AWMA Conference May 2002

Mike Harrell of Ash Grove Cement chaired session 1, entitled “Emissions and Compliance Monitoring”. The first speaker was Curtis Leslie of Trinity Consultants. Curtis talked about “Continuous Monitoring System Performance Reporting-Defining CMS Downtime”. He noted that there are 9 common monitoring systems between the MACT waste combustors and Portland cement non-haz combustors. There are also required reports, which include CM downtime report, excessive exceedance and CMS Performance Summary. Brian Graf raised a question at this point and pointed out that the malfunction definition (sudden, infrequent and reasonably not foreseeable) could be a problem.

The next speaker was Paul Peters of Lafarge Corporation. Paul talked about “Monitoring, Record Keeping and Reporting at a Cement Plant”. His talk included continuous monitoring, data flow and averages. He detailed and discussed all of the data collection and collaboration at the Fredonia plant and what the kiln operator sees. Paul’s presentation was well received and he received numerous requests following the session. Brian Graf, keeping the audience attentive, asked about whether or not drift correction was required as Essroc at Logansport flags the data when it does drift.

Steve Fryberger of AirSource Technologies ended the morning speakers by talking about “Trial Burn Projects: Issues Associated with Extended Sampling Run Times”. Steve talked about improving detection limits in the quest for area source versus major source. He also noted that FTIR is used to look for formaldehyde, a potential glitch in the area source designation. In reviewing a hypothetical overall run duration of five hours, Steve presented almost mind numbing detail of a compliance test. This presentation would have been better suited for a novice training session rather than an audience full of compliance testing veterans. Then it was time for the mid-morning break.

Following the break, Mike Harrell performed double duty as the continuing Session 1 Chairman and a presenter. Mike presented “Observations During a CPT of Mercury, Lead, Chromium and Chlorine Emissions from a Modern Cement Plant During Mill-on and Mill-off Operations”. (At the time of Mike’s presentation, only his Chanute facility and the Lafarge Paulding facility had met initial waste combustor MACT requirements). The new facility is a five stage preheater with a downdraft calciner, an in-line calciner and a baghouse. Test results indicated that the vertical mill keeps the SVM and LVM emissions comparable. They had to guesstimate clinker and CKD output so metals balance was based upon calculated output rather than actual. They did try to do long run times in order to decrease mercury detection limits, but mercury still was problematic because such a small amount was spiked and the raw mill on fed back into the raw mill off calculations.

Session 1 came to a close with a presentation by Ken Herstowski of Region VII. Ken gave an “Overview of Alternative Monitoring Requests Received by Region VII”. Some examples included the calculation of rolling averages for different modes of operation, minimum combustion chamber temperature limit methodology, alternative CEMs locations and using EEE testing for LLL compliance. The primary point of his talk from an operational and regulatory stand point was that alternative monitoring requests must be included in the Comprehensive Performance Test Plan and must contain supporting documentation.

Session 2, entitled “Facility Operations and Combustion MACT”, was chaired by Jim Woodford of Gossman Consulting, Inc., who did not present a paper this year for the first time in over a decade. The first speaker of Session 2 was Rex Coffman of TXI who spoke about “Operational Issues Facing TXI Cement”. TXI is in the process of building a new facility in Midlothian, TX, which will have hazardous waste cement kilns and non-hazardous waste cement kilns. He explained that it seemed to boil down to “how many balls can you keep in the air at one time.” Someone in the audience muttered something about size as well but we’ll let that go for the purposes of this conference summary.

The next speaker was Dave Constans of Gossman Consulting, Inc. He spoke on “The Difference in Implementation between the HWC MACT and PC MACT as it Affects HWC Cement Kilns Complying with Both Regulations”. The paper presented the differences between the PC MACT regulations and the HWC MACT regulations as these regulations impact the cement kilns attempting to comply with both regulations. In particular, highlighting the needed changes in implementation required to allow a hazardous waste burning cement kiln to comply with the kiln or kiln/inline raw mill provisions of the PC MACT regulation when not combusting hazardous waste. This paper is on the GCI website at http://gcisolutions.com/dlcawma02.htm.

Next Dave Gossman, also of Gossman Consulting, Inc., spoke about “The Use of a Comprehensive Facility Operations Review and HAZOP Study to Limit Liabilities and Risk in the Operations of Hazardous Waste Fuel Facilities”. Dave pointed out that both operating costs and large liabilities that some facilities have experienced can be avoided by having such a study performed.

