Draft   SEQ CHAPTER \h \r 1 Site Visit Report: 

Beryllium Site 1

 

Submitted to:

Office of Regulatory Analysis

Directorate for Evaluation and Analysis

Occupational Safety and Health Administration

Submitted as partial fulfillment of 

requirements for Task Order 41

Contract No. J-9-F-9-0010

Submitted by:

Eastern Research Group, Inc.

110 Hartwell Ave.

Lexington, MA 02421

March 3, 2003

TABLE OF CONTENTS

Introduction……………………………………………………
…………2

Facility and Process
Description…………………………………….3

Exposure
Assessment…………………………………………………5

Exposure Controls in
Place………………………………………….19

Discussion……………………………………………………
…………20

General
Recommendations………………………………………….22

Specific Recommendations…………………………………………
25

References……………………………………………………
………...29

Introduction

On December 2 – 3, 2002, investigators visited an aluminum-beryllium
metal alloy fabrication facility.  During this visit measurements were
made to characterize worker exposure to airborne and surface levels of
beryllium and gather data to assist in the development of some industry
guidance criteria.   

Beryllium has unique characteristics that make it a superior material
for certain specialized applications. Compared to other metals,
beryllium is very light, has a high melting point, low electrical
conductivity, superior strength and stiffness, high thermal
conductivity, and high resistance to corrosion. In addition, it is also
transparent to x-rays, absorbs neutrons, and is non-magnetic.  Beryllium
is used in several forms: as a pure metal, as beryllium oxide, and as an
alloy with copper, aluminum, magnesium, or nickel.  However, exposure to
beryllium compounds is associated with a chronic debilitating lung
disease--chronic beryllium disease (CBD)--that is the result of a
beryllium-specific, T lymphocyte-mediated hypersensitivity.  Symptoms of
the disease are exertional dyspnea and cough, followed by weight loss,
chest pain, arthralgias, and fatigue.  Progression of the disease may
lead to heart failure as a result of chronic respiratory insufficiency. 
 

Objective

The objective of this work is to develop guidance material that would
help business to reduce worker exposure during the processing of
beryllium from the best practices currently found in industry.  These
practices are for all users of beryllium regardless of the form the
beryllium is in; metal, oxide or alloy. 

The amount or length of exposure to beryllium necessary to cause a
specific individual to develop CBD is not known, but recent information
suggests that even short exposures to levels of beryllium below OSHA’s
Permissible Exposure Limit (PEL) of 2 µg/m3 averaged over an 8-hour day
may lead to CBD in some workers (OSHA, 1999).  CBD may develop within
months after initial exposure to beryllium or may have a very slow onset
and not develop for 25 years or more and may even develop after exposure
has ceased.  The prevalence of CBD among beryllium-exposed workers has
been reported to range from an average of about 2% to a high of
approximately 15% for workers involved in machining operations in the
manufacturing of beryllium products.

Measurement of exposure to total airborne beryllium dust may not be the
best predictor of CBD. Particle size, surface area, number of particles,
solubility, and the chemical form of beryllium involved may all be
relevant to the development of disease. It has also been suggested that
beryllium can enter through either intact skin or breaks in the skin to
initiate sensitization.  Recent studies have shown that particles less
than 1 (m in diameter can penetrate intact skin that has been flexed.
Therefore, it is essential that all possible pathways of exposure be
considered when assessing a work site for potential exposure to
beryllium.

Facility and Process Description

The company is located in Northeastern Pennsylvania in a 3-year old,
5000 square feet manufacturing facility and specializes in machining
high tolerance aluminum, beryllium, aluminum-beryllium alloy, and other
metals.  The company operates three 8-hour shifts per day, 5-6 days per
week.  The facility has eleven full-time employees. Eighty percent  of
the machining is performed on aluminum, 19.9% is on aluminum beryllium
and 0.1% on beryllium metal.

There are 5 major machining centers in this facility, which together
contain the following enclosed, automated machining equipment: 3 Fadal
“CNC’s”, 1 Makino CNC, and 1 Hardinge lathe.  All equipment is
operated by machinists, three of whom were evaluated during this site
visit.  Additionally, after machining the workers perform deburring on
certain products as a manual process, a process for which specialized
controls are used.    

Job Descriptions

Machinists

The three machinists who were processing beryllium were stationed at
Fadal CNC#1 (Subject 1) and at the Makino CNC (Subjects 2 and 3).  

The Fadal CNC was programmed to automatically machine parts with the
doors of the machine enclosure shut.  During the sampling period, the
machinist (Subject #1) operating the Fadal CNC received aluminum
beryllium plates, approximately 200 centimeters square, in a box brought
to the facility from a supplier.  A plate was taken from the box and
placed in position within the CNC.  The doors of the CNC were closed
with an interlock and the machining program initiated.  The machinist
observed the process from a work bench/desk approximately one meter from
the machine.  The program, which controlled the machining of the part
was completed in approximately two minutes.  The machinist then opened
the door of the CNC, dry wiped the machined part and removed it from the
machine.  The part was then placed on the workbench, inspected, then
placed in a box for shipment and a new plate taken from the box and
placed in the machine.  The task was repeated throughout the work shift.
 Approximately 10 to 15 plates per hour were completed in this fashion.

The Makino A55E CNC has a fixture on which parts could be mounted.  Once
the parts were mounted on the fixture, they were rotated into position
inside the CNC and automatically machined according to a preset program.
 Once machined, the parts were removed from the fixture, cleaned,
examined under a magnifying lens and deburred as necessary.  The
deburring could be done either by hand with a small abrasive pad or with
a powered brush.  The parts were then packed for shipment.  After
several cycles, chips built up in the machine and were removed by hand
into a scrap barrel at the rear of the machine.   Subject #2 was a
machinist who was instructing Subject #3 in the use of the Makino on the
sampling date.  Both machinists did the same tasks and worked
side-by-side for the entire duration.  A cycle for machining parts
lasted approximately 30 minutes from start to completion.  It could be
assumed that the training being done altered the work habits of the
machinist, although there was no chance to observe the normal work
habits.

At the end of the shift both machines (Fadel CNC and Makino)had shavings
removed by hand, were rinsed and wiped down with a clean, dry, paper
towel both inside and out.

Figure 1.  Plant diagram showing the equipment used for processing
beryllium during the survey in dark gray and equipment not using
beryllium, cross-hatched.  Locations of workers wearing air sampling
equipment are designated by numbers in circles. The “x’s” are
surface sampling locations described in Table II.  Double-sided arrows
designate doorways.Exposure Assessment

Sampling was conducted to determine the beryllium air and surface
contamination to which workers are potentially exposed and to derive
some simple measure of dose for these workers. The first day of the site
visit was spent meeting with company personnel (company management and
employees) to arrange sampling on the subsequent day, and to walk
through the plant to begin the industrial hygiene assessment of exposure
and control technology.  Employees with the highest potential beryllium
exposures in each process area or operation were the major focus of the
sampling.  Workers selected for sampling were briefed on the sampling
procedures to be conducted.  Because the goal of this study is to assess
the effects of engineering controls and work practices on beryllium
exposures, samplers were placed outside of any respiratory protective
equipment worn by the worker.  Two days of sampling were then conducted
on the only three employees present who worked with beryllium alloy.

