MEMORANDUM

TO:  Portland Cement NESHAP Docket

FROM:  Keith Barnett, USEPA/OAQPS/SPPD/MMG

DATE:  October 22, 2007

SUBJECT:  May 17, 2007, Meeting between US Environmental Protection
Agency (EPA) and Portland Cement Association (PCA) and cement industry
representatives at the Monaco Hotel, Portland Oregon.   

Introduction and Objectives

The purpose of the meeting was to describe EPA’s current regulatory
activities as they relate to the portland cement manufacturing industry
and to solicit discussion and input regarding these activities from the
industry representatives. Current regulatory activities include the
NESHAP for portland cement, the review of the NSPS for portland cement
manufacturing, and the cement sector project. 

Meeting Participants

Keith Barnett and Elineth Torres, USEPA

Andy O’Hare, Portland Cement Association

Numerous cement company representatives.

An agenda and meeting handouts are attached.

Meeting Summary

PORTLAND CEMENT NESHAP

A. Analysis of Section 114 Data

Confidential Business Information (CBI) – To avoid disclosing data
that has been claimed proprietary by the submitter, EPA plans to
randomly assigned kiln ID numbers that will not be traceable back to the
actual kiln. 

At this time, EPA is not planning to review and make determinations
regarding the claims of confidentiality, but will work with the data as
it is and take the necessary measures to protect confidentiality.

Format of emissions standards – Current plans are to use the mercury
concentrations and raw materials usage data to calculate emissions in
one of the following formats: 

Pounds of mercury per ton of raw feed, or

Pounds of mercury per ton of total raw material input (feed & fuel) to
kiln

EPA asked for feedback from the PCA on this.

Data Analysis – Once EPA has completed inputting the data into the
database (end of October), EPA plans to have each company review their
own data in order to confirm that the data have been correctly
interpreted and entered. It may also give each company the opportunity
to address any apparent anomalies in the data or calculated emissions.
The negative emissions calculated for one kiln was cited as an anomaly
that could be cleared up when each company reviews its own data.
Although, as PCA pointed out, the negative emissions may not be due to
faulty data but a result of the limits of a 30-day sampling period that
may not accurately reflect longer-term performance.

PCA suggested that EPA take advantage of mercury data that have been
collected for a longer period as part of the waste-derived fuels program
in the Office of Solid Waste.  (Note – these date were for hazardous
waste only and not suitable for this project)

PCA requested that EPA use the test data that was submitted as part of
the section 114 ICR and compare it with the calculated mercury
emissions.

 

Emission Standards – EPA is planning to look at various
subcategorization scenarios in order to properly characterize factors
that affect mercury emissions. For example, EPA is planning to consider
subcategorization based on the mercury content of the materials mined at
each kiln/plant quarry, i.e., on site. PCA also suggested that EPA
consider subcategorization based on the geographic or spatial
distribution of kilns.

Once an appropriate method of subcategorization is identified, EPA will
rank kilns within each subcategory by their emissions (e.g., lbs
mercury/ton of feed or raw material input) and identify the mean or
median value of the top 12% (or top 5 if there are fewer than 30 kilns
in the subcategory) as the emission limit for existing sources and the
lowest emission rate as the limit for new sources.

PCA suggested that EPA provide an equipment alternative to the numerical
emissions limit as was done in the December 2006 rule.

Non detects – For results reported as nondetects, the values entered
were equal to the detection limits (in most cases, 20 ppb for mercury
and 500 ppm for TOC). Typically, EPA uses one-half the detection limit.
PCA suggested that EPA do sensitivity analyses to determine the impact
on results.

Variability – PCA asked how EPA plans to use the variability in the
data. EPA will talk with OSW regarding their use of variability on the
hazardous wasted combustor rule.

Total Organic Carbon (TOC) - PCA recommended that EPA retain the use of
different emission limits for Greenfield and brownfield sources. 

PCA also noted that EPA be aware that the results from THC emissions
testing will differ from the THC CEM data. EPA will compare the two to
see how they track.

EPA noted that the TOC concentration data are being used for purposes of
developing subcategories; THC test and/or CEM data will be used to
develop emission limits.

Discussion of THC or VOC is a better indicator of the organic pollutants
of concern. THC is problematic in that there are limited options for
reducing organic emissions. PCA confirms that raw materials are source
of organic emissions.

Schedule - EPA is planning to propose amendments by March 2008 and
promulgate final rules by the end of 2008. The implementation period for
the amendments for existing sources will be 3 years and upon start up
for new sources.

B. Area Source Standards

EPA explained that under EPA’s urban air toxics program, emissions of
PM, THC, dioxin/furan, and mercury must be regulated. EPA is planning to
add area sources regulations when it proposes amendments to the NESHAP.
EPA believes that all area sources would be able to meet the PM limit of
0.3 lb/ton of feed.

