PTI/PTIO Application A0066143 Petmin USA Incorporated ... · 1. 2. 3.--Petmin USA Incorporated -...

PTI/PTIO Application A0066143Petmin USA Incorporated

0204012023April 09, 2020

1.

2.

3.

-

-

Petmin USA Incorporated - 0204012023 Page 1 PTI/PTIO Application - A0066143

Division of Air Pollution Control Apr 9 2020, 15:56:32

Application for Permit-to-Install or Permit-to-Install and Operate

Date application received: 04/09/2020

This section should be filled out for each permit to install (PTI) or Permit to Install and Operate (PTIO) application. A PTI isrequired for all air contaminant sources (emissions units) installed or modified after January 1, 1974 that are subject to OACChapter 3745-77. A PTIO is required for all air contaminant sources (emissions units) that are not subject to OAC Chapter3745-77 (Title V). See the application instructions for additional information.

For OEPA use only: x Installation Request Federally enforceable restrictions

Modification General Permit

Renewal Other

Please summarize the reason for this permit application. This text will be in the public notice that will appear in the newspaper

of the county where the facility is located.

New facility to convert iron ore pellets to merchant pig iron.

Is the purpose of this application to transition from OAC Chapter 3745-77 (Title V) to OAC Chapter 3745-31 (PTIO)?

No

Establish PER Due Date - Select an annual Permit Evaluation Report (PER) due date for this facility (does not apply to

facilities subject to Title V, OAC Chapter 3745-77). If the PER has previously been established and a change is now desired,

a PER Change Request form must be filed instead of selecting a date here.

Due Date: For Time Period:

Jan 1 - Dec 31, Due Feb 15 January 1 through December 31

Federal Rules Applicability

New Source Performance Standards (NSPS) New Source Performance Standards are listed under 40CFR 60 - Standards of Performance for New StationarySources.

Subject to subpart:

IIII - Stationary CompressionIgnition Internal CombustionEngines

National Emission Standards for Hazardous AirPollutants (NESHAP) National Emissions Standards for Hazardous Air Pollutantsare listed under 40 CFR 61. (These include asbestos,benzene, beryllium, mercury, and vinyl chloride).

Not affected

Maximum Achievable Control Technology (MACT) The Maximum Achievable Control Technology standardsare listed under 40 CFR 63 and OAC rule 3745-31-28.

Subject to subpart:

ZZZZ - Reciprocating InternalCombustion Engines

Prevention of Significant Deterioration (PSD) These rules are found under OAC rule 3745-31-10 throughOAC rule 3745-31-20.

Subject to Regulation

Greenhouse Gas Pollutant Prevention of SignificantDeterioration (PSD) These rules are listed under 40 CFR Parts 51, 52.

Subject to Regulation

Non-Attainment New Source Review These rules are found under OAC rule 3745-31-21 throughOAC rule 3745-31-27.

Not affected

112 (r) - Risk Management Plan Not affected

4.

5.

6.

7.

8.


These rules are found under 40 CFR 68.

Title IV (Acid Rain Requirements) These rules are found under 40 CFR 72 and 40 CFR 73.

Not affected

Express PTI/PTIO - Do you qualify for express PTI or PTIO processing?

No

Air Contaminant Sources in this Application - Identify the air contaminant source(s) for which you are applying below.

Attach additional pages if necessary. Section II of this application and an EAC form should be completed for each air

contaminant source.

Emissions Unit ID Company Equipment ID (company's name forair contaminant source)

Equipment Description (List all equipment thatare a part of this air contaminant source)

B001 Startup boiler

F001 Plant Roadways Plant Roadways dust control

P001 Process gas heater Catalytic thermal oxidizerconverting H2S into SO2

P002 Ladle preheater EAF and Casting baghouse

P003 Ladle preheat (backup) EAF and Casting baghouse

P004 Ladle drying station EAF and Casting baghouse

P005 Emergency Generator #1

P006 Emergency Generator #2

P007 Black Start Generator

P008 Quenching & pre-wastewatertreatment

Ammonia scrubber, Flare

P009 High Pressure Emergency DieselEngine

P010 Low Pressure Emergency DieselEngine

P901 EAF EAF and Casting baghouse

P902 Material Handling Screen building

The Emissions Unit ID would have been created when a previous air permit was issued. If no previous permitshave been issued for this air contaminant source, leave this field blank. If this air contaminant source waspreviously identified in STARShip applications as a “Z” source (e.g., Z001), please provide that identification anda new ID will be assigned when the PTI/PTIO is issued.

Trade Secret Information - Is any information included in this application being claimed as a trade secret per Ohio Revised

Code (ORC) 3704.08?

No

Permit Application Contact - Person to contact for questions about this application:

Palmira Farinha

Name Title

600 SuperiorAvenueFifth ThirdBuilding, Suite 1300

Cleveland, OH 44114

Street Address City/Township, State Zip Code

2164796876 [email protected]

Phone Fax E-mail

Application Attachments

AttachmentID

AttachmentType

Description EAC FormType

PublicDocument

Trade SecretDocument

Trade SecretJustification

Event Date

716948 PTE Calculatio X


Calculations

ns

716991 PTIApplication Report

Other X

716950 Appendix 2 Other X

716947 Appendix 1 Other X

1.

2.

3.

-

-

-

-

-


Section II - Specific Air Contaminant Source Information Facility ID: 0204012023Emissions Unit ID: B001

Company Equipment ID: Startup boiler

One copy of this section should be filled out for each air contaminant source (emissions unit) covered by thisPTI/PTIO application identified in Section I, Question 5. See the application instructions for additional information.

Air Contaminant Source Installation or Modification Schedule - Check all that apply (must be completed regardless of

date of installation or modification):

New installation (for which construction has not yet begun, in accordance with OAC rule 3745-31-33). When will you beginto install the air contaminant source?after installation permit has been issued

SCC Codes - List all Source Classification Code(s) (SCC) that describe the process(es) performed by this air contaminant

source (e.g., 1-02-002-04).

See Facility Profile

Emissions Information - The following table requests information needed to determine the applicable requirements and the

compliance status of this air contaminant source with those requirements. Suggestions for how to estimate emissions may

be found in the instructions to the Emissions Activity Category (EAC) forms required with this application. If you need further

assistance, contact your District Office/Local Air Agency representative.

If total potential emissions of HAPs or any Toxic Air Contaminant (as identified in OAC rule 3745-114-01) are greater

than 1 ton/yr, fill in the table for that (those) pollutant(s). For all other pollutants, if “Emissions before controls (max),

lb/hr” multiplied by 24 hours/day is greater than 10 lbs/day, fill in the table for that pollutant.

Actual emissions are calculated including add-on control equipment. If you have no add-on control equipment,

“Emissions before controls” will be the same as “Actual emissions”.

Actual emissions and Requested Allowable should be based on operating 8760 hr/yr unless you are requesting federally

enforceable operating restrictions to limit emissions. If so, calculate emissions based on requested operating restrictions

and describe in your calculations.

If you use units other than lbs/hr or ton/yr, specify the units used (e.g., gr/dscf, lb/ton charged, lb/MMBtu, tons/12-

months).

Requested Allowable (ton/yr) is often equivalent to Potential to Emit (PTE) as defined in OAC rule 3745-31-01 and OAC

rule 3745-77-01.

Pollutant Emissionsbefore controls(max)* (lb/hr)

Actualemissions

(lb/hr)

Actualemissions(ton/year)

RequestedAllowable

(lb/hr)

RequestedAllowable(ton/year)

Particulate emissions(PE/PM) (formerlyparticulate matter, PM)

0.11 0.11 0.49 0.11 0.49

PM # 10 microns indiameter (PE/PM10)

0.11 0.11 0.49 0.11 0.49

PM # 2.5 microns indiameter (PE/PM2.5)

0.11 0.11 0.49 0.11 0.49

Sulfur dioxide (SO2) 0.01 0.01 0.04 0.01 0.04

Nitrogen oxides (NOx) 0.63 0.63 2.78 0.63 2.78

Carbon monoxide (CO) 1.25 1.25 5.47 1.25 5.47

Organic compounds (OC) 0.08 0.08 0.36 0.08 0.36

Volatile organiccompounds (VOC)

0.08 0.08 0.36 0.08 0.36

Lead (Pb) 0 0 0 0 0

Total Hazardous AirPollutants (HAPs)

0 0 0 0 0

4.

5.

6.

7.

8.

9.

10.


Highest single HAP 0 0 0 0 0

Greenhouse Gas Pollutants:

Pollutant Emissionsbefore

controls(max)* (lb/hr)

Actualemissions

(lb/hr)


RequestedAllowable

(lb/hr)


CO2e(ton/year)

Carbon Dioxide 1784.1 1784.1 7814 1784.1 7814 7,814

Best Available Technology (BAT) - For each pollutant for which the Requested Allowable in the above table exceeds 10

tons per year, BAT, as defined in OAC 3745-31-01, is required. Describe what has been selected as BAT and the basis for

the selection:

Control Equipment - Does this air contaminant source employ emissions control equipment?


Process Flow Diagram - Attach a Process Flow Diagram to this application for this air contaminant source. See the

application instructions for additional information.

Process Flow Diagrams:

AttachmentID

AttachmentType


PublicDocument



Event Date

716954 Aux BoilerPFD

Processflowdiagram

X

Modeling information: (Note: items in bold in Tables 7-A and/or 7-B, as applicable, are required even if the tables do

not otherwise need to be completed. If applicable, all information is required An air quality modeling analysis is

required for PTIs and PTIOs for new installations or modifications, as defined in OAC rule 3745-31-01, where either the

increase of toxic air contaminants from any air contaminant source or the increase of any other pollutant for all air

contaminant sources combined exceed a threshold listed below. This analysis is to assure that the impact from the

requested project will not exceed Ohio's Acceptable Incremental Impacts for criteria pollutants and/or Maximum Allowable

Ground Level Concentrations (MAGLC) for toxic air contaminants. (See Ohio EPA, DAPC's Engineering Guide #69 for more

information.) Permit requests that would have unacceptable impacts cannot be approved as proposed. See the line-by-line

PTI/PTIO instructions for additional information.


Request for Federally Enforceable Limits - As part of this permit application, do you wish to propose voluntary restrictions

to limit emissions in order to avoid specific requirements listed below, (i.e., are you requesting federally enforceable limits to

obtain synthetic minor status)?

No

Continuous Emissions Monitoring - Does this air contaminant source utilize any continuous emissions monitoring (CEM)

equipment for indicating or demonstrating compliance? This does not include continuous parametric monitoring systems.


EAC Forms - The appropriate Emissions Activity Category (EAC) form(s) must be completed and attached for each air

contaminant source. At least one complete EAC form must be submitted for each air contaminant source for the application

to be considered complete. Refer to the list attached to the application instructions. Please indicate which EAC form

corresponds to this air contaminant source.


AttachmentID

AttachmentType


PublicDocument



Event Date

716953 Aux BoilerEAC

EAC 3101 Fuelburning

X


operation

1.

2.

3.

-

-

-

-

-


Section II - Specific Air Contaminant Source Information Facility ID: 0204012023Emissions Unit ID: F001

Company Equipment ID: Plant Roadways






source (e.g., 1-02-002-04).















months).


rule 3745-77-01.


Actualemissions

(lb/hr)


RequestedAllowable

(lb/hr)



0.99 0.05 0.22 0.05 0.22


0.99 0.05 0.22 0.05 0.22


0.10 4.9E-3 0.02 4.9E-3 0.02

Sulfur dioxide (SO2) 0 0 0 0 0

Nitrogen oxides (NOx) 0 0 0 0 0

Carbon monoxide (CO) 0 0 0 0 0

Organic compounds (OC) 0 0 0 0 0


0 0 0 0 0

Lead (Pb) 0 0 0 0 0


0 0 0 0 0

4.

5.

6.

7.

8.

9.

10.





the selection:






AttachmentID

AttachmentType


PublicDocument



Event Date

716958 RoadwaysPFD

Processflowdiagram

X














No









AttachmentID

AttachmentType


PublicDocument



Event Date

716957 Roads EAC EAC 3111RoadwaysandParkingAreas

X

1.

2.

3.

-

-

-

-

-


Section II - Specific Air Contaminant Source Information Facility ID: 0204012023Emissions Unit ID: P001

Company Equipment ID: Process gas heater






source (e.g., 1-02-002-04).















months).


rule 3745-77-01.


Actualemissions

(lb/hr)


RequestedAllowable

(lb/hr)



1.63 1.63 7.14 1.63 7.14


1.63 1.63 7.14 1.63 7.14


1.63 1.63 7.14 1.63 7.14






1.18 1.18 5.17 1.18 5.17

Lead (Pb) 0 0 0 0 0


0 0 0 0 0

4.

5.

6.

7.

8.

9.

10.






Actualemissions

(lb/hr)


RequestedAllowable

(lb/hr)


CO2e(ton/year)

Carbon Dioxide 70203 70203 307490 70203 307490 307,490



the selection:

See attached BACT review. Selected technology is good combustion practices and alow NOx burner.






AttachmentID

AttachmentType


PublicDocument



Event Date

716963 Petmin PFD Processflowdiagram

X














No









AttachmentID

AttachmentType


PublicDocument



Event Date


716967 PGH EAC EAC 3101 Fuelburningoperation

X

1.

2.

3.

-

-

-

-

-



Company Equipment ID: Ladle preheater






source (e.g., 1-02-002-04).















months).


rule 3745-77-01.


Actualemissions

(lb/hr)


RequestedAllowable

(lb/hr)



0 0 0 0 0


0 0 0 0 0


0 0 0 0 0






0.08 0.08 0.35 0.08 0.35

Lead (Pb) 0 0 0 0 0


0 0 0 0 0

4.

5.

6.

7.

8.

9.

10.






Actualemissions

(lb/hr)


RequestedAllowable

(lb/hr)


CO2e(ton/year)

Carbon Dioxide 1764.7 1764.7 7729.41 1764.7 7729.41 7,729.41



the selection:

Natural gas is used as fuel.






AttachmentID

AttachmentType


PublicDocument



Event Date


X














No









AttachmentID

AttachmentType


PublicDocument



Event Date

716962 Ladle EAC 3101 Fuel X


Preheat/Dry EAC

burningoperation

1.

2.

3.

-

-

-

-

-



Company Equipment ID: Ladle preheat(backup)






source (e.g., 1-02-002-04).















months).


rule 3745-77-01.


Actualemissions

(lb/hr)


RequestedAllowable

(lb/hr)



0 0 0 0 0


0 0 0 0 0


0 0 0 0 0






0.08 0.08 0.35 0.08 0.35

Lead (Pb) 0 0 0 0 0


0 0 0 0 0

4.

5.

6.

7.

8.

9.

10.






Actualemissions

(lb/hr)


RequestedAllowable

(lb/hr)


CO2e(ton/year)

Carbon Dioxide 1764.7 1764.7 7729.41 1764.7 7729.41 7,729.41



the selection:







AttachmentID

AttachmentType


PublicDocument



Event Date


X














No









AttachmentID

AttachmentType


PublicDocument



Event Date



Preheat/Dry EAC

burningoperation

1.

2.

3.

-

-

-

-

-



Company Equipment ID: Ladle dryingstation






source (e.g., 1-02-002-04).















months).


rule 3745-77-01.


Actualemissions

(lb/hr)


RequestedAllowable

(lb/hr)



0 0 0 0 0


0 0 0 0 0


0 0 0 0 0






0.08 0.08 0.35 0.08 0.35

Lead (Pb) 0 0 0 0 0


0 0 0 0 0

4.

5.

6.

7.

8.

9.

10.






Actualemissions

(lb/hr)


RequestedAllowable

(lb/hr)


CO2e(ton/year)

Carbon Dioxide 1764.7 1764.7 7729.41 1764.7 7729.41 7,729.41



the selection:







AttachmentID

AttachmentType


PublicDocument



Event Date


X














No









AttachmentID

AttachmentType


PublicDocument



Event Date



Preheat/Dry EAC

burningoperation

1.

2.

3.

-

-

-

-

-



Company Equipment ID: EmergencyGenerator #1






source (e.g., 1-02-002-04).















months).


rule 3745-77-01.


Actualemissions

(lb/hr)


RequestedAllowable

(lb/hr)



0.15 0.15 0.01 0.15 0.01


0.15 0.15 0.01 0.15 0.01


0.15 0.15 0.01 0.15 0.01

Sulfur dioxide (SO2) 1.2E-3 1.2E-3 6.0E-4 1.2E-3 6.0E-4





0.97 0.97 0.05 0.97 0.05

Lead (Pb) 0 0 0 0 0


0 0 0 0 0

4.

5.

6.

7.

8.

9.

10.






Actualemissions

(lb/hr)


RequestedAllowable

(lb/hr)


CO2e(ton/year)

Carbon Dioxide 3635 3635 181.75 3635 181.75 181.75



the selection:






AttachmentID

AttachmentType


PublicDocument



Event Date

716956 EmergencyEngine PFD

Processflowdiagram

X














No









AttachmentID

AttachmentType


PublicDocument



Event Date

716955 EmergencyEngine EAC

EAC 3862Stationary

X


InternalCombustion Engine

1.

2.

3.

-

-

-

-

-



Company Equipment ID: EmergencyGenerator #2






source (e.g., 1-02-002-04).















months).


rule 3745-77-01.


Actualemissions

(lb/hr)


RequestedAllowable

(lb/hr)



0.15 0.15 0.01 0.15 0.01


0.15 0.15 0.01 0.15 0.01


0.15 0.15 0.01 0.15 0.01






0.97 0.97 0.05 0.97 0.05

Lead (Pb) 0 0 0 0 0


0 0 0 0 0

4.

5.

6.

7.

8.

9.

10.






Actualemissions

(lb/hr)


RequestedAllowable

(lb/hr)


CO2e(ton/year)

Carbon Dioxide 3635 3635 181.75 3635 181.75 181.75



the selection:






AttachmentID

AttachmentType


PublicDocument



Event Date


Processflowdiagram

X














No









AttachmentID

AttachmentType


PublicDocument



Event Date

716964 EmergencyEngine EAC

EAC 3862Stationary

X


InternalCombustion Engine

1.

2.

3.

-

-

-

-

-



Company Equipment ID: Black StartGenerator






source (e.g., 1-02-002-04).















months).


rule 3745-77-01.


Actualemissions

(lb/hr)


RequestedAllowable

(lb/hr)



5.22E-03 5.22E-03 2.61E-04 5.22E-03 2.61E-04


5.22E-03 5.22E-03 2.61E-4 5.22E-03 2.61E-04


5.22E-03 5.22E-03 2.61E-4 5.22E-03 2.61E-4


Nitrogen oxides (NOx) 0.10 0.10 5.22E-03 0.10 5.22E-03


Organic compounds (OC) 0.05 0.05 2.44E-3 0.05 2.44E-3


0.05 0.05 2.44E-3 0.05 2.44E-3

Lead (Pb) 0 0 0 0 0


0 0 0 0 0

4.

5.

6.

7.

8.

9.

10.






Actualemissions

(lb/hr)


RequestedAllowable

(lb/hr)


CO2e(ton/year)

Carbon Dioxide 181.86 181.86 9.09 181.86 9.09 9.09



the selection:






AttachmentID

AttachmentType


PublicDocument



Event Date


Processflowdiagram

X














No









AttachmentID

AttachmentType


PublicDocument



Event Date

716971 BlackStart

EAC 3862Stationary

X


Engine EAC InternalCombustion Engine

1.

2.

3.

-

-

-

-

-



Company Equipment ID: Quenching & pre-wastewatertreatment






source (e.g., 1-02-002-04).















months).


rule 3745-77-01.


Actualemissions

(lb/hr)


RequestedAllowable

(lb/hr)



0.05 0.05 0.22 0.05 0.22


0.05 0.05 0.22 0.05 0.22


0.05 0.05 0.22 0.05 0.22






0.92 0.92 4.05 0.92 4.05

Lead (Pb) 0 0 0 0 0

Total Hazardous Air 0 0 0 0 0

4.

5.

6.

7.

8.

9.

10.


Pollutants (HAPs)





Actualemissions

(lb/hr)


RequestedAllowable

(lb/hr)


CO2e(ton/year)

Carbon Dioxide 777.46 777.46 3405.29 777.46 3405.29 3,405.29



the selection:






AttachmentID

AttachmentType


PublicDocument



Event Date


X














No









AttachmentID

AttachmentType


PublicDocument



Event Date

716966 Quench and EAC 3100 X


WW EAC Processoperation

1.

2.

3.

-

-

-

-

-



Company Equipment ID: High PressureEmergency DieselEngine






source (e.g., 1-02-002-04).















months).


rule 3745-77-01.


Actualemissions

(lb/hr)


RequestedAllowable

(lb/hr)



0.10 0.10 5.1E-3 0.10 5.1E-3


0.10 0.10 5.1E-3 0.10 5.1E-3


0.10 0.10 5.1E-3 0.10 5.1E-3

Sulfur dioxide (SO2) 0.02 0.02 8.7E-4 0.02 8.7E-4





0.03 0.03 1.3E-3 0.03 1.3E-3

Lead (Pb) 0 0 0 0 0


4.

5.

6.

7.

8.

9.

10.


Pollutants (HAPs)





Actualemissions

(lb/hr)


RequestedAllowable

(lb/hr)


CO2e(ton/year)

Carbon Dioxide 357.97 357.97 17.90 357.97 17.9 17.9



the selection:






AttachmentID

AttachmentType


PublicDocument



Event Date

716952 EmergencyPump PFD

Processflowdiagram

X














No









AttachmentID

AttachmentType


PublicDocument



Event Date

716970 High EAC 3862 X


PressureEmergencyPump EAC

Stationary InternalCombustion Engine

1.

2.

3.

-

-

-

-

-



Company Equipment ID: Low PressureEmergency DieselEngine






source (e.g., 1-02-002-04).















months).


rule 3745-77-01.


Actualemissions

(lb/hr)


RequestedAllowable

(lb/hr)



0.08 0.08 3.9E-3 0.08 3.9E-3


0.08 0.08 3.9E-3 0.08 3.9E-3


0.08 0.08 3.9E-3 0.08 3.9E-3

Sulfur dioxide (SO2) 0.01 0.01 6.6E-4 0.01 6.6E-4





0.05 0.05 2.3E-3 0.05 2.3E-3

Lead (Pb) 0 0 0 0 0


4.

5.

6.

7.

8.

9.

10.


Pollutants (HAPs)





Actualemissions

(lb/hr)


RequestedAllowable

(lb/hr)


CO2e(ton/year)

Carbon Dioxide 272.8 272.8 13.64 272.8 13.64 13.64



the selection:






AttachmentID

AttachmentType


PublicDocument



Event Date

716952 EmergencyPump PFD

Processflowdiagram

X














No









AttachmentID

AttachmentType


PublicDocument



Event Date

716951 Low EAC 3862 X


PressureEmergencyPump EAC

Stationary InternalCombustion Engine

1.

2.

3.

-

-

-

-

-



Company Equipment ID: EAF






source (e.g., 1-02-002-04).















months).


rule 3745-77-01.


Actualemissions

(lb/hr)


RequestedAllowable

(lb/hr)



12.43 12.43 54.44 12.43 54.44


12.43 12.43 54.44 12.43 54.44


9.94 9.94 43.55 9.94 43.55






1.38 1.38 6.06 1.38 6.06

Lead (Pb) 0 0 0 0 0


0 0 0 0 0

4.

5.

6.

7.

8.

9.

10.






Actualemissions

(lb/hr)


RequestedAllowable

(lb/hr)


CO2e(ton/year)

Carbon Dioxide 11209 11209 49095 11209 49095 49,095



the selection:

Emissions from the EAF, ladles, pouring and casting will be controlled by a fabricfilter. Prior to the fabric filter, the hot gases from the EAF are ducted through adrop-out box where larger particulates drop out and air is introduced to combustCO.






AttachmentID

AttachmentType


PublicDocument



Event Date


X














No









Attachment Attachment Description EAC Form Public Trade Secret Trade Secret Event Date


ID Type Type Document Document Justification716965 EAF EAC EAC 3100

Processoperation

X

1.

2.

3.

-

-

-

-

-



Company Equipment ID: Material Handling






source (e.g., 1-02-002-04).















months).


rule 3745-77-01.


Actualemissions

(lb/hr)


RequestedAllowable

(lb/hr)



0.30 0.30 1.29 0.30 1.29


0.30 0.30 1.29 0.30 1.29


0.20 0.20 0.89 0.20 0.89


Nitrogen oxides (NOx) 0 0 0 0 0

Carbon monoxide (CO) 0 0 0 0 0

Organic compounds (OC) 0 0 0 0 0


0 0 0 0 0

Lead (Pb) 0 0 0 0 0


0 0 0 0 0

4.

5.

6.

7.

8.

9.

10.





the selection:

Iron ore pellets have residual moisture from water sprays, as needed during loadingto the hopper (by third party). Conveyors and transfer points are protected fromwind, where located outside. The screening operation is ducted to a fabric filter.Indoor conveying and transfer operations are enclosed and air movement is inwarddue to the volume of air going to the fabric filter.






AttachmentID

AttachmentType


PublicDocument



Event Date

716959 MaterialHandlingPFD

Processflowdiagram

X

716960 OutdoorMaterialHandlingPFD

Processflowdiagram

X














No









AttachmentID

AttachmentType


PublicDocument



Event Date

716961 MaterialHandling

EAC 3113Material

X


EAC Handling

Appendix 1 – North American Steel Kentucky Permit Excerpt

Permit Number: V-15-019 R2 Page 53 of 194

SECTION B - EMISSION POINTS, EMISSION UNITS, APPLICABLE REGULATIONS, AND OPERATING CONDITIONS (CONTINUED)

c. All stainless steel scrap shall contain low concentrations of impurities. For the production of steel other than leaded steel, the permittee must not charge metallic scrap that contains scrap from motor vehicle bodies, engine blocks, oil filters, oily turning, machine shop borings, transformers or capacitors containing polychlorinated biphenyls, lead-containing components, chlorinated plastics, or free organic liquids. This restriction does not apply to any post-consumer engine blocks, post-consumer oil filters, or oily turnings that are processed or cleaned to the extent practicable such that the material donot include lead components, chlorinated plastics, or free organic liquids. For motor vehicle scrap that is charged to recover the chromium or nickel content the permittee must certify in the notification of compliance status to show compliance. [40 CFR 63.10685(a)(2)]

Compliance Demonstration Method: The permittee shall continue to comply with the provisions of the approved pollution prevention plan.

2. Emission Limitations:

a. The EAF1 baghouse shall not emit pollutants in excess of the following self- imposed/BACT limitations averaged over three heats:

1. PM emissions: 0.10 pounds per ton and 13.94 pounds per hour. 2. CO emissions: 2 pounds per ton and 265.76 pounds per hour. 3. NO2 emissions: 1.32 pound per ton and 175 pounds per hour. 4. VOC emissions: 0.150 pound per ton and 19.95 pounds per hour. 5. Lead emissions: 0.001 pound per ton and 0.167 pound per hour. 6. Graphite electrode sulfur content shall not exceed 0.02%. The EAF2 baghouse shall not emit pollutants in excess of the following self-imposed/BACT limitations averaged over three heats:

1. PM emissions: 0.193 pounds per ton and 25.71 pounds per hour.

2. CO emissions: 2 pounds per ton and 266 pounds per hour. 3. NO2 emissions: 1.00 pound per ton and 133 pounds per hour. 4. VOC emissions: 0.150 pound per ton and 19.95 pounds per hour. 5. Lead emissions: 0.002 pound per ton and 0.309 pound per hour. 6. Graphite electrode sulfur content shall not exceed 0.02%.

b. The AOD1 baghouse shall not emit pollutants in excess of the following self- imposed/BACT limitations averaged over three heats:

1. PM emissions: 0.13 pounds per ton and 16.98 pounds per hour. 2. CO emissions: 2.06 pounds per ton and 273.75 pounds per hour. 3. NO2 emissions: 0.578 pound per ton and 76.83 pounds per hour. 4. Lead emissions: 0.002 pound per ton and 0.204 pound per hour.

Permit Number: V-15-019 R2 Page 54 of 194

SECTION B - EMISSION POINTS, EMISSION UNITS, APPLICABLE REGULATIONS, AND OPERATING CONDITIONS (CONTINUED) The AOD2 baghouse shall not emit pollutants in excess of the following self- imposed/BACT limitations averaged over three heats:

1. PM emissions: 0.193 pounds per ton and 25.71 pounds per hour. 2. CO emissions: 2.06 pounds per ton and 273.98 pounds per hour. 3. NO2 emissions: 0.58 pound per ton and 76.87 pounds per hour. 4. Lead emissions: 0.002 pound per ton and 0.31 pound per hour.

c. Annual Emission Limitations: The EAF1 and EAF2 baghouses shall not emit pollutants in excess of the following self-imposed/BACT limitations: 1. PM emissions: 138.24 tons per year. 2. CO emissions: 1653.36 tons per year. 3. NO2 emissions: 1010.86 tons per year. 4. VOC emissions: 124.04 tons per year. 5. Lead emissions: 1.66 tons per year.

AOD1 and AOD2 Annual Emission Limitations: The AOD baghouses shall not emit pollutants in excess of the following self-imposed/BACT limitations:

1. PM emissions: 138.24 tons per year. 2. CO emissions: 1703 tons per year. 3. NO2 emissions: 477.87 tons per year. 4. Lead emissions: 1.70 tons per year.

d. Visible emissions from the EAF and AOD control devices shall not equal or exceed 3% opacity each. [40 CFR 60.272a(2)]

e. Visible emissions from the melting shop shall not exhibit 6% opacity or greater. [40 CFR

60.272a(3)] f. Visible emissions from dust handling equipment shall not equal or exceed 10% opacity.

[40 CFR 60.272a(b)]

Compliance Demonstration Method: See 3. Testing Requirements, 4. Specific Monitoring Requirements, and 5. Specific Recordkeeping Requirements, below. Also, see permit SECTION D SOURCE EMISSION LIMITATIONS AND TESTING REQUIREMENTS, item D.13, for CAM information.

g. Pursuant to 401 KAR 59:010, Appendix A, the emissions of PM matter shall not exceed

the allowable rate limit as calculated by the following equations using the process weight rate (in units of tons/hr).

