                                  Memorandum
                                       
TO:  Portland Cement Project Team

FROM: Keith Barnett, US EPA

SUBJECT:  Revised Floors without Kilns that Would have been CISWI Kilns Had the Solid Waste Definition Applied

DATE:  April 25, 2011 

Revision of Emission data base to remove possible CISWI kilns

The following table lists kilns that had been identified as potential CISWI kilns in the proposal of the CISWI rule.  This list was compiled based on responses to an EPA survey and was based on the proposed definition of non-hazardous solid waste.  

      Table 1.  List of CISWI Cement Kilns at Proposal of the CISWI Rule
                               ALCemexDemopolis
                             ALLafargeRobertaPlant
                             ALLehighCementCoLeeds
                           CACAPortlandCementColton
                         CALehighSouthwestReddingPlant
                              CAMitsubishiCement
                             CANationalCementofCA
                              FLCemexBrooksville
                                 FLCemexMiami
                                 FLFloridaRock
                              FLRinkerBrooksville
                              GACemexClinchfield
                               IALafargeBuffalo
                            IALehighCementMasonCity
                                IDAshGroveInkom
                             ILBuzziUnicemOglesby
                              ILLafargeJoppaPlant
                           KSMonarchCementCoHumboldt
                            MDESSROCFrederickPlant
                              OKLafargeTulsaPlant
                                OKLoneStarPryor
                               ORAshGroveDurkee
                       PAEssrocItalcementiGroupBessemer
                            PALafargeWhitehallPlant
                           PALehighCementEvansville
                              PALehighCementYork
                             SCLafargeHarleyville
                               TNCemexKnoxville
                             TXAshGroveMidlothian
                             TXCemexBalconesPlant
                            TXHolcimMidlothianPlant
                              TXLoneStarMaryneal
                             UTAshGroveLeamington
                              UTHolcimDevilsSlide
                               WAAshGroveSeattle
                               WALafargeSeattle


Follow up questions were sent by EPA to these facilities to gain additional information on some of their fuels, specifically used oil, tires, wood waste and spent seed corn.   These responses were then reviewed based on the final solid waste definition rule signed on February 21, 2011.  The final solid waste rule states that tires that come from an established tire collection program, clean wood, clean biomass, and used oil that is on spec are not solid waste.  We eliminated the kilns that burned only these alternative fuels from the CISWI list.  The remaining CISWI kilns are shown below in Table 2.  

 
                         Table 2.  CISWI Cement Kilns
                                  FacilityID
                                    UnitID
Lehigh Southwest, Redding CA
In-Line Kiln/Raw Mill (S260)
Cemex, Miami FL
Cement Kiln
Lafarge, Buffalo IA
Kiln 1
Lehigh Cement, Mason City IA
302.203
Lafarge, Whitehall PA
Kiln 2
Lafarge, Whitehall PA
Kiln 3
Lehigh Cement, Evansville PA
122
Lehigh Cement, Evansville PA
121
Lehigh Cement, York PA
200
Holcim, Midlothian TX
Kiln 2
Holcim, Midlothian TX
Kiln 1
Holcim, Devils Slide UT
Kiln 1
Ash Grove Cement, Midlothian TX
Kilns 1 through 3
California Portland Cement, Colton CA
Kilns 1 and 2
 
 
In addition to the kilns above, the recent ICR follow up on alternative fuels indicates the following kilns in Table 3 would also be CISWI kilns (though they were not in the CISWI data base).  

Table 3.  Additional Kilns identified as CISWI Units (but are not in CISWI data base)
Location
Number of Kilns
Why is it CISWI?
Lehigh, Union Bridge MD
1
Burns dried biosolids as a fuel
Lafarge, Alpena MI
5
Burns Plastics
 

Note that it is EPA's position that combustion does not occur in the cold end of a cement kiln.  Materials that enter the cold end of the kiln are gradually heated until they reach the temperature where clinker formation takes place.  This is in contrast to combustion where fuel is introduced into the kiln at a location where the temperature is sufficient to immediately commence the process of combusting the fuel.    (Fuels are never burned in the cold end of a cement kiln, and our interpretation reflects that understanding.)  This means that cement kilns' practices of using non-hazardous secondary materials in the cold end of the cement kiln would not make the kilns "solid waste incineration unit[s]" within the meaning of section 129 (g)(1) of the CAA since kilns would not "combus[t]" those materials.  Consequently, we do not include any cement kiln in this analysis based on utilization of non-hazardous secondary materials in the cold end of the cement kiln.

For purposes of this analysis, all of the kilns shown in Tables 2 and 3 were taken out of the Portland cement emissions data base and the floors recalculated as shown below.  That is, we recalculated the floors as if the recently promulgated definition of non-hazardous secondary materials which are solid wastes had been in place at the time of the cement kiln NESHAP rulemaking.  

Mercury Floor

The mercury floor was calculated based on a data base of 89 kilns where we had approximately 30 days of data on all mercury inputs.  Twenty-two of the 23 CISWI kilns shown above were part of the mercury data base.  Removing the 22 CISWI kilns leaves 67 kilns.  The best performing 12 percent would be 9 kilns.

Of the best performing 11 kilns that were used in the previous floor calculation, two would have been CIWSI, 1589, and 1248.  Removing these kilns from the MACT floor pool leaves the 9 best performing kilns.  The single best performing kiln now becomes kiln 1650.

The floors were recalculated (See attachment 1) based on the best performing 9 kilns and the new source floor was recalculated based on kiln 1650.  The results were as follows:

Existing Source floor:   58 lb/MM tons clinker (versus 55 in the current rule)
New Source Floor:  24 lb/M tons clinker (versus 21 in the current rule)

THC Floor

The THC Floor was re-estimated using the THC data set used in the final rule.  This data set (shown in Attachment 2) consisted of 15 kilns where we had continuous monitoring data.  Of these 15 kilns, four would have been CISWI kilns (including Union Bridge, the best performer on average).  This leaves 9 kilns (see attachment 3), and the best 12 percent would be two.  The best performing kilns on average become Suwannee, Branford and Holcim Trident. The resulting floor based on the 99th percentile of the data of the two floor kilns is 15 ppmvd and the new source floor would be 11 ppmvd (see attachment 4).   The corresponding organic HAP limits (under the alternative standard) would be 6 and 4 ppmvd respectively.

PM Floor

There were 45 kilns in the PM data set used in the final rule (see attachment 5).  Removing the CISWI kilns left 28 kilns (see attachment 6).  The best performing kilns would be the top 4.  New source MACT does not change; it is still the Mitsubishi kiln.

The existing source MACT floor using the four best performing kilns is 0.05 lb/ton clinker (See Attachment 7).   

HCl Floor

There were HCl test data for 20 major source kilns used in the final rule (see attachment 8).   Four of the 20 were CSWI kilns, leaving 16 NESHAP kilns (see attachment 9).  The best performing 12 percent would be 2.  Recalculation of the floor based on the two best performing kilns results in a floor of 1 ppmvd (see attachment 10), which is still below the minimum quantification limit of 3 ppmvd.   This standard thus would remain at 3 ppmvd for new and existing sources.