The final speaker of the Session and of the morning was Tim Weible of Lafarge Corporation. Tim spoke about an “HWC MACT Operator Training and Certification Plan for a Cement Plant Manufacturing Facility”. Tim’s presentation represented a step-by-step process for the Paulding facility and provided a ready template for other facilities in the process of meeting related regulatory requirements.

Session 3, entitled “Risk Assessments”, was co-chaired by Gary Pascoe of EA Engineering, Science & Technology and Benjing Sun of Exponent. The first speaker of the afternoon, speaking to a room full of people who were full from lunch, was Steve Zemba of Cambridge Environmental. While not part of his presentation, most recently, Steve had worked on a risk assessment for the Lone Star Industries, Inc. Greencastle facility in Region V. Steve spoke on “Particulate (Composition) Matters”. He talked about PM 2.5, asking “why is it harmful?” And then discussed other issues that are happening simultaneously with PM 2.5. As part of this discussion he included sulfate “toxicity,” pointing out that sulfate salts are often used for inhalers and that animal studies show that it does not cause cancer. So many factors, a lot to learn. He chastised EPA in that they are assessing the PM 2.5 as the lone culprit by pretending that PM 2.5 is all one entity.

Co-chair Benjing Sun gave the next presentation, entitled “Comparison of Site-specific vs. EPA Default Parameters for Particulate Size Distribution in Air Dispersion Modeling on Multi-Pathway Risk Assessment Results”. Unfortunately, after reading the title to his paper, he had used up all of his presentation time. But seriously folks… The one significant item that could be taken from his presentation was that facilities should collect their own particle size dust data.

Michael Peterson of Intertox spoke next about “Identification and Risk Assessment of Odorous Chemicals Associated with Combustion Processes”. The study analyzed odor(s) that found 40 sulfur compounds and 43 volatile organic compounds. Both upwind and downwind samples found benzene in similar concentrations. However, the benzene, toluene, ethylbenzene and xylenes found downwind matched gasoline. He pointed out that the “medical literature suggests that carbonyl sulfide acts like the body’s own neurotransmitter.” But the long and the short of it was that there was no increased risk.

At this point in the conference, since we were off schedule and other issues came into play, David Weeks gave his presentation originally scheduled for after the break, “Status of Multi-pathway Risk Assessment Post MACT Proposal”. All due respect to David Weeks, he once again pounded on risk assessments. He lamented that it had been eight years since the combustions strategy (again, anti-combustion strategy is a better descriptor) and since then there had been guidance documents, MACT rules were coming to a climax and there had been lots of conferences, but where was the emphasis on risk assessments? (These lamentations could easily paraphrase an old jazz song, “Is you is, or is you ain’t necessary.”) David continues to stress that the only insurance policy in all the testing is to do a comprehensive risk assessment. One can only take so much risk assessment discussion so at this point it was time for the afternoon break.

Craig Evans of PADEP magically appeared after the break to give his previously scheduled talk on “Pennsylvania’s Accumulative Exposure Project”, which was a follow-up presentation from a previous waste combustor conference. He presented results for the city of Chester, which had five facilities of concern, no NAAQS exceedances but did have arsenic exceedances from an incinerator. DEP had agreed to look at ambient air concentrations and ended up looking at 3 communities for VOC, PM-10, metals and HC. They performed an inhalation risk assessment where it turned out that benzene, 1,3-butadiene and carbon tetrachloride were risk assessment drivers. The risk hovers around one in 10,000.

The next speaker was Dr. Lucy Frasier of ENSR International. Her presentation addressed an “Acute Toxicity Benchmark for Nickel Oxide”. She pointed out that nickel sulfate is the current benchmark, yet nickel oxide is in combustion emissions and is actually less toxic than nickel sulfate, hmmmm. Her main point was that distinctions need to be made when evaluating toxicity.

The final speaker of Session 4 was the other co-chair, Gary Pascoe, who spoke on “Site-specific bioaccumulation Factors for Mercury: Impact on Health Risks from Fish Ingestion”. As anyone who has ever been involved in risk assessments for combustors knows, mercury is the driver in such assessments. Gary pointed out that the waste combustor MACT relies on four studies from northern lakes and consequently there are limitations when compared to other lakes in other parts of the country. In his example, site specific data resulted in a 0.1 hazard quotient versus a 0.5 hazard quotient based upon EPA guidance {say, didn’t David Weeks, formerly from USEPA Region VI help develop that guidance?}.

