September 2, 1997

Donald Sutton, Manager
Permits Section
Division of Air Pollution Control
Illinois Environmental Protection Agency
P.O. Box 19506
2200 Churchill Road
Springfield, Illinois  62794-9506

Dear Mr. Sutton:

The purpose of this letter is to provide further guidance on the major modification provisions of the federal rules for Prevention of Significant Deterioration (PSD), 40 CFR 52.21, as applied to construction of a proposed soybean extraction plant (Application 96100019) at the Archer Daniels Midland (ADM) plant in Decatur, Illinois.  Per electronic mail correspondence and telephone calls between USEPA and Illinois Environmental Protection Agency (IEPA) staff during the period August 21 to 28, 1997, the record is not complete regarding the BACT analyses for hexane and ethanol (VOM) emissions associated with the proposed project.  USEPA staff also highlighted to IEPA staff concerns raised in my letter to you of January 7, 1997, regarding ADM’s proposed use of fugitive roadway dust emission reductions as offsets for increases resulting from a proposed project.  In addition, it should be noted that, since it is our understanding that there are several proposed ADM projects that will need PSD review, if there is an integral relationship between any of these projects or sub-projects such that one depends upon another, then construction may not commence on any part of these inter-dependent projects until a PSD permit or permits has/have been issued for the whole project, that is, all of the integral portions of the project.

The following summarizes those concerns raised between USEPA and IEPA staff on the proposed permit which have yet to be addressed:

A Atop-down@ analysis of control technologies was not presented to support the BACT determinations of the proposed permit, nor have the BACT analyses themselves been provided.  My staff has expressed a request for these analyses but have yet to receive them.  The permitting record is incomplete without such analyses.

The VOM consumption limit for hexane, 10.4 pounds per ton of soybean processed, does not compare well with other similar limits from previous BACT determinations.  A BACT determination in 1981 for Boon Valley Co-op and Owensboro Grain established a 2.9 pounds per ton limit.  A PSD permit issued in the last several months to an ADM facility in North Kansas City, MO has a limit of 0.25 gallon per ton.  Since the density of hexane is 5.63 pounds per gallon, this limit is equal to about 1.4 pounds per ton.  Given that there is room for discussion whether pounds of hexane lost, consumed, and emitted are equivalent terms, the proposed limit seems well in excess of established limits.

Based on the information given to USEPA at this time, the proposed soybean extraction project appears to be a major modification pursuant to PSD for PM.  More data is needed to make an accurate PSD determination with regard to the creditable contemporaneous PM emission increases and decreases claimed for this project.  Unless ADM can prove through the netting of emissions that the overall increase in PM emissions from this project will not exceed major modification significance level, 15 tpy, PSD requirements do apply.

Leak detection and repair (LDAR) for the proposed project should be more well-defined.  For example, what is the meaning of "prompt repair" and "detailed inspection"?

The following are additional issues of concern the USEPA has regarding the proposed permit:

BACT is an emission limitation.  The proposed permit does not include an emission rate limit for VOM.  The BACT rate must be directly related to emissions to the atmosphere.

How will ADM maintain records of VOM usage?  Is the area of the proposed project sufficiently separated from the rest of the operations at the Decatur facility such that all VOM entering the area are readily recorded?

A detailed list of PM-emitting units/equipment at ADM and related maximum PM emission rates is attached to the proposed permit.  The same type of list should be attached for VOM.

The liquid to gas (L/G) ratio defined in the proposed permit is not a true L/G ratio.  The proposed permit sets a limit for the mineral oil scrubber of 5 gallons per minute of scrubber cross-sectional area.  This only relates to liquid flow and not to gas flow.  The proposed permit should include a true L/G ratio limitation, and scrubber gas and liquid flow rates need to be monitored.

I hope you will find this information useful.  If we can answer any questions regarding these comments, or if we can provide any further guidance, please contact John Kelly, of my staff, at (312) 886-4882.  Once again, thank you for your commitment to working with us to improve the permitting process.

Sincerely yours,


Cheryl L. Newton, Chief
Permits and Grants Section