Appendix 2 ‐ SCR Cost Estimation Petmin USA IncorporatedAshtabula, OH

For Oil‐Fired Industrial Boilers between 275 and 5,500 MMBTU/hour :

For Natural Gas‐Fired Industrial Boilers between 205 and 4,100 MMBTU/hour :

Total Capital Investment (TCI) = $3,628,964 in 2018 dollars

Direct Annual Costs (DAC) = $154,179 in 2018 dollars

Indirect Annual Costs (IDAC) = $401,286 in 2018 dollars

Total annual costs (TAC) = DAC + IDAC $555,465 in 2018 dollars

Annual Maintenance Cost = 0.005 x TCI = $18,145 in 2018 dollars

Annual Reagent Cost = qsol x Costreag x top = $21,533 in 2018 dollars

Annual Electricity Cost = P x Costelect x top = $70,006 in 2018 dollars

Annual Catalyst Replacement Cost = nscr x Volcat x (CCreplace/Rlayer) x FWF $44,495 in 2018 dollars

Direct Annual Cost = $154,179 in 2018 dollars

Administrative Charges (AC) = $2,846 in 2018 dollars

Capital Recovery Costs (CR)= CRF x TCI = $398,441 in 2018 dollars

Indirect Annual Cost (IDAC) = AC + CR = $401,286 in 2018 dollars

Total Annual Cost (TAC) = $555,465

NOx Removed = 46 tons/year

Cost Effectiveness = $12,070 per ton of NOx removed in 2018 dollars

0.03 x (Operator Cost + 0.4 x Annual Maintenance Cost) =

Total Annual Cost (TAC)

TCI = 80,000 x (200/BMW )0.35 x BMW x ELEVF x RF

per year in 2018 dollars

Annual Costs

IDAC = Administrative Charges + Capital Recovery Costs

Cost Effectiveness

Cost Effectiveness = Total Annual Cost/ NOx Removed/year

Direct Annual Costs (DAC)

DAC = (Annual Maintenance Cost) + (Annual Reagent Cost) + (Annual Electricity Cost) + (Annual Catalyst Cost)

Indirect Annual Cost (IDAC)

TAC = Direct Annual Costs + Indirect Annual Costs

Cost Estimate

Total Capital Investment (TCI)

TCI for Oil and Natural Gas Boilers

For Oil and Natural Gas‐Fired Utility Boilers >500 MW:

TCI = 60,670 x BMW x ELEVF x RF

For Oil‐Fired Industrial Boilers >5,500 MMBtu/hour:

For Natural Gas‐Fired Industrial Boilers >4,100 MMBtu/hour:

TCI = 7,082 x QB x ELEVF x RF

TCI = 5,275 x QB x ELEVF x RF

TCI = 9,760 x (1,640/QB )0.35 x QB x ELEVF x RF

For Oil and Natural Gas‐Fired Utility Boilers between 25MW and 500 MW:

TCI = 7,270 x (2,200/QB )0.35 x QB x ELEVF x RF

Prepared by: Matthew Ayer

Checked by: Joe Hollowell, PE Page 1 of 1

Construction of a New Facility

PSD PTI Application Report

September 2018

Updated December 2019

Prepared for:

Petmin USA Incorporated

Prepared by:

Matthew Ayer

Joseph N. Hollowell, PE

AYER Quality Engineering LLC

Cincinnati, OH

Petmin USA Inc. Ashtabula, OH Ashtabula Facility


Report – Petmin USA Incorporated

September 2018 (Updated December 2019)

Table of Contents

1.0 INTRODUCTION & SCHEDULE ................................................................................. 1

1.1 Updates from Previous Reports .................................................................................................... 2

1.2 Project Schedule ............................................................................................................................ 3

1.3 Key Contacts ................................................................................................................................. 3

1.4 Report Organization ...................................................................................................................... 4

2.0 PROJECT AND PROCESS DESCRIPTION ................................................................ 5

2.1 Project Location ............................................................................................................................ 5

2.2 Process Description ....................................................................................................................... 7

2.2.1 Pellet Unloading and Handling ........................................................................................................... 8

2.2.2 DRI Reactor and Process Gas Loop .................................................................................................... 9

2.2.3 CO2 Removal ..................................................................................................................................... 11

2.2.4 H2S Injection ..................................................................................................................................... 13

2.2.5 Reduction Process Start-up ................................................................................................................ 13

2.2.6 Reduction Process Shutdown ............................................................................................................ 14

2.2.7 Smelting in the Electric Arc Furnace and Pig Iron Casting ............................................................... 14

2.2.8 EAF Fume Capture System ............................................................................................................... 15

2.2.9 Auxiliary Boiler ................................................................................................................................. 16

2.2.10 Storage Silos ...................................................................................................................................... 16

3.0 FEDERAL & STATE RULE APPLICABILITY ........................................................ 18

3.1 Defining the Scope of the Project for Regulatory Analysis and Permitting ............................... 18

3.1.1 Kinder Morgan Terminal a Separate Stationary Source .................................................................... 18

3.1.2 Carbon Dioxide Plant a Separate Stationary Source ......................................................................... 19

3.1.3 Slag Processor a Separate Stationary Source ..................................................................................... 22

3.2 Federal Rule Applicability .......................................................................................................... 23

3.2.1 Prevention of Significant Deterioration (PSD) .................................................................................. 23

3.2.2 National Emission Standards for Hazardous Air Pollutants (NESHAP) ........................................... 25

3.2.3 Standards of Performance for New Stationary Sources (NSPS) ....................................................... 27

3.2.4 Preconstruction Permit Requirements ............................................................................................... 28

3.2.5 CAIR NOx Annual Trading Program – Negative Declaration ........................................................... 28

3.2.6 Title V Federal Operating Permit Program ....................................................................................... 29

3.2.7 Compliance Assurance Monitoring ................................................................................................... 29

3.2.8 Other Applicable Emission Limitations and Standards ..................................................................... 29

3.3 State of Ohio Regulations & Standards ...................................................................................... 29




3.3.1 General Requirements ....................................................................................................................... 29

3.3.2 Criteria of the Director and Best Available Technology ................................................................... 31

3.3.3 Particulate Matter .............................................................................................................................. 31

3.3.4 Sulfur Dioxide Emissions .................................................................................................................. 32

3.3.5 Control of Oxides of Nitrogen (NOx) ................................................................................................ 32

3.4 State of Ohio Regulations and Standards – By Emissions Unit .................................................. 32

3.4.1 Process Gas Heater ............................................................................................................................ 33

3.4.2 Quenching & Wastewater Treatment Waste Gas .............................................................................. 34

3.4.3 Ladle Preheat ..................................................................................................................................... 35

3.4.4 EAF, Pouring and Casting ................................................................................................................. 35

3.4.5 Material Handling .............................................................................................................................. 37

3.4.6 Roadways and Parking Areas ............................................................................................................ 38

3.4.7 Emergency Engines and Black Start Generators ............................................................................... 38

4.0 BEST AVAILABLE CONTROL TECHNOLOGY (BACT) ANALYSIS ................ 40

4.1 Introduction ................................................................................................................................. 40

4.2 Emissions Units .......................................................................................................................... 41

4.3 NOx BACT Review ..................................................................................................................... 42

4.3.1 Electric Arc Furnace .......................................................................................................................... 42


4.3.3 Ladle Preheaters ................................................................................................................................ 56

4.3.4 Small Ancillary Equipment – Startup Boiler, Flare, Emergency Engines, Black Start Generator .... 56

4.4 PM10/PM2.5 BACT Review ......................................................................................................... 57


4.4.2 Material Handling and Screening ...................................................................................................... 64

4.4.3 Roadways and Parking Areas ............................................................................................................ 68

4.4.4 Fuel Burning Equipment (Process Gas Heater, Ladle Preheat, Flare, and Startup Boiler ................. 69

4.4.5 Emergency Engines, Fire Pumps, and Black Start Generator ........................................................... 70

4.4.6 Storage Piles ...................................................................................................................................... 71

4.4.7 Cooling Tower ................................................................................................................................... 73

4.5 Greenhouse Gases (GHG) ........................................................................................................... 73


4.5.2 Other Emissions Units ....................................................................................................................... 74

4.6 CO BACT Review ...................................................................................................................... 75


4.6.2 Other Emissions Units ....................................................................................................................... 82

5.0 AMBIENT AIR QUALITY ANALYSIS ........................................................................ 83




5.1 Site Area Characteristics ............................................................................................................. 83

5.1.1 Wind Climatology ............................................................................................................................. 84

5.1.2 Land Use Classification ..................................................................................................................... 84

5.1.3 Terrain Analysis ................................................................................................................................ 85

5.2 Pre-Application Air Quality Monitoring ..................................................................................... 85

5.3 Ambient Air Quality Data ........................................................................................................... 86

5.4 Air Dispersion Modeling Parameters .......................................................................................... 88

5.4.1 AERMOD Modeling System ............................................................................................................ 89

5.4.2 Receptor Grid .................................................................................................................................... 90

5.4.3 Facility Building Data for Downwash Analysis ................................................................................ 91

5.4.4 Modeling Input – Emission Source Data ........................................................................................... 92

5.4.5 Option Selected for One-Hour NO2 Modeling .................................................................................. 95

5.5 Modeling Results – SIL .............................................................................................................. 99

5.6 Full Impacts Analysis ............................................................................................................... 100

5.6.1 Facility Emissions ........................................................................................................................... 100

5.6.2 Nearby Significantly Interacting Sources ........................................................................................ 100

5.7 Modeling Results – Full Impacts Analysis ............................................................................... 106

5.8 Modeling Results – PSD Increment .......................................................................................... 107

5.9 Ozone Analysis ......................................................................................................................... 107

5.10 Secondary PM2.5 Analysis ......................................................................................................... 111

5.11 Summary ................................................................................................................................... 112

6.0 ADDITIONAL IMPACTS ANALYSIS ...................................................................... 113

6.1 Growth Impact Analysis ........................................................................................................... 113

6.2 Ambient Air Quality Analysis .................................................................................................. 114

6.3 Soils and Vegetation Analysis .................................................................................................. 114

6.4 Visibility ................................................................................................................................... 114

6.5 Class I Area Impacts Analysis .................................................................................................. 115

List of Tables

Key Project Milestones .................................................................................................................................... 3

Key Project Contacts ....................................................................................................................................... 4

Merchant Pig Iron Chemical Components ...................................................................................................... 5

Project Potential to Emit & PSD Applicability ............................................................................................ 24

Applicable Ohio Administrative Code Rules in Chapter 3745 ...................................................................... 33

Project Maximum Emissions by Emissions Unit .......................................................................................... 41




Emissions Units Subject to PSD Review – NOx ............................................................................................ 42

RBLC Search Results – EAFs ....................................................................................................................... 48

Methods for Reducing NOx Emissions .......................................................................................................... 50

Control Technologies .................................................................................................................................. 51

Cost Effectiveness of NOx Controls ............................................................................................................ 54

BACT Recommendations for Ancillary Units ............................................................................................. 57

Emissions Units Subject to PSD Review – PM10, PM2.5 ............................................................................. 57

RBLC Search of BACT Controls for DRI Production ................................................................................ 67

Emissions Units Subject to PSD Review – GHG ........................................................................................ 73

Emissions Units Subject to PSD Review – CO ........................................................................................... 75

Summary of RBLC Search for CO Controls on EAFs ................................................................................ 81

Significant Monitoring Concentrations ....................................................................................................... 86

Background Concentrations ........................................................................................................................ 86

Season-Hour NO2 Background Data ........................................................................................................... 88

Point Source Emissions Units Modeled Parameters .................................................................................... 93

Volume Source Emissions Units Modeled Parameters ............................................................................... 94

Flare Modeling Parameters .......................................................................................................................... 95

Significant Impact Level Modeling Analysis Results ................................................................................. 99

Other Sources Included in the Model ........................................................................................................ 102

Stack Parameters for Other Sources Modeled ........................................................................................... 103

Sources Excluded from the Model ............................................................................................................ 104

NO2 Modeling Analysis Results ................................................................................................................ 106

PM2.5 Modeling Analysis Results .............................................................................................................. 106

PM10 Modeling Analysis Results ............................................................................................................... 106

PSD Increment Modeling Analysis Results............................................................................................... 107

List of Figures

Figure 1: Location of Proposed Petmin Facility ...................................................................................................... 6

Figure 2: Basic Process Flow Diagram ................................................................................................................... 7

Figure 3: Material Handling Flowchart ................................................................................................................... 8

Figure 4: Primary and Secondary EAF Hoods ...................................................................................................... 16




Figure 5: Electric Arc Furnace .............................................................................................................................. 43

Figure 6: SCR Process Diagram ............................................................................................................................ 53

Figure 7: Wind Rose Plot of 2013 to 2017 Meteorological Data from the Ashtabula, OH Airport and Upper Air

Data from Buffalo, NY ...................................................................................................................................... 84

Figure 8: Google Earth Image of the Location of the Proposed Petmin Facility .................................................. 85

Figure 9: Petmin Facility with Receptor Grid ....................................................................................................... 91

Figure 10: Petmin Facility Layout ........................................................................................................................... 92

Figure 11: Wind Rose Plot of Met Data in Ashtabula County .............................................................................. 109

Figure 12: Wind Rose Plot of Met Data in Tuscarawas County ........................................................................... 109

Figure 13: Wind Rose Plot of Met Data in Macomb County ................................................................................ 110




- 1 -

1.0 INTRODUCTION & SCHEDULE

Petmin USA Incorporated (Petmin) proposes a project (hereinafter referred to as the Project) to

construct and operate a new facility to produce merchant pig iron (MPI). The projected annual

production is 470,000 tons MPI per year. The maximum theoretical plant capacity (assuming 8,760

hours per year production) is 526, 738 tons MPI per year. The new facility will be located within

the Kinder Morgan (KM) Pinney Dock facility at 1149 East Fifth Street, Ashtabula, Ohio.

The proposed Project constitutes a new major stationary sourcea under the federal Prevention of

Significant Deterioration (PSD) preconstruction permitting regulations. Ohio EPA administers the

PSD preconstruction review process as part of its Permit to Install (PTI) program, approved in its

State Implementation Plan. This engineering report, and supporting forms and attachments,

comprise a complete PTI application package for the Project, including the requisite elements

described in Ohio Administrative Code (OAC) §3745-31-11 through 20.

Once constructed, the Petmin plant will also be required to apply for and obtain a Part 70 (also

known as “Title V”) federal operating permit.

As demonstrated in this report, with respect to the federal list of Hazardous Air Pollutants (HAPs),

the plant is classified an area source (i.e., below the thresholds for a major HAP source).

The principal raw material to produce MPI consists of iron pellets. KM will provide logistic

services to handle incoming iron ore pellets and outgoing pig iron ingots on a contract basis. The

ship unloading, storage and transfer of iron pellets to the new Petmin manufacturing facility will

be performed by KM.

The scope, or boundary of the Project described in this engineering report begins at the receipt of

iron pellets from KM at the new Petmin plant and ends with the production of cast nodular pig

iron shapes as its final product. Slag produced as a byproduct transfers ownership and

responsibility immediately upon being tapped from the electric arc furnace. Its handling is also

outside the boundary of the project. A waste gas (principally composed of carbon dioxide) is also

proposed to be transferred to a third-party for purification and sale as an alternative to Petmin

discharging the waste gas. Under this operational scenario, the environmental treatment of this gas

stream becomes the responsibility of the third part gas purification plant owner. These project

boundary limits are presented in detail in this engineering report.

a Please note that the use of italics in this report denotes a term with a specific regulatory definition.




- 2 -

This report and supporting information comprising a complete PTI application was submitted to

Ohio EPA Division of Air Pollution Control in September 2018. A DRAFT PTI was issued

November 14, 2018. A public notice was published November 16, 2018. A public meeting was

held December 20, 2018. Ohio EPA extended the requisite comment period by 19 days to January

4, 2019. Following its response to comments, Ohio EPA issued a FINAL PTI on February 6, 2019.

The permit application and technical report were based on the best available information. This

revision to the report updates a number of parameters associated with the project to reflect final

design values. The overall scope of the project, the applicable federal and state requirements, and

the air pollution control technology have not changed.

1.1 Updates from Previous Reports

The first technical report was submitted September 10, 2018, and was subsequently revised in

October 2018 as follows:

• Revision to the Ladle Heater, from a single heater rated at 19.15 MMBtu/hour to three

heaters, each rated at 9.0 MMBtu/hour. Total facility emissions are not changing from this

revision, only the emissions from the ladle heater emissions unit. This is because the ladle

heaters vent to the EAF baghouse and its emissions are calculated based on the maximum

production rate, which isn’t changing.

• Other various minor corrections to descriptions of emissions units throughout the report.

• Updated modeling, Section 5, due to requests from Ohio EPA.

This revision incorporates a number of final design aspects to the project. These updates are listed

as follows:

• CO emission rate for the EAF updated, triggering BACT review for CO emissions.

• Minor change to the Electric Arc Furnace production rate, with a corresponding change to

the NOx emission rate from the EAF.

• Particulate matter emission rate updated for fabric filters to use an absolute (i.e., outlet

concentration-based) efficiency, with corresponding expansion of BACT section.

• Addition of two additional emergency engines to the Project, for fire protection.

• Consideration of the third-party CO2 Processing Plant as a medium-term overall Project

objective in as a novel technology for a DRI manufacturing facility, but not required in the

PTI for operation of the plant. Sulfur control requires a thermal oxidizer and an absorption

tower (scrubber) to convert the sulfur by oxidation to SO2, then absorb the gas into an

aqueous phase for reaction with a neutralizing agent and subsequent offsite shipment.

• Addition of a wet scrubber to treat stripped wastewater – in order to absorb ammonia in

gas phase to an aqueous phase for offsite shipment and disposal.




- 3 -

• Refinement of PM calculations for material handling, based on final design details of

transfer operations.

• Clarification that the third-party slag processing company takes ownership of the slag at

the furnace.

• Corresponding updates to the ambient impact analysis (air dispersion modeling report).

• Minor corrections to technical descriptions based on review by the process engineering

design team.

1.2 Project Schedule

Key milestones in the current project schedule are as follows. Future permit issuance dates are

targets. Petmin is committed to providing complete and detailed responses to comments or

questions during the review period to facilitate timely and efficient review by Ohio EPA staff.

Key Project Milestones

Date Milestone

August 23, 2018 Submit Air Dispersion Modeling Protocol to Ohio EPA

September 5, 2018 Air Dispersion Modeling Protocol Approval from Ohio EPA

September 7, 2018 Submit PTI Application Package

January 2020 through June

2020 (Estimate)

Petmin Preconstruction Site Preparation Activities Approved Under

OAC 3745-31-33(E)

November 14, 2018 DRAFT PTI Issued

November 16, 2018 Public Notice

December 20, 2018 Public Meeting

February 6, 2019 FINAL PTI Issued

December 9, 2019 Revised Application Submitted to Modify Permit

April 2, 2020 DRAFT PTI Issued

April 6, 2020 Public Notice

May 7, 2020 Public Hearing

June 5, 2020 (Estimate) Final PTI Issued

Q1 2022 (Estimate) Complete Construction and Commissioning

Q2 2022 (Estimate) Plant Startup

1.3 Key Contacts

The report has been prepared by AYER Quality Engineering LLC (AYER) and reviewed by

Petmin. Principal contacts are as follows:




- 4 -

Key Project Contacts

Palmira Farinha

Project Director

Business Development


1003 Bridge Street (Upper)

Ashtabula, OH 44004

Cleveland, OH 44114

[email protected]

(216) 378 2947 X 2947 (O)

Matthew Ayer

Project Director AYER Quality Engineering LLC

3908 Pocahontas Avenue

Cincinnati, OH 45227

[email protected]

(513) 272-0506 (O)

(513) 335-0059 (M)

Joseph Hollowell, P.E.

Air Dispersion Modeling

[email protected]

(513) 470-8740 (M)

1.4 Report Organization

The report is organized as follows:

Section 1 Introduction & Schedule Section 4 Best Available Control Technology

Section 2 Project & Process Description Section 5 Ambient Air Quality Analysis

Section 3 Regulatory Review Section 6 Additional Impacts Analysis

As described in detail in this report, Petmin will meet state and federal applicable requirements,

and thereby prevent significant deterioration of regional air quality.

https://ayerquality.sharepoint.com/Shared%20Documents/PetMin/PTI%20Application/PTI%20Attachments/[email protected]

mailto:[email protected]

mailto:[email protected]




- 5 -

2.0 PROJECT AND PROCESS DESCRIPTION

The Project consists of the construction and operation of a facility to produce Merchant Pig Iron

(MPI) from iron pellets. More specifically, Petmin will produce nodular pig iron, the highest purity

grade of merchant pig iron.

Merchant pig iron contains at least 92% iron, for a degree of purity of around 96%. It is typically

separated into three main categories:

1. basic pig iron: used mainly in electric arc steelmaking

2. foundry pig iron (also known as haematite pig iron): used in mainly in the manufacture

of grey iron castings in cupola furnaces

3. high purity pig iron (also known as nodular pig iron): used in the manufacture of ductile

[also known as nodular or spheroidal graphite – SG] iron castings.1

Table 3 presents the proportion of chemical components contained in merchant pig iron.

Merchant Pig Iron Chemical Components

MPI Types Iron Carbon Silicon Manganese Sulfur Phosphorus

Basic Pig Iron 92% 3.5 – 4.5% <1.5% 0.5 – 1.0% <0.05% <0.12%

Haematite Pig Iron 92% 3.5 – 4.5% 1.5 – 3.5% 0.5 – 1.0% <0.05% <0.12%

Ductile/Nodular Pig Iron 92% 3.5 – 4.5% 0.05 – 2.0% <0.05% <0.05% <0.04%

Annual potential production is 526,739tons. The fundamental manufacturing process is an

established and proven technology – the direct reduction of iron pellets to form Direct Reduced

Iron (DRI). However, purification via smelting in an electric arc furnace (EAF) to produce MPI is

a novel technology. As described in this section, Petmin will be employing a technologically

advanced process for heat recovery and in-process recycling and reuse of process gases.

Natural gas is used as a reducer source (i.e., to produce reducing gases), as well as for fuel for

combustion to provide heat energy to the preheating furnace. A description of the process is found

in the following section.

2.1 Project Location

The Project will be located within the footprint of the Kinder Morgan Pinney Dock (KM) facility,

in the county and city of Ashtabula, as indicated by the arrow in the map in Figure 1. The site is

approximately 575 feet above sea level.




- 6 -

Figure 1: Location of Proposed Petmin Facility




- 7 -

2.2 Process Description

A general flow diagram of the process is shown in Figure 2.

Figure 2: Basic Process Flow Diagram




- 8 -

2.2.1 Pellet Unloading and Handling

Ships bearing iron ore pellets will arrive at the KM dock. Most will self-unload to a storage pile

on the dock. KM will transfer the pellets into a hopper discharging to a Petmin feed conveyor belt.

The inclined conveyor belt crosses above the railroad tracks and feeds the Petmin screening

building. As there is no pellet storage in the production facility, the pellet material handling will

be essentially a continuous process beginning with the first hopper.

As shown in the following Material Handling Flowchart, KM initially handles the iron pellets for

Petmin. With respect to operations and air quality permitting, KM has responsibility for the pellets

up to the transfer point into the pellet hopper. The Project emissions begin at this transfer point.

Figure 3: Material Handling Flowchart

The pellets are screened in the screening building to remove fines, which are transferred offsite.

Pellets are sprayed with a slurry made from cement and recycled process water to coat the pellets

before going into the reactor following a curing period.




- 9 -

2.2.2 DRI Reactor and Process Gas Loop

In the reactor, the direct reduction of iron oxides into metallic iron is accomplished by reacting the

iron pellets in a countercurrent fashion with reducing gases. Oxygen attached chemically to the

iron reacts with hydrogen (H2) and carbon monoxide (CO), therein producing highly metallized

pellets, called direct reduced iron (DRI).

Reducing gases are generated directly by in-situ reforming of natural gas inside the reactor. Once

treated, the reaction gases are reintroduced into the reactor with a fresh natural gas make-up. For

this process, called “Zero Reformer” (ZR), the reactor could be imagined as two reactors in series:

1. A top cylindrical zone is where the reduction reactions determine the product metallization;

and

2. An intermediate zone, which is the more active zone, for in-situ reforming, pellet reduction

and the carburization reactions that control the final carbon content of the product.

The bottom cone is used only for discharging the product into the hoppers before being transferred

to the EAF.

The iron ore pellets are fed to the top of the reactor, via hoppers, and flow by gravity through the

hot reducing gases, which are ascending the reactor bed in a counter-current fashion. Natural gas

is continuously added to process gas. This mixture of natural gas and process gas is heated to a

temperature of 950°C (1,740°F) in a process gas heater and then mixed with oxygen in order to

increase its temperature by partial and controlled combustion before being finally fed to the

reactor. The distribution of gas across the bed of the reactor is uniform and ensures intimate

contact between gas and solid pellets.

The metallic iron of the pellet acts as the catalyst of the methane cracking reaction, and the gas is

converted into reforming gas (H2 and CO). The operating temperature in the reactor is maintained

above 1,080°C (1,980°F).

The primary reforming reactions are:

CH4 + CO2 → 2CO + 2H2

CH4 + H2O → CO + 3H2

These reactions occur in the lower part of the reaction zone, which means that hydrogen and carbon

monoxide are produced in the intermediate zone of the reactor and move toward the upper part of

the reactor, where they encounter the iron ore pellets, primarily consisting of Fe2O3. The reaction




- 10 -

with hydrogen and carbon monoxide produces water and carbon dioxide, respectively. This

reduction reaction simultaneously converts the Fe2O3 to metallic iron (Fe). The primary reduction

reactions are:

Fe2O3 + 3H2 → 2Fe + 3H2O

Fe2O3 + 3CO → 2Fe + 3CO2

Reduction and carburization reactions take place simultaneously in the cylindrical portion of the

reactor, depending on the temperature (heat energy), which is the main driving force for the

kinetics of each particular reaction. Based on the operating temperatures of the ZR process, the

hydrogen-based dry reduction reactions are favored, allowing the production of DRI with a high

level of metallization.

The catalyst, in this case the iron (Fe) in the pellet, is continuously regenerated by the counterflow

of gases with the fall of the pellets by gravity into the reactor. The carburization reaction which is

favored in the ZR process has the advantage of producing DRI with more than 90% of the carbon

fixed as Fe3C.

A rotary valve, located at the bottom of the vessel, regulates the continuous gravity flow of the

charge downward through the reactor. DRI is distributed into two pressurized buffer bins (EAF

bins) by automated mechanisms. From these EAF bins, the hot DRI, now containing 94%

metalized iron, is transferred into the EAF feed hopper.

At the reactor exit, the process gas temperature is around 400°C to 450°C (750°F – 840°F). In

passing through a heat recovery unit, the thermal energy of this gas is recovered in the form of

steam. The process gas is then passed through a quench followed by a venturi, which effects both

the extraction of solids as well as cooling of the gas. Scrubbed gas is then passed through a process

gas separator where the quench water is separated from the process gas. The quench water is sent

to the plant’s wastewater treatment unit for solids recovery. Once cleaned, it is reused as

quenching water. The scrubbed gas exiting the separator is further cooled and condensed in a

quenching cooling tower. The gas is then compressed and aftercooled. One portion of the

quenched gas is fed to the process gas heater as a combustible, while another portion is sent to the

CO2 removal unit. Once the CO2 is removed, the compressed process gas is mixed with make-up

natural gas and is reheated by the process gas heater, thus closing the reduction gas cycle.

The condensed water is sent to the cooling tower and once cooled is sent back to the reduction unit

as a quenching and cooling agent.




- 11 -

During normal process conditions, a pilot flame will always be present in order to ensure rapid

lighting of the flare in the presence of gas. This flare also allows the process gas to be burned in

a safe manner during shutdown and start-up of the reactor and also during certain transitory

conditions (for example when the quality of the process gas is such that it cannot be used as a

combustible for the process gas heater). It is anticipated that during the first year of operation, the

startups and shutdowns could be more frequent while the plant is fully commissioned and

debugged. At that point, it is expected that typically two or three startups /shutdowns may be

encountered each year.

2.2.3 CO2 Removal

CO2 must be removed from gas exiting the reduction reactor in order to avoid being accumulated

in the process gas that is used as a reducing element. The system is composed of two columns

along with peripheral equipment. The first column is used to absorb the acid gases (CO2 and H2S)

and the second releases acid gases and regenerates amine. CO2 and H2S are extracted by

absorption. Gas circulates in the absorber and enters into contact while going countercurrent to

the methyl diethanolamine solution (MDEA at a concentration of around 50%) containing no

absorbed acid gases (regenerated amine).

The gas at the top of the absorber is washed with water in order to prevent amine build-up in the

treated gas. Demineralized water is continuously added in the absorber to make up for the water

lost in the purified process gas and in the acid gases of the regenerator.

From the absorber, the amine solution containing absorbed acid gases (enriched amine) is heated

via a heat exchanger and sent to the desorption column where it is further heated with a reboiler,

allowing carbon dioxide (CO2) and hydrogen sulfide (H2S) to be released. From the regenerator

exit, water vapor is extracted by a condenser.

Finally, the regenerated amine solution is returned to the absorber after being cooled to achieve

the required temperature while the process gas is returned to the reduction unit.

Part of the regenerated amine flow is filtered with activated carbon and centrifuged in order to

avoid the build-up of solids to extend the useful life. The centrifugation will generate a sludge,

containing mainly iron dust and a mix of water and MDEA. The sludge will be disposed of off-

site.

The gas containing CO2 and H2S may then be treated in either one of the following two options.

Petmin is requesting a permit that allows for operation under Scenario 2, which also covers




- 12 -

operation under Scenario 1 in order to account for uncertainties associated with the operation,

reliability, and long-term viability of a gas purification plant.

Scenario 1 – Gas Purified for Sale

In this scenario, the gas stream is sold to an independent third-party who processes the gas in a

CO2 Processing Plant onsite. This separate facility purchases the CO2 gas stream from the Petmin

facility as its feedstock and purifies the CO2. This process is described in more detail in Section

3.

The CO2 Processing Plant receives the CO2 that contains H2S. The sulfur is removed in the third-

party’s gas purification process, and the purified CO2 is sold on the commercial market.

Petmin recognizes that the CO2 processing plant will require planned maintenance activities.

Typically, these maintenance events will be planned and executed at the same time the DRI plant

is undergoing planned maintenance activities. Petmin also recognizes that the CO2 plant could

require some type of unscheduled maintenance or repair activity, or equipment breakdown of some

type, that would necessitate the temporary shutdown and “maintenance or safety bypass” of the

CO2 recovery plant.

Bypass of the CO2 Processing Plant requires a means to remove the hydrogen sulfide (H2S) from

the gas. For this reason, Petmin will have a thermal oxidizer in the bypass vent to oxidize the H2S

to sulfur dioxide, followed by a liquid, packed tower scrubber to remove the sulfur dioxide prior

to discharge of the resultant gas stream to the atmosphere via the PGH stack.

Scenario 2 – Gas Treated to Remove Sulfur and Discharged to Atmosphere

As a new venture with a degree of uncertainty, Petmin recognizes a number of possible situations

whereby CO2 purification/sale may not be practicable, or the CO2 Processing Plant may experience

a long-term shutdown. Under Scenario 2, the sulfur is removed from the CO2 offgas with the

sulfur-lean gas thereafter being discharged to atmosphere.

Just as in the “bypass mode” described in Scenario 1, the CO2 (with its contaminant H2S) is treated

in two steps:

• Oxidation of the H2S to SO2 with a thermal oxidizer, then

• Removal of the SO2 with an absorption column (i.e., packed tower scrubber), transferring

the sulfur, in dissolved solid salt form, to wastewater for pretreatment and discharge to the

City of Ashtabula. The SO2 scrubber will be designed to achieve a 98 percent removal

efficiency.




- 13 -

Addressing the Two Possible Operating Scenarios in the Permit to Install

Petmin is requesting authorization to operate under Scenario 2. This authorization also provides

Petmin the flexibility to operate under Scenario 1.

2.2.4 H2S Injection

Dimethyl disulfide (DMDS) is first heated by being mixed with a portion of the process gas and

then injected into a small, molybdenum and cobalt-based catalyst bed, where it is converted to H2S

and then added to make-up gas (maximum concentration of 25 ppmvd) upstream of the process gas

heater. The H2S allows the internal surfaces of the process gas heater to be protected from

corrosion. As noted above, the H2S exits the process gas loop and is oxidized to sulfur dioxide,

which is removed with a scrubber.

2.2.5 Reduction Process Start-up

The reactor takes approximately twenty-four hours to start up. Initially during start-up, the reactor

is empty, cold and pressurized with N2 gas. The feed rate of the package boiler is increased. The

reduction circuit is pressurized with nitrogen. The process gas heater and the reduction-sector

compressor are started, and iron pellets are fed to the reactor, which is full for around eleven hours

of start-up. The amine solution is preheated using steam from the package boiler. During this

time, only nitrogen is fed into the reduction circuit. Starting from the tenth hour of start-up, before

adding natural gas as a reducing agent, the preheated gases are sent to the gas quencher. The

temperature at the exit of the process gas heater gradually increases and reaches 850°C (1,560°F)

after eleven hours of start-up.

During the twelfth hour of start-up, injection of natural gas as a reducing agent commences, oxygen

is added downstream of the process gas heater, process gas is gradually formed, and, starting from

the thirteenth hour, reducing gas is fed to the reactor and the reduction of pellets commences. Over

a period of around four to six hours, part of the process gas exiting the reactor is sent to the flare

as it does not have the required quality to be used as fuel in the process gas heater. The other part

is sent to the CO2 removal unit.

