     0.24
                                     0.295
                                     2006
Maximum




                                    0.1465
                                     0.180
                                  2006, 2009
Median




                                     0.053
                                     0.065
                                     2009
Minimum
Motor Home Manufacturing
Trailer Cleaning
                                       1
                                       1
                                    0.0023
                                     0.003
                                     2002
1 Data Point
All Other Miscellaneous Wood Product Manufacturing
All Activities
                                       8
                                      18
                                     1.00
                                     1.229
                                     2005
Maximum




                                     0.144
                                     0.177
                                  2005, 2006
Median




                                     0.003
                                     0.003
                                     2003
Minimum


Table F-2 through Table F-12 present summary statistics of the TWA, personal monitoring data from OSHA IMIS by NAICS code for the CWP fabrication life cycle stage.  Not all NAICS codes within CWP fabrication are included in this summary; only NAICS codes for which data were obtained from OSHA IMIS.

Table F-2.  Summary Statistics for Pre-Baseline Monitoring Data for NAICS Code 321911
                                       
                     Monitored Formaldehyde Concentration
                           Number of Monitored Sites
                             Number of Data Points
                                     Year

                                     (ppm)
                                  (ug/m[3])



Maximum
                                     0.24
                                     295.0
                                       3
                                       3
                                     2006
Median
                                     0.053
                                     65.1


                                     2009
Minimum
                                    0.0044
                                      5.4


                                     2007
Mean
                                     0.099
                                     121.8


                                      N/A
Standard deviation
                                     0.124
                                     152.9


                                      N/A


Table F-3.  Summary Statistics for Pre-Baseline Monitoring Data for NAICS Code 321918
                                       
                     Monitored Formaldehyde Concentration
                           Number of Monitored Sites
                             Number of Data Points
                                     Year

                                     (ppm)
                                  (ug/m[3])



Maximum
                                     0.55
                                     676.0
                                       6
                                      10
                                     2008
Median
                                    0.1146
                                     140.8


                                      N/A
Minimum
                                    0.0011
                                      1.4


                                     2002
Mean
                                    0.17093
                                     210.1


                                      N/A
Standard deviation
                                     0.164
                                     201.6


                                      N/A

Table F-4.  Summary Statistics for Pre-Baseline Monitoring Data for NAICS Code 337110
                                       
                     Monitored Formaldehyde Concentration
                           Number of Monitored Sites
                             Number of Data Points
                                     Year

                                     (ppm)
                                  (ug/m[3])



Maximum
                                     0.69
                                     848.0
                                      16
                                      29
                                     2003
Median
                                     0.137
                                     168.4


                                     2004
Minimum
                                    0.0006
                                      0.7


                                     2004
Mean
                                     0.197
                                     242.6


                                      N/A
Standard deviation
                                     0.204
                                     250.9


                                      N/A

Table F-5.  Summary Statistics for Pre-Baseline Monitoring Data for NAICS Code 337121
                                       
                     Monitored Formaldehyde Concentration
                           Number of Monitored Sites
                             Number of Data Points
                                     Year

                                     (ppm)
                                  (ug/m[3])



Maximum
                                     0.111
                                     136.6
                                       3
                                       6
                                     2003
Median
                                     0.040
                                     49.2


                                      N/A
Minimum
                                     0.009
                                     11.1


                                     2003
Mean
                                     0.053
                                     65.6


                                      N/A
Standard deviation
                                     0.039
                                     48.5


                                      N/A


Table F-6.  Summary Statistics for Pre-Baseline Monitoring Data for NAICS Code 337122
                                       
                     Monitored Formaldehyde Concentration
                           Number of Monitored Sites
                             Number of Data Points
                                     Year

                                     (ppm)
                                  (ug/m[3])



Maximum
                                     0.590
                                     725.1
                                      11
                                      19
                                     2004
Median
                                     0.170
                                     208.9


                                     2008
Minimum
                                     0.028
                                     34.2


                                     2009
Mean
                                     0.222
                                     273.4


                                      N/A
Standard deviation
                                     0.182
                                     223.7


                                      N/A

Table F-7.  Summary Statistics for Pre-Baseline Monitoring Data for NAICS Code 337127
                                       
                     Monitored Formaldehyde Concentration
                           Number of Monitored Sites
                             Number of Data Points
                                     Year

                                     (ppm)
                                  (ug/m[3])



Maximum
                                     0.210
                                     258.1
                                       3
                                       6
                                     2005
Median
                                     0.060
                                     73.7


                                      N/A
Minimum
                                     0.010
                                     12.1


                                     2002
Mean
                                     0.076
                                     94.0


                                      N/A
Standard deviation
                                     0.073
                                     89.8


                                      N/A


Table F-8.  Summary Statistics for Pre-Baseline Monitoring Data for NAICS Code 337211
                                       
                     Monitored Formaldehyde Concentration
                           Number of Monitored Sites
                             Number of Data Points
                                     Year

                                     (ppm)
                                  (ug/m[3])



Maximum
                                     0.580
                                     712.8
                                       6
                                      18
                                     2004
Median
                                     0.029
                                     35.6


                                      N/A
Minimum
                                     0.005
                                      5.9


                                     2004
Mean
                                     0.082
                                     100.9


                                      N/A
Standard deviation
                                     0.140
                                     171.5


                                      N/A

Table F-9.  Summary Statistics for Pre-Baseline Monitoring Data for NAICS Code 337212
                                       
                     Monitored Formaldehyde Concentration
                           Number of Monitored Sites
                             Number of Data Points
                                     Year

                                     (ppm)
                                  (ug/m[3])



Maximum
                                     1.00
                                     1229
                                       2
                                       4
                                     2008
Median
                                     0.25
                                     312.2


                                      N/A
Minimum
                                     0.04
                                     49.2


                                     2008
Mean
                                     0.39
                                     475.6


                                      N/A
Standard deviation
                                     0.45
                                     550.0


                                      N/A


Table F-10.  Summary Statistics for Pre-Baseline Monitoring Data for NAICS Code 337215
                                       
                     Monitored Formaldehyde Concentration
                           Number of Monitored Sites
                             Number of Data Points
                                     Year

                                     (ppm)
                                  (ug/m[3])



Maximum
                                     0.23
                                      280
                                       3
                                       6
                                     2004
Median
                                     0.19
                                     228.0


                                      N/A
Minimum
                                     0.09
                                     108.2


                                     2004
Mean
                                     0.17
                                     211.6


                                      N/A
Standard deviation
                                     0.05
                                     65.6


                                      N/A

Table F-11.  Summary Statistics for Pre-Baseline Monitoring Data for NAICS Code 336213
                                       