Sampling Methods

Beryllium-related disease is assumed to be a result of the number of
deposited beryllium particles and perhaps the available surface area of
those particles in the lung.  Thus, parameters of interest include
surface contamination, skin exposure, airborne particle number
concentration (for particles of a size that deposits in lung), and the
surface area distribution of the beryllium particles, in addition to the
total gravimetric concentration of beryllium in air.  The methods used
to assess these parameters are listed below and briefly described in
this section.  

Parameter	Sampling Method  [unit of measure]

Surface contamination	NIOSH Method 9100, surface wipe samples [μg/100
cm2]

Skin exposure	Hand wipe samples [μg/100 cm2]

Total airborne beryllium concentration	Traditional air sampling for
total dust [μg/m3]

Airborne deposited particle number concentration 	Condensation particle
counter (CPC) measuring airborne concentration of particles proportional
to that which would deposit in the lung [number of particles between
0.01 and 1 μm per cm3 of air (p/cc)]

Particle surface area distribution	Scanning mobility particle sizer
(SMPS) [particle count (number) distribution by particle size (d vs
n/delta log d)] and Microorifice uniform deposition impactor (MOUDI)
[mass distribution by particle size(d vs m/delta log d)]



	Laboratory Analysis

Total mass	Gravimetric analysis [μg/sample]

Beryllium mass (in wipe and air samples)	NIOSH Method 7102, graphite
furnace atomic absorption [μg beryllium/sample]





Surface Samples

Surface contamination evaluation followed the well-established protocol
for lead surface contamination measurements, as did skin sampling. 
Surface sampling was done by marking off a 100 cm2 area using a plastic
template, with a square hole 10 cm on a side, which was placed on the
surface and the corners marked with the moistened towelette.  The
template was then be removed and the perimeter lined out with the same
moistened towel.  The inside of the square was then wiped according to
NIOSH Method 9100 and the towel placed in a screw-top glass vial for
analysis.  The template was cleaned before reuse.  Workers gloves were
also wiped, but at a separate time from the hand wiping so as not to
disrupt the ordinary wear of the glove immediately before skin sampling.

Hand Wipe Samples

Hand wipe samples were obtained by asking study participants to wipe
their hands before the end of their shift or during the shift, at least
two hours after their last hand washing.  They were instructed to lift a
fresh wet-wipe from its opened container and to thoroughly wipe both
hands (including the front and back, up to the wrists, and each finger),
removing as much visible dirt as possible.  The wipe was then placed in
a labeled screw-top glass vial.  The hand wiping exercise was supervised
and timed for 30 seconds by the investigators to ensure consistency from
subject to subject.  A tracing of both hands of each participant was
taken to estimate the total surface area of the participants’ hands. 
The concentration of beryllium on the workers’ hands is reported in
micrograms of beryllium per 100 cm2.  

Personal Air sampling (Airborne deposited particle number concentration
and Total airborne beryllium)

Each full-shift air-sampling subject wore a backpack containing a
condensation particle counter (CPC) (TSI, Model 3007) and two personal
air-sampling pumps [2 liters per minute (LPM)].One of the air sampling
pumps [nominal air flow rate] was attached to a personal impactor
[Anderson Marple impactor, 8th and final stages only] clipped outside
the backpack at the worker’s lapel. While stage 8 of the impactor
screened larger particles, a triple layer of polycarbonate filters in
the impactor base captured particles 0.01 to 1.0 μm, the size range
that would deposit in the lung.   A battery-operated valve allowed the
CPC to alternately measure the concentration of particles in two air
streams: 1) air drawn through the polycarbonate filters and 2) ambient
air from a tube also attached at the workers lapel.  By comparing the
particle concentration values recorded by the CPC for the filtered and
unfiltered air-streams, it was possible to estimate the concentration of
particles removed by the layered polycarbonate filters, which represents
the deposited submicrometer particle concentration (DSP).

 

The second personal air-sampling pump in the worker’s backpack was
used for collecting the “total” airborne beryllium sample.  This
pump was connected to a  37 mm cassette containing pre-weighed PVC
filters, clipped outside the backpack at the worker’s lapel.

When short-term samples were collected during specific tasks, the second
personal air-sampling pump was eliminated and instead a high-flow pump
(Gast, Model, 25 LPM) was connected to the 37 mm cassette assembly on
the workers lapel. The higher flow rate was required to ensure an
adequate volume of air was collected during the shorter sampling period.
A member of the investigative team held the pump and followed the worker
around to assure the tether did not interfere with the worker’s
activities.  The sampling pumps were calibrated both before and after
the shift using a Bios Dry-Cal meter.  Pumps were set within one percent
of the target flow rate.

The PVC filters were analyzed to obtain the total mass and beryllium
mass, from which the fractional beryllium content (% beryllium) was
determined.  This percentage was applied to the DSP number concentration
obtained from CPC results, to estimate the percent of that concentration
that was beryllium (Be DSP).  This assumes that the beryllium particles
were discrete from all other particles and that the percentage beryllium
multiplied by the number of particle thereby represents a discernable
value.  This final value, Be DSP, represents the airborne concentration
of beryllium particles that are of a size that would deposit in the lung
and is reported as particles per cubic centimeter of air (p/cc).

Finally, the surface area distribution of the airborne beryllium was to
be determined if sufficient beryllium mass was present.  This was to be
done using a Microorifice Uniform Deposit Impactor (MOUDI) and a
Submicrometer Particle Sizer (SMPS). This gave an estimate of both the
mass and count distributions. This method is purely investigational, but
the information may be relevant for future work.  The theory behind it
is that the surface area may be the normalizing method of dose
evaluation for various particle sizes, smaller particles having greater
surface area per unit mass and therefore greater possible toxicity.  The
sampling methods for determining surface area require evaluation of
either the mass or number distribution of the beryllium.  By determining
either distribution the surface area could be inferred.  

The SMPS separates particles using electrical charge characteristic that
influence particle trajectory. Only particles with the correct
trajectory pass through an open slit to be measured by a condensation
particle counter (TSI, P-Track). A controlled change in the SMPS test
voltage causes a different size particle to have the correct trajectory
and particle distribution can be measured as particle counts. During the
site visit, particles 0.01 to 0.4 μm were evaluated to provide the
particle count distribution by particle size.

 

The MOUDI uses vacuum pressure to classify particles in 8 stages ranging
from 0.06 to 10 μm aerodynamic diameter. This range is substantially
lower than conventional particle classifiers.  During MOUDI sampling,
particles were collected on pre-weighed substrate that was analyzed to
determine total mass and beryllium mass for particles in each stage. The
beryllium mass size distribution could then be calculated from these
results if beryllium is present in sufficient quantity in the air. 

Laboratory Analysis

OSHA’s Salt Lake City Laboratory performed the laboratory analysis.
Gravimetric results were reported to the nearest microgram. Beryllium
mass was determined for surface wipes, hand wipes, and air samples using
NIOSH Method 7102, graphite furnace atomic absorption spectrophotometry.
Field blanks made up 10 percent of the samples analyzed and another five
percent were media blanks. 