PCA asked what, if any, advantages there would be with being an area
sources once emissions standards were extended to area sources. They
asked if area source standards would apply to both kilns and coolers and
whether the General Provisions would also apply. PCA stated that not all
kilns could meet the 0.3 limit. EPA stated that the General Provisions
would apply to area sources and will review the 112(k) inventory to
determine if coolers at area sources need to be regulated. EPA also
stated that area sources would be subject to GACT and may not be subject
to all of the major source requirements (such as reporting and
recordkeeping).

C. Scrubber Testing

EPA reported that based on results from 2 tests, scrubbers had no
effects on total mercury with the raw mill on. During raw mill off
periods when mercury emissions are greater, the removal efficiency is
high. Citing information from a facility that reportedly operates a
scrubber to control SO2 emissions when the raw mill is off, EPA asked if
it was feasible to operate a scrubber only during periods when the raw
mill is off. PCA responded that it was not feasible to run a scrubber on
such a limited (as little as 10% of time) basis. Potential problems such
as clogged nozzles and sudden unexpected shut downs would require a
continuously operating scrubber. PCA suggested that the facility that
EPA was referring to was actually using lime injection and not a
scrubber. They doubted that lime injection could remove mercury.

EPA noted that as a result of the brick MACT ruling, it was unlikely
that EPA would consider a scrubber as new source MACT; instead, the best
performing source will be the new source MACT.

EPA stated that it was questioning whether the temperatures in the
finish mill were high enough to result in mercury emissions where gypsum
generated by the scrubber was added at the finish mill.

Scrubber tests are due to be completed in December 2007 with the reports
completed in 2008.

D. Implications of the Boiler MACT Decision

Based on the boiler MACT ruling, cement kilns burning solid waste might
eventually be regulated under section 129 of the CAA. The decision as to
what constitutes solid waste and whether tire-derived fuel would be
considered solid waste will determine whether kilns burning tires or
other waste will be regulated under section 129 or 112. All nonhazardous
waste kilns will be subject to section 112 rules until section 129 rules
are written, which are likely several years away.

EPA reviewed the differences in rules under section 112 and 129. A
decision by EPA on cement kilns is due by the end of 2007. 

PCA is interested in working with EPA on any future 129 rules that would
affect cement kilns. 

E. Residual Risk Rules

EPA explained that an ANPRM would be issued in December that contained
National Emissions Inventory data for the Portland cement industry and
that the notice would request that the industry review the information
for correctness. The NEI data would be used to determine residual risks
levels following the 1999. EPA said it would also update the NEI as a
result of new emission limits resulting from the amendments in response
to the reconsideration. EPA is also working on improving speciation data
for chromium and mercury using speciated chromium data from the OSW
rulemaking for hazardous waste combustors and speciated mercury data
submitted as part of the section 114 ICR.

NSPS REVIEW

EPA Data Needs – Data are needed for a limited number of kilns that
are well designed and that are using newer technologies to reduce
emissions of criteria pollutants. EPA has identified several kilns that
have been built or reconstructed in the last 10 years as the potential
kiln population of interest. EPA assumes that any new kilns will be
preheater/precalciner kilns, but is also interested in knowing if any
existing wet or dry kilns would ever be rebuilt and therefore become
subject to the NSPS. EPA will consider different regulations would be
required for new PH/PC kilns and reconstructed dry and wet kilns. EPA
distributed a list of the recently built or reconstructed kilns, data on
recent BACT determinations and associated NOx controls. 

EPA distributed a copy of a draft ICR and requested feedback on the
information requested. PCA was interested in determining if there was a
way to limit the amount of new information that would be collected. EPA
stated that at a minimum, test data and cost data were needed. Enough
data are needed to sufficiently reflect variability.

EPA plans to propose amendments in May 2008 and promulgate in 2009.

NEXT STEPS

The following areas were identified as needing follow-up by EPA or PCA:

Variability

Non detect limits

Subcategorization

Regulations consisting of technology and numerical limit options

Area sources – GACT vs MACT

Scrubbers and  mercury emissions from finish mills

Sectin 112 vs 129 rules – future meetings with OAQPA and OSW and PCA

Residual risk ANPRM

NSPS and ICR – simplified way of supplying data

AFTERNOON MEETING

Output-based rules - The NSPS for cement is an input-based rule, i.e.,
lb PM/ton of raw feed. EPA is considering an output based rule, i.e., lb
PM/ton of clinker. EPA thinks this format may be a more effective
format. Many permit limits, e.g., NOx limits, are an output based format
(lb/ton of clinker).

Sector based rules – PCA asked if existing state and federal
inventories could be used as part of EPA’s efforts to streamline
cement rules. PCA asked if sector-based rules would be in addition to
all other rules or would they integrate rules from EPA’s other
programs. It was suggested that the sector’s program sounded similar
to the title V program.

 PAGE   

 PAGE   5 