                     Monitored Formaldehyde Concentration
                           Number of Monitored Sites
                             Number of Data Points
                                     Year

                                     (ppm)
                                  (ug/m[3])



Single Data Point
                                    0.0023
                                      2.8
                                       1
                                       1
                                     2002

Table F-12.  Summary Statistics for Pre-Baseline Monitoring Data for NAICS Code 321999
                                       
                     Monitored Formaldehyde Concentration
                           Number of Monitored Sites
                             Number of Data Points
                                     Year

                                     (ppm)
                                  (ug/m[3])



Maximum
                                     1.00
                                     1229
                                       8
                                      18
                                     2005
Median
                                     0.14
                                     177.0


                                      N/A
Minimum
                                     0.003
                                      3.4


                                     2003
Mean
                                     0.20
                                     249.6


                                      N/A
Standard deviation
                                     0.26
                                     313.6


                                      N/A



Appendix G
SUMMARY OF OSHA INSPECTION REPORTS
ONE DATA GAP OF THE EXPOSURE ASSESSMENT IS THE USE OF PERSONAL PROTECTIVE EQUIPMENT (PPE) in manufacturing facilities.  To help resolve this data gap, OSHA inspection reports associated with the obtained OSHA IMIS data were obtained from OSHA.  This appendix summarizes the PPE information found in the reports received as of January 2011.
Table G-1 and the following sections present the findings of the review of the inspection reports received.
Table G-1.  Summary of OSHA Inspection Reports and Findings on PPE Use
                          Name of Inspected Facility
     Personal Protective Equipment (PPE) Worn by Workers During Inspection
                       Respirator Use During Inspection
Manufacturing
Carolina Curves
Inspection No. 303713325
Ear plugs, safety glasses/goggles, aprons
No
McKnight Plywood
Inspection No. 123445595
Safety glasses, rubber gloves, back braces
No
Marshfield Door Systems
Inspection No. 303864029
Safety glasses, gloves
No
Fiberesin Industries
Inspection No. 307061689
Safety glasses, gloves, eye protection
No
Unilin Flooring NC, LLC
Inspection No. 312479835
3M 6300 half mask with formaldehyde cartridges
Yes
Composite Wood Product Fabrication
Plymart Inc.
Inspection No. 308245372
Unknown
No
Bushline Inc.
Inspection No. 306132275
Half face air purifying respirator
Yes
England Inc.
Inspection No. 307674051
Safety glasses, smock, hearing protection
No
Commercial Furniture Group
Inspection No. 311144059
Safety glasses and hearing protection
No
Harris Tarkett
Inspection No. 307206987
Safety glasses, gloves, and hearing protection
No
Gunlocke Company
Inspection No. 307690578
Unknown
No
Armstrong Cabinets
Inspection No. 306021239
Gloves, apron
No
Wood Goods Industries, Inc.
Inspection No. 307041707
Respirator with organic vapor cartridge, nitrile gloves, safety glasses, hard hat, ear plugs, and steel toe shoes
Yes (not required)
Parenti & Rafaelli, Ltd
Inspection No. 305228652
Safety glasses, steel toe shoes, gloves, and half-face respirators. Dust masks are also available for use. 
Yes
Benvenuti and Stein
Inspection No. 312188972
Half-face cartridge respirator, safety glasses, safety shoes
Yes
Ameriwood Industries Inc.
Inspection No. 311606883
Safety glasses, ear plugs
No
Tru-Wood Cabinets, Inc.
Inspection No. 307013250
Safety glasses, earplugs
No
Jimson Manufacturing Co.
Inspection No. 303424014
Safety glasses
No
Newood Display Fixtures Manufacturing Company
Inspection No. 304064835
Respirator
Yes
Simple Designs Manufacturing
Inspection No. 306629171
Safety glasses, gloves, ear protection; dust masks are "available for use"
No

1.0	Manufacturing

Carolina Curves Inc., Inspection 303713325
The inspection occurred as a result of referral. Carolina is primarily engaged in the manufacture of curved plywood products for the furniture industry (e.g., chairs, bed posts, and other furniture items). These operations include receiving plywood, laminated veneer using an amino resin containing formaldehyde (0.1  -  0.5%), mold pressing of veneer (using pressure/Rf to curve plywood), wood cutting, sanding, shaping operations. Inspection of the curved wood products is the final phase of the curved plywood operation prior to shipping. According to the air sampling reports, workers may wear ear plugs, safety glasses/goggles and aprons as PPE. These reports did not indicate the use of a respirator during any of the inspected activity.
McKnight Plywood, Inspection 123445595
This inspection was a planned health inspection. McKnight Plywood makes laminated hardwood plywood products such as hardwood flooring using purchased veneers and formaldehyde-based glues (0.21% formaldehyde in the glue used in the old building and 0.32% in the glue used in the flooring plant). Some of the operations at McKnight Plywood include plywood sorting, glue application, pressing, cooling, cutting and sanding the finished products.
OSHA cited the facility for formaldehyde overexposure. However, based on the inspection summary and the air sampling reports, the only PPE worn at the facility are safety glasses, rubber gloves, and back braces. These PPE are only worn in some areas of the facility.
Marshfield Door Systems, Inc. Inspection 303864029
The inspection occurred as a result of a complaint. Workers at the inspected facility are potentially exposed to formaldehyde in the smoke emitted from the electronic press. Air sampling reports from the inspection indicated the use of PPE such as safety glasses and gloves by some workers. Respirators were not worn.
Fiberesin Industries, Inc., Inspection 307061689
The inspection occurred as a result of a complaint. Fibersin Industries manufactures particleboard. Air monitoring during the inspection indicated no overexposure to formaldehyde at the inspected facility. Because no overexposures were found, it is not mandatory for the employer to provide workers with respiratory protection. Eye protection is provided to workers, but the wearing of eye protection is not enforced. PPE for the press operator include the occasional use of gloves and the use of safety glasses and hearing protection.
Unilin Flooring NC, LLC., Inspection No. 312479835
OSHA conducted a partial inspection of the facility as a result of a non-formal complaint. The inspection did not cover all operations at the facility and OSHA did not identify any violation of OSHA standards or regulations. The inspected facility manufactures high density MDF used primarily in flooring material. The company provides respiratory protection training and fit testing for workers in the production area.  Workers were wearing 3M 6300 half masks with formaldehyde cartridges; they also had the option of using full face cartridge respirators and had been fit tested of both types of respirators. The company also conducts periodic formaldehyde sampling and has performed a ventilation study on-site.
2.0	Composite Wood Product Fabrication