The final Session of the day was Session 4, entitled Evaluation of Hazardous Waste Facilities. The chairman of this session was Bob Schreiber of Schreiber, Yonley and Associates. The first speaker was Brian Graf of Essroc in Logansport, Indiana. Brian talked about the event at his facility that anyone in the waste fuels at cement kilns industry knows about, the rail car explosion at his plant on 2-19-99. In fact, Brian started off his talk by saying that it was 12:01 a.m. on 2-19-99 when his facility had a BLEVE (boiling liquid expanding vapor explosion) and to make matters worse, it was toluene diisocyanate. Brian explained that the NTSB had released their findings (an internet search can readily find this report) and the MSHA report was still pending, but Essroc had also released their own findings. It happened at Midnight, only a skeleton crew was on hand. In the “lucky” department, 5 minutes earlier or 5 minutes later and there would have been somebody on that rail car. It took 2.5 hours to evacuate a 5-mile area. The mutual aid agreement that was in place definitely worked, especially the Air Force Base that had foamers. NTSB and MSHA butted heads over who had jurisdiction [we’ve all seen this in the movies right?] but in the end, NTSB won out. Draegger tube sampling helped them get their detection limits down to the ppb level.

At this point in the presentation, Brian did a quick tag team to Bob Schreiber, also on site for the response and cleanup. Of course, the first thought of the cleanup was, “How do you cleanup TDI which has an ILDH of 2.5 ppm. Also, without an MSHA 103 work order, they couldn’t do anything and when they finally did get the order, it was a Level B for personal protection equipment, even to perform sampling. They also needed to monitor down to the 200 ppb detection level and temperature was an issue. Forty degrees versus 60 degrees, ended up putting a badge in with a jar with a collected sample in a room that was 70 – 90 degrees F. Mostly there were non-detects for TDI but did get some contact amounts that were around 1000 ppb. There were lots of questions following this presentation, which concluded at 5:05 p.m. The next speaker was Ed Shepard of Schreiber and Yonley who talked about “Hazardous Waste Facility Closure”.

The final speaker of the day was Gary King of Safety Kleen who spoke on “Cement Kiln ESP Performance Under Varying Power Conditions”. The facility at Artesia is a long wet kiln with four parallel ESP fields. The four scenarios that were tested were; B field off, energy management system on & < % opacity, lowest opacity available @ 3-4% opacity and a high opacity of 14-16%. The B-field off scenario passed 20% opacity, the Energy Mgmt System was optimized at < 6% opacity, they operated at the lowest opacity but the high opacity had a major “burp” and the fuel got shut off. The opacity versus PM graph correlated pretty well. His conclusion was that waste fuels facilities should be given flexibility rather than agreeing to just one way. The reception in the exhibition area completed the day.

Day 2 of the conference began at 8:00 a.m. on Thursday, April 18^th. The first session of the day was entitled Insights Into Regulatory Developments Impacting HWCs and was chaired by Michelle Lusk.

The first presenter was Andy O’Hare of the American Portland Cement Alliance (APCA). Andy talked about the “Portland Cement NESHAP Settlement: An Overview”. He talked about crushers, NESHAP vs NSPS vs Subpart OOO applicability and because of this multiple regulatory scenario, crushers were no longer included in the PC MACT. Subpart Y (coal) was also modified to where only the transfer points from the coal mill to the kiln are covered. He also pointed out that while bins were included, there was no one definition that covered raw materials, clinker, or finished product except “enclosed.” Bulk loading and unloading now relies on Portland Cement NSPS and only covers the cement transfers. Concerning D/F bypass testing, operators can test with the raw mill on or the raw mill off. He also talked about the definition of significant change and how [on the surface anyway] it only applies if the plant feels that it will affect PM or D/F emissions. There is also a 360-hour grace period. Then finally, representative conditions originally set production limits by default but now revolved around more representativeness than it being maximum load. Andy also talked about areas that continued to be of concern such as PM CEMs (EPA to undertake separate rule making), raw mills and finish mills (COMs or broken bag detectors in lieu of Method 22), transfer points (now excludes totally enclosed transfer points AND partially enclosed located in buildings) and PM testing waiver (when demonstrating PM CEMs). The effective date of the rule, 6-14-02, has now passed and tests must be completed by December 2002.