Sampling Results

Surface Samples

Table I. Results of surface samples taken at locations shown in Figure
1.

Milling Room	 Total Be

 (ug)	Surface Area (cm2)	Be Concentration

(ug/100 cm2)

Computer desk surface	0.56	100	0.56

Table in front of Hardinge	0.07	100	0.07

Makino operator's chair	0.48	100	0.48

Exit door surface, Milling Room side	7.50	100	7.50

Floor in front of Fadal #1 near subject 1	33.00	100	33.00

Top of tool box near Hardinge	9.80	100	9.80

Computer in Milling Room	0.61	100	0.61

Top of tool box near Fadal #1	0.15	100	0.15

Desk used by subject 1	30.00	100	30.00

Floor in center of room	24.00	100	24.00

Floor by cabinets near Makino	3.10	100	3.10

Floor behind Fadal #1	52.00	100	52.00

On top of radio near Makino	20.00	100	20.00

Floor behind Makino	315.00	100	315.00

Top of tool chest, main corridor, near Fadal#2	3.70	100	3.70

Wall near exit door	0.79	100	0.79

Top of tool box near Fadal #3	1.05	100	1.05





	Office	 Total Be

 (ug)	Surface Area (cm2)	Be Concentration

(ug/100 cm2)

Computer desk surface	2.75	100	2.75

Work desk near window	0.71	100	0.71

Floor by door from office into milling room	0.46	100	0.46

Middle shelf in storage unit	3.27	100	3.27

Granite top work surface of Cordax 1808	<LOD	100	<LOD





	Lunch Room	 Total Be

 (ug)	Surface Area (cm2)	Be Concentration

(ug/100 cm2)

Table top	0.28	100	0.28

Floor in front of door leading into changing room	0.43	100	0.43

Top of microwave	0.38	100	0.38





	Changing Room	 Total Be

 (ug)	Surface Area (cm2)	Be Concentration

(ug/100 cm2)

Floor in middle of room	6.20	100	6.20

Surface of door leading to mill room	0.46	100	0.46



The highest surface contamination was found on floors in the milling
room.  Work surfaces were relatively lower, but nevertheless
contaminated.  The values do not necessarily represent what level can be
achieved after cleaning since the samples were taken during the middle
of a production shift.  Of interest is the highest level found, on the
floor behind the Makino CNC.  This is an area where chips are usually
emptied during the shift and possibly is a result of chip handling that
occurs during the operation.  Other floor contamination is possibly due
to the transport of materials from the CNC’s to work surfaces without
adequate cleaning of parts that may contain some beryllium residue.
Since there was little evidence of airborne beryllium, contamination on
other surfaces is more likely to have occurred via touch from the
workers’ hands or clothing, with the most notable contamination on
tool chests and the radio.  This is also supported by lower levels of
contamination found on the table in front of the Hardinge lathe.  This
lathe was not in use during the course of the study, workers were
generally not around this area and the surface concentration was
noticeably lower as a possible result.

There was no visible contamination on any surface except for the
shelving in the office which had a light layer of dust suspended on a
rough surface that probably presents some problem for cleaning and may
possibly be overlooked when cleaning is done. Some papers and other
items are brought into the office from the plant and may serve as the
route of this contamination.  The route of entry into the office is
usually not through the door directly into the plant, although this door
appears to be a tempting route for shortcutting the longer, proper route
through the changing room and lunchroom.

The widespread floor contamination in the milling room is probably the
reason why beryllium could be detected on the floor of the changing
room.  Workers take off booties and dispose of them before exiting the
milling room.  Even though there is a sticky mate at the door there may
be enough contamination remaining on workers shoes from even the short
walk to the door that beryllium is brought out of the work area.

There was also evidence that contamination continued into the lunchroom,
although the levels were much lower on the floor in that area.

Hand Samples

Table II. Wipe Samples Obtained from Both Hands Together

Gloves-On Samples	 Total Be

 (ug)	Surface Area (cm2)	Be Concentration

(ug/100 cm2)

Subject 1 (12/03/02)	66.00	723	9.13

Subject 3 (12/03/02)	63.00	680	9.26





	Bare-Hand Samples	 Total Be

 (ug)	Surface Area (cm2)	Be Concentration

(ug/100 cm2)

Subject 1 (12/03/02)	1.01	723	0.14

Subject 2 (12/03/02)	0.13	588	0.02

Subject 3 (12/03/02)	0.88	680	0.13

Subject 1 (12/04/02)	6.71	723	0.93

Subject 2 (12/04/02)	6.65	588	1.13

Subject 3 (12/04/02)	4.05	680	0.60

Supervisor (12/04/02)	1.82	648	0.28

The evidence in Table II seems to indicate that glove use provides a
substantial level of protection against skin contamination.  It also
indicates that beryllium contamination is being transferred to other
surfaces from the machined parts, in this case the gloves of the
workers.  As with the surface contamination in the previous section, the
most likely route of contamination is via inadequately cleaned parts. 
There is also a trend in the hand contamination data that was seen in
the air samples.  The levels were higher on the second day of the
survey.  All three machinists who were sampled randomly on both days
showed a significant increase in hand contamination on the second day. 
There was no noticeable increase in the activity or production levels on
the second day of the survey.  This may represent the normal variability
in the process.  It is unlikely to be related to the air levels since,
as it will be seen, the air levels were too low to add significantly to
skin or surface contamination.  Unfortunately, the final answer to this
question was beyond the scope of the work here.  

The supervisor spent most of the day outside the milling room and was
not considered a good candidate for exposure monitoring.  He was
expected to show little contamination because of that.  While his hands
were less contaminated than the beryllium workers’, he nevertheless
had measurable contamination on his hands.  Like the other workers in
the mill room he wore the required personal protective equipment while
in there.  Since surface contamination was found in the office it is
possible that the observed levels on the supervisor’s hands were from
contact with surfaces in that area and not as a result of anything in
the milling area.Air Samples

Table III. Total airborne beryllium concentration

g/m3.  Only one of six total beryllium measurements shown in Table
III had a detectable beryllium level.  This level was found on Subject
3, who was in training during the time of the plant visit.  For Subjects
1 and 2 total dust levels were higher on the second day of the visit. 
The lack of measurable beryllium does not allow the beryllium number
concentration to be accurately determined for anyone but Subject 3. 
What could be surmised was that the percentage of beryllium was less
than that given by dividing the reporting limit for beryllium by the
total mass.  Therefore, deposited beryllium number concentrations (Be
DSP) were lower than the resulting minimum product of that percentage
and the total deposited number concentration (Total DSP) as seen in
Table IV.