Plymart Inc. Inspection No. 308245372
Plymart is a high production cabinet shop with a series of large warehouse size buildings. The cabinets are mainly constructed of particleboard frames with hardwood doors. OSHA conducted a combined safety and health programmed inspection at the facility and cited Plymart for poor ventilation and housekeeping in the spray finish booth. However, air sampling throughout the facility indicated that concentrations of formaldehyde, as well as dust and solvent were all below the PEL and that there was no overexposure. Hence, the company had "decided not to require tight fitting respirators."
Bushline Inc. Inspection No. 306132275 
This inspection was a general schedule comprehensive occupational health inspection. Bushline manufactures upholstered furniture. During the OSHA inspection, the facility was cited for poor ventilation in the spray booth, the lack of PPE hazard assessment or employee training certification, lack of impervious gloves and clothing for workers handling corrosive liquids, as well as eye and face protection for employees potentially exposed to hazards from acids or caustic chemicals.
Formaldehyde glue is mixed and applied by hand to glue wood sections together in the rough mill area. The report stated that "proper PPE and an eyewash and shower were not provided in this area" and that formaldehyde sampling results were below the action level.
An employee wore the Cabot Safety half face air purifying respirator while mixing the adhesive in the chair line area, where wooden frame chairs are assembled using pneumatic guns and a formaldehyde glue. Past air sampling indicated that the formaldehyde concentration was below the PEL in this area.
England Inc. Inspection No. 307674051
This inspection was a general schedule comprehensive occupational health inspection. England Inc. manufactures furniture cushions and mattresses. During the inspection, the facility was cited for not providing the proper PPE where there is potential for exposure. The only worker PPE (worn in select areas of the facility) are safety glasses, smock, and hearing protection. It appears that no respirators were worn. The facility uses a water-based non-toxic glue in its processes, and sampling results show formaldehyde exposure below the OSHA PEL.
Harris-Tarkett Inc. Inspection No. 307206987
This inspection was a general schedule comprehensive occupational health inspection. During the inspection, workers were observed to wear a combination of safety glasses, gloves, and hearing protection in the wood processing area, and in the long strip area, where wood products are being sorted, cut, glued, pressed and finished. It appears that no respirator was worn in the areas with potential formaldehyde exposure. Respirators were also not worn by workers in the chemical room of the finishing department where they were exposed to methyl isobutyl ketone above the PEL and STEL levels.
The Commercial Furniture Group. Inspection No. 311144059
This inspection was a general schedule comprehensive occupational health inspection. The Commercial Furniture Group manufactures wooden chairs at the inspected facility. The operation here consists of a rough mill, a frame assembly area, a spray finishing area, a fabric cutting and storage area, and an upholstery area. The inspection report did not indicate the use of formaldehyde-based adhesives although formaldehyde could be emitted from the wood boards used to manufacture chairs. According to the report, workers wear safety glasses and hearing protection in certain areas of the facility. There was no indication for respiratory use. Air sampling results also indicate the formaldehyde concentration was below the OSHA PEL, therefore, the facility would not be required to provide respirators for formaldehyde exposures.
Gunlocke Company, LLC. Inspection No. 307690578
The inspection at Gunlocke Company was conducted as a result of a compliant. During the inspection, the observed workers were not wearing respirators.
Armstrong Cabinets Products. Inspection No. 306021239
This inspection was a planned health inspection. Armstrong Cabinets manufactures bath and kitchen wood cabinets. OSHA cited the company for not providing protective eye and face equipment to prevent chemical splashes. The facility has a written respiratory protection program but the use of a respirator is not mandatory. During the inspection, workers were not wearing respirators or dust masks even though the facility has seven spray booths. Occasional PPE worn include apron and gloves.
Wood Goods Industry Inc. Inspection No. 307041707
The inspection was conducted as a result of a complaint. The inspection summary suggests that this facility produced various wood products, such as tables, from rough lumber. Lumber was rip-sawed, chopped, and glued with non-hazardous wood glue (does not contain formaldehyde). The facility also had a spray operation where three chemical sprays, a water based stain, a vinyl sealer, and a clear top coat were sprayed onto the wood products. Employees rotated between spraying operations and hand sanding, wiping stains, and setting tables on the conveyor tracks. Air sampling for formaldehyde was conducted on an employee spraying the top coat. The other two coatings do not appear to contain formaldehyde.
The facility had an operational spray booth in place but did not require respirator use. During air sampling of spraying activities, only some workers were observed to wear a respirator with organic vapor cartridges.  Other PPE worn by workers during air sampling include nitrile gloves, safety glasses, hard hat, ear plugs, and steel toe shoes.
Parenti & Rafaelli, Ltd. Inspection No. 305228652
The inspection was conducted under the Site Specific Targeting program. A safety inspection had been initiated previously and the health inspection was initiated to address health issues such as spray booths, exposure to wood dust and formaldehyde. The company had no previous history of OSHA inspections.
The company required the use of half face respirators in the spray booth and dust masks are available to the rest of the employees. The company also required employees to wear safety glasses, steel toe shoes, gloves, and respirators in the spray booths.
The company made custom millwork and woodwork for commercial and residential sites. The types of wood used vary depending on customer specification but includes fir, pine, oak, maple, mahogany, cherry and particleboard. Air sampling for formaldehyde was below the PEL.
Benvenuti and Stein, Inc. Inspection No. 312188972
The health inspection was initiated due to a non-formal complaint regarding a chemical spill. The company made custom design wooden cabinets and the facility was divided into cabinet making, cabinet finishing, and the dock. The finish coating used during the spray operation contained formaldehyde.
The company provided a half-face cartridge respirator to reduce employee exposure to formaldehyde but did not implement a written respirator program. Other PPE worn by workers include safety glasses and safety shoes.
Ameriwood Industries Inc. Inspection No. 311606883
OSHA conducted this partial health inspection as follow-up to a previous inspection that was conducted in response to a complaint. OSHA determined that the formaldehyde exposure for workers operating the autodrill was well under the PEL based on personal monitoring. Workers only wore safety glasses and ear plugs during air sampling.
Tru-Wood Cabinets, Inc. Inspection No. 307013250
OSHA conducted this partial health inspection in response to a complaint. Tru-Wood Cabinets, Inc. manufactures wood cabinets (kitchen and bathroom) to be installed on new homes. At the time of inspection, the company employed 279 employees in two working shifts. The company had two buildings: frame door production and spray painting operations were conducted in building #1, while panel cutting and finishing occurred in building #2.
Monitoring data show exposure for formaldehyde was below the OSHA PEL. The only PPE listed in OSHA's air sampling worksheets were safety glasses and ear plugs. The inspection report indicates voluntary use of particulate masks by workers in building #2; however, it was uncertain how frequent these respirators were being worn.
Jimson Manufacturing Co. Inspection No. 303424014
OSHA conducted this partial health inspection in response to a complaint. This company manufactures furniture for retail stores. Sheets of MDF are received at the company (facility) and the wood is cut to size using panel saws.  Foil is applied to edges and flat pieces of wood via heat and pressure; and furniture is assembled with staple guns.
OSHA monitoring indicates that the formaldehyde exposure level was well below the OSHA PEL. The PPE listed in OSHA's air sampling worksheets included prescription glasses and safety glasses (with or without side shields). Dust masks and respirators were not worn.
Newood Display Fixtures Manufacturing Company. Inspection No. 304064835
Newood Display Fixtures Manufacturing is a large store fixture manufacturer with a staff of 60 employees. The health inspection was initiated based on an anonymous complaint alleging exposure to lacquer, paints, and formaldehyde in the west end of the mill. Some of the operations that occur at this facility include spray painting operation, glue-up, and packing. The spray area consists of three spray rooms and a spray painting prep area. The lacquer pump is electrically interconnected to the ventilation system, so that the spray guns will not operate if the ventilation system is inoperable. Sampling results indicated that formaldehyde TWA reached 0.299 and 0.276 ppm; however, the 8-hour TWA were quite low due to the shortness of the sampling period. The inspection did not result in a citation.
The only PPE listed in the air sampling report is 3M-N95-8210. This appears to refer to the N95 disposable particulate respirator model 8210.