Over the next eight hours, the flowrates of natural gas (as a source of reducing gas) and process

gas reach their desired values and the adjustment of operation parameters allows for the operation

to reach steady-state. At the end of the twenty-fourth hour, the DRI has the desired quality and

the normal production rate is attained.

Over several hours, the DRI metallization rate is less than the desired level. The off-spec DRI

produced during start-up is sent to a conveyor where its temperature is reduced through direct

contact with water in a quenching cooling conveyor. The off-spec DRI (also called “Remet”) is




- 14 -

recovered, temporarily stockpiled, and returned to the iron ore pellet conveyor feed hopper where

it is fed at a low input rate.

2.2.6 Reduction Process Shutdown

During the first five hours of reduction process shutdown, the production rate is gradually slowed,

until the injection of natural gas as a reducing agent is halted, the addition of combustion oxygen

is stopped (downstream of the process gas heater) and the CO2 removal unit is shutdown. Starting

from the sixth hour, the feeding of reducing gas to the reactor is replaced by nitrogen and reducing

gas is diverted to the gas washer. Process gas from the reactor is then sent to the flare. The reactor

is emptied. The amine solution is cooled.

The temperature of the process gas heater is gradually decreased until the fourteenth hour when it

is completely shut down. Over the next six hours, the reactor is completely emptied, and the

reducing gas compressor is shut down. The reduction circuit, including the reactor is finally

purged with nitrogen for the last two hours of shutdown.

2.2.7 Smelting in the Electric Arc Furnace and Pig Iron Casting

The hot DRI pellets are liquefied in an electric arc furnace (EAF) at around 1,500°C (2,730°F),

producing two distinct phases: metal and slag. The heat of melting of the pellets is achieved using

an electric current generated by three graphite electrodes.

A buffer silo (receiving hopper) is needed to couple the continuous reduction process to the semi‐

continuous melting process. The average feed rate to the EAF is 60.13 tons per hour. The EAF

operating cycle is known as a tap-to-tap cycle. The function of the feed rate and the electrical

supply as a function of time elapsed is described below.

• The total melting cycle lasts approximately 132 minutes.

• The hot DRI (approximately 675°C, 1,250° F) is fed to the EAF at a rate of 1.2 tons per

minute for around 60 minutes.

• The feed rate is then increased to around 1.4 tons per minute for approximately forty to

forty-five minutes and eventually diminishes to close to 1.2 tons per minute over the last 5

to 7 minutes.

• While the DRI is being loaded, the electric supply varies between 32 and 37 MW.

• The heat produced by the electric arc furnace melts the DRI at a temperature above 1,500°C

(2,730° F).

• During the next twenty minutes, the electric current ceases, and the loading of the EAF is

interrupted. The slag is deposited, the batch is sampled, and the metal is tapped.




- 15 -

• Once the tapping is finished, around 12 minutes are necessary to prepare the equipment for

the next cycle. During this period the level of the graphite electrodes is adjusted.

• Total yield per cycle is approximately 132.3 tons of liquid metal and 20.7 tons of slag.

In order to minimize heat losses during the transport of hot DRI between the reactor and the EAF,

the reactor is approximately thirty meters higher and placed near the EAF in order to feed by

gravity.

During the loading of DRI in the EAF, a mixture of lime and dololime is added (around 0.88 ton

per batch). Bauxite may also be added to adjust the characteristics of the slag. This flux agent

facilitates the separation of impurities contained in the pellets and is essential to controlling the

quality of the pig iron.

Metal, now containing more than 96% iron, is poured into ladles and transported to the pig caster,

which is cooled by direct contact with water. Water evaporates upon contact with hot metal and

make-up water is added to the circuit. Pig iron product drops from the pig caster to an outdoor

bunker before being moved to a storage pile awaiting shipment.

The slag byproduct is transferred to an independent third party at the point of its formation, i.e.,

pouring into the slag pots owned and transported by the third-party. The Slag Processing

Contractor (SPC) slag pot transport vehicle moves the pot containing molten slag to a slag

processing area operated by the SPC where the slag is decanted, cooled, and solidified. Once

cooled and solidified, the slag is broken into blocks which are then transported by the SPC to the

SPC’s processing facility.

2.2.8 EAF Fume Capture System

The primary fumes generated during DRI melting will be extracted through the hood around the

electrodes on the furnace roof, by a water-cooled duct. The connection between the furnace hood

and water-cooled ducts is obtained with a movable sliding skirt foreseen to allow the change of

the gap for the air inlet; the sliding skirt is located on the first fixed duct. After this connection,

before the fixed cooling duct line, the arrangement of a settling chamber – or drop out box (DOB)

– will allow for the combustion of CO and separation of the heaviest particles of dust from the off‐

gasses.

The main parameter to be controlled for optimum primary fume suction, according to the EAF

working conditions, is the pressure inside the furnace. For maintenance reasons, the pressure probe

is located inside the furnace duct elbow. The target is to have slightly positive pressure in the EAF




- 16 -

to avoid the direct suction and consequent infiltration of fresh air, which would unnecessarily

oxidize the EAF bath.

The secondary fume generated during the various furnace operating phases is controlled with a

canopy hood located on the main building roof. The canopy hood is designed according to the

EAF dimensions and levels to maintain the required ventilation inside the main melt shop building.

Figure 4: Primary and Secondary EAF Hoods

Figure 4 shows only the primary and secondary EAF hoods. There are additional hoods dedicated

to ladles, casting, bins, loading, etc. which ensure that the entire EAF building is evacuated to the

baghouse.

2.2.9 Auxiliary Boiler

A package boiler, with a heat input capacity of 15.2 million BTU per hour, will supply the steam

necessary to the CO2 removal unit as well as the energy necessary during the start-ups and

shutdowns of the facility. The vast majority of the time, the auxiliary boiler will function at a

minimal load.

2.2.10 Storage Silos

Lime, dololime and bauxite will be stored in silos (capacity of 120 m3 each) outside of the smelting

and casting building. The vent on each silo will be connected to the EAF baghouse duct system.




- 17 -

Cement will also be delivered by truck and stored in a silo, with a fabric filter on the vent, located

near the coating operation. Between 4,000 and 5,000 tons per year of cement will be required to

supply the facility.




- 18 -

3.0 FEDERAL & STATE RULE APPLICABILITY

This section presents a summary of program applicability, and emissions limitations and standards

that apply to the proposed Project.

The Project is located in Ashtabula, OH, which is designated as attainment or unclassified for all

air pollutants for which a National Ambient Air Quality Standard (NAAQS) has been established.

Therefore, the federal regulations for Prevention of Significant Deterioration (PSD) apply for all

relevant pollutants.

3.1 Defining the Scope of the Project for Regulatory Analysis and Permitting

A first step in evaluating regulatory applicability involves defining the scope, or battery limits of

the Petmin Project. Under its EPA-approved NSR rules, Ohio EPA retains the ultimate discretion

to make source determinations.

3.1.1 Kinder Morgan Terminal a Separate Stationary Source

As described in Section 2, KM will be contracted to provide iron pellets to the Petmin hopper. At

its Pinney Dock Terminal, KM routinely provides unloading, storage and material handling

services for various bulk commodities and general cargoes. The terminal encompasses 310 acres,

with 200 acres of open storage, with a storage capacity of 7 million tons. Transportation modes

are by water (Lake Erie), railroad (Norfolk Southern and CSX) and major roadways (Ohio State

Route 11, Interstate 90 and the Ohio Turnpike).

At plant capacity of 526,739tons MPI production, Petmin will require 874,200 tons per year of

iron pellets. As a conservatively high estimate, a three-month supply onsite would be 218,500

tons, or 2.5 percent of KM’s storage capacity. The Terminal typically handles on the order of 6 to

8 million tons per year of bulk commodities (limestone, iron ore, mineral sands, agri-products,

rock salt, coal, coke, etc.). Therefore, the iron ore pellets handled for transfer to Petmin will

represent on the order of 10 percent of the Terminal’s typical throughput.

On April 30, 2018, U.S. EPA issued clarifying guidance (referenced hereinafter as the “Common

Control Memorandum”) for purposes of determining when, for New Source Review purposes, a

facility is not under common control, and thus is its own unique “stationary source.2 Following

the principles of this guidance, KM and Petmin are clearly distinct, separate stationary sources for

purposes of NSR applicability. Key facts include:




- 19 -

1. No Common Control – According to the Common Control Memorandum, “the most relevant

considerations should be whether entities have the power to direct the actions of other entities to

the extent that they affect the applicability of and compliance with permitting requirements.”

• KM and Petmin share no common ownership. Petmin will have no control over, or “power to

direct” KM’s pollutant-emitting or permit compliance activities.

• KM has distinctly separate ownership, management, employees, and environmental staff.

• KM has its own, longstanding Ohio Permit to Install and Operate for its terminal operations,

with existing capacity for handling Petmin’s iron pellets, as described in Section 2 of this

report.

• KM is not dependent on Petmin for compliance with any portion of the requirements associated

with the material handling of iron pellets for delivery to Petmin.

• KM and Petmin will have a mutually beneficial contract for the supply of iron pellets to the

Petmin hopper. However, neither has the power or ability to direct the relevant activities of the

other.

2. Line of Demarcation – As described in Section 2, a clear line of demarcation sets the battery limits

for KM activities relative to Petmin. Petmin takes responsibility for the iron pellets at the feed

hopper supplying the Petmin plant.

3. Dissimilar operations, as shown by Standard Industrial Codes (SIC):

• Petmin SIC: 3312

• KM SIC: 4213 – Marine Cargo & Handling

In conclusion, Petmin does not control KM simply because KM will store Petmin’s iron pellets

and transfer those pellets to the hopper that feeds Petmin’s conveyor system. Second, there is no

indication that Petmin has any power or authority over other activities at KM. KM simply handles

a feedstock under a mutually beneficial arms-length arrangement between the two wholly-separate

business entities. Under the Common Control Memorandum, it appears clear that U.S. EPA would

not consider Petmin and KM part of the same stationary source.

3.1.2 Carbon Dioxide Plant a Separate Stationary Source

As described in Section 2, Petmin is seeking a mutually beneficial arrangement with a specialty

gases producer to sell its untreated CO2 from the DRI reactor gas recovery loop as feedstock to a

CO2 Processing Plant. If proven to be viable, CO2 capture and treatment provides the

environmental benefit of reducing greenhouse gas emissions from manufacturing MPI.

Plant Description

As described in Section 2, Petmin is requesting a Permit to Install that provides for, but does not

require, the operation of a CO2 Processing Plant. The CO2 Processing Plant, if constructed, will

require an approximately ¾ acre footprint, with a total of up to as much as 5 acres for material

handling (truck staging and/or rail siding). The exact site of the plant has not yet been determined




- 20 -

but will be sited as close to the Petmin equipment as possible in order to obviate a booster

compressor.

The CO2 feedstock stream will contain the following low-level contaminants that require removal

to produce 4.0 grade CO2: hydrogen sulfide, hydrogen, carbon monoxide and methane.

The precise design details of the gas purification process are under development by potential third-

parties. In one current proposal, the independent company will compress the feed stream raw

process gas, rich in CO2 from approximately 3 to 6 pounds per square inches gauge pressure (psig)

to approximately 320 psig. Hydrogen sulfide will be removed via a scavenger process. Water will

be removed from the stream in several stages of the process. Methane, hydrogen and carbon

monoxide will be removed via regenerative chemical adsorption processes. Nitrogen will be

removed in a distillation and boiling process.

Purified CO2 will be liquefied and subcooled to approximately (minus) -10 F° and then stored at

between 250 and 275 psig. After removal of the impurities, the plant will provide approximately

15 tons per hour (~30,700 pounds per hour) of 99.99+ percent purity 4.0 grade CO2.

The H2S will be removed and the sulfur-bearing adsorption media handled as follows (or

equivalent process):

The sulfur is adsorbed in a two-bed, lead / lag setup with a material such as Axens AxTrap 4513, which

is a patent-pending, high-capacity non-hazardous granular media specifically designed for use in the

removal of H2S. The media adsorbs contaminant level carbonyl sulfide, mercaptans and organic

sulfides, if present, from CO2 for total sulfur control. It is comprised of a porous mixed zinc-metal

oxide formed on a stable inert base along with an inorganic adsorption phase specific for heavier sulfur

compounds.

The beds are designed for an estimated 180 days on line (per bed) and then switched while the spent

bed is being replenished. The spent material is ether sent back to Axens or landfilled (non-hazardous).

Methane and CO are removed as separate streams of the Carbon Dioxide Plant (CDP) purification

steps. Depending on final costs and technical details, there are two possibilities:

• Methane and CO will be sold back to Petmin and reintroduced with the feed gas to the DRI; or

• Methane and CO will be flared, with a minimum removal efficiency of 98%,

The CDP will obtain its own PTIO from Ohio EPA, if one is required. (Note: It may be the case

that no PTIO is required). Similarly, the CDP will also have permitting and compliance




- 21 -

responsibility for wastewater discharges, solid waste, and other applicable environmental

programs.

The CDP will take ownership of the untreated gas stream. The CDP will also assume responsibility

for its plant’s equipment malfunctions and, if it were to occur, the associated release of

uncontrolled feedstock gas. The CDP may also be responsible for acceptance of feedstock gas

during Petmin startup and shutdown.

As noted above, CDP details have not been finalized. Petmin is seeking a PTI that provides for

operation of its plant with or without a CDP.

While not critical to the following analysis of “common control,” it is also noteworthy that the

determination of independent control for a CDP does not impact program applicability for Petmin.

The controlled emissions of CO and methane are minor source level emission rates.

• Controlled CO emissions are less than 10 tons per year.

• Controlled methane emissions are less than 10 tons per year.

• Even if Petmin did have ownership responsibility for these emissions (which, according to the

Common Control Memorandum, they would not) the Petmin Project would remain below the

significance threshold for CO. Petmin already triggers BACT review for Greenhouse Gas

Emissions. The additional methane emissions would be trivially small.

Analysis with Respect to the Common Control Memorandum

This business relationship closely matches the situation described in the Common Control

Memorandum. There, a collocated independent business entity is purchasing landfill gas and

purifying it to pipeline quality natural gas. In the case here, an independent business will be

purchasing untreated CO2 as feedstock and purifying it to high quality CO2.

In both cases, the ownership and management of the gas purchaser/purification plant is clear. The

company that produces the gas as a byproduct of its principal operation has no ownership,

management, or other nexus that would suggest “control” of the gas purification plant. Key facts

include:

1. No Common Control – According to the Common Control Memorandum, “the most relevant

considerations should be whether entities have the power to direct the actions of other entities to

the extent that they affect the applicability of and compliance with permitting requirements.”

• The Carbon Dioxide Plant (CDP) and Petmin would share no common ownership. Petmin

would have no control over, or “power to direct” the CDP’s pollutant-emitting or permit

compliance activities.




- 22 -

• The CDP would have distinctly separate ownership, management, employees, and

environmental staff.

• The CDP would acquire its own PTIO, if one is required. The CDP is not dependent on Petmin

for compliance with any portion of the requirements associated with the purification and sale

of the CO2.

• The CDP and Petmin would have a mutually beneficial contract for the purchase of feedstock,

then the CDP’s purification and resale, of high grade CO2.

2. Line of Demarcation – A clear line of demarcation sets the battery limits for the CDP activities

relative to Petmin. The CDP takes responsibility for the CO2 gas stream at a single designated point

in a closed vent system from Petmin that feeds the CDP.



• CDP SIC: 2813 – Industrial Gases

In conclusion, Petmin would not control the CDP. The CDP will purchase the untreated CO2 gas

stream and take responsibility at the transfer point. Second, there is no indication that Petmin

would have any power or authority over other activities at the CDP. Petmin simply would supply

a feedstock under a mutually beneficial arms-length arrangement between the two wholly-separate


not consider Petmin and the CDP part of the same stationary source.

3.1.3 Slag Processor a Separate Stationary Source

In addition to the untreated CO2 described in the previous section, the other principal byproduct

from the Petmin manufacturing process is metallurgical slag.

Petmin is negotiating with an independently owned third-party Slag Processing Company (SPC)

to take ownership and responsibility (using the SPC’s equipment) of the slag at its point of

generation, i.e., as it is poured from the EAF into the SPC-owned slag pots. The third-party vendor

will also manage the slag stockpile. No processing of slag is proposed to be performed onsite,

although the third-party vendor may need to break up the slag into a manageable size prior to

loading tri-axle dump trucks and moving the slag off-site.

As described in Section 2 of this report, slag is poured to the slag pot from the EAF; then, metal,

now containing more than 96% iron, is poured from the EAF into ladles and transported to the pig

caster, which is cooled by direct contact with water and by contact with the air. Water evaporates

upon contact with hot metal and make-up water is added to the circuit.

Each SPC-owned slag pot can hold several EAF loads. The SPC transports the pot containing

cooled and solidified slag to the storage area. Close to ground level, the transport vehicle angles




- 23 -

the pot and the block of slag falls to the ground. From there, the slag is transported via the

contractor to an outside facility.

The SPC and Petmin are clearly distinct, separate stationary sources for purposes of NSR

applicability. Key facts include:

1. No Common Control – According to the memorandum, “the most relevant considerations

should be whether entities have the power to direct the actions of other entities to the extent that

they affect the applicability of and compliance with permitting requirements.”

• The SPC and Petmin share no common ownership. Petmin will have no control over, or “power

to direct” the SPC’s pollutant-emitting or permit compliance activities.

• The SPC has distinctly separate ownership, management, employees, and environmental staff.

• The SPC will have its own, separate, minor source PTIO with Ohio EPA, if one is required.

• The SPC is not dependent on Petmin for compliance with any portion of the requirements

associated with the slag processing from receipt to its shipment offsite.

• The SPC and Petmin will have a mutually beneficial contract for the handling and removal of

slag. However, neither has the power or ability to direct the relevant activities of the other.

2. Line of Demarcation – A clear line of demarcation sets the battery limits for SPC activities relative

to Petmin. The SPC takes ownership and control of the slag upon its discharge from the furnace

into the SPC-owned slag pot.



• SPC SIC: 3295 – Minerals and Earths, Ground or Otherwise Treated

In conclusion, Petmin does not control the SPC simply because it transfers ownership of its slag

byproduct. Second, there is no indication that Petmin has any power or authority over other

activities at the SPC. Petmin simply supplies a byproduct which then becomes a feedstock for the

SPC under a mutually beneficial arms-length arrangement between the two wholly-separate


not consider Petmin and the SPC part of the same stationary source.

3.2 Federal Rule Applicability

This section summarizes the applicability of federal rules and standards to the Project and its

emissions.

3.2.1 Prevention of Significant Deterioration (PSD)

As confirmed with Ohio EPA Division of Air Pollution Control, the manufacture of MPI categorizes

Petmin into one of the source categories in OAC 3745-31(NNN)(2)(a). Therefore, the major source




- 24 -

threshold for PSD pollutants is 100 tons per year. Based on total Project emissions, Petmin will be a

major stationary source as defined in OAC 3745-31-01(NNN), as summarized in the following table.

Project Potential to Emit & PSD Applicability

Pollutant PTE

(tons/year)

Major Source

Threshold

(tons/year)

Major?

Significant

Emission Rate

(tons/year)

Significant?

PM10 63.97 100 No 15 Yes

PM2.5 52.40 100 No 10 Yes

NOx 484.57 100 Yes 40 Yes

SO2 3.63 100 No 40 No

CO 546.22 100 Yes 100 Yes

OZONE (VOC) 16.80 100 No 40 No

GHG/CO2e* 391,397 -NA-** -NA- 75,000 Yes

* CO2e and GHGs regulated under OAC 3745-31-34.

** In accordance with U.S. Supreme Court decision in Utility Air Regulatory Group, Petitioner V.

Environmental Protection Agency, et al. No. 12-1146, June 23, 2014, GHGs subject to PSD review only

if the proposed project is subject to PSD review for at least one non-GHG regulated NSR pollutant.

Project emissions of oxides of nitrogen (NOx), carbon monoxide (CO) and greenhouse gases (GHG)

exceed the major source threshold. Therefore, the Project triggers PSD review. The Project then

triggers PSD review for all PSD pollutants that exceed the significant emission rate. As shown in the

above table, the Project triggers PSD review for:

• Oxides of nitrogen (NOx)

• Carbon monoxide (CO)

• Particulate matter less than 10 microns (PM10)

• Particulate matter less than 2.5 microns (PM2.5)

• Greenhouse gases (GHG)

OAC 3745-31-11 through -20 are the regulations for PSD in Ohio. Any new major stationary source

or major modification of an existing major source located in an air quality attainment area3 must

undergo a PSD review. Consistent with Ohio PSD regulations and EPA guidance, the PSD review

consists of the following elements:

• An analysis of best available control technology (BACT) for emissions of those pollutants

for which the project results in a significant net emissions increase from the source.




- 25 -

• An air quality impact analysis to demonstrate that the proposed project will not exceed the

NAAQS for CO and NOx, and thus will not lead to significant deterioration in air quality;

• A Class I impact analysis to demonstrate the proposed project will not impact Class I

protected wilderness areas.

• An additional impacts analysis (e.g., impacts on soil, vegetation, visibility) to demonstrate

that additional impacts resulting from the proposed project are not expected.

Sections 4, 5, and 6 of this report present the results of these analyses.

3.2.2 National Emission Standards for Hazardous Air Pollutants (NESHAP)

MACT Subpart ZZZZ - Stationary Reciprocating Internal Combustion Engines

The Project will include two emergency generators, two emergency fire protection pumps, and one

black start generator. The emergency generators and fire protection pumps, excluding emergency

use, will be operated or limited hours for maintenance and readiness testing. Similarly, the black

start generator is used only at startup, along with limited maintenance and readiness testing.

Preliminary specifications anticipate ratings of 3,131 HP for each emergency generator; 311 HP

for one fire protection pump; 237 HP for the other fire protection pump; and 158 HP for the black

start generator. 40 CFR 63.6590(c)(1), states that such engines shall comply with MACT-ZZZZ

by complying with NSPS-IIII.

MACT-ZZZZ, at 40 CFR 63.6590(c)(1), states that new CI RICE at an area source of HAP

shall comply with MACT-ZZZZ by complying with NSPS-IIII.

40 CFR 60.4205(b) and 40 CFR 60.4211(c) (emergency, <30 L/cylinder displacement)

requires the emergency generators to meet Tier 4 standards.

40 CFR 60.4205(c) (emergency, <30 L/cylinder displacement) requires fire pumps to comply

with the emission standards listed in Table 4 of NSPS-IIII.

40 CFR 60.4204(b) and 40 CFR 60.4211(c) (non-emergency, <30 L/cylinder displacement)

requires the black start engine to meet Tier 4 standards.

40 CFR 60.4211(f) limits non-emergency engine operation, including readiness testing, to 100

hours per year.




- 26 -

MACT Subpart DDDDD - Industrial, Commercial, and Institutional Boilers and Process

Heaters – Negative Declaration

Petmin is not subject to this subpart because the facility is not a major source of HAP. The facility

has potential HAP emissions of less than 10 tpy for any single HAP and less than 25 tpy for all

HAPs.

MACT Subpart EEEEE - Iron and Steel Foundries – Negative Declaration

Petmin is not subject to this subpart because the facility is not a major source of HAP. Petmin is

also not subject to the rule because it will not meet the definition of 'iron and steel foundry' found

in 40 CFR 63.7765:

Iron and steel foundry means a facility or portion of a facility that melts scrap, ingot, and/or

other forms of iron and/or steel and pours the resulting molten metal into molds to produce

final or near final shape products for introduction into commerce.

MACT Subpart FFFFF - Integrated Iron and Steel Manufacturing Facilities – Negative

Declaration

Petmin is not subject to this subpart because the facility is not a major source of HAP. Petmin is

also not subject to the rule because it will not meet the definition of ‘Integrated iron and steel

manufacturing facility' found in 40 CFR 63.7852:

Integrated iron and steel manufacturing facility means an establishment engaged in the

production of steel from iron ore.

MACT Subpart YYYYY - Area Sources: Electric Arc Furnace Steelmaking – Negative

Declaration

Petmin is not subject to the rule because it does not produce carbon, alloy or specialty steel and

therefore is not included in the definition of ‘Electric arc furnace steelmaking facility' found in 40

CFR 63.10692:

Electric arc furnace (EAF) steelmaking facility means a steel plant that produces carbon, alloy,

or specialty steels using an EAF.

MACT Subpart ZZZZZ - Area Sources: Iron and Steel Foundries – Negative Declaration

Petmin is not subject to the rule because it will not meet the definition of 'iron and steel foundry'

found in 40 CFR 63.10906:




- 27 -

Iron and steel foundry means a facility or portion of a facility that melts scrap, ingot, and/or

other forms of iron and/or steel and pours the resulting molten metal into molds to produce

final or near final shape products for introduction into commerce.

MACT Subpart JJJJJJ - Industrial, Commercial, and Institutional Boilers Area Sources –

Negative Declaration

Petmin is not subject to the rule because it does not operate any boiler within a listed subcategory

listed in 40 CFR 63.11200:

The subcategories of boilers, as defined in §63.11237 are:

(a) Coal.

(b) Biomass.

(c) Oil.

(d) Seasonal boilers.

(e) Oil-fired boilers with heat input capacity of equal to or less than 5 million British thermal

units (Btu) per hour.

(f) Boilers with an oxygen trim system that maintains an optimum air-to-fuel ratio that would

otherwise be subject to a biennial tune-up.

(g) Limited-use boilers [burns solid or liquid fuels, by definition].

3.2.3 Standards of Performance for New Stationary Sources (NSPS)

NSPS - AAa: Steel: Electric Arc Furnaces and Ar-O2 Decarburization Vessels After Aug. 7,

1983 – Negative Declaration

Petmin is not subject to the rule because it does not produce carbon, alloy or specialty steel, as

required by the statement of applicability in 40 CFR 60.270a(a). In addition, the definition of

'electric arc furnace', in 40 CFR 60.271, specifically excludes processing of DRI:

Electric arc furnace (EAF) means a furnace that produces molten steel and heats the charge

materials with electric arcs from carbon electrodes. For the purposes of this subpart, an EAF

shall consist of the furnace shell and roof and the transformer. Furnaces that continuously feed

direct-reduced iron ore pellets as the primary source of iron are not affected facilities within

the scope of this definition.




- 28 -

NSPS - LL: Metallic Mineral Processing Plants – Negative Declaration

Petmin is not subject to the rule because it does not produce metallic mineral concentrates from

ore. Petmin is one of the "non-adjacent locations that will subsequently process metallic

concentrates into purified metals (or other products)", as described in 40 CFR 60.381:

Metallic mineral processing plant means any combination of equipment that produces metallic

mineral concentrates from ore. Metallic mineral processing commences with the mining of ore

and includes all operations either up to and including the loading of wet or dry concentrates or

solutions of metallic minerals for transfer to facilities at non-adjacent locations that will

subsequently process metallic concentrates into purified metals (or other products).

Because no Part 60 Subparts apply, 40 CFR Part 60, Subpart A, the so-called “General Provisions,”

also does not apply.

NSPS Subpart IIII - Stationary Compression Ignition Internal Combustion Engines

The project will include two emergency generators, two emergency fire protection pumps, and one

black start generator. Preliminary specifications anticipate ratings of 3,131 HP emergency

generators, one 311 HP and one 237 HP fire protection pump, and 158 HP for the black start

generator. NSPS-IIII is applicable to these engines. 40 CFR 60.4211(f) limits non-emergency

operation of the engines, including readiness testing, to 100 hours per year.

3.2.4 Preconstruction Permit Requirements

OAC 3745-31-02(A) provides that a Permit to Install (PTI) is required from Ohio EPA prior to

initiating construction or modification of a source of air contaminants unless specifically exempted

in OAC 3745-31-03. The PTI issued by Ohio EPA fulfills the preconstruction permitting

requirements under federal PSD rules. A construction permit is required for the Petmin project.

This report and its attachments are provided as a complete application package, compiled to

include all the various elements required for issuance of a PSD permit.

OAC 3745-31-05 requires all new sources to minimize emissions using the best available

technology (BAT), which is analyzed individually for each emission unit.

3.2.5 CAIR NOx Annual Trading Program – Negative Declaration

40 CFR 96 and OAC 3745-109 are the Clean Air Interstate Rules (CAIR) for NOx. OAC 3745-14

contains similar rules for Ohio’s NOx Budget Trading Program. These rules apply to generators,

i.e., producers of electricity, that meet certain size and other applicability criteria. The Petmin




- 29 -

project does not involve a generator. The CAIR are not applicable. Note that both cited state rules

were rescinded on January 29, 2018.

3.2.6 Title V Federal Operating Permit Program

Petmin is subject to rule 40 CFR 70 and OAC 3745-77, as a major facility under these rules. The

application process for a Title V operating permit for a new stationary source in Ohio begins after

the issuance of the final PTI and after the plant has begun operations. Petmin will be required to

submit an application for a Title V operating permit within one year of startup. The PTI authorizes

the operation of the plant until a final Title V permit is issued by Ohio EPA. For a new source, the

Title V permit typically mirrors the PTI in its facility and emissions-unit specific terms and

conditions.

3.2.7 Compliance Assurance Monitoring

The EAF baghouse is subject to the requirements of 40 CFR 64. A CAM plan is required by 40

CFR 64.2(a) because the EAF is subject to a PM standard and uses a control device to reduce

uncontrolled emissions that would exceed the major source threshold. A CAM Plan will be

submitted with the Title V permit application.

3.2.8 Other Applicable Emission Limitations and Standards

No other federal emission limitations or standards have been found to apply to the proposed

project.

3.3 State of Ohio Regulations & Standards

This section summarizes the applicability of Ohio-specific regulations and standards.

3.3.1 General Requirements

The following rules apply to all sources or air contaminants in Ohio for which a PTI is required.

Annual Emissions Fee

Ohio EPA will assess a separate fee based on the total annual emissions from the facility.

Emissions are self-reported in accordance with Ohio Administrative Code (OAC) Chapter 3745-

78. This fee assessed is based on a fee schedule in ORC section 3745.11 and funds Ohio EPA's

permit compliance oversight activities.




- 30 -

Scheduled Maintenance

Scheduled maintenance of air pollution control equipment must be performed in accordance with

OAC rule 3745-15-06(A). If scheduled maintenance requires shutting down or bypassing any air

pollution control equipment, the operator must also shut down the emissions unit(s) served by the

air pollution control equipment during maintenance, unless the conditions of OAC rule 3745-15-

06(A)(3) are met. Any emissions that exceed permitted amount(s) under the permit (unless

specifically exempted by rule) must be reported as deviations.

Malfunction Reporting

A reportable malfunction of any emissions unit(s) or any associated air pollution control system,

must reported to the Ohio EPA District Office in accordance with OAC rule 3745-15-06(B).

Malfunctions that must be reported are those that result in emissions that exceed permitted

emission levels. The permittee has the responsibility to evaluate control equipment breakdowns

and operational upsets to determine if a reportable malfunction has occurred.

Ohio EPA Authorization to Inspect, Require Tests, Request Information

Under Ohio law, the Director or his authorized representative may inspect the facility, conduct

tests, examine records or reports to determine compliance with air pollution laws and regulations

and the terms and conditions of the permit. The permittee is required to provide, within a

reasonable time, specific types of information Ohio EPA requests either verbally or in writing.

Prohibition of Air Pollution Nuisances

The final PTI and OAC rule 3745-15-07 prohibit operation of the air contaminant source(s)

regulated under this permit in a manner that causes a nuisance. Ohio EPA can require additional

controls or modification of the requirements of this permit through enforcement orders or judicial

enforcement action if, upon investigation, Ohio EPA determines existing operations are causing a

nuisance.