Table IV. Airborne deposited particle number concentration

Subject 1, 12/03/02





	 	 	Total DSP1	Be DSP2	 

average

1,765	p/cc	<1 p/cc

	median

2,075	p/cc	<1 p/cc

	standard deviation	2,913



	number of measurements	223



	%Beryllium	 	< 0.05	 	 	 



Subject 1, 12/04/02





	 	 	Total DSP1	Be DSP2	 

average

3,689	p/cc	<1 p/cc

	median

3,740	p/cc	<1 p/cc

	standard deviation	2,761



	number of measurements	372



	%Beryllium	 	< 0.04	 	 	 



Subject 2, 12/03/02





	 	 	Total DSP1	Be DSP2	 

average

1,688	p/cc	<2 p/cc

	median

1,649	p/cc	<2 p/cc	 

standard deviation	259



	number of measurements	15



	%Beryllium	 	< 0.14	 	 	 



Subject 2, 12/04/02





	 	 	Total DSP1	Be DSP2	 

average

4,342	p/cc	<1 p/cc

	median

4,368	p/cc	<1 p/cc	 

standard deviation	707



	number of measurements	460



	%Beryllium	 	< 0.03	 	 	 



Subject 3, 12/03/02





	 	 	Total DSP1	Be DSP2	 

average

1,475	p/cc	<1 p/cc

	median

1,727	p/cc	<1 p/cc	 

standard deviation	684



	number of measurements	247



	%Beryllium	 	< 0.05	 	 	 



Subject 3, 12/04/02





	 	 	Total DSP	Be DSP	 

average

3,483	p/cc	15 p/cc

	median

3,584	p/cc	15 p/cc	 

standard deviation	927



	number of measurements	333



	%Beryllium 	 	 0.42	 	 	 

1Total DSP - concentration of that fraction of all particles that would
deposit in the lung.

2 Be DSP – concentration of that fraction of the Total DSP that is
beryllium.

The lack of any detectable beryllium means that the conclusions that can
be drawn from the data are limited.  The size distribution of particles
that was determined by the two different methods is useful mostly for
comparison with previous work that has been done.  The particle number
distributions in Table V (page 15) therefore represent only the number
size distribution of the total airborne particulate and are not specific
for beryllium.  The number size distribution did not seem to vary much
regardless of the area in which sampling occurred as seen in Figures 2
and 3 (page 16-18).  The best fit to the data was obtained in several
cases using multiple modes when sampling was done outside the
manufacturing area.  This is most likely due to the size distribution of
the ambient air and was of greater magnitude than that contributed by
the manufacturing sources, which resulted in the observed size
distribution fit. The percent deposited in the final column of Table V
was calculated by double integration of the particle number distribution
and the total lung deposition curve to determine what percentage of the
particle counted would have been deposited in the lungs.

The particle mass distribution measurement shown in Table VI (page 18)
failed to detect any airborne beryllium despite the fact that it was run
for 32 hours, straight, at a flow rate of 30 liters per minute.  That
this measurement failed to detect the beryllium found in the single
personal sample should not be surprising.  It is quite likely that the
source of contamination was limited in extent.  The particle mass
distribution, due to the size and weight of the equipment required to
make the measurement, came from an area sample placed in the center of
the 50,000 cubic foot milling room.  Dilution or capture subsequent to
the personal exposure probably resulted in the lack of detection. The
particle mass size distribution was best represented as a multimodal
distribution.  At least two distinct modes could be fit to the data.  It
may be that the multiple modes are characteristic of multiple sources. 
Further conclusions about the source allocation of the particle sizes
are beyond the scope of this work.  The MOUDI sample for the mass
distribution was meant to be used to calculate the surface area
distribution of the beryllium should any be found.  In the limited time
available it was not possible to collect enough material, if it exists,
to make that determination.Table V. Particle number distributions from
the SMPS.

 	 	Geometric Mean Diameter	Geometric	% of Total	%

 	LOCATION	(um)	Standard Deviation	Particle Count1	Deposited2

A.	Five feet from front of the #1 CNC fadal mill.  





	Operator was running “Blanks or Cards” the material is AlBe	0.135
1.55	100	24.4

B.	Four feet from side of Makino A55 mill.    





	At end of run the chip drum was changed near the sampler port.	0.135
1.55	100	24.4







	C.	Four feet from the side of Makino A55 mill.  	0.135	1.53	100	24.4







	D.	5 ft from front of the #1 CNC Fadal mill	0.135	1.50	100	24.2







	E.	5 ft from front of the #1 CNC Fadal mill	0.15	1.50	100	22.7







	F.	Locker Room near changing bench, 	0.15	1.33	80



maintenance worker was cleaning in the area and doing laundry	0.23	1.30
20



	21.2

G.	Sample in lunch room during break	0.125	1.33	80



	0.23	1.30	20







23.4

H.	Sample in office, some minimal foot traffic 	0.132	1.40	90



	0.27	1.40	10







21.1

I.	Outside, ambient air sample, background	0.11	1.32	80



	0.2	1.40	20

	 	 	 	 	 	25.5

	1For multimodal distributions the % represents the fraction of the
total number of particles in a mode with the given mean and standard
deviation. If modes are close in size the graphic representation may not
appear to have multiple peaks.





	2Determined for each mode separately and then weighted by the fraction
in a given mode.











	(Figure 2a-i. Plot of particle number distributions from Table V.)

 

 

Figure 2a-i. Plot of particle number distributions from Table V.  The
letters in each figure correspond to the letters in Table V.  The light
gray line is the actual data from the SMPS and the dashed line is the
corresponding lognormal equation using as its parameters the stated
values of geometric mean and geometric standard deviation from Table V. 
A close match to the data was able to be achieved in all cases, allowing
the estimated deposition fraction to be reliably calculated for each
distribution and an estimate of the surface area distribution of the
beryllium to be made had any been found in the MOUDI sample.



 

Figure 3. Overlay of all SMPS size distributions shown separately in
Figure 2. The light-gray size distribution with its peak shifted to the
left of all the other distributions is the ambient air.  There is little
variation seen graphically between the distributions inside the plant. 
These distributions are much larger than those reported by other
investigators2, although the same piece of equipment was used to collect
both sets of data.

Table VI. Mass Size Distribution

g.

 

Figure 4. Mass size distribution (in the blue histogram), from the
MOUDI, for total dust with the resulting geometric mean and standard
deviation as given in Table VI represented by the red curve with two
modes.

Although the MOUDI used to measure the mass size distribution of
beryllium operated across 4 shifts for 32 hours in the plant, the sample
was divided into 8 stages.  This necessary division of the total sample
meant that the amount of beryllium on any single stage was only twelve
and half percent of the total that could have been collected. 
Nevertheless, the beryllium levels in the air are apparently very low on
average.  The mass size distribution is relatively small for the
dominant mode but further speculation about its source is not possible
with the data collected.  Had there been beryllium particles generated
by the same source this would have resulted in a large surface area to
mass ratio and potentially increased hazard.  Although no beryllium was
found in this sample it would be advisable to continue monitoring the
size of beryllium that is generated when and if possible.

Exposure Controls In Place

Access Control  

The facility has designed and implemented a modular, partitioned
approach to control access and beryllium migration from the machining
process area.  Safety zones are identified as minimum, medium and
maximum.  In the minimum safety zone workers enter the facility, take
breaks, eat meals, and perform general administrative business
activities without any special controls or personal protective
equipment.  The medium safety zone is described as the transition
between the minimum and maximum safety zone.  Worker locker room, shower
and hygiene facilities, personal protective equipment storage, and
clothes laundering capabilities are located in the medium safety zone. 
The maximum safety zone requires full protective gear to be donned
before entering and is where the machining processes take place.