The next speaker was Mike Galbraith of USEPA who spoke about “What’s Been Happening with the HWC MACT Rules?” Here are the web sites he gave to check things out; access.gpo.gove/ecfr and epa.gov/epaoswer/hazwaste/combust/preamble.htm; his e-mail address; galbraith.Michael@epa.gov - he did say that there was no plan to provide any formal guidance on SSMPs. Rick Colyer of USEPA OAQPS next spoke about “The MACT Hammer and Its Application to the Phase II Hazardous Waste Combustors”. The 10-year hammer date is May 15^th 2002. To find the most current information go to: http://www.epa.gov/ttn/atw/112j/112jaypg.html. Rick’s e-mail address is colyer.rick@epa.gov or you can call him at 919-541-5262.

The next speaker, John Richards of Air Control Techniques, talked about “Industry Perspectives on the Recently Published PM CEMs PS-11 and Procedure 2 Changes” (both of which are listed in Subpart LLL and EEE). They looked at PS-11 correlation relationships by doing 15 runs with paired trains over three distinct mass concentration ranges. For procedure 2 they examined quarterly response audits, response correlation audits and absolute calibration audits. While there are several issues, one is variable partition of condensibles associated with raw material changes. John advocates Method 17 over Method 5 as a reference method. His bottom line question was “Why can’t industry just keep using the Opacity Monitor?” With ISO 10550, which has been in place since 1994, there is a correlation coefficient of 0.95. APCA, CKRC, & PCA do not believe that PM CEMs are ready for the compliance environment.

Mike Benoit of CKRC was bumped up two speakers to become the first presentation following the break. He talked about “CKRC’s Concerns with EPA’s Site-Specific Risk Assessment Policy”. The SSRA was first introduced in a 1993 press release and implemented solely through guidance [marked as draft, do not cite or quote]. This process spawned millions of dollars of expense and delay of permits, none of which has roots in a promulgated rule. Mike referred to it as “rule making in drag” disguised as guidance.

Michelle Lusk then moved to the overhead projector to assist Mike (as a colleague and a session chair) but the projector did not want to work. Tom Blackwood, Mr. Technician of the conference, came to the rescue at which time Benoit acknowledged that it was an “operator training” issue. The overhead pointed out that there were no standards of “acceptable risk.” EPA has never applied their RA knowledge beyond each specific industry. D/F always drives the process no matter what else you find (and Hg). CKRC favors genuine guidance about bona fide rule but not “guidance from uncertain parentage.” Referring to “rule by guidance” he then pointed out that the Appalachian Power decision was “all over it” and directed everyone to talk with Dick Stohl for more details.

The next presenter, Deborah Henry of Black and Veach, talked about “CERCLA and EPCRA Reporting”. Operators must immediately report releases of certain hazardous substances that exceed applicable limits, 800 CERCLA cmpds and 356 extremely hazardous [EPCRA] substances. There are 222 EHS compounds that are not CERCLA. Exemptions can be claimed according to “federally permitted” releases. Some waste combustor concerns might be sulfur dioxide, which has a limit of 500 pounds/day, ammonia with 100 pounds/day, benzene at 10 pounds/day, mercury at 1 pound/day and lead at 10 pounds/day.

The final presentation of Session 5 was by Ajay Kumar of Holcim who filled in for Claude Culem of Holcim and talked about “EU Hazardous Waste Rules, Development and Implementation”. Ajay talked about the EU legal framework which included the waste framework Directive [Directive 75/442, amended in 1991] and the waste incineration Directive [Directive 2000/76]. Cement plants are assimilated as co-incinerators when burning hazardous waste. At the conclusion of his presentation, Ajay got to put on another hat and become co-chair of Session 6 with Carrie Yonley.

The first presentation of Session 6, by Mike Budin of RMT, Inc. was about “Comparative Analysis: RCRA Trial Burn versus HWC MACT CPT Plans”. This presentation was followed by Craig Doolittle of ENSR International who spoke on “Reducing Your Compliance Burden Through Pre-MACT Testing”. Craig also had an article in the May 2002 “Environmental Magazine” (EM). Craig discussed the Ontario Hydro method versus 0060 and recommended not spiking metals while gathering D/F data for risk assessment purposes. ENSR does agree with TNRCC guidance for particle size distribution by using a Method 5 train with polycarbonate filter, then shortening the run times and using XRF/XRD, which is better than default assumptions.