Simple Designs Manufacturing Inc. Inspection 306629171
This inspection was a partial health inspection conducted as a result of a complaint. The inspected facility makes furniture using plywood, hardboard, particleboard, and oriented strand board (OSB). The company safety process manual states that safety glasses and ear protection must be worn throughout the mill area. Gloves are also required in some areas of the plant. Dust masks are available for use. Air sampling reports also indicate the occasional use of an apron. However, none of the workers inspected wore a respirator.


Appendix H
SUMMARY OF ADDITIONAL EXPOSURE MONITORING DATA FROM MANUFACTURING


Table H-1. Formaldehyde Exposure Levels for Plywood, Particleboard, and Medium-Density Fiberboard Manufacture
                               Authors/Articles
                                  Publication
                      Air Concentrations of Formaldehyde 
                               Study Population
                               (Exposed Workers)
                          Abstract and Salient Points
The Workers' Compensation Board BC (WorkSafeBC)

Discussion Paper  -  Occupational exposure limit for formaldehyde. March 20, 2007
Not Applicable
Plywood manufacturing: predominantly below 0.3 ppm as 8-hour TWA
Pressed/panel particle/fiberboard manufacturing: 0.3 ppm as 8-hour TWA exceeded at times with an exposure range of 0.07 to 5.00 ppm
Workers in a variety of industries in British Columbia including those in the pressed/panel particle/fiberboard manufacturing facilities.
Abstract:  
None Available

Salient Points:
* Sampling data collected by WorkSafeBC over a period of 15 years (1990 to 2005) in British Columbia and other locations, including Ontario, Quebec, and the Netherlands indicate that worker exposure to formaldehyde is likely to occur primarily in the forest panel board manufacturing sector (plywood, particle/fiberboard, oriented strand board, laminated wood products manufacturing), resin manufacture, acute care hospitals, university and other laboratories, and funeral homes.

* The exposure data indicate that, overall, exposure levels have decreased over time as control and procedural measures have improved. This is particularly evident when one compares readings taken in the early 1980s with those in the late 1990s.

* The results indicate that industries in which exposure levels were predominantly below the current 0.3 ppm 8-hour TWA exposure limit include: 
  oo   Foundries; 
  oo   Resin manufacturing; and 
  oo   Plywood manufacturing. 

* Industries in which the current 0.3 ppm 8-hour TWA exposure limit (for British Columbia) was exceeded include (range of exposure shown):
  oo   Paperboard & cardboard manufacturing (0.04 to 0.38 ppm);
  oo   Laminated wood products manufacturing (0.24 to 0.38 ppm); 
  oo   Pressed/panel particle/fiberboard manufacturing (0.07  -  5.00 ppm); 
  oo   Oriented strand board manufacturing (0.03 to 0.32 ppm); 
  oo   Funeral homes (none detected to 0.74 ppm); 
  oo   University laboratories (0.52 to 1.15 ppm); 
  oo   Hospitals & medical clinics and laboratories (0.02 to 1.35 ppm); and 
  oo   Poultry hatcheries (0.1  -  0.35 ppm). 

* Formaldehyde exposure levels in a plywood manufacturing plant in 2002, as reported by the Ontario Ministry of Labour, ranged from 0.16 to 0.17 ppm as 8-hour TWA, with a 15-minute STEL range of 0.14 to 0.29 ppm and instantaneous ceiling limit values ranging from 0.33 to 0.44 ppm.

* Formaldehyde exposure levels in two particleboard manufacturing plants were also reported by the Ontario Ministry of Labour in 2002. The 8-hour TWA levels ranged from 0.07 to 0.20 ppm in one plant and from 0.08 to 0.11 ppm in the other. The range of the 15-minute STELs was from 0.15 to 0.54 ppm and 0.14 to 0.29 ppm, respectively, and for instantaneous ceiling limit values, from 0.18 to 7.47 ppm and 0.17 to 0.50 ppm, respectively.


Table H-2. Formaldehyde Exposure Levels for Plywood Manufacture
                               Authors/Articles
                                  Publication
                      Air concentrations of Formaldehyde 
                               Study Population
                               (Exposed Workers)
                          Abstract and Salient Points
Fransman W et al. 

Respiratory symptoms and occupational exposures in New Zealand plywood mill workers.

Ann Occup Hyg. 2003 Jun;47(4):287-95. 