Personal Protective Equipment (PPE) 

Donning

In the medium safety zone workers change out of personal shoes and don
work shoes and a half mask negative pressure air purifying respirators
equipped with high efficiency particulate air (HEPA) filters.  Loose
fitting powered air purifying respirators with HEPA filters are
available and optional to workers.  Prior to entering the maximum safety
zone workers, don disposable coveralls, caps and latex gloves.  Latex
glove are required at all times in the maximum safety zone when workers
are handling beryllium parts or equipment contaminated with beryllium.  


Doffing

To exit the maximum safety zone and enter the medium safety zones
workers remove and dispose of latex gloves in a designated beryllium
waste container, remove and hang disposable coveralls and caps, and
vacuum the tops and bottoms of work shoes with a HEPA vacuum cleaner.  
Workers dispose of coveralls at least every Tuesday and Friday or more
often as necessary.  After entering the medium safety zone workers
remove and store (in covered containers) respirators and are required to
wash hands and face.  

Work Practice Controls

Worker compliance with company established work practices and procedures
is excellent.  Management expectations regarding beryllium health and
safety are clear, certain, and well defined in documentation, training
and ongoing program audits.  Procedures are detailed and address
beryllium control expectations throughout the work place.  Workers
maintain excellent work practice discipline and attention to beryllium
control detail.  The Production Manager and the Vice President of
Operations and Corporate Development conduct weekly and monthly safety
audits of the facility and work practices being employed.  The results
of these audits are documented and tracked to communicate throughout the
organization where expectations are being met and where gaps in
performance are identified and corrected.  

Housekeeping

Housekeeping in all zones of the facility is well-organized.  The site
has established a daily, weekly, monthly and quarterly schedule of
housekeeping control expectations.  The plant was currently in the
middle of these longer cycles, although illness prevented the worker in
charge of cleaning to perform the daily cleaning for the day shift
during the course of this visit.  This schedule is documented and
activities are tracked to ensure housekeeping requirements are
completed. The site employs one full-time maintenance worker who is
dedicated to and responsible for completing the housekeeping schedule. 
Two large (2 to 3 blower/motors) type, HEPA rated vacuum systems are
available for cleaning. 

Local Exhaust Ventilation Controls

Each machining center is completely enclosed with steel framed/Plexiglas
and under negative pressure.  A central 6000 cubic feet per minute (CFM)
exhaust ventilation unit is used to provide local exhaust ventilation
continuously whether the machines are open or closed to all of the
machining centers.  This is a ceiling mounted unit located in a minimum
safety zone which discharges through a HEPA filter into the ambient air.
 The differential pressure across the filters is checked on a daily
basis, by reading a pressure gauge.  The ventilation system is serviced
and maintained by an independent contractor on an annual basis.  The
contractor personnel are trained in hazardous materials and reportedly
understand the control and protection requirements for beryllium.

Discussion

The overall appearance of the plant is one of a well-maintained and
clean working environment.  The lack of any visible settled dust or dry
residue from liquid spills is a credit to good housekeeping practices. 
The lack of any measurable beryllium in most of the air samples is very
likely due to good housekeeping and the full enclosure of the machines
using beryllium, as well as good work practices on the part of the
machinists.  Although the layout of the plant is well-considered, the
overall design is not optimal to retain the beryllium in the working
area.  An optimal design would be one in which beryllium would not be
detected in areas outside the mill room.  

Detection of beryllium on surfaces, particularly in the office, is an
indication that closer scrutiny could be given to practices of moving
materials between the milling room and the office. The door from the
office directly into the milling room offers the temptation of easily
moving a few items between the rooms without going through any
decontamination process.  Paper that is moved between the milling room
and the office should also not be overlooked as a potential source of
cross-contamination.

Changing out from protective gear in the milling room itself presents
another potential problem. The change out area, in the milling room,
next to the door to the change room is not apart from general work areas
in the milling room.  The machinist at Fadal #1 had to move through that
area to access the rear of his CNC mill.  This could result in floor
contamination that is then transferred to work shoes after the workers
remove their shoe covers before leaving the room.

The sticky mat in front of the exit from the milling room is a good idea
for limiting contamination but it may not be sufficient for the number
of people who exit the room at break times.  The mat seemed to be fully
loaded with dirt before the last person exited the room.  Not only could
this result in insufficient decontamination it could be a source of
contamination transfer from the mat to the shoes of the later exiting
workers.

Workers in the milling room also hang their protective outer garments on
hooks on the wall in the milling room before leaving the room.  Again,
as with their shoes, this results in a potential exposure of their work
uniforms before they exit the work area.  Also, because the protective
outer garments are bunched together on the hooks, there is the potential
for contamination to be transferred from the outside of the garment to
the inside.  This could occur within the same garment or between
separate garments.  The contamination could then be transferred to the
personal clothing or the skin of the workers, many of whom wear
short-sleeved shirts under their protective outer garments.

The protective garments themselves, although of a lightweight material,
seemed to increase the wearer’s thermal load.  This may be the reason
why workers commonly wore short-sleeved shirts to work, even though it
was snowing outside. Several of the workers were observed to be working
either with the sleeves of their protective garments pushed up to their
elbows or with the sleeves cut short.  This defeats the purpose of the
long-sleeved features of the protective garments.

Most of the workers did not shower before leaving the facility, although
there were showers available.  They did wear hats in the facility,
presumably to limit hair contamination.  However, the same hats were
always worn.  These could become contaminated over time due to handling,
since beryllium was detected on the bare hands of workers after they
removed their gloves.  This could then result in the hats becoming
sources of contamination rather than protection.

This contamination could also be due to the removal of inadequately
cleaned parts from the CNC mill.  It should be assumed that machined
parts could always contain some residual contamination.  

The Makino CNC mill presents another problem.  The parts must be removed
using tools to loosen screws holding the pieces in place.  The tools
themselves can become contaminated by coming in contact with machined
parts.  The tools are moved back and forth to a cart and could spread
contamination while this is done. 

Removal of contaminated debris from the CNC’s is also a potential
problem.  Chips and shavings potentially contain small debris that could
be easily transferred during the removal operation.  The highest surface
contamination level found, 315 g/100cm2, behind the Makino CNC, was
in the area where this debris was commonly removed.  The next highest
level, 52 g/100cm2, was found behind Fadal #1 in the area where chip
removal was done.

That there was almost two orders magnitude difference between the
outside surface of the gloves and the contamination level on the
workers’ hands is testimony to the effectiveness of glove use.  The
variation between the first and second days of the visit, when there was
more than a five-fold increase in the surface contamination levels on
the bare hands, is a problem.  Workers do change gloves during the
course of the day as they move in and out of the work area.  Gloves can
be removed and put on in the milling room. There may be insufficient
hand cleaning before gloves are put on, which means there is also
inadvertent contamination occurring while the gloves are off.  