Brad Phillips of Schreiber and Yonley was the final speaker before lunch. Brad talked about “Mercury Emissions: A Fine Line Between Non-detects and the Regulatory Limit. In talking about testing for Mercury and determining SREs”, Brad could not remember that anyone performed Tier III or Tier II for Mercury during all the BIF testing. Among some of the issues of concern were feedstream sampling, analytical capabilities, and spiking variations. He presented some calculations on a hypothetical kiln that exhibited a variation of 13.63, as compared to the limit of 56 ug/dscm.

Session 6 concluded after lunch, Jerry Drake of Compliance Strategies and Solutions, Inc., spoke about “Demonstration of Similarity for RCRA, BIF and MACT Performance Tests”.

Session 7 was entitled “Legal Aspects of MACT” and was chaired by Terry Satterlee of Lathrop and Gage. The first speaker, Dick Stohl, started off talking about the “Judicial Review Process Associated with the HWC MACT Litigation Case. The D.C. Circuit Court has eleven judges that sit in panels of 3 judges. They review what the agency has done on the record and then submitted briefs and listen to oral arguments that focus on the highlights. In 2000 they wiped out part 1 of the waste combustor MACT that included cessation of burning, NIC and progress reports. In 2002 they wiped out part 2 as many parties challenged it and the Sierra Club prevailed. When EPA issues complex regulations, it makes picking litigation issues difficult because the court will not allow you to litigate all of it and in some cases then the court will not hear arguments on certain issues, such as “emission limitations,” because they are not raised before the oral arguments or had “not raised it with sufficient specificity before EPA.” In the end the Sierra Club agreed to allow interim standards as long as EPA issued final rules under court order. The decision was also made that the “best performing 12%” does not have to be technology based (e.g. cleaner raw materials & cleaner fuels). To make this session more of a point/counter point they moved Steve Silverman, who was the team leader for the Government, up to immediately follow Dick’s presentation. To get the best performing 12%, 3 data points do not give an accurate picture of performance over time but the one sure thing that the court would bless is to take the best 12% and divide by two. So any future exercise must include the best performing 12%, period! Steve also had two observations. The D.C. Court knocked out the 6-year rule and then approved the interim standards with no public comment. Dick’s previous point about “emission limitations” becomes a dead issue since the rules must be based upon the best performing 12%, no matter why they are the best performing 12%. He closed by saying that we will just have to wait and see “what the great savants on the court” have to say.

Even though Dick and Steve were a tough act to follow, Tom Gerver of Lathrop and Gage jumped into the breach as the next speaker. He talked about “HWC MACT Permitting and Implementation Issues”. The conference then went to the final break in the exhibition area.

Terry Satterlee kicked off after break by filling in for Brad Hiles. She talked about “Closure & Post Closure and Remedial Action Plans”. She told the assembled group that EPA was coming out with “a series of guidance that is results oriented” and that it was clear the EPA was primarily going to be concerned with action oriented corrective action plans, so be prepared.

Session 8 was entitled “Fuel Blending and BIF” and was chaired by Gary King of Safety Kleen. The first speaker was Jim Cudahy of Focus Environmental and his presentation was entitled “The HWC Phase II MACT and Energy Recovery from Waste”. He asked the question as to whether the BIF rule actually encourages energy recovery. He pointed out that MACT did not incorporate resource recovery. Ultimately, liquids will have no place to go. A comment from the crowd was that a company that owns the greatest number of BIF boilers in the world has just purchased one of the largest distillation/recovery operations in Europe and product reclamation could come into play.

Tom Busman, also of Focus Environmental, followed Jim. His presentation was entitled “Applying the Compliance Evaluation Strategy Developed for the HWC MACT to the HWC Phase II MACT”. Tom had compared the Phase II database to the HWC MACT database and found that Phase II units would pass all Phase I interim limits except for PM. However, he pointed out that this conclusion was based upon an insufficient database that makes up the Phase II database. The Phase II database will ultimately set the limits very low.

Hazardous Waste Combustors Specialty Conference