Emissions Unit-Specific Exemptions

Certain emissions units are permanently exempt from obtaining a PTI, as listed in OAC rule 3745-

31-02. Certain emissions units are exempt from air pollution control requirements and from PTI

requirements based on de minimis emission rates of regulated air pollutants and HAPs, pursuant

to OAC rule 3745-15-05. However, the potential to emit from these otherwise exempt units must

be included in the facility-wide potential to emit analysis for purpose of major source applicability

criteria.




- 31 -

Opacity

OAC 3745-17-07(A) limits opacity from stacks to 20% as a six-minute average and OAC 3745-

17-07(B) limits opacity from non-stack (fugitive) sources to 20% as a three-minute average. For

new sources, a lower limit is often established as part of the BAT determination.

3.3.2 Criteria of the Director and Best Available Technology

OAC rule 3745-31-05 stipulates that a PTI shall be approved and issued based on an application

satisfactorily demonstrating that the installation and operation will employ Best Available

Technology (with specific exceptions noted).

(Note: State BAT should not be confused with the generally more stringent federal requirement

for a Project to meet BACT for the pollutants that exceed a significant emission rate).

"Best available technology" or "BAT" means any combination of work practices, raw

material specifications, throughput limitations, source design characteristics, an evaluation

of the annualized cost per ton of air pollutant removed, and air pollution control devices

that have been previously demonstrated to the director of environmental protection to

operate satisfactorily in this state or other states with similar air quality on substantially

similar air pollution sources.

In general, BAT determinations are made by Ohio EPA in accordance with an Ohio EPA

Interoffice Memo issued February 7, 2014.4

3.3.3 Particulate Matter

OAC 3745-17-08 requires control measures be implemented for sources of fugitive dust in areas

of the state defined by Appendix A. The city of Ashtabula, including the Petmin site, are included

in Appendix A.

OAC 3745-17-10(B)(1) limits particulate emissions from gas-fired boilers and heaters to not more

than 0.020 pound per million BTU actual heat input.

OAC 3745-17-11(A)(2) limits particulate emissions from processes by applying two different

calculations, one based on maximum throughput of processed materials, and the other based on

the uncontrolled emission rate. This rule is rarely relevant to new emissions units, as a lower limit

or control requirement is established in its place under the BACT or BAT requirement.




- 32 -

3.3.4 Sulfur Dioxide Emissions

OAC 3745-18-06(A) exempts natural gas combustion from the sulfur dioxide limits. OAC 3745-

18-06(E) calculates an emission limit from process sources in Ashtabula County based on their

maximum process weight rate, in tons per hour, with the equation: 30 P0.67. As with the calculated

particulate matter limits, the allowable emission rate calculation for sulfur dioxide often yields an

unreasonably high value that is replaced with a lower BAT limit.

3.3.5 Control of Oxides of Nitrogen (NOx)

OAC 3745-110 requires that reasonably available control technology (RACT) be applied to any,

“very large boiler, large boiler, mid-size boiler, small boiler, stationary combustion turbine,

stationary internal combustion engine, or reheat furnace”, unless exempted by OAC 3745-110(K),

which includes:

(1) Any industrial boiler having a maximum heat input of less than or equal to twenty mmBtu/hr.

(2) Any emergency standby boiler, stationary internal combustion engine, or stationary combustion

turbine which operates less than five hundred hours during any consecutive twelve-month period.

However, the owner or operator of the emergency standby engine, boiler, or turbine shall maintain

for a period of not less than three years, in a bound log book, or other format acceptable to the

director, a list of the dates and number of hours the emergency standby engine operated.

Petmin has no equipment or processes to which this rule applies. Neither the process gas heater

nor the EAF are included in the regulated units described in the applicability statements. The

miscellaneous natural gas-fired equipment is exempted (and below the defined input capacity for

‘small boiler’) and the emergency generators are also exempted.

NOx-emitting sources will, however, be limited as a result of the BACT analysis as part of the PSD

review.

3.4 State of Ohio Regulations and Standards – By Emissions Unit

This section provides a summary of the Ohio emissions limitations and standards applicable to

each specific emissions unit. The table lists the state rules applicable to each emissions unit. The

subsequent discussion will concentrate on the best available technology (BAT) approach being

requested.




- 33 -

Applicable Ohio Administrative Code Rules in Chapter 3745

Source of Emissions \ OAC Rule 3745- 17-07

(A)

17-07

(B) 17-08

17-10

(B)(1) 17-11

18-06

(E)

31-05

(A)(3)

Process gas heater (preheater) X X X

Quenching & wastewater treatment (flare,

NH3 scrubber) <10 tpy

Ladle preheat and dry <10 tpy

EAF X (A)(2) X

Material Handling - Pellets X X X (A)(2) X

Roadways/Parking X X X

Emergency Generators & Pumps (100 hrs) X (B)(5) X <10 tpy

Black Start Generator (100 hrs) X (B)(5) <10 tpy

Cooling Tower (de minimis)

Remet Storage Pile (de minimis)

Material Handling – Remet, (de minimis)

Bulk Storage (de minimis)

Auxiliary Boiler X X <10 tpy

3.4.1 Process Gas Heater

The reduction gas, consisting of H2 and CO, is heated by the process gas heater combusting natural

gas supplemented by a lesser amount of CO and H2 recovered in the reconditioning loop. The process

gas heater emits only the products of combustion, utilizing a total of 218.9 million BTU per hour

input capacity low-NOx burners.

Opacity Standard

The process gas heater is defined as ‘fuel-burning equipment’ and exhausts through a stack.

Therefore, the standard of 20% opacity as a 6-minute average and associated exceptions in OAC rule

3745-17-07(A) apply.

PM Standard

The limit of 0.020 pounds PM per million BTU of gaseous fuels combusted in OAC rule 3745-17-

10(B) is also applicable. The unit burns a combination of natural gas and process gas. The unit can

be expected to be inherently compliant with this limit based on the use of these clean gaseous fuels.




- 34 -

SO2 Standard

The county-specific sulfur dioxide rule (OAC 3745-18-10) does not list Petmin and the general

emission limitations in OAC 3745-18-06 are not applicable to this combustion unit.

BAT

The emissions unit is subject to the PM and opacity limits described above. BAT for NOx is equivalent

to the selected technology discussed in the BACT review.

Petmin requests that the BAT for NOx be expressed as a source design characteristic. The proposed

process gas heater burner is designed to emit no more than 0.064 pounds NOx per million BTU

heat input.

3.4.2 Quenching & Wastewater Treatment Waste Gas

The quenching and wastewater treatment system normally operates at a de minimis level except

during startup and shutdown, when process gas is vented to the flare in order to combust the hydrogen

and carbon monoxide. There is only one planned shutdown and startup per year.

PM Standards

The process gas contains only combustion-related PM. Therefore, OAC rule 3745-17-11 does not

apply.

Opacity Standards

The visible emission standards do not apply to an “air contaminant source which is not subject to any

mass emission limitation in paragraphs (B)(3) and (B)(4) of rule 3745-17-08 of the Administrative

Code, or rule 3745-17-09, 3745-17-10 or 3745-17-11 of the Administrative Code.”

SO2 Standard

The county-specific sulfur dioxide rules (OAC 3745-18-10) do not list Petmin and the process

limitations in OAC 3745-18-06(E) do not apply to this source because OAC 3745-18-06(C) exempts

sources with process weight rates less than 1,000 pounds per hour.

BAT

Potential emissions of all criteria pollutants do not exceed 10 tons per year; therefore, the BAT

requirement is not triggered.




- 35 -

3.4.3 Ladle Preheat

The ladle preheat consists of three lances burning natural gas which are used to bring the pouring

ladles to a temperature where the refractory lining will not be damaged by thermal shock during

subsequent tapping.

PM Standards

The ladle preheat contains only combustion-related PM. Table I of OAC rule 3745-17-11 relates

process weight rate (PWR) to an allowable emission rate. Only natural gas is introduced into the

process and gaseous fuels are excluded from the definition of PWR in both OAC rule 3745-17-01 and

17-11(A)(4). Table I imposes an emission limit of 0.551 pounds PM per hour for processes with a

PWR equal to zero. Figure 2 of OAC rule 3745-17-11 does not apply because the UMRE is less than

10 pounds per hour.

Opacity Standards

The ladle preheat emissions are captured and exhausted through the EAF baghouse. Therefore, the

OAC rule 3745-17-07(A) standard of 20% opacity as a 6-minute average, and associated

exceptions, applies to the ladle preheat emissions as they exit that baghouse.

SO2 Standard

The county-specific sulfur dioxide rules (OAC 3745-18-10) do not list Petmin and the process

limitations in OAC 3745-18-06(E) do not apply to this source because OAC 3745-18-06(C) exempts

sources with process weight rates less than 1,000 pounds per hour.

BAT

Potential emissions of all criteria pollutants do not exceed 10 tons per year; therefore, the BAT


3.4.4 EAF, Pouring and Casting

DRI from the reactor is stored in bins before batch-loading to the EAF. A complete cycle of the

EAF is designed to process 153 tons of DRI over approximately 132 minutes, consisting of 112

minutes of smelting and 20 minutes for tapping to the ladle, yielding 132.3 tons of metal and 20.7

tons of slag. Slag is transferred to a slag pot for removal. Metal is transferred to the ladle and

subsequently poured into a water-cooled ingot casting line to produce the final product.

The EAF, ladle preheat, slagging, pouring and casting processes are located in the same building,

which is designed to fully capture the associated emissions and direct them to a baghouse. The




- 36 -

main EAF evacuation duct leads to a knock-out box where flow is reduced to settle large particles

and air is introduced to combust the CO in the hot gases and provide cooling prior to entering the

baghouse.

PM Standard

Table 1 of OAC rule 3745-17-11 requires an allowable PM limit to be calculated using the process

weight rate (P), defined for P > 30 as E = 55.0(P)0.11 – 40.0. The EAF will be able to produce 153

tons of metal and slag during a 132-minute batch. Allowable emissions for a 69.5 ton per hour

process weight rate are therefore 47.70 pounds per hour.

Figure II contains a second equation to calculate allowable emissions using the uncontrolled mass

rate of emissions (UMRE). There are no available estimates of UMRE for the EAF: AP-42 only

shows a controlled emission factor and the supplier of Petmin’s equipment provides only a design

outlet concentration for the fabric filter. However, an examination of Figure II indicates that the

proposed design outlet concentration would necessarily comply with Figure II or else be associated

with a UMRE of less than 10 pounds per hour and not be subject to Figure II.

Therefore, both methods of calculating an allowable emission rate from OAC rule 3745-17-11

yield results that are less restrictive than that resulting from BAT and BACT.

Opacity Standard

The EAF and associated processes are subject to OAC rule 3745-17-11 and exhaust through a

stack. Therefore, the OAC rule 3745-17-07(A) standard of 20% opacity as a 6-minute average,

and associated exceptions, applies.

BAT

The fumes generated during the smelting in the EAF are extracted from the roof of the furnace via

a flexible suction pipe cooled with water leading to a sedimentation chamber (“drop-out box”).

The overall ventilation system is designed to maintain a negative pressure inside the building, and

vent it to the EAF baghouse. This feature, along with the capture system for the EAF, provides a

capture efficiency that is presumed to be 100 percent. The configuration of the sedimentation

chamber makes it possible to separate coarse particles from the gas and the addition of air to the

hot gas (800° C) further combusts remaining CO.

The BAT limit will be equivalent to the BACT limit for CO and NOx.




- 37 -

Petmin requests that the BAT for NOx be expressed as a monthly allowable emission rate. The

proposed emissions from the EAF are based on the manufacturer’s NOx emission factor of 1.4

pounds per ton of metal times 60.13 tons of metal per hour, operating 8,760 hours per year, and

converted to tons: 1.4 x 60.13 x 8,760 ÷ 2,000 = 368.72 tons per year. As a monthly value, 368.72

÷ 12 = 30.7 tons NOx per month.

Petmin requests that the BAT for CO be expressed as a source design characteristic.

Petmin requests that the BAT for PM be expressed as a source design characteristic. The proposed

EAF baghouse is designed to emit no more than 0.0025 grain PM per standard cubic foot of exhaust

gas.

3.4.5 Material Handling

Material handling operations controlled by Petmin include the transfer, conveying, and screening of

iron ore pellets that have been loaded into the conveying system hopper by a third party.

Materials produced in the EAF are not expected to contain fines. Partially reduced iron ore pellets

(Remet) diverted from the reduction reactor during startup or shutdown have already had fines

removed during processing. Slag is a solid that is not processed on site. (Note: It may be necessary to

break large pieces to facilitate truck loading, but this operation is performed by an independent third

party). Ingots (“pigs”) are solid cast metal. Handling pigs for shipment is not expected to result in

particulate emissions as there would be negligible fines or dust.

Fluxes are pneumatically conveyed to storage silos and fed to the process through closed feeders

inside a building vented to the EAF baghouse. These material handling operations are considered to

be de minimis.

PM Standards

The transfer, conveying, and screening of iron ore pellets has the potential to generate fugitive dust

and is subject to OAC rule 3745-17-08. The rule prescribes control measures that must be taken to

minimize or eliminate such emissions, i.e., application of water or other dust suppressants to open

sources or the capture of fugitive emissions with venting to fabric filters.

Opacity Standards

The visible emission standards in OAC rule 3745-17-07(B)(1) are applicable because the site is located

within an Appendix A area. Visible emissions of fugitive dust are limited to 20% opacity as a 3-minute

average.




- 38 -

BAT

Potential emissions of PM10 exceed 10 tons per year and therefore the BAT requirement is

applicable. The BAT limit is equivalent to the BACT requirement.

The conveyors and transfer points will be designed to minimize or eliminate visible emissions of

fugitive dust. Where appropriate, e.g., screens, emissions will be captured and vented to a

baghouse with a maximum designed outlet concentration of 0.0025 gr/dscf.

3.4.6 Roadways and Parking Areas

Petmin leases land within an existing facility that has its own permitted roadways. Contractually,

Petmin is responsible for roadways only on its own leased property. The calculations provided

with this application provide detailed information concerning these roadway segments and

associated traffic.

PM Standards

The operation of vehicles on plant roads has the potential to generate fugitive dust and is subject to

OAC rule 3745-17-08. The rule prescribes control measures that must be taken to minimize or

eliminate such emissions, i.e., application of water or other dust suppressants to open sources.

Opacity Standards

The visible emission standards in OAC rule 3745-17-07(B)(1) are applicable because the site is

located within an Appendix A area. Visible emissions of fugitive dust are limited to 20% opacity

as a 3-minute average.

BAT

Potential emissions all criteria pollutants do not exceed 10 tons per year and therefore the BAT


3.4.7 Emergency Engines and Black Start Generators

The emergency generators and fire protection pumps are sized to maintain vital functions in the

event of a power outage. They do not have the capacity to provide the power needed to sustain

production operations. The black start generator will provide power only during startup from a

cold condition. These startups are expected to occur, ideally only one time, but, accounting for

unplanned shutdowns, a few times per year.




- 39 -

All five diesel engines will be current models meeting the Tier 4 requirements and subject to

MACT subpart ZZZZ and NSPS subpart IIII, as described in 3.2 and below. The federal rules for

limit operation to less than 100 hours per year for maintenance and readiness testing. There is no

limit on the number of hours allowed during emergencies.

PM Standards

The emergency generators are “stationary large internal combustion engines” and will comply with

OAC rule 3745-17-11(B)(5)(b). The fire protection pumps and the black start generator are

“stationary small internal combustion engines” and will comply with OAC rule 3745-17-

11(B)(5)(a).

Opacity Standards

The emergency engines and black start generator exhaust through stacks. Therefore, the standard

of 20% opacity as a 6-minute average, and associated exceptions, applies to each stack.

SO2 Standard

The sulfur dioxide limit for the emergency generators is specified in OAC 3745-18-06(G) because

it will be greater than ten MM Btu per hour total rated capacity; the black start generator will be

less than ten MM Btu per hour total rated capacity and exempted by OAC 3745-18-06(B).

BAT

Potential emissions all criteria pollutants do not exceed 10 tons per year and therefore the BAT





- 40 -

4.0 BEST AVAILABLE CONTROL TECHNOLOGY (BACT) ANALYSIS This section presents the BACT analysis for the Project.

4.1 Introduction

The source is located in Ashtabula County, which is currently designated as attainment or

unclassifiable for PM/PM10/PM2.5, NOx, CO, SO2, and VOC. Therefore, these pollutants were

reviewed pursuant to the PSD Program.

As summarized in Section 3.2, the Project is subject to PSD review for the following pollutants:

• Oxides of nitrogen (NOx)

• Carbon monoxide (CO)

• Particulate matter less than 10 microns (PM10)

• Particulate matter less than 2.5 microns (PM2.5)

• Greenhouse Gases (GHG)

BACT is an emission limitation based on the maximum degree of reduction of each pollutant

subject to the PSD requirements. In accordance with the “Top-Down” Best Available Control

Technology Guidance Document outline in the 1990 draft U.S. EPA New Source Review Workshop

Manual (the Manual), this BACT analysis takes into account the energy, environmental, and

economic impacts on the source. The BACT analysis addresses the above four pollutants.

Reductions may be determined through the application of available control technologies, process

design, and/or operational limitations. Such reductions are necessary to demonstrate that the

emissions remaining after application of BACT will not cause or contribute to significant

deterioration of air quality, thereby protecting public health and the environment.

The “Top-Down” approach in the Manual is summarized as the following 5-step process:

Table 1: Identify all control technologies.

Table 2: Eliminate technically infeasible options.

Table 3: Rank remaining control technologies by control effectiveness.

Table 4: Evaluate the most effective controls and document results.

Table 5: Select BACT.

BACT must also, at a minimum, meet any applicable New Source Performance Standard (NSPS)

under 40 CFR, Part 60 and Maximum Achievable Control Technology (MACT) standard under

40 CFR Part 61 – 63. For this Project, as described in Section 3, NSPS standards apply only to

the emergency engines and black start generators; no MACT standards are applicable.




- 41 -

4.2 Emissions Units

The following table summarizes Project emissions by emissions units.

Project Maximum Emissions by Emissions Unit

Source of Emissions NOx CO SO2 PM10 PM2.5 VOC HAPs NH3 CO2e

PGH - Process gas heater (preheater) 82.71 48.92 3.46 7.14 7.14 5.17 0.00 0.00 307,490

Startup Boiler (8,760 hrs basis) 2.78 5.47 0.04 0.49 0.49 0.36 0.00 0.00 7,814

Flare (Quench, WW treatment, NH3

Scrubber) 1.97 8.97 0.01 0.22 0.22 4.05 0.00 0.01 3,405

Ladle Preheat 9.29 2.26 0.04 * * 0.35 0.00 0.00 7,729

Ladle Preheat (backup) 9.29 2.26 0.04 * * 0.35 0.00 0.00 7,729

Ladle Dry 9.29 2.26 0.04 * * 0.35 0.00 0.00 7,729

EAF and Casting 368.72 474.06 0.00 54.44 43.55 6.06 0.00 0.00 49,095

Emergency Generator #1 0.17 0.90 0.00 0.01 0.01 0.05 0.00 0.00 181.75

Emergency Generator #2 0.17 0.90 0.00 0.01 0.01 0.05 0.00 0.00 181.75

Black Start Generator 0.01 0.06 0.00 0.00 0.00 0.00 0.00 0.00 9.09

High Pressure Emergency Pump 0.07 0.07 0.00 0.01 0.01 0.00 0.00 0.01 17.90

Low Pressure Emergency Pump 0.07 0.07 0.00 0.00 0.00 0.00 0.00 0.01 13.64

Material Handling – Iron Pellets 0.00 0.00 0.00 1.29 0.89 0.00 0.00 0.00 0.00

Roadways/Parking 0.00 0.00 0.00 0.22 0.02 0.00 0.00 0.00 0.00

Remet Storage Piles (de minimis) 0.00 0.00 0.00 0.05 0.05 0.00 0.00 0.00 0.00

Fines Handling (de minimis) 0.00 0.00 0.00 0.06 0.01 0.00 0.00 0.00 0.00

Bulk Flux Silos (de minimis) 0.00 0.00 0.00 0.02 0.00 0.00 0.00 0.00 0.00

Cooling Tower (de minimis) 0.00 0.00 0.00 0.02 0.00 0.00 0.00 0.00 0.00

Total Project-Wide PTE 484.57 546.22 3.63 63.97 52.40 16.8 0.11 0.01 391,397

* Included in Electric arc furnace PM10 and PM2.5 emissions.




- 42 -

In each BACT section, a table lists the emissions units (EU) associated with the Project that require

BACT review for that pollutant. The units are listed in the decreasing order of maximum emission

rates; the BACT analysis will also follow this sequence, as “returns on investment” in controls (in

terms of possible emissions reductions) tend to diminish with the smallest emissions units. The

maximum emission rates shown in the tables take into account the controls described in this section

and selected as BACT.

4.3 NOx BACT Review

BACT for NOx is analyzed by emissions unit.

Emissions Units Subject to PSD Review – NOx

EU ID Description Maximum Emission

Rate (tpy)

Electric arc furnace Smelting, tapping, pouring, casting 368.72

Process gas heater Natural gas-fired combustion unit 82.71

Ladle preheat and dry Natural gas-fired combustion units 27.87

Startup boiler Natural gas-fired combustion unit 2.78

DRI gas reconditioning Flare used to control CO emissions at start up 1.97

EG1 Diesel-fired generator, emergency use 0.17


High Pressure Emg. Pump High pressure emergency diesel fire pump 0.10

Low Pressure Emg. Pump Low pressure emergency diesel fire pump 0.07

Black Start Gen Diesel-fired generator, black start 0.01

4.3.1 Electric Arc Furnace

The operation of the EAF was described in Section 2. An EAF can be described as a furnace

heating charged materials by the way of an electric arc.

An electric arc furnace used for steelmaking consists of a refractory-lined vessel, usually water-

cooled in larger sizes, covered with a retractable roof, and through which one or more graphite

electrodes enter the furnace. The furnace is primarily split into three sections:

• The shell, which consists of the sidewalls and lower steel bowl;

• The hearth, which consists of the refractory that lines the lower bowl; and




- 43 -

• The roof, which may be refractory-lined or water-cooled, and can be shaped as a section

of a sphere. The roof also supports the refractory delta in its center, through which one or

more graphite electrodes enter.

A typical alternating current furnace has three electrodes. The arc is formed between the charged

material and the electrode. The charge is heated both by current passing through the charge and

by the radiant energy evolved by the arc. The electric furnaces can be categorized as direct arc or

indirect arc. Both types of units are suited for the melting of high melting point alloys such as

steels.

The picture below shows an electric arc furnace.

Figure 5: Electric Arc Furnace

In the EAF, NOx will be produced at the electric arc. There is a substantial difference between

conventional EAF operations for steelmaking and the Petmin EAF designed for MPI production.

The principal difference between an EAF used in steelmaking and the Petmin EAF is the arc

exposure. In conventional steelmaking the arc is protected by a foamy slag layer that minimizes

the contact with the surrounding atmosphere. For MPI production, the arc is totally exposed to the

furnace atmosphere, which increases nitrogen oxidation (formation of thermal NOx). This EAF

operation is similar to an EAF designed for stainless steel-making, where the molten bath operates

with no foamy slag.




- 44 -

The Petmin EAF is a unique, first-of-its-kind design for MPI production. NOx formation is

complex, and difficult to predict for a new technology. Emission factors for NOx were developed

based on a proprietary model by its design engineers.

Steps 1 & 2: Identify all Control Technologies and Eliminate Technically Infeasible Options

US EPA has compiled the following general set of potential options for reducing NOx5. Potential

application to the Petmin EAF is described for each option, quoted directly from the EPA manual.

Method 1. Reducing Temperature – Reducing combustion temperature means avoiding the

stoichiometric ratio (the exact ratio of chemicals that enter into reaction). Essentially, this

technique dilutes calories with an excess of fuel, air, flue gas, or steam. Combustion controls use

different forms of this technique and are different for fuels with high and low nitrogen content.

This option generally applies to combustion processes. It is infeasible for the Petmin EAF. The

temperature cannot be reduced.

Method 2. Reducing Residence Time – Reducing residence time at high combustion

temperatures can be done by ignition or injection timing with internal combustion engines. It can

also be done in boilers by restricting the flame to a short region in which the combustion air

becomes flue gas. This is immediately followed by injection of fuel, steam, more combustion air,

or recirculating flue gas. This short residence time at peak temperature keeps the vast majority of

nitrogen from becoming ionized. This bears no relationship to total residence time of a flue gas in

a boiler. This option generally applies to combustion processes. It is infeasible for the Petmin

EAF. The residence time cannot be reduced.

Method 3. Chemical Reduction of NOx – This technique provides a chemically reducing (i.e.,

reversal of oxidization) substance to remove oxygen from nitrogen oxides. Examples include

Selective Catalytic Reduction (SCR) which uses ammonia, Selective Non-Catalytic Reduction

(SNCR) which use ammonia or urea, and Fuel Reburning (FR). Non-thermal plasma, an emerging

technology, when used with a reducing agent, chemically reduces NOx. All of these technologies

attempt to chemically reduce the valence level of nitrogen to zero after the valence has become

higher. Some low-NOx burners also are based partially on this principle. According to the process

engineering design firm applying this EAF technology to MPI production, it is not possible due to

conditions of the furnace.

More specifically, SCR is not feasible since it requires clean and hot exhaust gases where the NOx

waste gas is introduced. In the EAF for MPI production, the waste gas stream is hot and it carries

dust, then it is cleaned only when it is cold. Operating a SCR system under its hot condition would

simply clog the scrubbing system.




- 45 -

When evaluating technical feasibility, it should be recognized that, with enough engineers, time,

and money, a system could, in theory, be designed and built as long as it does not violate the laws

of thermodynamics. In this sense, it could be theoretically possible to construct an exhaust gas

reheat system and SCR control downstream of the EAF baghouse. With a starting point of 171°F

and 356,000 acfm, the waste gas would first need to be heated to the 480°F to 800°F range where

SCR reactions occur.6 At 500°F, the waste gas stream would increase to 542,000 acfm, plus

additional combustion gases required for reheating (plus additional NOx formation). A technically

feasible control strategy is an “available technology” that has been demonstrated to function

efficiently on an emissions unit that is identical or similar to the unit under review7. Applying

SCR technology in this manner would be impractical, has not been demonstrated on a similar unit,

and is ruled out as technically infeasible.

Selective Non-Catalytic Reduction (SNCR) uses a reducing agent, typically urea or ammonia,

injected directly into the exhaust of combustion processes such as coal-fired boilers at a point

where the temperature is between 1,800° and 2,100°F. Conditions in the EAF exhaust do not

match SNCR applicability. This technology is technically infeasible.

Other chemical reduction technologies are also infeasible:

• Non-Selective Catalytic Reduction (NSCR), the NOx reduction technology used on fuel

rich combustion systems such as automobiles (with downstream oxidation catalyst) is only

applicable to combustion systems such as reciprocating engines. There are no known

applications of NSCR to a source similar to the EAF.

• Non-thermal plasma is an emerging technology under development on a laboratory scale

and has not been proven for large-scale application.

Method 4. Oxidation of NOx – This technique intentionally raises the valence of the nitrogen ion

to allow water to absorb it (i.e., it is based on the greater solubility of NOx at higher valence). This

is accomplished either by using a catalyst, injecting hydrogen peroxide, creating ozone within the

air flow, or injecting ozone into the air flow. Non-thermal plasma, when used without a reducing

agent, can be used to oxidize NOx. A scrubber must be added to the process to absorb N2O5

emissions to the atmosphere. Any resultant nitric acid can be either neutralized by the scrubber

liquid and then sold (usually as a calcium or ammonia salt), or collected as nitric acid to sell to

customers. These technologies are generally not suitable for use on the EAF and are technically

infeasible. One emerging technology, discussed below, is not considered to be an “available

technology,” as it has not yet been proven in similar sources.

EMx™ (SCONOX) is an add-on control device that catalytically oxidizes NO to NO2 and then

adsorbs the NO2 onto a potassium carbonate coated surface. The potassium carbonate is




- 46 -

regenerated using steam injected hydrogen, to react with the nitrates and nitrites to form water and

elemental nitrogen. This technology has been demonstrated on relatively small combined-cycle

gas turbines. Use on pre-cleaned EAF gas with iron fines would foul or plug the catalytic oxidation

process, and this technology would be infeasible. Moreover, this technology has not been applied

to an EAF or similar source.

LoTOx™ is a technology where ozone is injected into the waste gas stream to oxidize insoluble

NOx to soluble N2O2, which predominantly reacts in moisture to form nitric acid. The nitric acid,

along with N2O2 and nitrous acid, are removed using a caustic scrubber, i.e., absorption tower.

Despite being ruled out as economically infeasible at a cost of $84,000 per ton of NOx removed, a

demonstration project was planned for a DRI-fed indurating furnace at the former Essar Steel

facility in Minnesota.8 After bankruptcy proceedings, the plant is still considered to be under

construction by Mesabi Metallics LLC, with a construction completion target date unknown at this

time. Further, it is unknown at this time whether the demonstration project will be pursued.

Therefore, this technology remains unproven for an EAF, and is eliminated as technically

infeasible.

Method 5. Removal of nitrogen from combustion – This is accomplished by removing nitrogen

as a reactant either by: (1) using oxygen instead of air in the combustion process; or (2) using ultra-

low nitrogen content fuel to form less fuel NOx. This option is not feasible for an EAF; there is no

fuel or combustion process.

Method 6. Sorption, both adsorption and absorption – Treatment of flue gas by injection of

sorbents (such as ammonia, powdered limestone, aluminum oxide, or carbon) can remove NOx

and other pollutants (principally sulfur). There have been successful efforts to make sorption

products a marketable commodity. This kind of treatment has been applied in the combustion

chamber, flue, and baghouse. The use of carbon as an adsorbent has not led to a marketable

product. The sorption method is often referred to as using a dry sorbent, but slurries also have

been used. The use of dry sorbents and slurries has been proven for sulfur dioxide treatment but

is not proven for NOx control. This technology is infeasible for the EAF.

Method 7. Combinations of these methods – Many of these methods can be combined to achieve

a lower NOx concentration than can be achieved alone by any one method. With none of the above

approaches considered feasible for the EAF, a combination method is also not feasible.

As additional research to identify available technologies, EAFs were searched via the EPA

RACT/BACT/LAER Clearinghouse database. No new control technologies were discovered from

this search.




- 47 -

Step 3: Rank Remaining Control Technologies by Control Effectiveness

No feasible add-on control technology was found for this emissions unit. This EAF is a one-of-a-

kind design for smelting DRI, in which the arc is directly exposed to the furnace atmosphere, with

no foamy slag layer. The emission factor for the EAF was developed based on a similar design

used in making stainless steel.

Step 4: Evaluate the Most Effective Controls and Document Results

A “no add-on controls” case is the most effective control for the Petmin EAF.

BACT Emission Limitations at Other Facilities

The EPA database system, RACT/BACT/LAER clearinghouse (RBLC), provides emission limit

data for industrial processes throughout the United States. The US EPA Workshop Manual

stipulates that a BACT analysis should also include a technology review of similar sources. The

units found, based on a search of permits issued in the last 10 years, were all carbon steel processes;

no stainless steel EAF was listed in the database.

Based on the emission factor provided by the EAF design team, at capacity the Petmin EAF will

emit 368.72 tons NOx annually to manufacture 526,739tons MPI, which translates to 1.4 pounds

of NOx emitted per ton of finished MPI. This summary table confirms what was previously known,

that the use of a direct arc, with no foamy slag layer between the arc and metal, results in higher

temperatures and NOx formation than EAFs in conventional steel-making.

The unique Petmin EAF is not directly comparable to any of these units listed in the table. Further,

the Petmin EAF provides other environmental benefits, in producing high-purity, nodular grade

merchant pig iron in an energy efficient manner that can compete with producers on a global

market. These benefits are not reflected in the NOx emissions values. Furthermore, the emission

factor is based on an emissions model developed by the furnace manufacturer. NOx formation is

highly complex. As typical for unique, emerging-technology processes, the emission factor can

be re-evaluated upon source measurement (stack testing) when the unit achieves its design

production rate.