General Recommendations

OSHA has developed the following general recommendations for all
beryllium workplaces (OSHA, 1999):

Engineering Controls 

Employers should use appropriate engineering controls and work practices
to ensure that worker exposures to beryllium are maintained below the
current OSHA PELs to the extent feasible. The following engineering
controls and practices should be used by employers: 

enclose processes; 

design and install appropriate local exhaust ventilation; 

use vacuum systems in machining operations; 

use pellets instead of powders wherever possible; 

use product substitution where possible; 

minimize the number of workers who have access to areas where there is a
potential for beryllium exposure; 

monitor employee exposures to airborne beryllium dust and fume, using
personal sampling techniques, on a regular basis to ensure that
exposures are below the PELs and that proper respiratory protection is
being used where necessary. 

Work Practices to Reduce Beryllium Exposure 

Employers should ensure that employees use the following safe practices
to reduce their exposure to beryllium: 

use high-efficiency particulate air (HEPA) vacuums to clean equipment
and the floor around their work areas; 

do not leave a film of dust on the floor after the water dries if a wet
mop is used to clean; 

do not use long vacuum hoses and do not loop the hoses that are used; 

do not disconnect or disable the vacuum system during any machining
operation; 

never use compressed air to clean parts or working surfaces; 

avoid prolonged skin contact with beryllium particulate; and 

do not allow workers to eat, drink, smoke, or apply cosmetics at their
work stations. 

Hygiene and Personal Protective Clothing 

OSHA is aware of CBD cases that have occurred among family members of
beryllium-exposed workers. To reduce "carry-home" exposures, employers
should provide showers, clean work clothes, and clean areas for storing
street clothes. Protective clothing should be provided to employees who
work in areas where beryllium-containing powders are used and where
there is a potential for spills. In addition, employers should ensure
that employees: 

change into work uniforms before entering their work area; 

place their uniforms in a labeled bin with a cover at the end of the
work shift; 

shower and change into street clothes prior to leaving the facility; 

wash their face, hands, and forearms before eating, smoking, or applying
cosmetics; 

keep their work clothes as clean as possible during the workshift; 

wipe off their shoes before leaving the work area; and 

do not wear their work uniform (including their work shoes) outside of
the facility. 

Respiratory Protection 

Recent data suggest that exposures to beryllium even at levels below the
2 micrograms/m3 PEL may have caused CBD in some workers. Therefore,
employers should consider providing their beryllium-exposed workers with
air-purifying respirators equipped with 100-series filters (either N-,
P-, or R-type) or, where appropriate, powered air-purifying respirators
equipped with HEPA filters, particularly in areas where material
containing beryllium can become airborne. 

Training 

Employers should give employees exposed to beryllium training and
information about the following items: 

material safety data sheets (MSDSs) for beryllium; 

the fatal lung disease that may occur as a result of exposure; 

the availability of the BeLPT blood test to determine whether an exposed
worker has become sensitized to beryllium; 

the potential for developing lung cancer as a result of exposure; 

the importance of avoiding skin contact; 

the engineering controls the employer is using to reduce worker
exposures to beryllium; 

specific work practices that can be used to reduce exposure to
beryllium; 

the use of appropriate protective equipment, including the use of
respirators; 

the results of any industrial hygiene sampling for levels of beryllium
in the workplace; and 

a copy of all pertinent Hazard Information Bulletins from OSHA. 

Health Screening Methods for Beryllium Sensitization and Chronic
Beryllium Disease 

Employers should consider sending beryllium-exposed employees to a
physician or other licensed health care professional to be evaluated for
beryllium sensitization or the presence of CBD. The screening
examination for CBD usually begins with a chest x-ray and a blood test
for beryllium sensitization, namely, the BeLPT, plus any further
evaluation considered appropriate by the health care professional. The
blood test can detect an adverse health response to beryllium exposure
earlier than breathing tests or chest x-rays can. The BeLPT is not
routinely done in most medical laboratories; however, the health care
professional may order this test from any laboratory that has overnight
courier service to one of the Medical Research Centers listed below. If
a worker is sent to a health care professional for health screening, a
copy of OSHA’s Hazard Information Bulletin 19990902, Preventing
Adverse Health Effects From Exposure to Beryllium on the Job (OSHA,
1999) should accompany the employee.

Employees who work in a place where beryllium is used and have developed
any of the symptoms listed below, should inform their health care
professional of their past beryllium exposure, or seek information from
a health care professional who specializes in occupational lung diseases
to determine whether they may have developed CBD: 

unexplained cough, 

shortness of breath, 

fatigue, 

weight loss or loss of appetite, 

fevers, and/or 

skin rash. 

If they do not have any of the above symptoms but are concerned that
they may have become sensitized to beryllium, they should inform their
health care professional that they would like to be tested with the
blood BeLPT. They should also take a copy of the above-mentioned Hazard
Information Bulletin1 with them.

Specific Recommendations

The current OSHA Permissible Exposure Limit (PEL) for beryllium in air
is 2.0 µg/m3.  OSHA considers this level to be insufficient to protect
all workers.  OSHA has not yet decided on a level that will be
sufficient.  In the interim OSHA advises that levels be maintained as
low as feasible.  Other federal agencies such as the Department of
Energy (DOE) have requirements for 0.2 µg/m3  for beryllium in a total
dust sample (10CFR850).  They also require protective clothing and
equipment where surface contamination levels are above 3 µg/100 cm2. 
Housekeeping efforts must maintain removable surface contamination at or
below 3 µg/100 cm2 during non-operational hours. Removable
contamination on equipment surfaces must not exceed 0.2 µg/100 cm2 when
released to the public or for non-beryllium use. Removable contamination
on equipment surfaces must not exceed 3 µg/100 cm2 when released to
other beryllium handling facilities. This may not be sufficient to
adequately protect workers.  

 

The cause for this level of concern is the most recent information on
prevention of sensitization, which seems to indicate that skin exposure
and its source, contaminated surfaces, may be important.  Beryllium
sensitization is thought to be a precursor for chronic beryllium disease
in the presence of sufficient ultrafine particulate dose. However, it
has been hypothesized that routine dermal contact with beryllium
contamination results in beryllium sensitization in the unprotected
genetically susceptible worker. In a poster presentation at the American
Industrial Hygiene Conference and Exhibition (ref. 4) it was shown that
wipe samples, collected to measure the quantity of total beryllium on
the hands and surfaces of production and support personnel in a copper
beryllium rolling mill, had an average concentration (geometric mean) of
beryllium on the hands of production personnel of 0.23 (g/100cm2 while
support personnel had significantly less at 0.06 (g/100cm2  (p ( 0.005).
 However, the difference between skin integrity for support personnel
versus production personnel was not significantly different (p = 0.86). 
Personnel were also evaluated for number of times their hands were
washed.  The average number of hand washings for production personnel
was not a statistically significant difference (p = 0.31) at 4.1 for
support vs. 3.7 for production personnel.  The average beryllium air
sample level was 0.033 (g/m3 for production and 0.030 (g/m3 for support,
not significantly different at p=0.05.