0.08 ppm to 0.60 ppm (0.01 to 0.74 mg/m[3]) [Geometric mean of this range = 0.06 ppm (0.08 mg/m3 )]
112 plywood mill workers.
Abstract:
OBJECTIVES: To study work exposure and respiratory symptoms in New Zealand plywood mill workers. METHODS: Personal inhalable dust (n = 57), bacterial endotoxin (n = 20), abietic acid (n = 20), terpene (n = 20) and formaldehyde (n = 22) measurements were taken and a respiratory health questionnaire was administered to 112 plywood mill workers. RESULTS: Twenty-six percent of the dust exposures exceeded 1 mg/m(3); however, none of the samples exceeded the legal limit of 5 mg/m(3) [geometric mean (GM) = 0.7 mg/m(3), geometric standard deviation (GSD) = 1.9]. Workers in the composer area (where broken sheets are joined together) were significantly (P < 0.01) more highly exposed. Endotoxin levels were low to moderate (GM = 23.0 EU/m(3), GSD = 2.8). Abietic acid levels ranged from 0.3 to 2.4 micro g/m(3) (GM = 0.7 micro g/m(3), GSD = 1.8) and were significantly (P < 0.05) higher for workers in the composer area of the process. Geometric mean levels of alpha-pinene, beta-pinene and Delta(3)-carene were 1.0 (GSD = 2.7), 1.5 (GSD = 2.8) and 0.1 (GSD = 1.4), respectively, and alpha-pinene and beta-pinene levels were significantly (P < 0.001) higher for workers in the 'green end' of the process, up to and including the veneer dryers. Formaldehyde levels ranged from 0.01 to 0.74 mg/m(3) [GM = 0.08 mg/m(3) (= 0.06 ppm.), GSD = 3.0]. Asthma symptoms were more common in plywood mill workers (20.5%, n = 112) than in the general population [12.8%, n = 415, adjusted OR (95% CI) = 1.5 (0.9-2.8)]. Asthma symptoms were associated with duration of employment and were reported to lessen or disappear during holidays. No clear association with any of the measured exposures was found, with the exception of formaldehyde, where workers with high exposure reported more asthma symptoms (36.4%) than low exposed workers [7.9%, adjusted OR (95% CI) = 4.3 (0.7-27.7)]. CONCLUSIONS: Plywood mill workers are exposed to inhalable dust, bacterial endotoxin, abietic acid, terpenes and formaldehyde, and they appear to have an increased risk of developing work-related respiratory symptoms. These symptoms may be due to formaldehyde exposure, although a potential causal role for other exposures cannot be excluded.




Salient Points:
* Personal measurements for inhalable dust (n = 57), bacterial endotoxin (n = 20), abietic acid (n = 20), terpene (n = 20) and formaldehyde (n = 22) were taken and a respiratory health questionnaire was administered to 112 New Zealand plywood mill workers.

* Formaldehyde levels ranged from 0.08 ppm to 0.60 ppm (0.01 to 0.74 mg/m[3]), the geometric mean of this range was determined to be 0.06 ppm (0.08 mg/m[3])

* Asthma symptoms were more common in plywood mill workers (20.5%, n = 112) than in the general population (12.8%, n = 415). No clear association with any of the measured exposures was found, with the exception of formaldehyde, where workers with high exposure reported more asthma symptoms (36.4%) than low exposed workers.

* Plywood mill workers appear to have an increased risk of developing work-related respiratory symptoms. These symptoms may be due to formaldehyde exposure, although a potential causal role for other exposures cannot be excluded.
Ballarin C et al. 

Micronucleated cells in nasal mucosa of formaldehyde-exposed workers.

Mutat Res. 1992 Jul; 280(1):1-7.
0.07-0.08 ppm (0.1 mg/m[3]) in the sawmill and shearing-press departments 

0.32 ppm (0.39 mg/m[3]) in the warehouse area.
15 nonsmoking workers in a plywood factory
Abstract:
The frequency of micronuclei (MN) and cytology of respiratory nasal mucosa cells were evaluated in 15 nonsmokers exposed to formaldehyde in a plywood factory. Each subject was paired with a control matched for age and sex. Mean levels of exposure to formaldehyde ranged from about 0.1 mg/m3 in the sawmill and shearing-press departments to 0.39 mg/m[3] in the warehouse area. There was a contemporary exposure to low levels of wood dust (inspirable mass ranged from 0.23 mg/m[3] in the warehouse to 0.73 mg/m[3] during sawing operations). Nasal respiratory cell samples were collected by an otorhinolaryngologist near the inner turbinate using a brush for endocervical cytology. After staining (Feulgen plus Fast Green and Papanicolaou's method for MN analysis and cytology, respectively), about 6000 cells were screened for micronuclei and scored in parallel for cytology according to a histopathological scale. A higher frequency of micronucleated cells was observed in the exposed group than in the controls (0.90+-0.47 vs. 0.25+-0.22, Mann-Whitney U test: p less than 0.01). Cytological examination indicated chronic phlogosis in the nasal respiratory mucosa of plywood factory workers, with a high frequency of squamous metaplasia cells (mean score 2.3+-0.5 vs. 1.6+-0.5 in the control group, Mann-Whitney U test: p less than 0.01).

Salient Points: 
* The authors examined smears of nasal respiratory mucosa cells sampled from the inner turbinate of 15 nonsmokers who were exposed to formaldehyde released from a urea-formaldehyde glue used in a plywood factory. 

* Estimates of formaldehyde air concentrations ranged from: 0.21 to 0.60 mg/m[3] (mean 0.39+-0.20 mg/m[3]) in the warehouse where seven subjects worked, 0.08 to 0.14 mg/m[3] (mean 0.1+-0.02 mg/m[3]) in the shearing press where six subjects worked, and 0.09 mg/m[3] (only one sample taken) in the sawmill area where two subjects worked. 

* Mean wood dust concentrations for the three areas were 0.23+-0.1mg/m[3], 0.41+-0.21 mg/m3, and 0.73 mg/m3, respectively. Exposed subjects worked at the factory for 2 to 19 years (mean 6.8+-5.0 years). 

* The authors found a statistically significantly higher percentage of micronucleated mucosal cells in the exposed group compared with the control group (0.91%+-0.47 versus 0.25%+-0.22).
Malaka T and Kodama AM. 

Respiratory health of plywood workers occupationally exposed to formaldehyde.

Arch Environ Health. 1990 Sep-Oct; 45(5):288-94
0.28 ppm  to 3.48 ppm 

Average personal exposure was 1.13 ppm
186 male plywood workers
Abstract: 
This study was undertaken to enlarge our understanding of the adverse health effects of formaldehyde exposure in the workplace and community environment. The respiratory health status of 186 male plywood workers was evaluated by spirometric tests, respiratory questionnaires, and chest x-rays. Area concentrations of formaldehyde were measured in the work environment and found to range from 0.28 to 3.48 ppm. The average personal exposure was to 1.13 ppm of formaldehyde. Exposure to formaldehyde was associated with decrements in the baseline spirometric values (i.e., forced expiratory volume in 1 sec (FEV1.0), forced expiratory volume/forced vital capacity (FEV/FVC), and FEF25%-75%) and with several respiratory symptoms and diseases, including cough, phlegm, asthma, chronic bronchitis, and chest colds. The results of the study support the hypothesis that chronic exposure to formaldehyde induces symptoms and signs of chronic obstructive lung disease.