Not listed in the RBLC database, a recent PSD permit for a stainless steel EAF in Alabama was

modified to account for high thermal NOx formation that had been mischaracterized in the original

permit application. The original permit, issued for 0.35 pounds NOx per ton metal production, was

modified to 1.0 pounds NOx per ton metal, based on stack test results.9

Another stainless steel facility, North American Stainless, in KY, has two EAFs that have

undergone PSD review. One has an emission rate of 1.32 pounds NOx per ton plus 0.578 pounds

per ton from the Argon Oxygen Decarburization step, for a total emission rate from smelting




- 48 -

operations of 1.898 pounds NOx per ton of metal produced. The other EAF is 1.0 plus 0.58 for a

total emission rate of 1.58 pounds NOx per ton of metal produced. The excerpt from this permit

is included as Appendix 1.

RBLC Search Results – EAFs

RBLC

ID/Date Facility Description

NOx in

Lbs per

Ton Metal

NE-0063

11/07/2017 Nucor Steel Corp Electric Arc Furnace - scrap steel to billets and bars 0.42

AL-0309

3/02/2016

Nucor Steel Decatur,

LLC Two (2) Electric Arc Furnaces - galvanizing plant 0.42

OK-0173

1/19/2016 CMC Steel Oklahoma Electric Arc Furnace - scrap metal to structural steel 0.3

LA-0309

6/4/2015

Benteler Steel / Tube

MFG Corp

Electric Arc Furnace - Steel mill - scrap steel to billets

to steel pipe 0.35

MI-0417

10/27/2014 Gerdau Macsteel, Inc.

Melt shop: EAF w/6 oxyfuel burners, ladle metallurgy

station, 2 vacuum degassers. - scrap steel to billets 0.2

TX-0705

7/24/2014 Structural Metals Inc

Electric Arc Furnace (EAF) and Ladle Metallurgy

Station (LMS) - scrap steel to billets 0.2159

NE-0055

10/09/2013 Nucor Steel

Twin-shell EAF with 3 sidewall burners (each 18

MMBH) and 1 sump burner (10 MMBH - scrap steel to

billets and bars

0.28

TX-0651

10/02/2013 Nucor Corporation

Electric Arc Furnace - Steel mill - scrap steel to steel

shapes 0.9

AR-0140

9/18/2013 Big River Steel LLC

Two electric arc furnaces to melt scrap iron and steel -

to produce galvanized steel coils 0.3

OH-0350

07/18/2012 Republic Steel

Electric Arc Furnace 150 T/HR steel production. - scrap

metal to cast steel 0.5

GA-0142

12/29/2010 Osceola Steel Co.

60 TPH Electric Arc Furnace (with low-NOx burners) -

Steel mill - scrap steel to billets 0.35

CO-0066

11/30/2011

Evraz Rocky Mountain

Steel ERMS Pueblo

steel mini-mill with electric arc furnace to produce steel

from scrap 0.28

OH-0339

12/29/2010

Timken - Harrison

Steel

2 EAF’s: increase annual steel production; replace all or

part of the coke with scrap tires - scrap steel to billets 0.2

OH-0342

12/29/2010

Timken - Faircrest

Steel

EAF: increase annual steel production; replace all or

part of the coke with scrap tires - scrap steel to billets 0.2

OH-0341

12/23/2010

Nucor Steel Marion,

Inc.

EAF: charging, melting, tapping, slag skimming;

casting, torch cutting; ladle preheaters; and tundish

preheaters. - scrap steel to billets

0.43

TX-0576

4/19/2010 TPCO America Inc EAF - 149 tons steel/hr - scrap steel to pipe 0.3

OH-0331

1/11/2010

AK Steel Corp.

Mansfield Works

Melt shop: 2 EAFs; 6 Preheaters, AOD, LMF - scrap

steel to billets to steel coil 1.43




- 49 -

Step 5: Select BACT

Based on the information presented above, the NOx BACT option selected is no add-on controls,

and a NOx emission factor of 1.4 pounds NOx per ton of MPI production. Therefore, BACT for

this unique piece of equipment consists of:

• NOx emissions of 1.4 pounds NOx per ton of MPI produced;

• Direct Evacuation Control (DEC) for capture of air contaminants, as described in Section

2 of this report.


The Process Gas Heater (PGH) is a natural gas-fired heater that has been custom engineered for

Petmin to supply the process gas at the optimum temperature, flow and pressure to the DRI reactor.

Some cleaned process gas from the reconditioning loop, containing hydrogen, methane, and carbon

monoxide, is also burned. The PGH has been specified with low-NOx burners.


Seven methods for reducing NOx emissions were identified in the previous section and are listed

again in the following table.




- 50 -

Methods for Reducing NOx Emissions

Ambient or Emission

Control Principle or Method Successful Technologies

Pollution Prevention Method

(P2) or Add-on Technology (A)

1. Reducing peak

temperature

Flue Gas Recirculation (FGR)

Natural Gas Reburning

Low NOx Burners (LNB)

Combustion Optimization

Burners Out of Service (BOOS)

Less Excess Air (LEA)

Inject Water or Steam

Over Fire Air (OFA)

Air Staging

Reduced Air Preheat

Catalytic Combustion

P2

P2

P2

P2

P2

P2

P2

P2

P2

P2

P2

2. Reducing residence time at

peak temperature

Inject Air

Inject Fuel

Inject Steam

P2

P2

P2

3. Chemical reduction of NOx Fuel Reburning (FR)

Low NOx Burners (LNB)

Selective Catalytic Reduction

(SCR)

Selective Non-Catalytic

Reduction (SNCR)

P2

P2

A

A

4. Oxidation of NOx with

subsequent absorption

Non-Thermal Plasma Reactor

Inject Oxidant

A

A

5. Removal of nitrogen Oxygen Instead of Air

Ultra-Low Nitrogen Fuel

P2

P2

6. Using a sorbent Sorbent in Combustion Chambers

Sorbent in Ducts

A

A

7. Combinations of these

Methods

All Commercial Products P2 and A

Listed by the above sections, the following technologies are sorted as feasible or infeasible. Note

that the following technologies, described in the previous section, are not included in the table as

they would not be considered “successful technologies” for this combustion unit.

• Non-Selective Catalytic Reduction (NSCR)

• SCONOX

• LoTOx




- 51 -

Control Technologies

Abatement Principle or

Control Method Successful Technology

Technically

Feasible? Reason

1. Reducing Peak

Temperature

Flue Gas Recirculation No Does not fit with process and/or custom PGH

design basis.

Natural Gas Reburning No Does not fit with process and/or custom PGH

design basis.

Low NOx Burners - LNB Yes

Combustion Optimization - LNBs are optimized at startup for efficiency

and NOx level.

Burners Out of Service No Does not fit with process and/or custom PGH

design basis.

Less Excess Air No Does not fit with process and/or custom PGH

design basis.

Inject Water or Steam No Does not fit with process and/or custom PGH

design basis.

Less Overfire Air No Does not fit with process and/or custom PGH

design basis.

Air Staging No Does not fit with process and/or custom PGH

design basis.

Reduced Air Preheat No Does not fit with process and/or custom PGH

design basis.

Catalytic Combustion No Does not fit with process and/or custom PGH

design basis.

2. Reducing Residence

Time at Peak

Temperature

Inject Air No Does not fit with process and/or custom PGH

design basis.

Inject Fuel No Does not fit with process and/or custom PGH

design basis.

Inject Steam No Does not fit with process and/or custom PGH

design basis.

Fuel Reburning No Does not fit with process and/or custom PGH

design basis.

3. Chemical Reduction

of NOx

Selective Catalytic

Reduction

Yes

Selective Non-Catalytic

Reduction

No Waste gas is outside the temperature range for

this technology.

4. Oxidation of NOx

with Subsequent

Absorption

Non-Thermal Plasma

Reactor

No Unproven on this scale and type of unit.

Inject Oxidant No Does not fit with process and/or custom PGH

design basis.

5. Removal of nitrogen Oxygen instead of Air

No Does not fit with process and/or custom PGH

design basis.

Ultra-Low Nitrogen Fuel No A substitute for natural gas is not feasible.

6. Using a sorbent Sorbent in Combustion

Chambers

No Use of dry or slurried sorbents does not fit

with process and/or custom PGH design basis




- 52 -

Control Technologies

Abatement Principle or

Control Method Successful Technology

Technically

Feasible? Reason

Sorbent in Ducts Not applicable; this option fits specialized

processes where sorbent can be introduced in

ductwork.

7. Combination of these

methods

Combinations of the

above.

Yes LNB and SCR can be combined.


Low NOx Burners

“Low NOx burners reduce NOx by accomplishing the combustion process in stages. Staging

partially delays the combustion process, resulting in a cooler flame which suppresses thermal NOx

formation. NOx emission reductions of 40 to 85 percent (relative to uncontrolled emission levels)

have been observed with low NOx burners.”10 It is important to note that the actual reduction

achievable is also highly dependent on the specific application.

The use of Low NOx burners is a minimum standard in the industry and is considered the

“baseline” for the Project. Low NOx burner technology applied to the custom-designed PGH

achieves a NOx emission rate of 40 ppmv, which translates to an emission factor of 0.064 pounds

NOx per million BTU of heat input.

SCR

The SCR process chemically reduces the NOx molecule into molecular nitrogen and water vapor.

A nitrogen-based reagent such as ammonia is injected into the ductwork, downstream of the

combustion unit. The waste gas mixes with the reagent and enters a reactor module containing

catalyst. The hot flue gas and reagent diffuse through the catalyst. The reagent reacts selectively

with the NOx within a specific temperature range and in the presence of the catalyst and oxygen.11




- 53 -

Figure 6: SCR Process Diagram

Preliminary design data indicate that combining SCR would reduce the NOx emission rate from

40 to 10 ppmv. Therefore, SCR control would provide a 75% NOx reduction from the baseline.

Ranking the feasible controls by effectiveness:

• Low NOx with SCR: 75% reduction

• Low NOx: Baseline


Cost Evaluation - SCR

A cost estimate for SCR was developed using U.S. EPA’s Office of Air Quality Planning &

Standards “Air Pollution Control Cost Estimation Spreadsheet for Selective Catalytic Reduction,

May 2016.” The worksheet is provided as Appendix 2.

Source-specific data input to the model include the fuel type, NOx emission rate, expected

reduction, air flow, exhaust temperature, and site elevation. A urea-based reagent system for the

supply of ammonia was selected for safety reasons at the heavy industrial-type site.

The default interest rate of 7 percent was used, consistent with OAQPS regulatory analyses12.

In this context, n is the control system economic life, which, as stated above, typically varies from

10 to 20 years. The interest rate (i) used in this Manual is a pretax marginal rate of return on

private investment of 7% (annual). This value, which could also be thought of as a “real private

rate of return”, is used in most of the OAQPS cost analyses and is in keeping with current OAQPS




- 54 -

guidelines and the Office of Management and budget recommendation for use in regulatory

analyses.

The Chemical Engineering Plant Cost Index (CEPCI) was updated from the 2012 value of 584.6

to June 2018 of 601.17. The current CEPCI is not available in the public domain. As an unbiased

estimate, the value was estimated from the January 2018 CEPCI, then scaling the value based on

the June-to-January increase in the Consumer Price Index. (Note: this was an extremely minor

adjustment and had an insignificant impact on final, annualized costs).

The “average” life listed in Ohio Engineering Guide 46 Table 5-3 is 10 years, and the “high”

equipment life 20 years for either a catalytic incinerator or for an absorber. As this unit would be

well maintained, but is also in a severe service environment in the metals industry, a midpoint 15-

year equipment life was selected.

Otherwise, default values were selected in the analysis.

The total Capital Cost (dollar values presented to the nearest three significant figures) for SCR for

the PGH is estimated to be $3.63 million dollars. The total annualized cost is estimated to be

$555,000. 46 tons of NOx would be removed, at a cost of $12,100 per ton of NOx removed.

The cost effectiveness is presented in the following summary, as recommended by the NSR

Workshop Manual.

Cost Effectiveness of NOx Controls

Control

Alternative

Emissions Economic Impacts Environmental

Impacts

Emissions Emissions

Reduction

Installed

Capital

Cost

Total

Annualized

Cost

Cost

Effectiveness

Over Baseline

Toxics

Impact

Adverse

Impact

(lbs/hr) (tpy) (tpy) ($) ($/yr) ($/ton) (Yes/No) (Yes/No)

SCR 3.50 15.34 46.02 $3,630,000 $555,000 $12,100 NO YES

Baseline 14.01 61.36

Other Impacts

The implementation of SCR results in other environmental impacts:

• A source of ammonia must be transported to the site and stored in bulk;

• Ammonia slip and fugitive equipment leaks result in ammonia emissions to the

atmosphere;




- 55 -

• The additional electrical energy required translates to NOx and other pollutants formed in

power generation; and

• Spent catalyst requires disposal.

Comparison to Most Recent PSD Source

A recent PSD project in Ohio13 presented a BACT Control Technology Economic Impact Analysis

for a much larger natural gas-fired (1,687 v. 218.9 MM BTU/hr) DRI preheater/reformer of a

different design. A pro forma total capital cost estimate created using the OAQPS SCR estimator

using the source-specific input data for this DRI facility is $13.7 million, v. the $24.4 million

estimate presented by the permittee in the application. Given these data, the OAQPS estimator

provided a value that is only 56% of the total capital cost presented by the applicant. This

comparison suggests that the OAQPS estimator used in this analysis may be biased low for

developing a total, installed capital cost for SCR on a DRI preheater.

Similar Source

The HYL Zero Reformer and its Process Gas Heater and associated gas reconditioning loop

comprise a proprietary, custom, unique design. An earlier version of this technology has been

installed for one other U.S. project, at Consolidated Environmental Management, a division of

Nucor, at a plant near Convent, LA. No other similar sources were identified.

In that case, SCR was selected, along with Low NOx burners, as BACT. That BACT also

mentioned the use of low-nitrogen fuel, as the reactor top gas containing CO and H2 that is used to

supplement natural gas in the PGH would be considered “low nitrogen,” thereby partially

eliminating the NOx component that comes from fuel-bound nitrogen. This is also the case for the

Petmin Project. This feature is integral to the Zero Reformer process.

A significant difference exists in controls cost effectiveness between the LA project and Petmin’s

due to economies of scale. The LA plant is designed for annual production of 6.6 million tons pig

iron, from two DRI reactors. At 3.3 million tons capacity, each PGH/reactor has seven (7) times

the capacity of the Petmin Project. The cost of add-on controls is much lower, in dollars per ton,

at this scale. By increasing the input capacity of the PGH by 7, and re-running the SCR cost

estimator, the cost effectiveness changes from $12,100 per ton to $7,500 per ton (keeping other

key parameters – interest rate, equipment life, reagent and energy costs, etc. the same).

The EPA cost estimator indicates that total air pollution control (APC) capital cost, when

increasing pig iron production capacity by six, from 526,739tons to 3.3 million tons, increases

from $3.63 million to $12.9 million. This economy of scale tendency in APC equipment is also

described in Ohio EPA Engineering Guide 46, as the “six tenths factor.” For a first-pass estimate

of total capital cost, to scale the capital cost for a seven-fold increase in size, the multiplier would




- 56 -

be 7 raised to the power of 0.6, which equals 3.2. Scaling the estimated cost for the Petmin SCR

system by 3.2 yields an estimate of $11.6 million, which is in reasonable agreement with the $12.9

value from the OAQPS cost estimator.

In other words, it is expected and predictable that APC becomes significantly more cost effective

with economies of scale. This known relationship also explains how SCR technology can be cost

effective at the much larger LA plant’s Zero Reformer process, but not cost effective for the much

smaller Petmin project.

Step 5: Select BACT

Natural gas-fired Low-NOx burners (supplemented with integral low-nitrogen fuel) are cost

effective.

SCR add-on controls are technically feasible. However, $12,100 per ton of NOx removed is not

economically feasible.

BACT consists of low NOx burners, use of natural gas supplemented with an integral low-nitrogen

fuel source, and good combustion practices.

4.3.3 Ladle Preheaters

After tapping from the EAF, metal is kept molten by preheating the ladles used for pouring the

finished MPI into ingots, also known as “pigs.” Natural gas torches are used for this purpose, as

is common in the metals industry. Combustion of natural gas results in NOx emissions, with a

calculated maximum annual emission rate of 27.87 tons per year. The ladles are equipped with a

hood that is directed to the EAF baghouse.

A review of the RBLC database, a study of PSD permits obtained for other sources, and generally

available technical literature indicate that no additional controls, beyond the use of good

combustion practices and the use of natural gas, have been successfully applied to reduce NOx

emissions from ladle preheating. BACT is the use of natural gas and good combustion practices.

4.3.4 Small Ancillary Equipment – Startup Boiler, Flare, Emergency Engines, Black Start Generator

BACT recommendations for these small NOx-generating emissions units are presented in the

following table. Controls beyond the listed technologies are not technically feasible.




- 57 -

BACT Recommendations for Ancillary Units

Emissions Unit

NOx

Emissions

(tpy)

Description BACT

15.17 MM

BTU/hr Startup

Boiler with Low

NOx Burners

2.78

The boiler uses low-NOx burner technology, guaranteed

to achieve a NOx emission rate of 18 g/GJ (0.0418 lb/MM

BTU). Further, while the emission rate is “potential”

operation at 8,760 hours per year, this unit is actually kept

at low fire except during startup (estimated <1% of the

time). Controls beyond low-NOx burners are not feasible.

Low-NOx

Burners

Good combustion

DRI Gas

Reconditioning

Loop APC

Device (Flare) &

Thermal

Oxidizer

1.97

NOx emissions are from operation of a control device for

CO and for H2S under certain conditions as described in

Section 2. Add-on controls are not feasible.

Good air

pollution control

practice.

2 Emergency

Generators

Fire Pump 1

Fire Pump 2

0.17(ea)

0.10

0.07

These are emergency-only engines, limited to 100 hours

per year for maintenance and readiness testing (excluding

emergencies). This federally enforceable limit on hours of

operation is equivalent to a 98.8% reduction compared to

8,760-hour operation. The unit is required to meet Tier 4

NSPS standards by the manufacturer. Add-on controls

for emergency-only engines are not considered feasible.

Emergency use

and readiness

testing only.

Tier 4 NSPS

standards certified

by engine

manufacturer.

Black Start

Generator 0.01

This is a small, black start generator. Otherwise, same as

above.

Black start and

readiness testing

only. Certification

same as above.

4.4 PM10/PM2.5 BACT Review

BACT for PM10/PM2.5 is analyzed by emissions unit.

Emissions Units Subject to PSD Review – PM10, PM2.5

EU ID Description

Maximum Emission

Rate (tpy)

PM10 PM2.5

Electric arc furnace Smelting, tapping, pouring, casting 54.44 43.55

Ladle preheat & dry Natural gas-fired combustion units * *

Process gas heater Natural gas-fired combustion unit 7.14 7.14

Pellet Material Handling Handling, screening iron ore pellets 1.29 0.89




- 58 -

Emissions Units Subject to PSD Review – PM10, PM2.5

EU ID Description

Maximum Emission

Rate (tpy)

PM10 PM2.5

Startup boiler Natural gas-fired combustion unit 0.49 0.49

DRI gas reconditioning Flare used to control CO emissions at start up 0.22 0.22

Roadways/parking Fugitive emissions from traffic 0.22 0.02

Fines Handling Handling of iron ore fines 0.06 0.01

Storage piles Petmin remet piles 0.05 0.05

3 Generators Diesel generators – emergency only 0.02 0.02

Cooling tower Drift from non-contact cooling tower 0.02 0.00

High Pressure Emg. Pump High pressure emergency pump 0.01 0.01

Low Pressure Emg. Pump High pressure emergency pump 0.00 0.00

* Included in Electric Arc Furnace PM10 and PM2.5 emissions.


The operation of the EAF was described in Section 2.


Method 1. Fabric Filter – Fabric filters are the standard in the iron and steel industry for most

PM10/PM2.5 control applications. Baghouses often are capable of 99.9% removal efficiencies, and

baghouse removal efficiency is relatively level across the particle size range.

The main operating limitation of a baghouse is that its operating temperature is limited by the bag

material. Most filter materials are limited to 200°F to 300°F. Some materials like glass fiber or

Nomex™ may be operated at 400°F, but are more expensive. Baghouses also do not operate well

when there is a significant amount of moisture in the stream being treated. The moisture causes

the particulates to form a very thick, wet and heavy cake that plugs the bags and cannot be

removed. The plugging significantly reduces or blocks the airflow increasing the pressure drop

across the bags or completely making the unit inoperable. High moisture levels can also weaken

some bag materials (e.g., polyester) due to hydrolysis, resulting in reduced bag life.

Baghouse control efficiency under normal loading conditions typically is in the 99+ percent range.

Reduced efficiencies will occur when the inlet particle concentration is low. Baghouse controls




- 59 -

do not significantly affect the temperature of the stream being treated; therefore, condensable PM

control is limited to particulates present in a filterable form in the baghouse when it is being treated.

Another disadvantage of baghouses is high energy use. High efficiency filter media in baghouses

require higher pressure drops than electrostatic precipitators (ESPs). On high-flow volume

sources, the increase in pressure drop for a baghouse vs. ESP can result in significant increases in

electrical energy use. Therefore, baghouses are most effective controlling sources with filterable

particulates and less effective controlling hot exhaust streams with condensable particulates.

Baghouses are the most effective means of controlling directly emitted filterable PM2.5 because

filters are best at controlling small particles.

Method 2. Wet Scrubber – Wet scrubbers, also termed particulate scrubbers, remove particles

from waste gas by capturing the particles in liquid droplets (usually water) and separating the

droplets from the gas stream. The droplets transport the particulate out of the gas stream.

Scrubbers may capture particulates through the following mechanisms:

• Impaction of the particle directly into a target droplet.

• Interception of the particle by a target droplet as the particle comes near the droplet.

• Diffusion of the particle through the gas surrounding the target droplet until the particle is

close enough to be captured.

Scrubbers are generally classified according to the liquid contacting mechanism used. The most

common scrubber designs are spray-chamber scrubbers, cyclone spray chambers, orifice and wet

impingement scrubbers, and venturi scrubbers.

Operating conditions inside of a scrubber can be very corrosive if acid gases are present in the

waste gas, and highly abrasive PM can cause erosion problems. These conditions lead to reduced

equipment operating life and/or increased capital cost for materials of construction. Scrubber

control efficiency under normal loading conditions typically is in the 95 to 99+ percent range.

Scrubber efficiency is typically a function of pressure drop across the scrubber. Consequently,

higher collection efficiencies will consume more electrical power to operate the scrubber blower.

The temperature of an exhaust gas may be reduced in a wet scrubber due to evaporative cooling

by the scrubber water. Since wet scrubbers use water, they may also capture PM composed of

water-soluble salts. Therefore, wet scrubbers are more likely to control condensable particulates

than a baghouse or dry ESP. Reduced efficiencies will occur when the inlet particle concentration

is low. Outlet concentrations achieved will depend on the size range and nature of the particles.

Wet scrubbers are not as effective as baghouses and ESPs in controlling directly emitted filterable

PM2.5.




- 60 -

Method 3. Electrostatic Precipitators (ESP) – An ESP applies electric forces to separate

suspended particles from the flue gas stream. In an ESP, an intense electrostatic field is maintained

between high-voltage discharge electrodes, typically wires or rigid frames, and grounded

collecting electrodes, typically plates. A corona discharge from the discharge electrodes ionizes

the gas passing through the precipitator, and gas ions subsequently ionize the particles. The

electric field drives the negatively charged particles to the collecting electrodes. Periodically, the

collecting electrodes are rapped mechanically to dislodge collected PM, which falls into hoppers

for removal. Collector dust is removed from the precipitator for disposal, recycling or

reprocessing.

Since ESPs use electrical forces for particle collection, the electrical properties of the particles can

adversely impact ESP operation. Particles with high resistivity may not readily accept an electric

charge and will be difficult to collect. Particles with high conductivity or magnetic properties,

such as iron, will strongly adhere to the collection plates and be difficult to remove.

ESP control efficiency under normal loading conditions typically is in the 98 to 99+ percent range.

Reduced efficiencies will occur when the inlet particle concentration is low. ESPs are very

effective at removing particles larger than 1 micron. ESPs use electrostatic force to move

particulates to the collection plates. This increases the effectiveness of impaction and interception

collection mechanisms. However, electrostatic forces diminish as particle sizes decrease.

Therefore, they are not as effective as baghouses on particles 1 micron and smaller where diffusion

is the primary particulate control mechanism. Travel distances are larger in ESPs than in fabric

filters; therefore, they are not as effective as filtration in capturing sub-micron particles.

Method 4. Wet Electrostatic Precipitators (Wet ESP) – A wet ESP operates in the same manner

as a dry ESP; it applies electric forces to separate suspended particles from the flue gas stream.

However, particle removal in a wet ESP is accomplished with water sprays instead of mechanical

cleaning methods. In a wet ESP, an intense electrostatic field is maintained between high-voltage

discharge electrodes, typically wires or rigid frames, and grounded collecting electrodes, typically

plates. A corona discharge from the discharge electrodes ionizes the gas passing through the

precipitator, and gas ions subsequently ionize the particles. The electric field drives the negatively

charged particles to the collecting electrodes. As a result of using water sprays, wet ESPs generate

wastewater which must be treated to remove suspended particles and dissolved solids.

Wet ESP water sprays are effective at removing material collected on the ESP plates, but may

require more maintenance than dry ESPs in order to keep the water spray system working properly.

Wet ESP control efficiency under normal loading conditions typically is in the 98 to 99+ percent

range. Reduced efficiencies will occur when the inlet particle concentration is low. Outlet




- 61 -

concentrations achieved will depend on the size range and nature of the particles being filtered.

Wet ESP control efficiency for PM2.5 is similar to dry ESP controls for the reasons noted above.

Method 5. Mechanical Collectors – Mechanical collectors use a variety of mechanical forces to

collect PM:

• Inertial separators use inertia and gravity to remove the larger particles from the smaller

ones.

• Cyclones use centrifugal force to separate PM from gas streams.

Drop-out boxes are typically used as inertial separators. Larger particles are trapped in drop-out

boxes as the inertia they contain forces them to go straight as the rest of the gas stream turns to

flow into and out of the drop-out box. Particles are also removed by gravitational settling in the

drop-out box. Inertial separators can only remove the larger dust particles (>75 microns). They

are typically used upstream of other control devices in high inlet dust loading cases.

Cyclone separators are designed to remove particles by inducing a vortex as the gas stream enters

the chamber, causing the exhaust gas stream to flow in a spiral pattern. Centrifugal forces cause

the larger particles to concentrate on the outside of the vortex and consequently slide down the

outer wall and fall to the bottom of the cyclone, where they are removed. The cleaned gas flows

out of the top of the cyclone.

There are two principal types of cyclones: tangential entry and axial entry. In tangential entry

cyclones, the exhaust gas enters an opening located on the tangent at the top of the unit. In axial

flow cyclones, the exhaust gases enter at the middle of one end of a cylinder and flow through

vanes that cause the gas to spin. A peripheral stream removes collected particles while the cleaned

gas exits at the center of the opposite end of the cylinder.

Overall cyclone control efficiencies range from 50 to 99 percent with higher efficiencies being

achieved with large particles and low efficiencies for smaller particles (<PM10 and PM2.5).

Mechanical separators are often used upstream of other PM control devices to reduce the loading

on the primary control device. This improves overall control efficiency and may reduce the overall

cost of the control system when the waste gas is heavily laden with PM.


1. Fabric Filter (Baghouse) – 99+%*

2. Electrostatic Precipitator (ESP) – 99+*%

3. Wet ESP – 99+*%

4. Wet (Venturi) Scrubber – 99+*%




- 62 -

5. Mechanical Collector – 50% - 80%

(*nominal efficiencies listed for PM10 control)

Potential control alternatives were ranked for effectiveness in controlling PM10/PM2.5 emissions

from the EAF. As noted, in general, fabric filters, if feasible, are the most effective APC method

in the metals industry.


Fabric filters or baghouses are an industry standard for PM10/PM2.5 control in many applications.

Baghouses are capable of 99+% removal efficiencies. Baghouse removal efficiency is relatively

level across the particle size range so that excellent control of all particle sizes can be obtained.

Baghouses can be effectively applied to most dry dust sources, but are not typically effective at

very high temperatures that may burn the filter material.

Electrostatic precipitators are common dust control devices that are good at trapping dusts within

certain applications. ESPs applied to materials within the correct resistivity range can, in certain

cases, achieve very high removal efficiencies, up to 99.5%. ESPs typically have a higher initial

capital cost than baghouse controls. The presence of iron in the exhaust stream makes it likely that

the collected particulate would be difficult to remove from the plates. In addition, a wet ESP

would create a wastewater stream that would cause other environmental issues.

High-energy wet scrubbers are technically feasible and achieve good control efficiencies.

However, scrubber systems have very high pressure drops that result in high system operating

costs. They also require water treatment and sludge disposal that are not required for other PM

control options.

Mechanical collectors are effective at removing large dust particles using centrifugal forces.

However, fine dusts are typically not as effectively removed, due to the high gas stream velocity

that must be established, often keeping smaller particles entrained in the stream.

Step 5: Select BACT

An important distinction in this Project is that the particulate matter is estimated to be as much as

80 percent by weight of PM2.5. Given the expected particle size distribution of the EAF exhaust

gas, a fabric filter baghouse is the top control technology.14




- 63 -

Based on the top-down BACT analysis, the best available control technology is a mechanical

collector (drop-out box) followed by a fabric filter baghouse.

Efficiency for a baghouse in the iron and steel industry is specified on an absolute basis, i.e., at a

continuous, guaranteed minimum performance in units of outlet concentration, in units of mass

per unit volume of gas. Petmin has specified a baghouse guaranteed to achieve an absolute

efficiency of 0.0025 grains per dry standard cubic feet.

Dust Capture, Baghouse Operation

The baghouse will be “pulse‐jet negative pressure” type, designed for high filtering velocity with

bag cleaning performed by means of compressed air. The filter will be constructed of separate

cells that can be isolated by dampers. Inside the filter, the flow of gases is upward, crossing the

polyester felt bags from the outside to the inside. The bags have mechanical and physical

characteristics to withstand abrasion and chemical aggression, and can operate continuously to a

maximum working temperature of 130°C. The dust particles remain attached to outside the bag

surface and are cleaned periodically by a compressed air pulse‐jet. The cleaning cycle time is

adjustable and depends on the operating phase and will be set during start up and operation.

The primary fumes generated during DRI melting will be extracted through the hood around the

electrodes on the furnace roof, by a water-cooled duct. The connection between the furnace hood

and water-cooled ducts is obtained with a movable sliding skirt to allow the change of the gap for

the air inlet; the sliding skirt is located on the first fixed duct. On the first part of the first water

cooled duct there will be a sliding cooled elbow to avoid dust deposit and to allow EAF roof

opening and maintenance works on the furnace. After this connection, before the fixed cooling

duct line, the arrangement of a settling chamber – or drop out box (DOB) – will facilitate additional

oxidation of CO and will separate the heaviest particles of dust from the off‐gasses.

Further cooling is achieved in the remaining part of the water‐cooled hot gas line. The main

parameter to be controlled for a correct primary fume suction, according to the EAF working

conditions, is the pressure inside the furnace; for maintenance reasons, the pressure probe is

located inside the furnace elbow. The target is to have slightly positive pressure in the EAF and

Technology

Expected Control Efficiency

at a Given Size

5µm 2µm 1µm

Fabric Filter 99.8 99.5 99

Venturi Scrubber 99.7 99 97

ESP 99 95 85




- 64 -

that is the reason why there is no water-cooled elbow on the EAF roof, so to avoid the direct

suction and consequent infiltration of fresh air, which would unnecessarily oxidize the EAF bath.