A total of 255 surface wipe samples were also collected from the
production, support, and common areas of the plant.  Work surfaces
touched by production personnel had higher concentrations of beryllium
(3.76 (g/100cm2) than surfaces touched by support personnel (1.03
(g/100cm2 ). This was a statistically significant difference between the
geometric means of surface wipe samples collected in production areas
and support areas (p = <0.005).  There was no sensitization among the 47
support workers while there were 10 sensitized workers among 87
production workers.  It would appear that surface contamination and
subsequent hand contamination may be responsible for sensitization.  A
housekeeping standard of less than 1.0 (g/100cm2 on surfaces would
therefore seem appropriate to better prevent sensitization.  This is
less than the value currently enforced by DOE, but the workers examined
(ref. 4) were not wearing personal protective equipment such as gloves
or disposable coveralls, a requirement in areas above 0.2 (g/100cm2  in
DOE facilities.  

Air levels should include some measurement of the deposited
submicrometer beryllium particle number concentration.  It has been
hypothesized that these may be related to the risk of disease. (refs.
20, 21)  Concentrations above 1 particle per cubic centimeter of air
have been found in areas where workers have been found with CBD, so
maintaining exposure levels below that value could be prudent.  It is
possible, though still somewhat unwieldy to make that determination
using the procedures employed by this study.  A simpler alternative is
to begin to focus on those areas that might have the greatest potential
for problem.  To do this most efficiently it is possible to use logic to
arrive at a simpler screening tool.  The use of this screening tool is
meant only for rapid determination of the most problematic areas and is
not meant as a long-term solution to take the place of actual particle
number sampling.  The logic for this method is:

The target concentration above which action should be taken is 0.1 p/cc
of deposited submicrometer beryllium, one-tenth the level in areas where
disease has been found.

The lower the total particle number concentration the less likely the
target will be exceeded.

The lowest reasonable particle number concentration that should be
expected, based on experience, is approximately 1000 p/cc.

The lower the beryllium content of the dust, the less likely the target
value will be exceeded.

The lowest reasonable beryllium concentration for processes that are
emitting beryllium, based on experience, appears to be approximately
0.01% (1 part per ten thousand).

The higher the total mass of all dust the lower the beryllium content of
the dust will be and subsequently the less likely the target value will
be exceeded.

In a well controlled environment, such as that in which beryllium work
should be done, the highest total submicrometer mass would likely be
less than 100 (g/m3 .

An analytically quantifiable lower limit of 0.01(g of beryllium per
sample seems feasible in most cases of filter sampling.

At least one cubic meter of air can be sampled during an eight-hour work
shift.

Given the above assumptions, one can calculate that the detectable limit
of 0.01g/m3 (h) is 0.01% (e) of 100 g/m3  (g).  Thus the
detectable level (if sampled in one cubic meter of air) divided by the
highest total submicrometer mass also gives the lowest percentage likely
to be expected for a beryllium operation.  Further, this percentage,
however arrived at, applied to the lowest total particle number
concentration that is expected, 1000p/cc (c), gives the target value of
0.01 p/cc.  Therefore any submicrometer beryllium concentration higher
than the detectable level is likely to result in a value that exceeds
the target value if particle number concentrations are measured
directly. A commercially available, single stage personal impactor can
be used to separate the particles into a fraction less than one
micrometer using a personal sampling pump.  If beryllium is detected in
that fraction further effort should be expended as soon as possible to
determine the actual deposited submicrometer beryllium particle number
concentration and appropriate remediation actions taken to alleviate the
problem.  If beryllium is not detected in that fraction, further study
could be postponed until those areas with detectable levels have been
identified.

g of total dust may require sampling several consecutive shifts.  No
final determination of an area as being beryllium free in the air should
be made without collecting at least enough sample to give a total weight
change of 100g.  This could take 40 hours of sampling in a relatively
clean environment.  Without that mass of material, however, it may be
possible to be under the detectable limit of 0.01g if the beryllium
content is as low as 0.01%.]

This site visit was not intended to serve as a comprehensive industrial
hygiene review. Thus, investigators did not evaluate all possible
hazards at Site 1.  However, during the course of the visit,
investigators noted several opportunities for reducing surface
contamination. These are presented below. 

To optimize migration control, the design layout and sequence of work
practices are important during transitions.  The following
considerations might be incorporated into current or future facility
designs:

Company work uniforms should be provided and used.

Ideally, there should only be one designated/routine entry into the
milling area (maximum safety zone).  

The transition zone should be clearly marked and if possible enclosed. 
It should only be used only for the purpose of preparing workers,
material or equipment for entry or exit.

Tack mat floor covering should be designed for 3 foot falls per foot to
increase effectiveness.

Disposable protective gloves should on before handling any potentially
contaminated articles.

Disposable gloves should be on at all times inside the transition area
and maximum safety zones.  Gloves should only be removed just before
exiting the transition area into the medium safety zone.

Glove disposal containers should be located at the transition zone.

Containers of clean disposable gloves should be available at the
entrance into the transition area and at operator workstations. 

Hand wipes should be available wherever clean gloves are stored to
facilitate hand washing prior to donning clean gloves.

Benches for seating should be provided at the transition area to
facilitate shoe cleaning and shoe cover donning and doffing.

An alternative to cleaning and covering work shoes is to institute a
shoe change capability.  Dirty work shoes would be stored in covered
containers in the transition area and interim company provided footwear
would be used in the medium and minimum safety zones.

Trashcans should have lids with foot pedals.  This will alleviate the
temptation of workers to throw potentially contaminated trash from any
distance and will also make it unnecessary to touch the lid to open it.

The door between the office and the milling room or between the office
and the lunchroom should be removed and either walled over or opened to
the plant.  The plant should decide whether the office is part of the
beryllium work zone or the outside and act accordingly.

There is space in the lunchroom to use for expansion of the changing
room.  This expansion could include separate entrance and exits into the
milling room.  A keyed entry into the milling room from the locker room
would be preferable with the door only able to be opened from the locker
room side.  The exit door would lead a person through a change room and
then the showers before exiting into the locker room (Figure 5).  If
this space is not reconstructed as shown there should at least be a
separate, walled and well ventilated area in which the outer garments
are removed and stored, as mentioned above.  This area should not be
part of the Milling Room.  More space should be made for the garments to
hang so that there is no crowding of garments.

All porous surfaces should be eliminated to the extent feasible, unless
they are disposable and are disposed of regularly, in that they may
accumulate beryllium dust.  This especially includes fabric-covered
chairs.

Tools used on the Makino CNC should be rinsed and dried after use.

Foot cross over designs and techniques should be considered at the bench
to control transfer of shoe contamination from the mill area to the
transition area. 

All reusable PPE should be placed in covered storage containers.

Showers should be taken at the end of the work shift.

Airflow should be modeled between the minimum safety zone, medium safety
zone, transition area, and maximum safety zone to evaluate appropriate
airflow patterns.

Workers and management should closely study the work practice sequences
at the transition zone.  Identify all articles having a potential for
contamination and develop work practice sequences that minimize
migration of beryllium particles from the maximum safety zone into the
transition area and in turn into the medium safety zones.

Because there is the possibility that the company involved in this visit
may expand current operations into a new facility it is recommended that
an alternative to the current layout of the factory be considered.  The
new design would consist of four zones which, for ease of referral, will
be coded by color; red, orange, yellow and green.  

The red zone will be that area in which active beryllium machining work
is done.  This zone will require the strictest controls and limited
access to all but essential personnel.  Company work clothes will be
required in this area.  