Salient Points:
* In this cross-sectional study of 186 plywood workers, area concentrations of formaldehyde ranged from 0.28 to 3.48 ppm. The average personal exposure was 1.13 ppm. 

* Reported average respirable and total wood-dust concentrations in workplace air were 0.60 and 1.35 mg/m[3], respectively. 

* Exposure was associated with increased complaints of cough, phlegm production, asthma, chronic bronchitis and colds. 

* Decrement in FEV1/FVC ratio in exposed subjects compared to controls was small but significant, although the authors noted that the clinical significance of the differences was unclear. 


Table H-3. Formaldehyde Exposure Levels for Particleboard and Medium-Density Fiberboard Manufacture
(Includes Wood Workers and Oriented Strand Board Manufacture)
                                Authors/Article
                                  Publication
                      Air concentrations of Formaldehyde 
                      Study Population (Exposed Workers)
                          Abstract and Salient Points
The Workers' Compensation Board BC (WorkSafeBC)

Discussion Paper  -  Occupational exposure limit for formaldehyde. March 20, 2007
Not Applicable
0.03-0.032 ppm for oriented strand board
Workers in a variety of industries in British Columbia including those in the plywood, particle/fiberboard, and oriented strand board manufacturing facilities.
Abstract:  
None available

Salient Points:
* Sampling data collected by WorkSafeBC over a period of 15 years (1990 to 2005) in British Columbia and other locations, including Ontario, Quebec, and the Netherlands indicate that worker exposure to formaldehyde is likely to occur primarily in the forest panel board manufacturing sector (plywood, particle/ fiberboard, oriented strand board, laminated wood products manufacturing), resin manufacture, acute care hospitals, university and other laboratories, and funeral homes.

* Industries in which the current 0.3 ppm 8-hour TWA exposure limit (for British Columbia) was exceeded include (range of exposure shown):
  oo   Paperboard & cardboard manufacturing (0.04 to 0.38 ppm);
  oo   Laminated wood products manufacturing (0.24 to 0.38 ppm); 
  oo   Pressed/panel particle/fiberboard manufacturing (0.07 to 5.00 ppm); 
  oo   Oriented strand board manufacturing (0.03 to 0.32 ppm); 
  oo   Funeral homes (none detected to 0.74 ppm); 
  oo   University laboratories (0.52 to 1.15 ppm); 
  oo   Hospitals & medical clinics and laboratories (0.02 to 1.35 ppm); and 
  oo   Poultry hatcheries (0.1 to 0.35 ppm). 

For other details of this study, as they relate to exposure levels in plywood and particle/fiberboard manufacturing, see Table H-1.
Lavoue J et al. 2005

Investigation of Determinants of Past
and Current Exposures to Formaldehyde in the Reconstituted
Wood Panel Industry in Quebec
Annals of Occupational Hygiene. 2005 49(7):587-602
Geometric means (GMs) of personal and area levels between 0.1 and 0.4 ppm for MDF and PB.
Highest estimated ambient GM 0.43 ppm in the main production area of MDF manufacturing facilities. 
Highest personal estimated GM 0.22 ppm for workers in close proximity to the press in the PB manufacture process.
Employees in the 12 plants  (number not specified)
Abstract:
OBJECTIVES: Past and present formaldehyde measurements made in facilities manufacturing reconstituted wood panels in Quebec have been collected to assess formaldehyde exposure and its determinants in this industry.
METHODS: All 12 plants manufacturing OSB, MDF, and PB in Quebec were visited by a research team which took area and personal measurements. Past measurements taken by governmental occupational health teams in these plants were also collected. Log-transformed formaldehyde concentrations were analyzed with extended linear mixed-effects models.
RESULTS: During 2001 - 2002, 275 measurements were taken by the research team, while 590 measurements dating back to 1984 were collected from governmental files. The area measurements had a global GM of 0.28 ppm [geometric standard deviation (GSD): 3.1]. The GM of the personal measurements was 0.17 ppm (GSD: 2.3). The fixed-effects of the models for personal and area measurements explained 61 and 57% of the variance, respectively. Job (working area for area concentrations), process (PB, MDF, OSB), season of sampling, origin of the data (research, governmental) and year of sampling were significant determinants of exposure. Proximity to the press, winter conditions, PB and MDF processes and governmental data resulted in the highest exposures. Significant within-sampling campaign correlation was found for both personal and area models. The final models include different residual variances by process for personal measurements and by working area for area measurements.




CONCLUSIONS: Several determinants of exposure to formaldehyde in the reconstituted wood panel industry were successfully identified. Higher levels found in governmental data as compared to research data may be explained by a `worst-case' strategy bias. The observed intra-sampling campaign correlation supports existing results suggesting that measurements taken in a small time frame tend to be correlated. Exposures in this sector are low compared to the 8-hour TWA OELs of most Canadian jurisdictions (e.g., 0.75 ppm) but close to the most demanding ones (e.g., 0.3 ppm, as for British Columbia).

Salient Points:
* Twelve plants manufacturing OSB, MDF, and PB in Quebec were visited from June 2001 to March 2002. The visits were conducted by a team of 2 - 4 industrial hygienists and technicians and lasted 1 to 2 days.
* The research team took area and personal measurements. Past measurements taken by governmental occupational health teams in these plants were also collected. The visits performed by the research team yielded 275 measurements while 590 measurements were available from governmental files.
* Exposures to formaldehyde in the OSB manufacturing industry were mostly below 0.1 ppm with some work zones/jobs associated with somewhat higher levels according to the governmental data. 
* Formaldehyde exposure levels in the MDF and PB manufacturing industries were similar, with observed and estimated GMs of personal and area levels between 0.1 and 0.4 ppm depending on the source of data. The highest estimated ambient GM was 0.43 ppm in the main production area of MDF manufacturing facilities (this estimate is increased to 1.28 ppm when governmental data are used). The highest personal estimated GM is 0.22 ppm for workers in close proximity to the press in the PB manufacture process (this estimate is increased to 0.62 ppm if the governmental data are considered).