The secondary fume generated during the various furnace operating phases is controlled with a

canopy hood located on the main building roof. The canopy hood is designed according to the

EAF dimensions and operating parameters. The canopy hood is also used to maintain the required

ventilation inside the main melt shop building. In summary, BACT for PM10 and for PM2.5 for the

EAF and the pig casting operation consists of:

• Capture designed for full capture of particulate matter from the EAF

• Drop-out box pre-collector for removal of large particles, followed by

• Fabric filter (baghouse) designed to achieve an absolute efficiency of 0.0025 grain per dry

standard cubic foot percent control of filterable particulate matter

4.4.2 Material Handling and Screening

Iron ore pellets are loaded onto Petmin’s covered conveyor system by an independent third party.

Pellets are conveyed to a screening operation to remove fines before continuing to the production

building. Pellets are coated with a cement slurry prior to charging to the reactor.

Control of fugitive emissions is dependent upon the proximity of dust generation points to the

control equipment. Iron ore pellets are roughly spherical in shape, requiring conveyor belts to

have only gradual slope. Collection of dust is simplified in Petmin’s design for the screening and

production buildings, with a central dust collection system. At the tip area where pellets are loaded

into the hoppers, there is also dust suppression at the transfer points, utilizing a suppressing spray

to blanket and suppress airborne particulate matter.


Method 1. Fabric Filter – Fabric filters are the standard in the iron and steel industry for most

PM10/PM2.5 control applications. Baghouses often are capable of 99.9% removal efficiencies, and

baghouse removal efficiency is relatively level across the particle size range.

Local collection hoods routed to fabric filters with enhanced filter media are the most efficient

means of removing PM10/PM2.5 from dusty sources. The advantage of local collection hoods and

bag filters is that air flows from individual collection points in the system can be adjusted to

increase or decrease collection rates, increasing the effectiveness of the system. The fabric filter

is largely insensitive to variability in dust loading that could occur during iron oxide screening,

transfers and other operations. For these reasons, local collection hoods and baghouse filters are

the industry standard for the control of particulate from iron ore pellet handling.




- 65 -

Method 2. Wet Scrubber – Wet scrubbers, also termed particulate scrubbers, remove particles

from waste gas by capturing the particles in liquid droplets (usually water) and separating the

droplets from the gas stream. The droplets transport the particulate out of the gas stream.

Scrubbers were described in the previous subsection.

Method 3. Electrostatic Precipitators (ESP) – An ESP applies electric forces to separate

suspended particles from the flue gas stream. In an ESP, an intense electrostatic field is maintained

between high-voltage discharge electrodes, typically wires or rigid frames, and grounded

collecting electrodes, typically plates. A corona discharge from the discharge electrodes ionizes

the gas passing through the precipitator, and gas ions subsequently ionize the particles. The

electric field drives the negatively charged particles to the collecting electrodes. Periodically, the

collecting electrodes are rapped mechanically to dislodge collected PM, which falls into hoppers

for removal. Collector dust is removed from the precipitator for disposal, recycling or

reprocessing.

ESPs were described in the previous subsection.

Method 4. Centrifugal Collectors – Centrifugal collectors use cyclonic action to separate

particles from the gas stream. In a typical cyclone, the gas stream enters a vessel at an angle and

is spun rapidly. The centrifugal force created by the circular flow throws the particles toward the

wall of the cyclone. After striking the wall, these particles fall into a hopper located beneath the

cyclone. Single-cyclone separators create a dual vortex to separate coarse particles from fine. The

main vortex spirals downward and carries most of the coarser dust particles. The inner vortex

created near the bottom of the cyclone spirals upward and carries finer dust particles. Multiclones

consist of a number of small diameter cyclones, operating in parallel and having a common gas

inlet and outlet. Multiclones operate on the same principle as cyclones by creating a main

downward vortex and an ascending inner vortex. Multiclones are more efficient than single

cyclones because they are longer and smaller in diameter. The longer length provides longer

residence time while the smaller diameter creates greater centrifugal force. These two factors

result in better separation of dust particulates. The pressure drop of multiclone collectors is higher

than that of single-cyclone separators. Typical removal efficiency is 80%.

Method 5. Wet Suppression – Fine mists of water applied to dust generating sources, such as

bulk material drop points, reduce dust emissions by impacting small particulates with water. The

wetted particulate becomes heavier and quickly settles out of the air, reducing airborne dust.

Alternatively, material may be thoroughly wetted prior to handling, which suppresses the

generation of dust when the material is disturbed. Typical removal efficiency is 50% - 90%.




- 66 -

Method 6. Covered Conveyors and Enclosed Transfer Points – Covered conveyors are an

application of partial enclosures and wind screens that are specific to conveyor systems. The

conveyor hoods work to help prevent wind from lifting dust particles from materials being

transported on the conveyor. Similarly, enclosed transfer points work to isolate material drop

points between conveyors from the surrounding weather conditions. Enclosed transfer points are

typically designed with minimized material drop heights to reduce dust generated by materials

being transferred.


1. Fabric Filter (Baghouse) - 99.9%

2. Electrostatic Precipitator (ESP) - 99.5%

3. Wet Scrubber - 99%

4. Wet Suppression - 90%

5. Centrifugal Collectors - 80%

6. Covered Conveyors and Enclosed Transfer Points - 95% (Conveyance Only)

Potential control alternatives were ranked for effectiveness in controlling PM10/PM2.5 emissions

from the material handling and screening operations. The baghouse is the top control option.


Fabric filters or baghouses are an industry standard for PM10/PM2.5 control in many applications.

Baghouses are capable of up to 99.9% removal efficiencies. As described in the previous

subsection, a dust collection system is specified to achieve a specific outlet concentration, i.e.,

absolute efficiency. Baghouse removal efficiency is relatively level across the particle size range

so that excellent control of all particle sizes can be obtained. Baghouses are not sensitive to

variations in loading rates.

Review of the EPA RACT/BACT/LAER Controls (RBLC) database confirms that the use of a

fabric filter constitutes BACT. An absolute control efficiency of 0.0025 grains per dry standard

cubic foot compares favorably to the most recent DRI project, where a venturi scrubber option

achieving 0.0075 grains per dry standard cubic foot was determined to meet BACT.




- 67 -

RBLC Search of BACT Controls for DRI Production

Date Company Process Area Controlled Technology Employed Control Efficiency

5/24/2010

NUCOR STEEL

LOUISIANA, Convent,

LA

PIG-101 - Pig Iron Desulfurization Station Baghouse Vent Fabric Filter 99.50%

PIG-102 - Pig Iron Solidification Baghouse Vent Fabric Filter 99.50%

2/29/2018

IRONUNITS LLC -

TOLEDO HBI, Toledo

OH

Direct Reduced Iron reactor shaft furnace (P001) Venturi Scrubber 0.0075 gr/dscf

Iron briquetting machine (P002) Venturi Scrubber 0.0076 gr/dscf

Iron briquette cooling system (P003) Venturi Scrubber 0.0076 gr/dscf

Oxide Handling, Bins, Screens (P901) Baghouse efficiency not specified

Baghouse efficiency not specified

4/24/2014 MAG PELLET LLC,

Reynolds, IN

Mixing Area Material Handling System Baghouse 0.002 gr/dscf

Furnace Hood Exhaust Baghouse 99%

Baghouse 99%

Furnace Windbox Exhaust Baghouse 99%

Baghouse 99%

Coke Breeze Conveyance & Storage Bin Baghouse 0.002 gr/dscf

Oxide Pellet Storage System Baghouse 0.002 gr/dscf

Oxide Pellet Loadout System Baghouse 0.002 gr/dscf

5/10/2011

THYSSENKRUPP

WAPUPACA, INC.,

Indianapolis, IN

16 Ton Iron Bath Desulfurization Ladle Operation Baghouse 99.90%

1/27/2011

DIRECT REDUCTION

IRON PLANT, Convent,

LA

DRI Oxide Day Bins Dust Collection Baghouse 99.50%

DRI Oxide Screen Dust Collection Baghouse 99.50%

DRI Furnace Feed Conveyor Baghouse Baghouse 99.50%

DRI Coating Bin Filter Baghouse 99.50%




- 68 -

Step 5: Select BACT

Based on the top-down BACT analysis, the best or “top” available control technology is the use

of covered conveyors for material transfer, and a fabric filter baghouse achieving 0.0025 grain per

dry standard cubic foot control of PM10/PM2.5. The screening building will be vented to a baghouse

for collection of fines. Once pellets have entered the production building, emissions are captured

and plant air exits through the EAF baghouse.

Covered conveyors and enclosed transfer points will be installed to limit emissions from material

handling activities outside of buildings where individual dust generation sources are small and

scattered across a wide area. In summary, BACT consists of:

• Covered conveyors and transfer points

• At the tip area where pellets are loaded into the hoppers, dust suppression at the transfer

points, utilizing a suppressing spray to blanket and suppress airborne particulate matter.

• Fabric filter (baghouse) with a design efficiency of 0.0025 grain per dry standard cubic

foot on PM10 and PM2.5.

4.4.3 Roadways and Parking Areas

Petmin’s facility is located on a leased site within the perimeter of an existing, permitted industrial

facility. Facility roadways are therefore contiguous with other controlled roads.


Method 1. Commercial Dust Suppressants – Chemical surface stabilizers act to agglomerate

particles on road surfaces. These larger particles are less likely to become airborne due to wind or

passing vehicles. Surface stabilizers must be reapplied periodically as they are broken down by

traffic and rain events. Efficiencies of 95% can be achieved.

Method 2. Wet Suppression – The use of water suppression can control PM10/PM2.5 emissions

by up to 95%, although it must be applied more frequently than chemical dust suppressants. Water

is typically applied using a water truck, i.e., a truck-mounted sprayer.


1. Commercial Dust Suppressants - 95%


Potential control alternatives were ranked equally effective in controlling PM10/PM2.5 emissions

from the roadways and parking areas.




- 69 -


Commercial dust suppressants and Wet suppression are equally effective in controlling fugitive

dust from roadways. The main differences between the options are the required frequency of

reapplication. Water costs less and has lower potential for soiling vehicles and contaminating

nearby surface water.

Step 5: Select BACT

Based on the top-down BACT analysis, the best available control technology is wet suppression,

due to having comparable control efficiency, fewer possible environmental issues, and lower cost.

In summary, BACT consists of applying wet suppression to roadways, if required, based on routine

visual assessments.

4.4.4 Fuel Burning Equipment (Process Gas Heater, Ladle Preheat, Flare, and Startup Boiler

The listed equipment burns natural gas as the primary fuel. Natural gas is recognized as a “clean”

fuel with respect to potential emissions of particulate matter. The process gas heater and flare also

burn some process gas, consisting principally of hydrogen and carbon monoxide. All will emit

particulate matter as the products of incomplete combustion. Larger carbon compounds due to

incomplete combustion are typically filterable. Smaller organics that are not completely

combusted can be gaseous at typical flue gas temperatures, and later condense after being emitted

by the source. Control technologies that rely upon direct filtration or capture of solid particles may

be ineffective at controlling condensable particulate matter. In the case of natural gas combustion,

half or more of total particulate is generally assumed to be condensable.


Method 1. Good Combustion Practices – Good combustion practices are used to reduce

emissions of PM10/PM2.5, as well as other pollutants, by optimizing conditions in the combustion

zone of a fuel burning source. Particulate emissions from the burning of fuels are usually due to

the incomplete combustion of hydrocarbon fuel but may also be due to inorganic particles present

in the fuel as an impurity. Good combustion practices typically entail introducing the proper ratio

of combustion air to the fuel, maintaining a minimum temperature in the firebox of the combustor,

or a minimum residence time of fuel and air in the combustion zone. By employing good

combustion practices, both the filterable and condensable fractions of particulate matter normally

emitted may be greatly reduced.


1. Good Combustion Practices - 50%




- 70 -

A review of the potential control options, and combination of options, was conducted to identify

the most effective strategy for controlling PM10/PM2.5 emissions from the combustion of natural

gas. Petmin has identified good combustion practices as the most effective and technically feasible

control option.


Good combustion practices can be effective at reducing PM10/PM2.5 emissions generated from

the incomplete combustion of hydrocarbon fuels, because the technique limits both the filterable

and condensable fractions of particulate emissions. Because of this property unique to the control

technology set, and the fact that the majority of particulates generated by natural gas combustion

are condensable, good combustion practices ranks the highest of the identified control

technologies.

Step 5: Select BACT

Based on the top-down BACT analysis, the best available technology for controlling PM10/PM2.5

from the fuel burning equipment is the use of natural gas, supplemented by hydrogen and carbon

monoxide from recycled gas, and good combustion practices. Petmin will maintain good

combustion practices to control the generation of PM10/PM2.5 emissions that could otherwise occur

from incomplete combustion of natural gas and process gas.

4.4.5 Emergency Engines, Fire Pumps, and Black Start Generator

The listed equipment burns ultra-low sulfur diesel fuel as the primary fuel. Larger carbon

compounds due to incomplete combustion are typically filterable. Smaller organics that are not

completely combusted can be gaseous at typical flue gas temperatures, and later condense after

being emitted by the source. Control technologies that rely upon direct filtration or capture of

solid particles may be ineffective at controlling condensable particulate matter.


Method 1. Good Combustion Practices – Good combustion practices are used to reduce

emissions of PM10/PM2.5, as well as other pollutants, by optimizing conditions in the combustion

zone of a fuel burning source. Particulate emissions from the burning of fuels are usually due to

the incomplete combustion of hydrocarbon fuel, but may also be due to inorganic particles present

in the fuel as an impurity. Good combustion practices typically entail introducing the proper ratio

of combustion air to the fuel, as well as proper design and maintenance of equipment. By

employing good combustion practices, both the filterable and condensable fractions of particulate

matter normally emitted may be greatly reduced.




- 71 -

Method 2. Emergency Generator Classification – The proposed diesel generators and fire

protection pumps are classified as emergency generators. The black start generator is not defined

as an emergency generator but will operate only during startup from a cold condition. Each of the

units is limited by federal requirements to operate no more than 100 hours per year for maintenance

and readiness testing. Otherwise, operation is limited to emergency conditions or a black start

condition. These limitations effectively minimize PM10/PM2.5 emissions.


1. 1. Limitation on Hours for Maintenance and Readiness Testing 98.8%

(100 hours out of 8,760)

2. Good Combustion Practices – 50%

A review of the potential control options, and combination of options, was conducted to identify

the most effective strategy for controlling PM10/PM2.5 emissions from the combustion of diesel

fuel. Petmin has identified good combustion practices, in combination with very limited operation,

as the most effective control option.


Good combustion practices can be effective at reducing PM10/PM2.5 emissions generated from

the incomplete combustion of hydrocarbon fuels, because the technique limits both the filterable

and condensable fractions of particulate emissions. Operating fewer than 100 hours per year for

maintenance and readiness testing results in zero emissions approximately 99% of the time.

Step 5: Select BACT

Based on the top-down BACT analysis, the best available technology for controlling PM10/PM2.5

from the generators is good combustion practices, combined with limited operation. Petmin will

maintain good combustion practices in all fuel burning equipment to control the generation of

PM10/PM2.5 emissions due to incomplete combustion of diesel fuel.

4.4.6 Storage Piles

Petmin will have storage piles of two materials: Remet (partially-reduced iron ore pellets, which

are taken back to the pellet loading hopper; and pig iron castings. As described previously in this

report, Remet is an off-specification, partially reduced iron that is produced in extremely limited

quantity, ideally only during plant startup/shutdown. As shown in the calculations provided with

this application, the emissions from storage piles are minimal, and qualify in Ohio rules as “de

minimis” units.




- 72 -


Method 1. Reduction of Fines – Fugitive dust is a result of fine particulate in the stored material

being entrained in the air as a result of pile disturbance and/or high wind conditions. If there are

no particles light enough to be windborne, there will be no emissions.

For the materials stored by Petmin, Remet has had fines removed from the pellets before

processing; pig iron castings are cooled/solidified metal. Remet is only produced during upset and

startup/shutdown conditions, which will be minimized to the extent possible.

Method 2. Commercial Dust Suppressants – Chemical surface stabilizers act to agglomerate

particles on storage piles. These larger particles are less likely to become airborne due to wind or

disturbances. Surface stabilizers must be reapplied periodically after rain events and as stock is

moved from the piles. Efficiencies of 95% can be achieved.

Method 3. Wet Suppression – The use of water suppression can control PM10/PM2.5 emissions

by up to 95%, although it must be applied more frequently than chemical dust suppressants. Water

is typically applied using a sprayer or sprinkler.


1. Reduction of Fines – 99%

2. Commercial Dust Suppressants - 95%


A review of the potential control options, and Petmin’s stockpiled materials, was conducted to

identify the most effective strategy for controlling PM10/PM2.5 emissions from the storage piles.

All listed options are comparable.


Reduction of fines is the first option, if it can be achieved, due to its simplicity. Commercial

dust suppressants and Wet suppression are equally effective in controlling fugitive dust from

storage piles. The main differences between the latter two options are the required frequency of

reapplication. Water costs less and has lower potential for contaminating nearby surface water.

Step 5: Select BACT

Based on the top-down BACT analysis, the best available control technology for storage piles is

the reduction of fines in the stored material. As a contingency, wet suppression is preferred to

other dust suppressants due to having comparable control efficiency, fewer possible environmental

issues, and lower cost. In summary, BACT consists of:




- 73 -

• Minimal fines in stored materials

• Wet suppression is not expected to be required based on the materials and quantities in

storage piles.

4.4.7 Cooling Tower

The cooling towers are equipped with drift eliminators. Emission calculations show that total PM

emissions are 0.019 tons per year. Control beyond the use of drift eliminators is not feasible.

BACT is the use of drift eliminators.

4.5 Greenhouse Gases (GHG)

BACT for GHG emissions is analyzed by emissions unit.

Emissions Units Subject to PSD Review – GHG


Rate (tpy)

Process gas heater Natural gas-fired combustion unit 307,490

Electric arc furnace Smelting, tapping, pouring, casting 49,095

Ladle preheat & dry Natural gas-fired combustion units 23,188

Startup boiler Natural gas-fired combustion unit 7,814

DRI gas reconditioning Flare used to control CO emissions at start up 3,405



High Pressure Emg. Pump High pressure emergency diesel fire pump 17.9




GHG are minimized from the PGH by use of clean-burning natural gas as fuel; good combustion

practices, by optimizing the air-to-fuel ratio; operating the burners in a manner to obtain the

maximum heat output per unit of carbon combusted; and recovering/recycling unburned reduction

gases in the reducing gas reconditioning loop, including methane.

The technology used in the reducing gas reconditioning loop, chemical absorption, could, in

theory, be applied to the PGH waste gas stream. However, unlike the reconditioning loop, the




- 74 -

exhaust gas from the PGH does not contain the recoverable CO and H2 that can be reused in the

process, which made CO2 separation an economically viable possibility. Chemical absorption is

not an “available technology” for a simple combustion unit.

Calcium cycle separation uses quicklime (CaO) to capture CO2 to form limestone, which can be

reheated to release CO2 in a concentrated stream and regenerating the quicklime for reuse. This

technique is in the R&D phase and is not considered an “available” technology.

A remaining technology to consider is carbon capture and sequestration (CCS). In theory, CCS

involves capturing and concentrating CO2 from its source by separating it from the PGH gas

stream, then transporting it, then storing it away from the atmosphere for a long period of time,

such as in underground geological formations or in the deep ocean. CCS is not an “available

technology” for use to Petmin to capture and sequester the CO2 from the PGH.

Petmin’s end product, nodular pig iron, is purified to a degree that requires less processing by its

end users. Fundamentally, the production of NPI reduces GHG downstream.

With no technically feasible options available BACT consists of:

• Good combustion practices

• Production of NPI, a high-purity product that results in lower GHG emissions by end users

It should be noted that the Petmin project planners and designers maintain a long-range objective

to sell the CO2-rich waste gas from the gas conditioning loop to a gas processing plant owned and

operated by an independent third party. This novel, innovative approach is not a demonstrated

technology in DRI production, and is, therefore, above and beyond BACT-level controls.

However, if and when demonstrated to be practicable, it is possible that a 50% GHG reduction

could be achieved through innovative separation of CO2 sold as a feedstock to a third-party

processor.

4.5.2 Other Emissions Units

Carbon present in the DRI melt or from sublimation/oxidation of the electrodes is inherent to the

process. GHG formation is minimized by the use of electrical energy as the heat source;

conservation of energy by well-designed refractory and good operating practices. No add-on

controls or other technology (as described in the previous section) are technically feasible.

The remaining emissions units are relatively small combustion units that are either:




- 75 -

• Fueled with natural gas, and operated with good combustion practices; or

• Emergency or black start-up engines with minimal expected run time and

correspondingly minimal GHG emissions.

No additional control measures are technically feasible for these units.

4.6 CO BACT Review

BACT for CO is analyzed by emissions unit. This section focuses on the principal source of CO

emissions from the Project, the Electric Arc Furnace (EAF).

Emissions Units Subject to PSD Review – CO


Rate (tpy)

Electric arc furnace Smelting, tapping, pouring, casting 474.06

Process gas heater Natural gas-fired combustion unit 48.92

DRI gas reconditioning Flare used to control CO emissions at start up 8.97

Ladle preheat & dry Natural gas-fired combustion units 6.78

Startup boiler Natural gas-fired combustion unit 5.47



High Pressure Emg.

Pump

High pressure emergency diesel fire pump 0.09




The Tenova EAF design engineering team provided the following information to establish BACT

for CO emissions.

Description

The EAF is a cylindrical vessel with a diameter of 21 feet. The walls of the furnace are completely

refractory lined. The hearth or bottom of the furnace is thick refractory, and the roof is water-

cooled. Carbon electrodes pass through openings at the top of the furnace and are 24 inches (610

mm) in diameter. Electric current is carried from a transformer, through support arms, and into the




- 76 -

electrodes where the metallic bath completes the electric circuit. The electrodes deliver the power

to the furnace in the form of an electric arc between the electrode and the furnace charge and bath.

The arc itself is a plasma of hot, ionic gasses in excess of 6,000°F (3,300°C).Unlike most EAFs,

which are charged with steel scrap and sometimes with Direct Reduced Iron (DRI) to produce

liquid steel, this novel unit is designed to be fed with high carbon DRI to produce hot pig iron.

Pig iron is an alloy of iron and carbon with higher carbon content than steel. Therefore, the process

is much higher in carbon than an EAF producing steel. Unlike most EAFs producing steel, there

is no oxygen injection; by contrast, the Petmin EAF uses a reducing atmosphere. Under this

condition CO is the combustion product of carbon because of the low amount of oxygen. CO is

emitted as a byproduct.

CO Emissions

In the EAF, possible sources of the carbon are from the DRI, possible additions of carbon, and the

furnace electrodes. More specifically, the EAF generates CO as a result of oxidation of carbon that

is introduced into the furnace charge to refine the metal bath. Sublimation and subsequent

oxidation of the carbon electrode is also expected to occur. Several operating parameters impact

CO emissions, including the amount of carbon, oxygen, temperature within the EAF, composition

of the DRI charged, and operating practices.

Step 1: Identify all Control Technologies

Possible alternatives for control of CO emissions from the EAF were identified as follows:

• Flare CO Emissions

• DEC Controls

• Post Combustion Reaction in DEC Circuit and Chamber

• Oxygen Injection

• CO Oxidation Catalyst and Catalytic Incineration

• EFSOP™ (Expert Furnace System Optimization Process)

Step 2: Eliminate Technically Infeasible Options

Flaring

Flaring of blast furnace gases and Basic Oxygen Furnace (BOF) gases is quite common in the iron

and steel industry. Flare stacks are used to burn off non-recoverable flammable gases. However,

the flaring option is used when the heating value and the auto-ignition temperature of the off-gases

are within their flammability limit. In the exhaust gas from the Petmin EAF, the CO concentration

is a few thousand ppm, much lower than the concentration required for flaring. (The lower




- 77 -

flammability limit for CO is 125,000 ppm.) Direct flaring is not a technically feasible option in

this case. Further, based on review of available information, there is no known application of

flaring raw EAF exhaust gases. This technology was eliminated from further consideration.

DEC Controls

A Direct Shell Evacuation Control (DEC) system is comprised of a water-cooled duct connected

to the EAF via the furnace roof’s “fourth hole”. The duct is connected to the melt shop canopy

collector system. During melting and refining, a neutral or slightly negative pressure is maintained

in the furnace to withdraw exhaust gases through the DEC. Where the DEC duct meets the “fourth

hole”, an adjustable gap exists that allows combustion air to enter, providing oxygen to oxidize

CO and any other combustible species. The DEC system allows excellent emissions capture,

promotes combustion of CO and combustibles to a certain extent, and requires the lowest air

volume compared to other EAF capture devices.

The capability is limited due to the cyclic operating schedule (i.e., hot-cold cycling). Thermal

oxidation of CO to CO2 requires temperatures above 1,200°F (650°C). In the Petmin EAF, exhaust

gas temperatures will fluctuate during each melt and often drop below 1,200°F (650°C). It is

estimated that this would occur for several minutes during each melt. The minimum temperature

that would be encountered is estimated to be approximately 350°F (180°C). Thus, during these

periods, the thermal destruction efficiency would be expected to decrease, resulting in significant

CO emissions. Consequently, this control alternative is considered reasonably efficient for

oxidation of residual concentrations of CO but with limitations due to the temperature cycling

intrinsic to the melt cycle.

Post Combustion Reaction in DEC Circuit and Chamber

The principle of destruction in this technology is to raise the EAF exhaust gases in the primary

suction circuit with auxiliary fuel firing to a sufficiently high temperature for a minimum time to

facilitate oxidation through post combustion. Combustion chambers are usually utilized, and their

configuration must provide effective mixing in the chamber with an acceptable residence time.

Post combustion naturally takes place while combustion air is drawn into the circuit and reacts

with CO and other combustibles contained in the offgas.

Again, the reaction depends on various parameters – most importantly, gas temperature and

combustible content. (See also the discussion of “Oxygen Injection” in the next subsection).

Effectively controlling the reactions would require post–combustion system, including burners and

associated controls that can assure ignition of the gas mixture whenever necessary. Based on a

review of available resources, this technology has not been applied on EAFs in the United States;

also, the feasibility of these units to significantly reduce CO emissions, without resulting in severe




- 78 -

operation problems, is unknown. Such units are expected to consume large quantities of natural

gas and oxygen, resulting in excessive annual operating costs and increased NOx emissions.

Precise performance guarantees are difficult to estimate because of the lack of operating

experience. Further, due to the heat and particulate loading, the burners that supply the auxiliary

fuel would have a short life expectancy and may sustain severe maintenance and reliability

problems.

Additionally, a single or multiple duct burner system would not be able to heat the relatively cool

gases from the EAF during cold cycling.

Potentially, there are three locations where post combustion chambers can be installed, (i.e,

upstream or downstream the mixing point of the DEC duct with the main duct, or downstream

of the baghouse).

• Locating a post combustion chamber upstream of the mixing point of the DEC circuit with

the main duct would take advantage of the elevated temperatures in the gas stream.

However, at this location, the post combustion chamber would be subject to extremely high

particulate loading. The units would be exposed to foul frequently from the particulate

accumulation, resulting in severe maintenance and reliability concerns. This option would

be technically difficult to implement and operate, and is considered infeasible.

• Locating a post combustion chamber downstream of the mixing point of the DEC duct with

the main duct would be at a much lower temperature in the exhaust gas stream.

Nevertheless, at this location, the post combustion chamber would still be subject to high

particulate loading, with the same fouling, maintenance and reliability concerns. Moreover,

the higher gas flow at lower temperatures would greatly increase the auxiliary fuel

requirements. Combustion of additional fuel will result in increases in NOx emissions.

Finally, after post-combustion the gas would need to be cooled down again, in order to be

filtered in the baghouse. This option is technically infeasible.

• The post combustion chamber could be installed downstream of the EAF baghouse.

However, at this location fuel consumption and NOx emission concerns would be

considered extreme to a point of practical infeasibility.

Oxygen Injection

A theoretical means of reducing CO would be oxygen injection at the entrance of the ductwork to

increase oxidation of the available CO to CO2. Oxygen injection directly into the furnace to reduce

CO emission levels is an experimental operating practice in Europe used to increase the heat input

to the melt, but the practice has not been demonstrated to reduce CO emissions.




- 79 -

Typically, as already described, the DEC system draws air into the duct, creating an oxygen-rich

mixture of EAF exhaust gases after the gap, where CO is naturally oxidized. The addition of

oxygen is expected to convert little if any CO to CO2. The capability is also limited due to the

cyclic operating schedule (i.e., hot-cold cycling). Thermal oxidation of CO to CO2 requires

temperatures above 1,200°F (650°C). Exhaust gas temperatures will fluctuate during each melt

and often drop below 1,200°F (650°C). It is estimated that this would still occur for several minutes

during each melt, although with a lower frequency and duration than in the DEC control case. The

minimum temperature that would be encountered is estimated to be approximately 350°F (180°C).

Thus, during these periods, the thermal destruction efficiency would decrease significantly,

resulting in elevated CO emissions.

This control alternative is not considered efficient and is eliminated from further consideration in

this analysis.

CO Oxidation Catalyst and Catalytic Incineration

Catalytic incinerators or oxidizers use a bed of catalyst that facilitates the overall combustion of

combustible gases. The catalyst increases the reaction rate and allows the chemical conversion at

lower temperatures than a thermal incinerator. The catalyst is typically a porous noble metal

material, which is supported in individual compartments in the unit. An auxiliary fuel-fired burner

ahead of the bed heats the entering exhaust gases to approximately 600°F (316°C) to maintain

proper bed temperature.

The temperature of the exhaust gases from the EAF will vary significantly during the various

stages of the heat (212° to 1,300° F, or 100° to 700° C). Thus, the temperature will be below the

minimum 500°F (260°C) threshold for effective operation of the catalysts during much of its

operation. In addition, the wide variation in temperature would make operation of a catalyst

impractical. Lastly, the particulate loading in the exhaust gas stream is anticipated to be too high.

Masking effects, such as plugging and coating of the catalyst surface, would almost certainly result

in impractical maintenance requirements, and would significantly degrade the performance of the

catalyst. Trace metals that could occasionally be present in the exhaust stream are generally

considered poisons to catalysts and deactivate the available reaction sites on the catalyst surface.

PM can also build up on the catalyst, effectively blocking the porous catalyst matrix and rendering

the catalyst inactive. In cases of significant levels of poisoning compounds and particulate loading,

catalyst replacement costs are significant.

As in the thermal incineration (post-reaction chamber) discussion, there are two potential locations

where the catalyst bed could be installed, i.e. upstream or downstream of the EAF baghouse. For

the same reasons discussed earlier (e.g., fouling due to PM), the upstream location is considered

technically infeasible.




- 80 -

Alternatively, the catalyst could be installed downstream of the EAF baghouse. However, even at

this location, fouling due to PM could occur, and further, the exhaust would be at a much lower

temperature and with highly increased gas flow. These factors would greatly increase the auxiliary

fuel requirements. The associated combustion of additional auxiliary fuel would result in an

overwhelming increase in natural gas consumption, resulting in increases in emissions (especially

NOx) to the atmosphere.

Based on a review of the available information resources, there is no application of a catalytic

incineration system or an oxidation catalyst to control CO emissions from an EAF. This control

option, while theoretically feasible, is considered to be infeasible as a practical matter to install a

catalytic thermal oxidizer on a variable exhaust stream of several hundred thousand cubic feet per

minute exhaust flow, for CO control. With no known applications for use on an EAF, this option

is eliminated from consideration.

EFSOP™ (Expert Furnace System Optimization Process)

EFSOP® is a burner optimization label that Tenova uses to sell their burner and control systems.

It is an off-gas based process control system which measures off-gas from the melting process on

a continuous basis and uses the output in conjunction with a computer model to optimize furnace

operations and reduce overall conversion costs. A conditioning system cleans the offgas sample

and a portion of it is analyzed for carbon monoxide (CO), carbon dioxide (CO2), hydrogen (H2)

and oxygen (O2).

