MEMORANDUM  

DATE:		2/28/2023

TO:	Docket for Proposed Amendments to New Source Performance Standards: Electric Arc Furnaces (EAFs) and Argon-Oxygen Decarburization Vessels (AOD): 40 CFR 60 Subparts AA, AAa, and AAb (Docket ID: EPA-HQ-OAR-2002-0049)

FROM:	Donna Lee Jones, U.S. Environmental Protection Agency, Office of Air Quality Planning and Standards (OAQPS), Research Triangle Park, NC; Gabrielle Raymond, RTI International 

SUBJECT:	Conference call with Electric Arc Furnace Industry Representatives
--------------------------------------------------------------------------------


SUMMARY

      A Microsoft(TM) TEAMS conference call was held on February 27, 2023, between representatives from the EPA, the Steel Manufacturers Association (SMA), the American Iron and Steel Institute (AISI), the Specialty Steel Institute of America (SSINA), and various Electric Arc Furnace (EAF) company representatives to discuss the proposed rule changes as a result of the technology review of the EAF New Source Performance Standards (NSPS) (40 CFR part 60, subparts AA, AAa, and new AAb). 
      
Pound/ton steel limit for AAb 

Chuck: Welcome. We are here to discuss a possible subcategory for stainless steel. The highest number 0.16 lb/ton steel was from a stainless facility. 

Joe: First, we have deep objections to a lb/ton limit. But, if there was going to be a lb/ton limit, then we need a subcategory for stainless. There is no correlation between concentration and lb/ton. In the 80's there was not found any correlation of mass emissions and steel production. The concentration limit can be met by large and small specialty or carbon steel facilities. We have concerns with how the 0.16 lb/ton limit was developed. 

The denominator varies dramatically between steel making. Specialized steel have heat times 2-5 times as long as carbon. Length of production cycle; grade produced; alloys added all change the correlation between PM and tonnage. Would be unfair for specialty steel. 

No evidence on record that changing from concentration limit is appropriate.

Vendor guarantees: we can't get those for lb/ton limit because they have no control over that equation.

Wayne: We have comments on correlation between lb/ton to gr/dscf for the cost baseline. It was based on NAS and that data was not the best based on EPA's record. Why does EPA believe this is a standard that can be met? Is this correlation analysis the best EPA has?

DLJ: EPA is not accusing facilities of ramping air flow. We just see there is a lot of flexibility in air flow. We prefer lb/ton where tons are tons and pounds are pounds. Highest lb/ton is stainless. The limit proposed was based on stainless. In regard to the vendor guarantees, the NESHAP is still out there. Still have to meet that gr/dscf. so can still use that information as appropriate and needed. This NSPS limit is for new sources and modified /reconstructed sources. There are two types of facility changes that trigger NSPS. We understand the angst from the overlap with brand new sources. 

Wayne: New standards not adequately justified. We just don't understand how you got there. We see that you removed the non-0% opacity facilities, then used the air-to-cloth ratios from IIS facilities, wholly different from what's in the record, and plotted emissions by production rank. Presumed that NAS was 70% of the standard. 

Ron: EPA added concentrations.

DLJ: We know you can't add gr/dscf from baghouses; we are fixing that. We decided lb/ton would be better. There is no reason why we need to correlate lb/ton to gr/dscf for the rule. We have removed the baseline in the analysis. We plotted gr/dscf and lb/ton, had same shape curve. Used highest level as BSER. 

Ron: conceptual problem with lb/ton for batch processes. Sometimes there is extended heat or have to hold the heat. The pounds keep changing, but tons steel are not. They are a fixed denominator. Delays in batches and heats, safety delays. How is EPA thinking about that? 

DLJ: Hopefully, this doesn't impact the 3-5 hour test, but facilities could retest.

Ron: There is variability in batch heat times. 

DLJ: It would be good for us to see how delays would impact emissions.

Ron: We don't know.

Joe: What are the benefits of lb/ton? EPA has made speculative assumptions of why concentration is a problem, not based on factual evidence. We have a system that works. 

DLJ: EPA staff in field noticing nonoptimal operation of baghouse because of air flow. Secondary baghouse, low PM, this seems to imply it's not operating efficiently. Can be operated well or not well and have the flexibility to meet the gr/dscf. lb/ton puts all the emissions in one place. With concentration, can't see if  a facility is doing what it needs to do. Facility can split flow into two baghouses. lb/ton combines emissions. 