* Governmental measurements were consistently found higher than those measured by the research team, pointing to the probable existence of a `worst-case' bias in governmental data.
* Area measurements were consistently higher than personal measurements. This is consistent with the fact that the most exposed workers are those who spend the most time in the production area unprotected by ventilated booths. While none of the personal measurements in this study was greater than 2 ppm, 4% and 6% of the research and governmental ambient concentrations, respectively, were greater than this value.
* Although the results of this study point to generally low personal full-shift exposure to formaldehyde in the industrial sector studied, the potential for short-term high exposures associated with specific and occasional tasks cannot be ruled out on the basis of our study.
* Levels reported prior to 1985 are consistently higher than those in this study database. According to the authors, this maybe explained by the generalized use, around 1985, of low formaldehyde emission resins.
* The measured and estimated TWA levels of exposure to formaldehyde in this sector can be considered low compared to the 8 hour TWA OELs of most Canadian jurisdictions (e.g. 0.75 ppm) but close to the most demanding ones (e.g., 0.3 ppm, as for British Columbia).
Chung KY et al.
A study on dust emission, particle size distribution and formaldehyde concentration during machining of MDF.

Ann Occup Hyg. 2000 Sep; 44(6): 455-66. 

Less than 0.14 ppm (0.17mg/m[3])
N/A
Abstract:
A study to characterize the quantity, particle size distribution, and morphology of dust created during the machining of MDF was carried out. Four different types of MDF boards were included in this study, including a 'zero-formaldehyde' board that contains isocyanate-based resin, rather than urea-formaldehyde resin. In addition, natural softwood (pine) and natural hardwood (oak) were included in the study, for comparison with MDF. The results show that in general, the dust generated by machining MDF is comparable in terms of particle size distribution and morphology with the dust generated by similarly machining hardwood or softwood. The quantity of dust generated during sanding is higher for sanding MDF compared with sanding either hardwood or softwood. However, for sawing, there is no significant difference between MDF and natural woods, in terms of the quantity of dust generated. Free formaldehyde in the air was less than 0.17mg/m[3] during machining of the Class B (higher formaldehyde potential) MDF board. There was no measurable isocyanate in the dust generated from the boards.
Herbert FA et al.
Respiratory consequences of exposure to wood dust and formaldehyde of workers manufacturing oriented strand board.
      

Arch Environ Health. 1994 Nov-Dec; 49(6):465-70.
Up to 0.27 ppm
99 workers employed in the manufacture of oriented strand board.
Abstract:
A cross-sectional study was performed at a plant in which 99 workers were employed in the manufacture of oriented strand board. This group was compared with 165 unexposed workers from the petroleum industry. Both groups were assessed, using a questionnaire, spirometry, and skin prick tests to common environmental antigens. Environmental studies showed a low dust level of 0.27 mg/m[3], consisting of particles of a mass median aerodynamic equivalent diameter of 2.5 microns. There were parameter concentrations of formaldehyde, up to 0.27 ppm. A significant difference between the oriented strand board workers and oil field workers was noted for the forced expiratory volume in 1s/forced vital capacity ratio, without significant differences in either the forced expiratory volume in 1 s or the forced vital capacity. Oriented strand board workers who were current smokers were three times as likely to have a forced expiratory volume in 1s/forced vital capacity ratio of less than 75% of that found in the currently smoking oil field workers. Significant reductions in forced expiratory volume in 1s (p = .044) and forced vital capacity (p = .022) in oriented strand board workers were noted across the work shift. The oriented strand board workers complained of self-reported asthma and of lower respiratory tract symptoms significantly more frequently than did oil workers for all of the symptoms examined. The prevalence of atopy was not different in the two groups. Lung function was significantly better in oriented strand board workers who had no symptoms, compared with oriented strand board workers who were symptomatic. (ABSTRACT TRUNCATED AT 250 WORDS)

Salient Points: 
* In this cross-sectional study, 99 workers employed in the manufacture of oriented strand board were compared with 165 unexposed workers from the petroleum industry.




* Parameter concentrations of formaldehyde, up to 0.27 ppm, were noted.
* Environmental studies showed a low dust level of 0.27mg/m[3], consisting of particles of a mass median aerodynamic equivalent diameter of 2.5 microns.
* Significant reductions in forced expiratory volume and forced vital capacity were noted across the work shift in oriented strand board workers. 
* Oriented strand board workers complained of self-reported asthma and of lower respiratory tract symptoms significantly more frequently than did oil workers for all of the symptoms examined.
Edling C et al.

Occupational exposure to formaldehyde and histopathological changes in the nasal mucosa.


Br J Ind Med. 1988 Nov; 45(11):761-5.

0.08 to 0.9 ppm. (midpoint of this range is 0.49 ppm). 

Peaks of up to 4.1 ppm were measured during the 8-year period
75 workers from two particle board processing plants and a laminate plant
Abstract:
To study the cytotoxic effect of formaldehyde on the human nasal mucosa, 75 men with occupational exposure to formaldehyde or to formaldehyde and wood dust were examined, looking particularly at early signs of irritative effects and histopathological changes in the nasal mucosa. All men underwent a medical examination, and a nasal biopsy specimen was examined by a pathologist and graded from 0-8 according to the morphological changes. A high frequency of nasal symptoms, mostly a running nose and crusting, was related to exposure to formaldehyde. Only three men had a normal mucosa; the remainder had loss of cilia and goblet cell hyperplasia (11%) and squamous metaplasia (78%); in six cases (8%) there was a mild dysplasia. The histological grading showed a significantly higher score when compared with unexposed contents (2.9 v 1.8). There was no dose response relation, no malignancies, and no difference in the histological score between those exposed to formaldehyde or to formaldehyde and wood dust.




Salient Points:
* Measurements taken by in-house hygienists during an 8-year period before the study (between 1975 and 1983) were in the range 0.08 to 0.9 ppm with peaks reaching 4.07 ppm. A mean TWA concentration was not reported, but the midpoint of this range is 0.49 ppm. Air concentrations were qualitatively assessed as being "somewhat higher" during earlier periods. 
* Wood dust air concentrations in the particle board plants ranged from 0.6 to 1.1 mg/m[3]; air in the laminate plant was reported to be without wood dust. 
* Employment durations ranged from 1 to 39 years with a mean of 10.5 years. 
* Runny nose, nasal crusting, and runny eyes when at work were reported by 60 and 75% of the exposed subjects, respectively. Gross clinical examination showed that 25% of exposed workers had either swollen nasal mucosa or dry nasal mucosa. 
* The authors reported that the average histological score for the exposed group (2.8) was statistically significantly greater than the control score (1.8) and that there was no difference in average histological scores between the exposed workers from the particle board plants, where confounding exposure to wood dust occurred, and those from the laminate plant (without wood dust exposure). This observation supports the hypothesis that the observed nasal epithelial lesions were caused by formaldehyde and not by an interaction between formaldehyde and wood dust.
Horvath EP et al. 1988. 