This technology has been established and Tenova has several installations operating worldwide,

which have measured and analyzed data from millions of heats. In steel producing EAFs, typical

combustion efficiency of CO conversion to CO2 varies from 25-70% for the heat cycle.

The EFSOP® system provides on-line measurements in real-time of what is occurring as the

furnace emissions are exhausted. An added advantage is that the system can be used to control

post-combustion systems installed into the furace. This optimizes furnace combustion, increases

production and saves energy. Steelmakers have reported energy savings of 20kWh/ton of steel and

tap-to-tap time reductions of 2 to 3 minutes with EFSOP®.

The EFSOP® system analyzes the furnace off-gas just before the combustion gap to quantify the

amount of carbon monoxide (CO) in the off-gas. The CO results from the incomplete combustion

of oxygen and fuel in the furnace shell. Some furnace practices and charge mixes also cause high

levels of hydrogen (H2) in the off-gas streams. Together, these combustible gases can make up

over 30 per cent of the furnace offgas and they represent a tremendous loss of energy in the

steelmaking process.




- 81 -

The EFSOP® system allows the steelmaker to tailor the operation of the DEC system to match the

actual requirements of the process. In steel producing EAFs, usually provided with an oxygen

injection system and burners, the continuous fume analysis of the furnace off-gas allows the

steelmaker to conduct controlled post-combustion into the furnace, before it escapes into the DEC

system, thus capturing some of the energy that would be otherwise lost in the off-gas.

In the Petmin EAF, designed for DRI consumption, neither oxygen injection nor oxy-fuel burners

are present or desirable. While the EFSOP® system could be used to analyze the off-gas, there are

no corresponding parameters for optimizing key parameters for CO control. Thus, this technology

would not be effective for the Petmin EAF, and is eliminated from consideration.


DEC Controls is the only feasible control option identified.


DEC Controls is the most effective option.

A review of permits issued for EAFs in the last 10 years is summarized in the following table.

Summary of RBLC Search for CO Controls on EAFs

Facility

Name/State

Issuance

Date

Control Technology Determined to be

BACT

CO

Limit

(lbs/ton)

Annual CO

Emissions Scaled to

Petmin Production

Rate (tpy)

Big River Steel

LLC/AR 09/13

Scrap management plan, good operating

practices 2.0 478

ERMS/ CO 11/11 Process controls 2.0 478

OSCEOLA

Steel/GA 12/10

DEC system including scrap management

program, oxyburners, oxygen lancing,

increased water-cooled duct length,

dampers and actuators, dilution air cross

connection to allow canopy air to be mixed

with DEC emissions, fourth hole (airgap)

2.0 478

NUCOR

JEWETT/TX 10/13 Good combustion practice 2.27 542

FAIRCREST

Steel/OH 10/13

DEC system with adjustable air gap,

elbow, and water-cooled ductwork for

enhanced burnout of CO

3.50 836

Charter Steel/WI 06/16 Not listed 3.50 836

TPCO/TX 04/10 Not listed 3.99 1,829.82

HARRISON

Steel/OH 12/10 Not listed 4.80 1,147




- 82 -

The right-hand column indicates what the ton-per-year limitation would be for Petmin, if it had

the pound CO per ton metal produced allowable value in the listed permit.

AP-42 lists several emission factors for CO from EAFs that range from 1.0 to more than 3.8 pounds

CO per ton. This wide variability is not explained, but is likely a result of factors that include:

• Amount of carbon in the raw melt, and/or charged;

• The amount of arc drawn into the EAF

• The effectiveness of the slag practice

• The number of charges

• EAF temperature, and

• Production rate

Step 5: Select BACT

The use of a DRI fed EAF in hot metal production is a novel technology. Proprietary emissions

modeling by the design engineering team indicate that the value corresponding to 1.8 pounds per

ton of product as indicated in the AP-42 rating “B” can be achieved with the use of a DEC Control

system. This emission limitation clearly indicates satisfactory environmental performance when

compared to the “similar” sources listed above.

Petmin also suggests a term in the PTI that the DEC system must be operating at all times the EAF

is in operation.

4.6.2 Other Emissions Units

The remaining emissions units are either:

• Combustion units using natural gas as fuel. CO formation is minimized following good

combustion practices. No add-on CO controls are in the realm of feasibility for these clean-

burning units; or

• Emergency use only or black startup engines with operation limited to 100 hours per year

for maintenance and readiness testing, as well as their emergency or black-start only

functions. Federal rules limit operation time and corresponding emissions of CO to

minimal annual values. Controls beyond these limitations would not be feasible.




- 83 -

5.0 Ambient Air Quality Analysis

As described in previous sections of this report, Petmin is proposing to construct and operate a

new facility located in Ashtabula County. The new facility will be located within the Kinder

Morgan (KM) Pinney Dock facility at 1149 East Fifth Street, Ashtabula, Ohio.

The Petmin facility is a new major stationary source. The Ashtabula area is presently designated

as in attainment for all pollutants. The potential emissions of NOx and CO associated with the

proposed new facility are greater than the major source threshold (100 tpy for a source in the 28

categories listed) in Prevention of Significant Deterioration (PSD) rules; therefore, the Project will

be subject to PSD requirements.

The potential emissions of SO2 and VOC will be below the significance thresholds listed in the

PSD rules and Ohio EPA Modeling Guidance document. PM10, and PM2.5 exceed the modeling

significance thresholds. These pollutants will be modeled in addition to NOx and CO. No air toxic

(listed in OAC §3745-114) exceeds the modeling threshold.

As part of the PSD permitting process, an air quality impact analysis is required to demonstrate

that the proposed emissions from the project (i.e., that includes the above-listed emissions units),

in conjunction with emissions from other existing off-site sources, will not cause any exceedances

of applicable National Ambient Air Quality Standards (NAAQS) for NO2, CO, PM10, and PM2.5.

Further, the study will demonstrate that the emissions associated with this project will not exceed

the available PSD increment.

The modeling study was performed by AYER Quality Engineering LLC (AYER) of Cincinnati,

OH. Joseph N. Hollowell, P.E. performed the analysis, with review by Matthew Ayer, Senior

Consultant.

The modeling study was performed following the procedures in Guideline for Air Quality Models15

(Supplement W to 40 CFR §51) and Engineering Guide #69: Air Dispersion Modeling Guidance,16

Ohio EPA Division of Air Pollution Control, Revised November 14, 2018. As described in more

detail in subsequent sections to this report, the U.S. EPA AERMOD model was selected for this

study.

5.1 Site Area Characteristics

Certain site area characteristics, including wind climatology, land use classification, and terrain

analysis, are necessary components of the dispersion modeling analysis.




- 84 -

5.1.1 Wind Climatology

The Petmin MPI manufacturing plant will be located in Ashtabula (Ashtabula County), in

northeastern Ohio, on the Pinney Dock Terminal property on Lake Erie in the City of Ashtabula.

The facility is approximately 48 miles (77 km) northeast of downtown Cleveland, and

approximately 34 miles (54 km) southwest of Erie, PA International Airport. Terrain in the

vicinity is flat to rolling, with elevations ranging from 525 feet (160 meters) to 650 feet (198

meters) within 5 km of the facility.

Figure 5 is a five-year composite wind rose generated from meteorological data provided by Ohio

EPA.17 The figure graphically summarizes the frequencies of wind speeds and directions.

Figure 7: Wind Rose Plot of 2013 to 2017 Meteorological Data from the

Ashtabula, OH Airport and Upper Air Data from Buffalo, NY

5.1.2 Land Use Classification

Air dispersion models require that the land use in the vicinity of the source be classified as either

URBAN or RURAL so as to assign the proper dispersion coefficients. In modeling studies, a

binary typing scheme can be used to delineate urban versus rural land use in the vicinity of the

project. The predominant land use is then selected in the model (RURAL or URBAN). In this

particular case, a detailed analysis was not necessary. RURAL dispersion coefficients were




- 85 -

selected because it is immediately apparent that the land surrounding the site is predominantly

undeveloped rural areas and Lake Erie. Figure 8 shows a Google Earth image18 of the where the

facility will be located.

Figure 8: Google Earth Image of the Location of the Proposed Petmin Facility

5.1.3 Terrain Analysis

National Elevation Data (NED) digital terrain data were used in the model, obtained from the

United States Geological Survey (USGS) website as 1/3 arc second series topographic map data.

AERMAP is used with the NED digital terrain data files to establish receptor coordinates,

elevations, and “hill heights” (height scale) for model input.

5.2 Pre-Application Air Quality Monitoring

Ohio EPA collects continuous ambient monitoring data and has concurred that it is representative

for the site and fulfills the pre-application monitoring requirement for this PSD analysis. The Ohio

EPA has discretionary authority to exempt an applicant from performing onsite meteorological

monitoring if either (1) the predicted ambient impact caused by the modification, i.e., the highest




- 86 -

modeled concentration for the applicable averaging time, or (2) the existing ambient pollutant

concentrations are less than the prescribed significant monitoring value (Table 18). As shown in

Section 5.5 and Table 24, the modeled annual NO2 and 8-hour CO concentrations from Petmin are

both below the Significant Monitoring Concentrations. The NO2 data collection site (Monitor ID

Number 39-035-0073) is located in the city of Cleveland approximately 77 kilometers (48 miles)

distant from Petmin. The ozone data collection site (Monitor ID Number 39-007-1001) is located

in the city of Conneaut in Ashtabula County at the JQ Conneaut Water Plant approximately 19

kilometers (12 miles) away.

Significant Monitoring Concentrations

Pollutant Significant Monitoring Concentration

Averaging Time Concentration (g/m3)

Nitrogen dioxide Annual 14

Carbon monoxide 8-hour 575

PM10 24-hour 10

5.3 Ambient Air Quality Data

Background concentrations were obtained from Ohio EPA.19 The monitors and concentrations

used (except for NO2 and ozone as described below) are shown in Table 19.

Background Concentrations

Pollutant

Background Concentration Monitor Location and ID

Averaging

Time

Concentration

(g/m3)

PM10 24-Hour 36.3 Lake, Fairport Harbor water treatment plant (39-

085-1001)

PM2.5 Annual 7.8 Trumbull, Warren Water Dept. (39-155-0014)

PM2.5 24-Hour 18 Trumbull (39-155-0014)

The NO2 data is from the closest monitor in the area, near Cleveland, approximately 77 kilometers

(48 miles) from the facility. The ozone data is also from a nearby monitor, located in the city of

Conneaut in Ashtabula County at the JQ Conneaut Water Plant. Ozone is a regional pollutant.

This monitor is not only the closest to the proposed location of the facility, but is also highly

representative of the site, as it is only approximately 19 kilometers (12 miles) away. Five years

(2013-2017) of ozone data were obtained. However, since the Ashtabula monitor does not record

ozone during January-March, and November-December, data for these missing months were




- 87 -

obtained from a nearby monitor in Cleveland (GT Craig NCore, Monitor ID Number 39-035-

0060).

Missing data were substituted using the averages of the surrounding hourly values for gaps of less

than five hours. Gaps of greater than five hours were filled in using one of two ways. For the five

ozone data gaps greater than five hours, the following substitutions were used:

• 85 hours in 2015 from August 21 at 15:00 to August 25 at 3:00. These values were filled

in using the data from the same hours from 2016.

• 22 hours in 2017 from March 8 at 14:00 to March 9 at 11:00. These values were filled in

using the data from the same hours from 2016.

• 13 hours in 2017 from March 23 at 1:00 to March 23 at 13:00. These values were filled

in using the data from the same hours from 2016.

• 19 hours in 2017 from November 8 at 21:00 to November 9 at 15:00. These values were

filled in using the data from the same hours from 2016.

• 27 hours in 2017 from November 27 at 10:00 to November 28 at 12:00. These values were

filled in using the data from the same hours from 2016.

For each NO2 gap greater than five hours, the 98th percentile value for the particular season-hour

was used to fill in the missing data. For example, if Hour 6 in March were missing, then the 98th

percentile concentration of Hour 6 for all days in the Spring season was used. Continuing this

example, if there were 92 Hour 6 values in the Spring season (March through May) for each year,

the 98th percentile value was calculated from these 92 values using the Microsoft Excel “percentile

function.” This procedure is followed for each of the four years. Then, the average of these four

98th-percentile values is input to the model to fill the gap.

Data gap-filling and calculation of season-hour concentrations were performed in Microsoft

Excel.™ These files have previously been provided to Ohio EPA as supplements to the air

dispersion modeling study protocol.

AERMOD20 has several options for inputting background concentration data. For ozone, the

hourly concentration data file was used. AERMOD then merges (or “pairs”) the hourly data with

the meteorological data.

For NO2 background data, EPA guidance stipulates the use of average concentration data, with

several averaging options. For this study, the hourly background data for NO2 was converted to a

season-hour-average table for use in the model. In this approach, the hourly data is converted by:




- 88 -

• Calculating the 98th percentile value for each hour of the day for each season of each year,

and then

• Averaging over the four-year period, 2014-2017, of data to obtain a single value for each

hour in each season.

The season-hour values used are presented in Table 20. The raw background data and spreadsheets

showing this computation can be provided upon request.

Season-Hour NO2 Background Data

Hour of Day Concentration (ppb)

Winter Spring Summer Fall

1 32.47 35.06 17.14 22.15

2 33.87 33.32 16.98 23.05

3 34.30 35.13 16.18 22.29

4 33.93 33.58 15.99 24.10

5 33.78 33.10 19.12 25.59

6 34.91 36.82 23.52 27.60

7 36.08 37.57 25.56 33.59

8 36.24 34.76 23.79 32.80

9 34.08 28.14 21.87 31.00

10 28.76 25.79 19.46 25.46

11 26.21 24.52 19.02 22.12

12 23.67 20.94 18.13 19.80

13 21.72 19.28 18.73 19.19

14 22.08 19.83 18.20 18.95

15 22.87 20.51 18.86 20.85

16 24.17 21.65 18.44 22.45

17 25.37 24.05 18.62 24.44

18 27.52 24.35 16.93 25.69

19 28.37 24.98 17.82 28.88

20 29.02 28.26 17.94 26.39

21 34.06 27.18 18.85 25.43

22 32.22 25.97 15.44 25.24

23 32.12 31.49 16.20 25.83

24 32.73 32.76 15.75 25.54

5.4 Air Dispersion Modeling Parameters

This section provides details concerning the selection of the model and input parameters. Air

dispersion modeling was used to perform the preliminary analysis in order to determine whether




- 89 -

there is a significant impact to air quality. If the modeling indicates a significant impact, the impact

area will be determined. The impact area is defined as the circular area with a radius extending

from the source to (1) the most distant point where approved dispersion modeling predicts a

significant ambient impact will occur, or (2) a modeling receptor distance of 50 km, whichever is

less. If the preliminary analysis indicates a significant increase may occur, then a full impacts

analysis is required. The full impact analysis requires modeling all sources within 50 km that may

impact ambient concentrations within the significant impact area.

5.4.1 AERMOD Modeling System

The AERMOD computer model was selected for the air dispersion modeling study. The American

Meteorological Society/Environmental Protection Agency Regulatory Model Improvement

Committee (AERMIC) was formed to introduce state-of-the-art modeling concepts into the EPA's

air quality models. Through AERMIC, the AERMOD modeling system was selected to

incorporate air dispersion based on planetary boundary layer turbulence structure and scaling

concepts, including treatment of both surface and elevated sources, in both simple and complex

terrain.

AERMOD is a steady-state plume dispersion model for assessment of pollutant concentrations

from a variety of sources. AERMOD simulates transport and dispersion from multiple point, area,

or volume sources based on an up-to-date characterization of the atmospheric boundary layer.

Sources may be located in rural or urban areas, and receptors may be located in simple or complex

terrain. AERMOD accounts for building wake effects (i.e., plume downwash) based on the Plume

Rise Model Enhancements (PRIME) building downwash algorithms. The model employs hourly

sequential preprocessed meteorological data to estimate concentrations for a broad range of

averaging periods, including one hour, one year, and multiple years. AERMOD is designed to

operate in concert with two pre-processor codes: AERMET, which processes meteorological data

for input to AERMOD, and AERMAP, which processes terrain elevation data and generates

receptor information for input to AERMOD.

AERMOD, contained in the commercially available Beeline Software package (BEEST),21 was

used for this analysis. This software is updated with the most recent version of the AERMOD

dispersion model and input data preprocessors, as follows:

• AERMOD version 19191

• AERMAP version 18081

• AERMET version 9191

• AERMINUTE version 15272

• AERSURFACE version 13016




- 90 -

The following options were used:

• Regulatory default options with the Tier 3 OLM option selected for NO2

• Terrain heights with intermediate terrain processing

• Building downwash

5.4.2 Receptor Grid

The entire KM Pinney Dock site is restricted from public access, by a combination of Lake Erie

docks, with the land perimeter entirely fenced, with a security gate at the main entrance. This

fenceline is used for the modeling study. A receptor grid was established, with receptors placed

along the property line spaced 25 meters apart. The NED data was obtained in GeoTIFF22 files

that make up a 50 km box surrounding the Petmin facility. Receptors for modeling against the

Significant Impact Levels (SIL) were then established onto a rectangular (Cartesian) grid as

follows:

• Spaced 50 meters apart outside the property line, out to a distance of 1 km from the property

line;

• Spaced 100 meters apart up to a distance of 2 km from the property line;

• Spaced 250 meters apart up to a distance of 5 km from the property line;

• Spaced 500 meters apart up to a distance of 10 km from the property line; and

• Spaced 1,000 meters apart up to a distance of 40 km from the property line.

AERMAP was used with the NED digital terrain data files to establish receptor coordinates,

elevations, and “hill heights” (height scale) for model input. Figure 9 shows this receptor grid.




- 91 -

Figure 9: Petmin Facility with Receptor Grid

5.4.3 Facility Building Data for Downwash Analysis

A Google Earth image was input into the BEEST software, which allows for the fenceline and

building outlines to be traced. Building tiers were also traced, and the height of each tier was input

from Petmin-provided data. Stacks were located, and their heights were input from Petmin data.

The model’s graphical output shows the outlines of the buildings and tiers as lines, and the stack

location as points, thus assuring that the facility data is correctly aligned to the satellite image and

the Cartesian grid.

A layout of the facility is shown in Figure 10.




- 92 -

Figure 10: Petmin Facility Layout

5.4.4 Modeling Input – Emission Source Data

Hourly emission rates (annual emissions are based on 8,760 hours per year of operation) for the

modeled pollutants (NOx, PM10, PM2.5, CO) and the stack parameters for the emissions units were

obtained from the engineering design firm contracted by Petmin and cross-checked by AYER.

Details concerning emission rates are shown in the emission calculations section of the report and

discussed in the BACT section. These values are shown in Table 21. Each point source emissions

unit has one stack, with unobstructed vertical flow. The transfer points are modeled as volume

sources with the parameters and emission rates shown in Table 22.




- 93 -

Point Source Emissions Units Modeled Parameters

Parameter Process Gas

Heaterb Flarec

EAF/

Casting

Baghouse

Startup

Boiler

Emer.

Gen.d

Blackstart

Gen.

HP Fire

Pump

LP Fire

Pump

Source OEPA ID P001 P008 P002/3/4,

P901

B001 P005/6 P007 TBD TBD

Stack Height (ft) 200e 167.54 108.6 50 11.1 4.5 11.1 11.1

Stack Equivalent Diameter (ft) 5 5.66 15.09 2 1 0.25 0.5 0.42

Stack Temperature (ºF) 392 3,310 205.7 370 829 1,065 842 986

Stack Velocity (ft/sec) 74.51 65.62 47.08 48.81 392.2 244.5 158.5 145.3

Stack Flowrate (acfm) 87,780 99,163 505,229 9,200 18,481 720 1,867 1,189

Heat Release, Q (MMBtu/hr) N/A 96.68 N/A N/A N/A N/A N/A N/A

GEP Stack Height (ft)a 528.8 403.6 404.2 513.0 404.5 404.5 303.9 303.9

Modeled NOx Rate (lbs/hour) 18.88 0.45 90.55 0.635 0.039 1.19e-3 0.022 0.017

Modeled PM10 Rate (lbs/hour) 1.63 0.049 12.43 0.11 1.73 e-3 5.96e-5 1.20e-3 8.90e-4

Modeled PM2.5 Rate (lbs/hour) 1.63 0.049 9.94 0.11 1.73e-3 5.96e-5 1.20e-3 8.90e-4

Modeled CO Rate (lbs/hour) 11.17 2.05 109.78 1.249 0.205 0.015 0.020 0.016

a) This value is provided to demonstrate that the actual stack heights are below GEP stack height (calculated as specified in Appendix W to Part 51 – Guideline

on Air Quality Models).

b) The thermal oxidizer vents through the process gas heater so its emissions are included here.

c) The flare parameters shown above (height-equivalent, diameter-equivalent, and flowrate) are the actual parameters. The modeled parameters, height-

equivalent and diameter-equivalent, are calculated using the equations from Ohio EPA Engineering Guide #69 and shown in Table 22.

d) The values shown are for each emergency generator (both are identical).

e) This is the minimum stack height for the Process Gas Heater. The design has not been finalized at this time. A taller stack will improve air dispersion

modeling results, therefore the overall conclusion for the modeling will not change with a taller stack if Petmin decides to build the stack taller than 200 feet.




- 94 -

Volume Source Emissions Units Modeled Parameters

Model

ID Source Description

Release

Height

(ft)

Init. Horiz.

Dimension

(ft)

Init. Vert.

Dimension

(ft)

PM10

(lbs/hr)

PM10

(tpy)

PM2.5

(lbs/hr)

PM2.5

(tpy)

TP1 Loader A to Receiving Bin A 18.25 3.609 4.232 0.010674 0.04675 0.003016 0.013212

TP2 Loader B to Receiving Bin B 18.25 3.609 4.232 0.010674 0.04675 0.003016 0.013212

TP3 Receiving Bin A to Conveyor 1A 14.25 1.280 3.314 0.010674 0.04675 0.003016 0.013212

TP4 Receiving Bin B to Conveyor 1B 14.25 1.280 3.314 0.010674 0.04675 0.003016 0.013212

TP5 Conveyor 1A to Conveyor 2-Main Conveyor 9.5 0.755 2.198 0.010674 0.04675 0.003016 0.013212

TP6 Conveyor 1B to Conveyor 2-Main Conveyor 3.25 0.755 2.198 0.010674 0.04675 0.003016 0.013212

TP7 Conveyor 2 to Feed Bin 92.08 3.839 42.815 0.021347 0.0935 0.006033 0.026424

TP8 Feed Bin to Conveyor 3 66.17 1.280 30.774 0.021347 0.0935 0.006033 0.026424

TP9 Conveyor 3 to Screening Bin 64.17 0.919 29.856 0.021347 0.0935 0.006033 0.026424




- 95 -

The equations and calculations for the modeled flare parameters are shown below. Per Ohio EPA

Engineering Guide #69, the temperature for flares should be set to 1,273 K (1,831.7ºF) and the

velocity should be set at 20 meters/sec (65.62 ft/sec). The modeled flare parameters are shown in

Table 23.

𝐻𝑒𝑓𝑓 = 𝐻𝑎𝑐𝑡𝑢𝑎𝑙 + 0.945 𝑄𝑡0.478 = 30 𝑚 + 0.945 (6.61 𝑀𝑀 𝐵𝑇𝑈/ℎ𝑟) = 32.33 𝑚

𝑑𝑒𝑓𝑓 = 0.261𝑄𝑡 = 0.261 (6.61 𝑀𝑀 𝐵𝑇𝑈/ℎ𝑟) = 0.508 𝑚

Flare Modeling Parameters

Parameter Flare

Modeled NOx Rate (lbs/hour) 0.45

Modeled PM10 Rate (lbs/hour) 0.049

Modeled PM2.5 Rate (lbs/hour) 0.049

Modeled CO Rate (lbs/hour) 2.05

Heat Release, Qt (MMBtu/hr) 6.61

Modeled Stack Height (ft) 106.07

Modeled Stack Diameter (ft) 5.66

Modeled Stack Temperature (ºF) 1,831.7

Modeled Stack Velocity (ft/sec) 65.62

Modeled Stack Flowrate (acfm) 99,062.7

GEP Stack Height (ft) 403.6

5.4.5 Option Selected for One-Hour NO2 Modeling

The Tier 3 Ozone Limiting Method (OLM) was selected in AERMOD for modeling NOx

emissions. This method uses hourly ozone background data, aligned with the corresponding

season-hour NO2 background data.

As a result of change to 40 CFR Part 51, Appendix W - Guideline on Air Quality Models, the

OLM method is now available as a regulatory-default option within the EPA-preferred AERMOD

dispersion model. However, it must still be approved by U.S. EPA; therefore, a further justification

of the use of the OLM is included below.

In order to justify the use of OLM for the purpose of determining compliance with the Federal 1-

hour NO2 standard in this modeling study, the five requirements noted in Section 3.2.2.(e) are met

as follows:




- 96 -

3.2.2 (e)(i). The model has received a scientific peer review;

As noted in a memorandum from EPA23 “Since AERMOD is the preferred model for dispersion

for a wide range of applications, the focus of the alternative model demonstration for use of the

OLM/PVMRM options within AERMOD is on the treatment of NOx chemistry within the model

and does not need to address basic dispersion algorithms within AERMOD.” Therefore, the

following will address the basic chemistry of each of the non-regulatory options.

Basic OLM Chemistry:

To provide some background, the following is a simplified explanation of the basic chemistry

relevant to the OLM. First, the relatively high temperatures typical of most combustion sources

promote the formation of NO2 by the following thermal reaction:

2 NO + O2 → 2 NO2

For “in-stack” formation of NO2, the OLM assumes a default 10% of the NOx in the exhaust is

converted to NO2 by this reaction, and no further conversion by this reaction occurs once the

exhaust leaves the stack. The remaining percentage of the NOx emissions is assumed to be nitric

oxide (NO).

As the exhaust leaves the stack and mixes with the ambient air, the NO reacts with ambient ozone

(O3) to form NO2 and molecular oxygen (O2):

NO + O3 → NO2 + O2 Oxidation of NO by ambient O3

The OLM assumes that at any given receptor location, the amount of NO that is converted to NO2

by this reaction is proportional to the ambient O3 concentration. If the O3 concentration is less

than the NO concentration, the amount of NO2 formed by this reaction is limited. If the O3

concentration is greater than or equal to the NO concentration, all of the NO is assumed to be

converted to NO2.

In the presence of radiation from the sun, ambient NO2 can be destroyed:

NO2 + sunlight → NO + O Photo-dissociation of NO2

As a conservative assumption, the OLM ignores this reaction.




- 97 -

Another reaction that can form NO2 in the atmosphere is the reaction of NO with reactive

hydrocarbons (HC):

NO + HC → NO2 + HC' Oxidation of NO by reactive HC

The OLM also ignores this reaction. This may be a non-conservative assumption with respect to

NO2 formation in urban/industrial areas with relatively large amounts of reactive HC emissions.

The chemistry for the OLM model has received scientific peer review as noted in "Sensitivity

Analysis of PVMRM and OLM in AERMOD".24 Also, the following documents posted on EPA’s

Support Center for Regulatory Air Modeling (SCRAM) website indicate that both models (OLM

and PVMRM) have been peer-reviewed:

Cole, H.S. and J.E. Summerhays, 1979. A Review of Techniques Available for Estimation

of Short-Term NO2 Concentrations, Journal of the Air Pollution Control Association,

29(8):812-817.

Hanrahan, P.L., 1999a. The Plume Volume Molar Ratio Method for Determining

NO2/NOx Ratios in Modeling - Part I: Methodology. Journal of the Air & Waste

Management Association, 49, 1324-1331.

Hanrahan, P.L., 1999b. The Plume Volume Molar Ratio Method for Determining NO

/NOx Ratios in Modeling - Part II: Evaluation Studies. Journal of the Air & Waste

Management Association, 49, 1332-1338.

MACTEC, 2005. Evaluation of Bias in AERMOD-PVMRM. Final Report, Alaska DEC

Contract No. 18-9010-12. MACTEC Federal Programs, Inc., RTP, NC.

These documents demonstrate that the OLM model has received appropriate scientific peer review.

3.2.2 (e)(ii). The model can be demonstrated to be applicable to the problem on a

theoretical basis;

The OLM model has been reviewed and the chemistry has been widely accepted by EPA as being

appropriate for addressing the formation of NO2 and the calculation of NO2 concentration at

receptors downwind. Additionally, the "Sensitivity Analysis of PVMRM and OLM in AERMOD"

report would indicate OLM/PVMRM provides a better estimation of the NO2 impacts compared

to other screening options.




- 98 -

The requirement of 3.2.2 (e)(ii) is: Can the model (OLM or PVMRM) be demonstrated to be

applicable to the problem on a theoretical basis? As noted in the document entitled “Sensitivity

Analysis of PVMRM and OLM in AERMOD”, “This report presents results of a sensitivity

analysis of the PVMRM and OLM options for NOx to NO2 conversion in the AERMOD dispersion

model. Several single source scenarios were examined as well as a multiple-source scenario. The

average conversion ratios of NO2/NOx for the PVMRM option tend to be lower than for the OLM

option and for the Tier 2 option or the Ambient Ratio Method (ARM2) which uses 0.5 as the

default minimum ratio and 0.9 as the default maximum ratio for both the annual average and the

hourly average. The sensitivity of the PVMRM and OLM options to emission rate, source

parameters and modeling options appear to be reasonable and are as expected based on the

formulations of the two methods. For a given NOx emission rate and ambient ozone concentration,

the NO2/NOx conversion ratio for PVMRM is primarily controlled by the volume of the plume,

whereas the conversion ratio for OLM is primarily controlled by the ground-level NOx

concentration.

Overall the PVMRM option appears to provide a more realistic treatment of the conversion of NOx

to NO2 as a function of distance downwind from the source than OLM or the other NO2 screening

options (Hanrahan, 1999a; Hanrahan, 1999b). No anomalous behavior of the PVMRM or OLM

options was identified as a result of these sensitivity tests.”

Based on this report for both OLM/PVMRM it appears to be applicable to the problem at hand,

i.e., the prediction of NO2 formation. As noted by the author, these options provide a more accurate

estimation of NO2 impacts compared to other screening options.

3.2.2 (e)(iii). The data bases which are necessary to perform the analysis are available

and adequate;

Five years (2013-2017) of both hourly processed meteorological data (Erie, PA/Buffalo, NY) and

concurrent hourly ozone monitoring data are available for this modeling application. Hourly ozone

concentrations from the Ashtabula monitoring station (19 km from the facility) will be input to

AERMOD for each year modeled (2013-2017). These data sets have been processed into

AERMOD-ready formats and are adequate for use with AERMOD-OLM.

The in-stack ratio that will be used for the sources in the model is the default ratio of 0.5 for all

sources, including the offsite sources.

3.2.2 (e)(iv). Appropriate performance evaluations of the model have shown that the

model is not biased toward underestimates;




- 99 -

It has been shown in the sensitivity analysis (discussion on item 3.2.2 (e)(ii) above), that the OLM

provides similar but more conservative results than PVMRM. Therefore, it is reasonable to assume

that the OLM would also provide an unbiased estimate of concentration. (A formal, agency-

reviewed assessment of bias has not yet been published for the OLM option.)

3.2.2 (e)(v). A protocol on methods and procedures to be followed has been established.

The methods and procedures for conducting an assessment for determining compliance with the

federal 1-hour NAAQS are contained within this document. Specific OLM inputs are discussed

here. The in-stack ratio of NO2/NOx for all of the sources will be set to the default value of 0.5.

The default value of 0.9 will be used for the ambient equilibrium ratio in OLM.