⨀ ␃ഀ䃆

옍)

@

kd0

kdð

y欀灤

＀혛８￿￿￿￿￿￿￿￿￿￿￿᳿ᣖ￿￿￿￿￿￿￿￿
￿￿￿￿혝８￿￿￿￿￿￿￿￿￿￿￿㋿ۖĀ̋

ᨀᣖ￿￿￿￿￿￿￿￿￿￿￿￿혛８￿￿￿￿￿￿￿￿
￿￿￿᳿ᣖ￿￿￿￿￿￿￿￿￿￿￿￿혝８￿￿￿￿￿
￿￿￿￿￿￿㋿ۖĀ̋

ᨀᣖ￿￿￿￿￿￿￿￿￿￿￿￿혛８￿￿￿￿￿￿￿￿
￿￿￿᳿ᣖ￿￿￿￿￿￿￿￿￿￿￿￿혝８￿￿￿￿￿
￿￿￿￿￿￿㋿ۖĀ̋

ᨀᣖ￿￿￿￿￿￿￿￿￿￿￿￿혛８￿￿￿￿￿￿￿￿
￿￿￿᳿ᣖ￿￿￿￿￿￿￿￿￿￿￿￿혝８￿￿￿￿￿
￿￿￿￿￿￿㋿ۖĀ̋

ᨀᣖ￿￿￿￿￿￿￿￿￿￿￿￿혛８￿￿￿￿￿￿￿￿
￿￿￿᳿ᣖ￿￿￿￿￿￿￿￿￿￿￿￿혝８￿￿￿￿￿
￿￿￿￿￿￿㋿ۖĀ̋

ᨀᣖ￿￿￿￿￿￿￿￿￿￿￿￿혛８￿￿￿￿￿￿￿￿
￿￿￿᳿ᣖ￿￿￿￿￿￿￿￿￿￿￿￿혝８￿￿￿￿￿
￿￿￿￿￿￿㋿ۖĀ̋

ᨀᣖ￿￿￿￿￿￿￿￿￿￿￿￿혛８￿￿￿￿￿￿￿￿
￿￿￿᳿ᣖ￿￿￿￿￿￿￿￿￿￿￿￿혝８￿￿￿￿￿
￿￿￿￿￿￿㋿ۖĀ̋

＀혝８￿￿￿￿￿￿￿￿￿￿￿㋿ۖĀ̋

＀혝８￿￿￿￿￿￿￿￿￿￿￿㋿ۖĀ̋

＀혝８￿￿￿￿￿￿￿￿￿￿￿㋿ۖĀ̋

＀혝８￿￿￿￿￿￿￿￿￿￿￿㋿ۖĀ̋

＀혝８￿￿￿￿￿￿￿￿￿￿￿㋿ۖĀ̋

＀혛８￿￿￿￿￿￿￿￿￿￿￿᳿ᣖ￿￿￿￿￿￿￿￿
￿￿￿￿혝８￿￿￿￿￿￿￿￿￿￿￿㋿ۖĀ̋

＀혝８￿￿￿￿￿￿￿￿￿￿￿㋿ۖĀ̋

਀&䘋

ꐓdꐔd⑛封Ĥ਀&䘋ꐓdꐔd⑛封Ĥ਀&䘋

਀&䘋

ny material from the red zone into the orange zone.  Nothing should
leave the red or orange zones without thorough decontamination.  Paper
and porous materials may not be able to be adequately decontaminated. 
Therefore, its use in the red zone should be minimized.  Preferably
there should be a barrier separating these zones.  Coming from the red
zone into the orange zone should require the donning of some protective
covering, including shoe covers to limit migration of beryllium from the
red zone. This protective covering should be removed and remain in the
orange zone.  While one time disposable use of the outer garment would
be best, cost effectiveness may require some reuse. Care should be taken
if this garment needs to be stored not to transfer contamination from
the interior of the garment to the exterior.  Windows and intercom
systems should be available between all zones.  As little as possible
should be physically transferred between them.  The level of personal
protective equipment used in these zones should be a matter of
confidence based on the engineering controls and a history of industrial
hygiene sampling.  A lunchroom and restrooms in the orange zone should
be provided for the red zone workers.  

The yellow zone should be that area by which workers leave the orange
zone.  In the yellow zone, showers should be required before exiting, as
well as a change of work clothes and shoes. 

The green zone can contain the workers’ lockers with their street
clothes and clean work clothes into which they can change at the
beginning of the shift. 

Offices can be placed in either the green or orange zones depending on
the level of personal protective equipment that is preferable.  It is
assumed that airborne levels of beryllium can be made negligible in the
orange zone and that the primary concern will be surface contamination
and skin protection.  

To encourage workers to use the shower and change facilities, privacy
should be considered.  The number of workers involved in beryllium
machining is small enough that private shower and changing cubicles
might be made available for most.  Workers should be given time as part
of their shift to make the transition between areas.  Company clothes
can be worn without disposable covering in the red zone, but laundry
facilities used to clean these clothes should be made aware of the
potential for beryllium contamination.  Alternatively, a cover garment
could be worn over the company clothes to limit the amount of
contamination that occurs, but the company should be aware that this
covering might not be completely effective and some contamination may
remain on the clothes underneath.  In either case, the clothes that are
worn should be company clothes so that the workers are required to
change before leaving the plant.

 

Figure 5. Four-zone concept applied to current facility. Arrows
designate doors.  One-sided arrows designate doors able to be opened
only from one side.

References

(OSHA, 1999)  Occupational Safety and Health Administration (OSHA) - 
Health Intelligence Bulletin. Preventing Adverse Health Effects From
Exposure to Beryllium on the Job. September 2, 1999.

McCawley, M.A., M.S. Kent and M.T. Berakis, 2001. Ultrafine beryllium
number concentration as a possible metric for chronic beryllium disease
risk.  Appl. Occupational and Environ. Hyg. 16(5): 631-638.

Preliminary Draft – Do not quote or cite.                        
March 3, 2003                       Page #    PAGE  1 

ABRASIVE 

BLASTING

GLOVE BOX

SANDING

GLOVE BOX

STORAGE

LUNCH ROOM

MILLING ROOM

OFFICE

CHANGING ROOM

FUTURE

PRODUCTION

EXPANSION

CEILING MOUNTED

EXHAUST

VENTILATION

HARDINGE

LATHE

FADAL#3

FADAL#2

MAKINO

FADAL#1

X

X

X

X

X

X

X

X

X

X

3

2

1

X

X

X

X

X

 X

X

X

X

X

X

X

X

X

X

X

 

Entrance

Lunch Room and Break Area

Parts

Storage

Locker Room

Laundry

Hygiene

Facilities

Future

Production

Expansion

Hardinge

Lathe

Transition

Fadels

Makino

Ceiling

Mounted

Exhaust 

Ventilation

Sanding

Glove

-

box

Abrasive

Blasting

Glove

-

box

Entrance

Lunch Room and Break Area

Parts

Storage

Locker Room

Laundry

Hygiene

Facilities

Future

Production

Expansion

Hardinge

Lathe

Makino

Ceiling

Mounted

Exhaust 

Ventilation

Sanding

Glove

-

box

Abrasive

Blasting

Glove

-

box