Effects of formaldehyde on the mucous membranes and lungs: A study of an industrial population. 

JAMA. 1988 Feb 5; 259(5):701-7.
0.17 to 2.93 ppm (mean of 0.69 ppm)
109 workers in a particle board and molded plastics plant
Abstract:
One hundred nine workers and 254 control subjects were studied to evaluate the effects of formaldehyde on the mucous membranes and lungs. A modified, respiratory symptom questionnaire and spirometry were administered to all study participants before and after their work shift, and formaldehyde levels were determined for each test subject. Over the course of the monitored work shift, test subjects demonstrated a dose-dependent excess of irritant symptoms and a statistically significant decline in certain lung function parameters. Analysis of test and control subject data combined revealed a correlation between formaldehyde exposure and these pulmonary changes. Baseline spirometry values were not significantly different between test and control groups, and formaldehyde-exposed workers did not report an excess of respiratory symptoms. Formaldehyde is a dose-dependent irritant of the eyes and mucous membranes at low-level exposures. It can exert a small, across-shift effect on airways but after a mean exposure of 10 years does not appear to cause permanent respiratory impairment.

Salient Points:
* The duration of exposure among exposed workers ranged from <1 year to 20 years, with a mean and median of 10.3 and 10 years, respectively.
* Estimates of formaldehyde air concentrations ranged from 0.17 to 2.93 ppm with a mean of 0.69 ppm. Nuisance particles (predominantly softwood dust) were also detected in the particle board area. 




* The percentages of particle board workers reporting a number of symptoms of respiratory irritation over a work shift were statistically significantly greater than work shift reporting percentages for a nonexposed group of 264 food-processing workers: cough (34.9 versus 18.9%), chest pains (9.2 versus 2%), phlegm production (26.6 versus 9.8%),burning nose (28.4 versus 2%), stuffy nose (33.9 versus 14.2%), burning or watering eyes (39.5 versus 9.1%), itchy nose (21.1 versus 7.9%), and sore/burning throat (22 versus 3.9%).
* The authors concluded that the effects noted were possibly not only due to formaldehyde alone but could also be attributed to the presence of respirable particles of wood dust.
Kauppinen TP and Niemelä RI

Occupational exposure to chemical agents in the particleboard industry.

Scand J Work Environ Health. 1985 Oct; 11(5):357-63.
Before 1975, greater than 2 ppm during many work phases.

Means below 2 ppm during most work phases during 1975 and 1984.

For area measurements, means between 0.4 ppm and 2.3 ppm (depending on individual locations in the plant).
Not available 
Abstract:
Most of the measurements concerned formaldehyde and wood dust. The other substances measured included terpenes, solvents, and heptachlor. Before 1975, the formaldehyde concentration regularly exceeded 2 ppm during many work phases. Considerable improvements in ventilation and the composition of glues have occurred since then, and today the exposure level is below 2 ppm during most work phases. High peak concentrations, 20 to 30 ppm at the highest, were characteristic of exposure in earlier years. The concentrations of wood dust have also decreased (e.g., from over 5 mg/m[3] to 1 mg/m[3] or below during forming). These data have been used to evaluate past exposures in an epidemiologic study on cancer risks in the particleboard, plywood, and sawmill industries.
Alexandersson R et al.

Exposure to formaldehyde: Effects on pulmonary function.

Arch Environ Health. 1982 Sep-Oct; 37(5):279-84.
0.37 ppm (0.45 mg/m[3])
47 woodworkers
Abstract:
Forty-seven subjects exposed to formaldehyde (mean air concentration 0.45 mg/m[3]) and 20 unexposed subjects, all of whom were employed at a carpentry shop, were studied with regard to symptoms and pulmonary function. Symptoms involving eyes and throat irritation as well as chest oppression were significantly more common in the exposed subjects than in the unexposed controls. Spirometry and single breath nitrogen washout were normal Monday morning before exposure to formaldehyde. A reduction in forced expiratory volume in 1 sec by an average of 0.2 L (P = 0.002), percent forced expiratory volume by 2% (P = 0.04), maximum midexpiratory flow by 0.3 L/sec (P = 0.04), and an increase in closing volume in percentage of vital capacity by 3.4% (P - 0.002) were seen after a day of work and exposure to formaldehyde, suggesting bronchoconstriction. Smokers and nonsmokers displayed similar changes in spirometry and nitrogen washout.

Appendix I
MODEL RESULTS FOR INDOOR AIR CONCENTRATIONS AT CWP FABRICATION SITES USING PRE-BASELINE EMISSION RATES


Table I-1. Results for the Calculation Trials for the What-If Low- and High-End Pre-Baseline Indoor Air Concentrations
                               Varied Parameter
                Calculated Concentration for Pre-Baseline, ppm
                                       
                               Intermediate [4]
                                  Low End [5]
                                 High End [6]
                                       
                                     HWPW
                                      PB
                                      MDF
                                     HWPW
                                      PB
                                      MDF
                                     HWPW
                                      PB
                                      MDF
Vary Emitting Surface Area [1]
                                     0.019
                                     0.024
                                     0.044
                                     0.006
                                     0.007
                                     0.009
                                     0.142
                                     0.197
                                     0.399
Vary Effective Ventilation Rate [2]
                                       
                                       
                                       
                                     0.007
                                     0.008
                                     0.011
                                     0.414
                                     0.580
                                     1.186
Vary Emission Rate [3]
                                       
                                       
                                       
                                     0.012
                                     0.018
                                     0.018
                                     0.022
                                     0.044
                                     0.090
  [1] For area, the intermediate value is based on the aggregate average per-site throughput calculated using domestic composite wood product consumption data from the EPA market profile. The high-end and low-end values are calculated assuming 1,000% and 10% of the aggregate average throughput, respectively.
  [2] Effective ventilation is varied based on values shown in Table 3-3. The low-end, typical, and high-end values are 50, 1,500 and 10,000 cfm, respectively.
  3 Emission rate is varied based on values shown in Table 3-4.
  [4] Intermediate concentration is calculated using intermediate value for surface area, typical value for effective ventilation rate, and weighted-average pre-baseline emission rate for each CWP type.
  [5] Low-end values calculated according to Table 3-2.
  [6] High-end values calculated according to Table 3-2.

Table I-2.  Combined (for All CWP) Results: Estimated Range of Values for the What-If Low- and High-End Indoor Air Concentrations for Pre-Baseline
                                     Case
                              Intermediate (ppm)
                                 Low End (ppm)
                                High End (ppm)
Pre-Baseline 
                                0.019  -  0.044
                                0.006  -  0.018
                                0.022  -  1.186