Wayne: not realistic of how they operate.

Ron: can't divert flow on a whim.

Steve Fruh: is there a concern with not being able to correlate concentration to lb/ton? Is there a concern with lb/ton limit based on stringency? We have lb/ton limits in metals industry with batch operations. 

Wayne: Vendor guarantees, won't do for lb/ton standards. Can facilities meet lb/ton? we don't know.

Ron: If there is interest in the mass of emissions from a stack test, you can calculate that. What is the incremental benefit of lb/ton limit? 

DLJ: one number for entire facility.

Joe:  NAS value isn't representative of smaller mills. EPA has no data from smaller mills. Everyone can meet concentration limit. 

DLJ only new sources have to meet the lb/ton limit.

Zero Percent Opacity Limit for AAb

Wayne: zero percent opacity can't be met on a continuous basis. EPA has 13 facilities with 0% opacity during a stack test, during melting and refining. Stack test is a snap shot of time. We reached out to those facilities and asked are you meeting 0% at all times, and they said no. What would it take to do so, no one way. Not capable. Large scale movement, materials movement. People can't make a 0% opacity on continuous basis. Continuous would include charging and tapping. We don't now how to meet. So many variables. No data. Not feasible.

The Finkl permit, not a 0% opacity limit. If they go over 0% in 30 minutes it triggers a response, and they check again. Not a limit. Infeasible. Zero evidence. 

Unless you have a Consteel or shaft EAF, the DEC is off during charging and tapping, and emissions are going into the shop and through the hoods. 

Broadly speaking, 0% opacity is infeasible across the board.

Steve Fruh: what is the current opacity

Wayne: 6% opacity daily once per shift

Steve: we are not requiring CEMS right.

DLJ: we proposed same frequency; added charging and tapping or time frame with greatest potential for emissions.

Wayne: Finkl does more on forging side. It's a BACT standard not a numerical limit. It wasn't cited in the proposal.

Steve : What data was collected for bl/ton results?

Wayne: It's the 2010 test data from ICR for Hg and PM.

DLJ: Yes, and we looked at the 2005 data. Couldn't always calculate lb/ton with 2005 data. But things are still looking good for 0.16 lb/ton steel limit. 

Ron: There is a difference in highest facility observed and what everyone can meet. There was no statistics or variability added.

DLJ: For NSPS, its the best system of emission reductions and ratcheted down to where its cost effective. 

DLJ: If 0% opacity is not doable, what if there was a hybrid situation with a different limit for charging and tapping. Two EAF facilities have a permit limit with 3% at all times. 

Wayne: I don't think we could say across the board if that was feasible. 0% wasn't feasible for melting and refining. BACT in RBLC shows 0% not feasible. They show 6% is the number to hit.

Changes to AA and AAa 

Wayne: I want to bring up the changes to existing sources for monitoring that result in significant costs. 

DLJ: We understand the issue.

Wayne: We need something more concrete.

Chuck: We are not going to final for those items that cause burden and costs.

Steve: We are making changes to the rule for this issue; we will let you know if this changes.

Wayne/Rich: See our supplemental comments: changes to bag leak detector, flow monitoring; pressure monitoring.  

Wayne: Could EPA issue a note of understanding that the changes to existing monitoring are not going to be finalized?

DLJ: This is it. We are also adding 6 months of compliance for what stays in the rule.

Wayne: What is the 6 months for? We keep getting calls from facilities.

Chuck: We will get with our people and get back to you. We can have a follow up meeting if necessary. 

Steve Fruh: We are removing these costly items from the rules. If someone higher up says we can't do that, we will let you know.  


      
ATTENDEES
      
EPA
Donna Lee Jones (EPA/OAQPS)
Chuck French (EPA/OAQPS)
Steve Fruh (EPA/OAQPS)
Kevin McGinn (EPA/OAQPS)
Dakota Prentice, EPA/Region V
Gabrielle Raymond (RTI International) 

SMA
Eric Stuart (Vice President) 

Kelley Drye & Warren, LLP
Wayne D'Angelo 
Joe Green 
Ron Sahu (consultant) 

American Iron and Steel Institute
Paul Balserak

Industry Representatives 
Rich Zavoda, Cleveland Cliffs, Inc. 
Joshua Cain, CORP
Tom Sanicola