5.5 Modeling Results – SIL

The model was executed for the Petmin emissions units. As shown in Table 24, the analysis

indicated that the predicted offsite impacts for just these emissions units were below the Significant

Impact Levels (SIL) for CO (1-hour and 8-hour) and annual PM10. NO2 (1-hour and annual), PM2.5

(24-hour and annual), and PM10 (24-hour) were all above the respective SIL. Since these impacts

exceeded the modeling significance level, i.e., the project results in a significant impact for these

pollutants. Therefore, a full impacts (a.k.a. cumulative or interactive) analysis is required to

demonstrate compliance with the 1-hour and annual NO2, the 24-hour and annual PM2.5, and the

24-hour PM10 NAAQS, including secondary formation of PM2.5.

Significant Impact Level Modeling Analysis Results

Pollutant Averaging Period SIL (μg/m3) Maximum Concentration

(μg/m3) Below SIL?

NO2 1-hour 7.5 147.8 No

NO2 Annual 1 2.17 No

PM10 24-hour 5 4.41 No

PM10 Annual 1 0.33 Yes

PM2.5 24-hour 1.2 4.29 No

PM2.5 Annual 0.2 0.23 No

CO 1-hour 2,000 159.92 Yes

CO 8-hour 500 83.58 Yes




- 100 -

The full impacts analysis for all of the pollutants and averaging periods was performed using the

40 km grid developed as described in Section 5.4.2. Secondary impacts of PM2.5 from NOx and

SO2 will be addressed later in the report in Section 5.10.

5.6 Full Impacts Analysis

A cumulative analysis for NO2 (1-hour and annual), PM2.5 (24-hour and annual), and PM10 (24-

hour only) was performed in accordance with Ohio EPA Modeling Guidance:

If maximum predicted impacts of a pollutant due to proposed emission increases from the

existing facility are at or above applicable SILs and there are nearby sources of that pollutant

that could significantly interact with emissions from the facility’s proposed modification, the

predicted air quality impacts from the existing facility as modified, along with predicted air

quality impacts from nearby significantly interacting sources, should be added to

representative background levels and compared to applicable NAAQS. If the results are below

applicable NAAQS, the facility, as proposed to be modified, is considered to be in compliance

with NAAQS for that pollutant.

5.6.1 Facility Emissions

The modeled emission rates for the new Petmin facility are shown in Tables 21 and 22 above. The

emission values are based on each emissions unit’s potential to emit (PTE) for both hourly and

annual emissions.

5.6.2 Nearby Significantly Interacting Sources

Ohio EPA was consulted to obtain the data for nearby sources relevant to the significant impact

area. A database of sources was obtained from the Ohio EPA25 and was filtered to extract those

sources located in Ashtabula, Geauga, Lake, and Trumbull Counties. The database included

emission rates and stack parameters for most of the sources.

For those sources which will be included in the model that had missing information, an additional

request was made to Ohio EPA and the missing information was provided via email26.

As confirmed with Ohio EPA, the only PSD source in the vicinity of Ashtabula County is the

Cristal USA Inc., Ashtabula Complex Plant 1 (Facility ID: 0204010200). As a PSD source, the

emissions units at this facility were included in the model.




- 101 -

All sources within the Significant Impact Areas (SIA) with potential allowable emissions rates

greater than the PSD significant emission rates (40 tpy NOx, 10 tpy PM2.5) must also be included.

All sources that could potentially interact with the facility were included based on the ‘20D’

approach. Under this approach, the excluded sources are those whose potential allowable

emissions in tpy are less than 20 times the distance between the closest of the modeled units at

Petmin, in kilometers. Any source located greater than 25 km from Petmin were automatically

excluded from the model per The Guideline on Air Quality Models (Appendix W of 40 CFR 51).

Sources modeled are shown in Table 25.




- 102 -

Other Sources Included in the Model

Facility Name Facility ID OEPA Source ID Pollutant

Cristal USA Inc., Ashtabula Complex Plant 1 0204010200 P001 NOx, PM2.5




Cristal USA Inc., Ashtabula Complex Plant 1 0204010200 P008 PM2.5

Cristal USA Inc., Ashtabula Complex Plant 1 0204010200 P010-Preheater NOx, PM2.5

Cristal USA Inc., Ashtabula Complex Plant 1 0204010200 P010-Vaporizer NOx, PM2.5

Cristal USA Inc., Ashtabula Complex Plant 1 0204010200 P011-Preheater NOx, PM2.5




Cristal USA Inc., Ashtabula Complex Plant 2 0204010193 B013 NOx, PM2.5





Cristal USA Inc., Ashtabula Complex Plant 2 0204010193 P001-BA-815 NOx, PM2.5

Cristal USA Inc., Ashtabula Complex Plant 2 0204010193 P001-BA-816 NOx, PM2.5














Koski Construction Co 0204010008 P902 NOx

Painesville Municipal Electric Plant 0243110008 B001 NOx



Carmeuse Lime, Inc - Grand River Operations 0243030257 P001 NOx

Carmeuse Lime, Inc - Grand River Operations 0243030257 P002 NOx

USA Waste Geneva Landfill, Inc. 0204030303 P001-P005 NOx

Ashta Chemicals Inc. 0204010056 B001 PM2.5

Ashta Chemicals Inc. 0204010056 B002 PM2.5

MFG, Composite Systems 0204010230 B003 PM2.5

MFG, Composite Systems 0204010230 B004 PM2.5




- 103 -

The stack parameters input into the model for the other sources are shown in Table 26.

Stack Parameters for Other Sources Modeled

Model ID Source Description

Stack

Height

(ft)

Temp.

(°F)

Exit

Velocity

(ft/s)

Stack

Diameter

(ft)

NOx

(lbs/hr)

PM2.5

(lbs/hr)

C1P001 Cristal Plant 1 P001 212 95 29.18 2 0.212 0.17

C1P002 Cristal Plant 1 P002 60 245 51.44 3.8 0.686 0.33



C1P008 Cristal Plant 1 P008 74.4 86 37.71 2.2 0 0.46

C1P010P Cristal Plant 1 P010 Preheater 100 350 4.72 3 1.43 0.14

C1P010V Cristal Plant 1 P010 Vaporizer 100 350 6.79 2.5 0.69 0.29

C1P011P Cristal Plant 1 P011 Preheater 45.8 350 2.89 3.83 0.68 0.14

C1P011V Cristal Plant 1 P011 Vaporizer 62.4 350 10.61 2 1.42 0.29

C1P901 Cristal Plant 1 P901 30 70 117.89 0.3 0 0.65


C2B013 Cristal Plant 2 B013 50 300 72.95 4 12.53 0.85





C2P001P815 Cristal Plant 2 P001 Heater 815 131 700 9.43 3 2.13 0.19

C2P001P816 Cristal Plant 2 P001 Heater 816 130 700 4.72 3 0.21 0.044

C2P001V Cristal Plant 2 P001 Vaporizer 87 100 15.95 1.3 3.86 0.336

C2P012 Cristal Plant 2 P012 63 200 58.95 3 2.268 4









C2P903 Cristal Plant 2 P903 70 68 1.41 3 0 2.01



KP902 Koski Construction P902 43.25 199 50.74 4.663 32.4 0

PV_B001 Painesville B001 157 331 37.62 8 114.92 0

PV_B003 Painesville B003 171 350 25.94 11.33 114.92 0

PV_B004 Painesville B004 171 350 25.94 11.33 114.92 0

CL_P001 Carmeuse P001 146 432 54.00 10.77 147 0

CL_P002 Carmeuse P002 146 432 54.00 10.77 147 0




- 104 -

Stack Parameters for Other Sources Modeled

Model ID Source Description

Stack

Height

(ft)

Temp.

(°F)

Exit

Velocity

(ft/s)

Stack

Diameter

(ft)

NOx

(lbs/hr)

PM2.5

(lbs/hr)

GN_GENS 5 Geneva Generators - 1 source 37.17 800 72.04 1.33 18.3 0

ACB001 Ashta Chemicals B001 89 104 16.31 1.1 0 0.018

ACB002 Ashta Chemicals B002 89 104 16.31 1.1 0 0.05

MCSB003 MFG, Composite B003 36.7 430 21.76 1.8 0 0.0699

MCSB004 MFG, Composite B004 36.7 430 21.76 1.8 0 0.0695

A list of those sources excluded using the ‘20D’ approach is shown in Table 27.

Sources Excluded from the Model

Facility Name Facility ID OEPA Source ID

Ashta Chemicals Inc. 0204010056 B013

Avery Dennison MFD, Bldg 7 0243111361 K002

Avery Dennison STD, Bldg 5 0243111362 K003




CFF of Avery Dennison 0243081207 K002

CFF of Avery Dennison 0243081207 K003

Continental Structural Plastics of Ohio LLC aka CSP OH 0204020245 B001

Continental Structural Plastics of Ohio LLC aka CSP OH 0204020245 B005

Continental Structural Plastics of Ohio LLC aka CSP OH 0204020245 F002

Continental Structural Plastics of Ohio LLC aka CSP OH 0204020245 K001

Detrex Corporation, Ashtabula Plant 0204010192 P200

Eastlake Substation 0243160009 B006

General Aluminum Manufacturing Co 0204020428 P019

General Aluminum Manufacturing Co 0204020428 P020

Grand River Asphalt Co 0243000491 P904

Kokosing Materials Inc Plant 520 0228002002 P901

Lake County Solid Waste Facility 0243111198 F002

Leather Resource of America, Inc. 0204020430 R001



Marking Films Div. of Avery Dennison Building #11 0243001188 K002

Marking Films Div. of Avery Dennison Building #11 0243001188 K003

Masco Cabinetry Middlefield LLC (KraftMaid Plant 3) 0204000360 K018








- 105 -






















Masco Cabinetry Middlefield LLC (KraftMaid Plant 3) 0204000360 P001

Masco Cabinetry Middlefield LLC (KraftMaid Plant 3) 0204000360 RTO

Nova Chemicals Inc 0243000012 B001

Nova Chemicals Inc 0243000012 B002

The Lubrizol Corporation 0243000024 B001





The Lubrizol Corporation 0243000024 N001

The Lubrizol Corporation 0243000024 P020



The Lubrizol Corporation - Wickliffe Facility 0243150025 B001



The Lubrizol Corporation - Wickliffe Facility 0243150025 P009

Trumbull Energy Center 0278112011 P001

Trumbull Energy Center 0278112011 P002

USA Waste Geneva Landfill, Inc. 0204030303 F001

USA Waste Geneva Landfill, Inc. 0204030303 F002

USA Waste Geneva Landfill, Inc. 0204030303 P901

Veitsch-Radex America - Saybrook Plant 0204000450 B001




- 106 -



Veitsch-Radex America - Saybrook Plant 0204000450 P001



Vesuvius USA Corp. - Conneaut Plant 0204020036 B005

Vesuvius USA Corp. - Conneaut Plant 0204020036 P055

5.7 Modeling Results – Full Impacts Analysis

Table 28 presents the results of the full impacts analysis for the 1-hour and annual NO2 modeling.

Table 29 presents results of the full impacts analysis for the 24-hour and annual PM2.5 modeling.

Table 30 presents the results of the full impacts analysis for the 24-hour PM10 modeling. As

shown in the tables, there is no exceedance of the 1-hour or annual NO2 NAAQS, the 24-hour or

annual PM2.5 NAAQS, or the 24-hour PM10 NAAQS as a result of the proposed Petmin Project.

NO2 Modeling Analysis Results

NAAQS

(μg/m3)

Maximum Concentration,

Including Background (μg/m3)

Meets

NAAQS?

Averaging

Period

188 102.4 Yes 1-hour

100 30.87 Yes Annual

PM2.5 Modeling Analysis Results

NAAQS (μg/m3)

Maximum

Concentration,

(μg/m3)

Background

Concentration

(μg/m3)

Meets

NAAQS?

Averaging

Period

35 6.62 18 Yes 24-hour

12 1.48 7.8 Yes Annual

PM10 Modeling Analysis Results

NAAQS (μg/m3)

Maximum

Concentration,

(μg/m3)

Background

Concentration

(μg/m3)

Meets

NAAQS?

Averaging

Period

150 8.98 34.6 Yes 24-hour




- 107 -

5.8 Modeling Results – PSD Increment

Next, modeling was performed to determine what fraction of the PSD increment the new emissions

units consume. The baseline concentration is defined for each pollutant (and relevant averaging

time) and it is the ambient concentration existing at the time of the trigger date (October 1, 2008

for NO2). Significant deterioration is said to occur when the amount of new pollution would

exceed the applicable PSD increment. This analysis is also necessary to verify that the new Petmin

facility and all of the PSD projects in a 50 km area around the facility do not consume the entire

increment (or more than the increment). The PSD increments for Class II Areas are shown in

Table 31. Ohio EPA has a policy that a new source cannot consume more than one-half of the

available PSD increment. There is no increment for the 1-hour NO2, 1-hour CO, or 8-hour CO

standards. The 24-hour PM2.5 and PM10 increments are the high-second-high values and the annual

increments are the maximum values over five years.

In addition to the Petmin facility, there is only one other facility which triggered PSD in the area,

the Cristal Complex Plant 1 with their final PSD permit issued on October 1, 2008. Therefore,

these are the only sources modeled against the PSD increment. Available increment varies in time

and space. The predicted maximum annual impact at the single highest-case receptor from the

combination of the Project and the existing PSD sources (in this case one facility) is shown in the

following table. Given the minimal impact relative to the increment, it was not necessary to

evaluate what specific fraction of available increment is consumed by the Project at this or any

other receptors.

PSD Increment Modeling Analysis Results

Pollutant Averaging

Period

PSD

Increment

(μg/m3)

One-Half PSD

Increment

(μg/m3)

Maximum

Concentration

(μg/m3)

Less Than One-

Half PSD

Increment?

NO2 Annual 25 12.5 2.21 Yes

PM2.5 Annual 4 2 0.41 Yes

PM2.5 24-hour 9 4.5 3.06 Yes

PM10 Annual 17 8.5 0.41 Yes

PM10 24-hour 30 15 3.53 Yes

5.9 Ozone Analysis

On April 30, 2019, U.S. EPA issued guidance on the development of Modeled Emission Rates for

Precursors (MERPs) as a Tier 1 demonstration tool for ozone and fine particulate matter (PM2.5)

under PSD.27 This guidance addresses the MERPs and how to use them to demonstrate that the

Project will not result in quantifiable ozone formation.




- 108 -

Under this guidance, the Petmin proposed NOx emissions increase of 484.57 tons per year is above

the lowest 8-hr ozone MERP value of 126 tons per year of any source modeled in the Ohio Valley

Climate Zone (see Table 4-1 of the MERPs guidance). Proposed VOC emissions are less than the

PSD Significant Emission Rate of 40 tons per year and, therefore, do not require evaluation. Since

the NOx emissions are above the MERP, ozone impacts would be expected to exceed the critical

ozone air quality threshold of 1 ppb. Therefore, further analysis is required.

Since the NOx emissions are above the lowest (most conservative) NOx MERP for 8-hour ozone

in the Ohio Valley Climate Zone of the US, Appendix A of the guidance was used to identify a

comparable hypothetical source that is considered most representative of this source.

The selection of the hypothetical site used for this analysis was straightforward. First, the two

nearest sites, Tuscarawas County, OH (Source 12 in Appendix A of the guidance) and Macomb,

MI (Source 11) are essentially identical in distance from the Project (180 and 186 km,

respectively).

A glance at regional topography and the wind roses for the sites indicate that both terrain and wind

pattern (speeds and directions) are extremely similar between the Macomb, MI site and the Project

site in Ashtabula, OH. The Macomb site is characteristic of the Great Lakes region in key

parameters that influence pollutant travel and ozone formation – wind speed, wind direction,

terrain and general climatology. In contrast, the Tuscarawas County site is located in an area of

complex terrain, with a strong wind frequency from the Southeast and dissimilar speed

distribution, presumably from river valley influences. Figure 11 shows a wind rose plot28 of

meteorological data from a site in Ashtabula County. Figure 12 shows a wind rose plot of

meteorological data from a site in Tuscarawas County, and Figure 13 shows a wind rose plot of

meteorological data from a site in Macomb County.




- 109 -

Figure 11: Wind Rose Plot of Met Data in Ashtabula County

Figure 12: Wind Rose Plot of Met Data in Tuscarawas County




- 110 -

Figure 13: Wind Rose Plot of Met Data in Macomb County

The Macomb site was selected as the most representative for this analysis in evaluating predicted

ozone formation as a function of NO2 emission rate.

With the elevated emissions releases shown in Appendix A of the guidance, the source derived

NOx MERPs for the 8-hour ozone standard are calculated for source 11 using the equation from

Section 5 of the MERP guidance. These values range from 532 to 875 tons per year as shown in

the calculations below.

• MERP for Source 11 EUS Region Short Release (tpy) = 1.0 ppb x (500 tpy/0.941 ppb) = 531 tpy

• MERP for Source 11 EUS Region Elevated Release (tpy) = 1.0 ppb x (500 tpy/0.936 ppb) = 534 tpy



Because the lowest calculated MERPs shown above is larger than the proposed Project NOx

emission increase, the ozone air quality impacts from the Projects would be expected to be less

than the critical air quality threshold. No further analysis is required based on this screening

procedure.




- 111 -

5.10 Secondary PM2.5 Analysis

The same guidance from April 2019 that was used for the ozone analysis was used for the PM2.5

analysis. Direct PM2.5 emissions were modeled using AERMOD. Both the NOx and SO2

emissions are well below the lowest (most conservative) daily and annual PM2.5 MERP values of

any source modeled in the Ohio Valley or Upper Midwest Climate Zones in the continental US.

However, the NOx and SO2 precursor contributions along with direct PM2.5 contributions are

considered together to determine if the project source’s air quality impact of PM2.5 would exceed

the critical air quality threshold. Because the modeling results for direct PM2.5 emissions were

above the PM2.5 SIL, the secondary analysis for comparison with the SIL is not needed. Scenario

B in Section 4.1.3 of the guidance was followed for a cumulative impact analysis for PM2.5

NAAQS. In this case, the analysis for the secondary formation of PM2.5 from NOx and SO2 are

added to the direct-modeled PM2.5 emissions, including the relevant surrounding sources and also

the background. The calculated total concentration is then compared with the NAAQS for

compliance demonstration purposes. The equations used for the analysis of the primary and

secondary impacts of daily and annual PM2.5 along with the calculations are shown below.

Equations for Primary and Secondary PM2.5 Impacts

Eq. 2 𝑃𝑟𝑜𝑗𝑒𝑐𝑡 𝐼𝑚𝑝𝑎𝑐𝑡 = 𝑃𝑟𝑜𝑗𝑒𝑐𝑡 𝐸𝑚𝑖𝑠𝑠𝑖𝑜𝑛 𝑅𝑎𝑡𝑒 ×𝑀𝑜𝑑𝑒𝑙𝑒𝑑 𝑎𝑖𝑟 𝑞𝑢𝑎𝑙𝑖𝑡𝑦 𝑖𝑚𝑝𝑎𝑐𝑡 𝑓𝑟𝑜𝑚 ℎ𝑦𝑝𝑜𝑡ℎ𝑒𝑡𝑖𝑐𝑎𝑙 𝑠𝑜𝑢𝑟𝑐𝑒

𝑀𝑜𝑑𝑒𝑙𝑒𝑑 𝑒𝑚𝑖𝑠𝑠𝑖𝑜𝑛 𝑟𝑎𝑡𝑒 𝑓𝑟𝑜𝑚 ℎ𝑦𝑝𝑜𝑡ℎ𝑒𝑡𝑖𝑐𝑎𝑙 𝑠𝑜𝑢𝑟𝑐𝑒

Eq. 3 𝑃𝑟𝑜𝑗𝑒𝑐𝑡𝑒𝑑 𝐷𝑒𝑠𝑖𝑔𝑛 𝑉𝑎𝑙𝑢𝑒 𝑤𝑖𝑡ℎ 𝑃𝑟𝑜𝑗𝑒𝑐𝑡 = 𝑃𝑟𝑜𝑗𝑒𝑐𝑡 𝐼𝑚𝑝𝑎𝑐𝑡 (𝐸𝑞. 2) +

𝑆𝑜𝑢𝑟𝑐𝑒 𝑀𝑜𝑑𝑒𝑙𝑒𝑑 𝐼𝑚𝑝𝑎𝑐𝑡 + 𝑀𝑜𝑛𝑖𝑡𝑜𝑟𝑒𝑑 𝐷𝑒𝑠𝑖𝑔𝑛 𝑉𝑎𝑙𝑢𝑒

Daily PM2.5

Source Nitrate Impact =

484.57 𝑡𝑝𝑦 ×0.126 𝜇𝑔/𝑚3

500 𝑡𝑝𝑦= 0.122 𝜇𝑔/𝑚3

Source Sulfate Impact =

12.43 𝑡𝑝𝑦 ×0.297 𝜇𝑔/𝑚3

500 𝑡𝑝𝑦= 0.00738 𝜇𝑔/𝑚3

Projected Design Value with Project =

0.122 𝜇𝑔/𝑚3 + 0.00738 𝜇𝑔/𝑚3 + 6.62 𝜇𝑔/𝑚3 + 18 𝜇𝑔/𝑚3 = 24.75 𝜇𝑔/𝑚3 < 35 𝜇𝑔/𝑚3




- 112 -

Annual PM2.5

Source Nitrate Impact =

484.57 𝑡𝑝𝑦 ×0.0072338 𝜇𝑔/𝑚3

500 𝑡𝑝𝑦= 0.00701 𝜇𝑔/𝑚3

Source Sulfate Impact =

12.43 𝑡𝑝𝑦 ×0.00838426 𝜇𝑔/𝑚3

500 𝑡𝑝𝑦= 0.000208 𝜇𝑔/𝑚3

Projected Design Value with Project =

0.00701 𝜇𝑔/𝑚3 + 0.000208 𝜇𝑔/𝑚3 + 1.48 𝜇𝑔/𝑚3 + 7.8 𝜇𝑔/𝑚3 = 9.29 𝜇𝑔/𝑚3 < 12 𝜇𝑔/𝑚3

As shown above, the primary and secondary PM2.5 impacts for both the 24-hour and annual

averaging periods are each less than their respective NAAQS as shown above. These results

indicate that the critical air quality threshold would not be exceeded when considering the

combined impacts of these primary and secondary precursors on both daily and annual PM2.5. No

additional modeling is required beyond this screening-level procedure.

5.11 Summary

AERMOD predicts that the construction of the new Petmin facility results in offsite impacts that

exceed the NO2 (1-hour and annual), PM2.5 (24-hour and annual), and PM10 (24-hour) significant

impact levels (SIL). Therefore, a full impacts analysis was conducted to evaluate the maximum

ambient impacts from those pollutants for comparison to the NAAQS. The results of this full

impacts analysis indicate that the construction of the new Petmin facility and other interactive

sources result in maximum offsite impacts below the NAAQS for the modeled pollutants.

The results also indicate that less than one-half of the available PSD increment is consumed from

the Petmin facility and the only other PSD source (Cristal Complex Plant 1) in the area. The offsite

impacts are below the SILs for the PM10 annual averaging period and CO for both the 1-hour and

8-hour averaging periods; therefore, no further analysis was required for PM10 annual or CO. Since

NOx and SO2 emissions create secondary PM2.5, an analysis was performed following U.S. EPA

guidance. This analysis indicates that the critical air quality threshold would not be exceeded when

considering the combined impacts of these primary and secondary precursors on both daily and

annual PM2.5.

Electronic files associated with this study have been transmitted to Ohio EPA Division of Air

Pollution Control (DAPC). The list of files is included in Appendix 3.




- 113 -

6.0 ADDITIONAL IMPACTS ANALYSIS

All PSD applications must include an additional impacts analysis that assesses the impact of air,

ground, and water pollution on soils, vegetation, and visibility caused by the proposed emission

increases and associated growth. The additional impacts analysis generally has four parts:

• growth;

• ambient air quality impact analysis;

• soils and vegetation impacts; and

• visibility impairment.

The additional impacts analysis is performed consistent with the requirements of OAC §3745-31-

17 (Attainment Provisions – Additional Impacts Analysis). The only pollutants for which the

project will exceed the PSD significant emission rates are NOx, CO, and Greenhouse Gases

(GHG). Per the U.S. EPA’s guidance, the additional impacts analyses requirements do not apply

to the emissions of GHGs for a project.29 Therefore, only the impacts of NOx and CO emissions

are addressed in this analysis. The following sections address these four aspects to the project and

demonstrate “no additional impacts.”

6.1 Growth Impact Analysis

The purpose of the growth analysis is to quantify the associated industrial, commercial, and

residential growth that will occur as a result of the project, and to evaluate secondary emissions

that will occur from that projected growth.

The growth analysis addresses only permanent economic growth attributable to the proposed

project. Short-term or temporary impacts, such as construction, are not considered permanent

growth and, therefore, are not addressed as additional impacts.

The proposed project is expected to result in approximately 110 new permanent, full-time positions

at the site. The Pinney Dock site has been previously developed for industrial and commercial use,

with supporting transportation infrastructure (water, rail and roadway). Petmin anticipates that the

workforce will be drawn largely from the surrounding communities. As a result of the relatively

self-contained nature of Petmin’s manufacturing operations, no related industrial growth is

expected to accompany the operation of the plant. In summary, secondary emissions associated

with the incremental growth as a result of this project will be insignificant.




- 114 -

6.2 Ambient Air Quality Analysis

The ambient air quality analysis projects the air quality which will exist in the area of the proposed

source or modification from the growth analysis after it begins operation. No sources of secondary

air emissions were identified that could warrant adding to the ambient air quality analysis

presented in modeling section of this study.

6.3 Soils and Vegetation Analysis

Gaseous emissions of air pollutants can lead to deterioration of soil quality and subsequently

impact vegetation. For most types of soils and vegetation, ambient concentrations of criteria

pollutants below the secondary national ambient air quality standards (NAAQS) will not result in

harmful effects.

Modeled NO2 concentrations from the proposed project are well below the well below the NAAQS

for NO2 (100 g/m3 as an annual average).

As shown in the modeling section, the predicted annual average NO2 concentration at the worst-

case receptor is 5.10 μg/m3 (2.83 x 10-3 ppmv NO2), including background.

A study from EPA Region 7 concluded:

In periods of two weeks or greater duration with intermittent exposures of several hours

per day, adverse effects on growth and yield start to appear when the concentration of NOx

reaches the range of 0.1 to 0.5 ppm, depending on the species of plant, nature of effect, and

conditions of exposure.30

The predicted concentration is approximately two orders of magnitude below the range at which

adverse effects have been observed for NOx-sensitive plants. Therefore, no impact on soils and

vegetation is predicted.

In the case of CO, no secondary NAAQS has been established. There is no known impact on soils

and vegetation associated with CO emissions that meet the primary NAAQS.

6.4 Visibility

A Level I visibility analysis for the proposed project emissions was conducted using the U.S. EPA

VISCREEN model. Maximum annual emission rates of NOx and particulate were input. Typically,

the distances used in this model are the nearest and furthest Class I Areas. In this case, there are

no Class I areas within 300 km. According to the model, “although VISCREEN is designed

primarily for assessing plume visual impacts in Class I areas, it can also be applied in PSD Class




- 115 -

II areas. In such cases these distances can be specified arbitrarily.”31 For this analysis, per the

protocol response from Ohio EPA32, the distances to the State Game Lands #314 in Erie County,

PA, (24 km and 27 km for the minimum and maximum distances, respectively), were used. The

results, presented in Appendix 4, indicate that the project impacts will not exceed the visibility

screening criteria. Therefore, a refined analysis of visibility impacts is not warranted.

Furthermore, there are no scenic vistas, airports, or other areas near the site that would be affected

by trace reductions in visibility.

6.5 Class I Area Impacts Analysis

There are no Class I areas located within 300 km of the proposed project. Based on the Federal

Land Manager (FLM) guidance:

Therefore, the Agencies will consider a source locating greater than 50 km from a Class I area

to have negligible impacts with respect to Class I AQRVs if its total SO2, NOx, PM10, and

H2SO4 annual emissions (in tons per year, based on 24-hour maximum allowable emissions),

divided by the distance (in km) from the Class I area (Q/D) is 10 or less. The Agencies would

not request any further Class I AQRV impact analyses from such sources.33

The nearest Class I area is Otter Creek Wilderness, which is 348 km from Petmin. The Q/D for

the project is approximately 1.5, well below the FLM’s guidance threshold of 10. Therefore, no

further AQRV analysis is required for the Project.

1 https://www.metallics.org/pig-iron.html

2 Memorandum, April 30, 2018, U.S. EPA Office of Air and Radiation, William L. Wehrum Assistant

Administrator to Patrick McDonnell, Secretary of the Pennsylvania Department of Environmental

Protection.

3 Ashtabula County is listed as “Attainment/Unclassifiable” in the prepublication version of “Additional

Air Quality Designations for the 2015 Ozone National Ambient Air Quality Standards”, signed by EPA

Administrator E. Scott Pruitt on April 30, 2018. See https://www.epa.gov/sites/production/files/2018-

04/documents/placeholder.pdf, page 86/114.

4 Interoffice Memo, “BAT Requirements for Permits Issued February 7, 2014 or After,” Mike Hopkins,

Ohio EPA, February 7, 2014.

http://www.epa.ohio.gov/Portals/27/sb265/Final20140207Post090803BATv11.pdf

5 Nitrogen Oxides, How and Why They Are Controlled, Technical Bulletin, U.S. EPA Office of Air

Quality Planning and Standards, EPA 456/F-99-006R, November 1999.

https://www.metallics.org/pig-iron.html

https://www.epa.gov/sites/production/files/2018-04/documents/placeholder.pdf

https://www.epa.gov/sites/production/files/2018-04/documents/placeholder.pdf




- 116 -

6 “Air Pollution Control Technology Fact Sheet, Selective Catalytic Reduction,” U.S. EPA, EPA-

452/F-03-032.

7 Prevention of Deterioration Workshop Manual, EPA-450/2-80-081, October 1980, pp.I-B-6 to I-B-7.

8 Air Emission Permit issued to Essar Steel Minnesota LLC, Minnesota Pollution Control Agency,

amended October 14, 2011.

https://www.pca.state.mn.us/sites/default/files/06100067-004-aqpermit.pdf

9 Permit Modification Application and technical report submitted to Alabama Department of

Environmental Management, Outokumpu Stainless Steel USA, LLC, June 19, 2014.

http://www.adem.state.al.us/newsEvents/notices/may17/pdfs/5outokumpu.pdf

10 Compilation of Air Pollutant Emission Factors, AP-42, Fifth Edition, U.S. EPA, Section 1.4.4.

11 Air Pollution Control Technology Fact Sheet, Selective Catalytic Reduction, EPA-452/F-03-032.

12 OAQPS Cost Control Manual, Fifth Edition, EPA 453/B-96-01. This interest rate value is also used

as standard for government analyses, as explained in detail in the Sixth Edition, January 2002.

13 Technical report provided in support of PTI application for Iron Units LLC PSD permit application,

RTP Environmental Associates, September 2017.

14 Air Pollution, Its Origin and Control, Wark, K. and Warner, C., Harper & Row, Second Edition, 1981. Table 5-

13, p. 236. “Typical Fractional Collection Efficiencies of Particulate Control Equipment.”

15 Guideline on Air Quality Models (Appendix W of 40 CFR 51), May 22, 2017.

16 Ohio Environmental Protection Agency, Engineering Guide #69: Air Dispersion Modeling Guidance,

Revised November 14, 2018.

17 The met data for this study was pre-processed by the Ohio EPA and was obtained from Chris Beekman,

Air Quality Modeling Specialist at Ohio EPA on September XX, 2019.

18 Google Earth V 7.3.2.5491 (64-bit). (September 14, 2015). Ashtabula, OH, United States UTM Zone

17, 51786 m E, 4639406 m N, elevation 575 ft. Google 2018. http://www.earth.google.com [Viewed

August 24, 2018].

19 Email correspondence with William Kenny at Ohio EPA, on December 14, 2017 (2010-2014 ozone

and NO2 data), email correspondence with Chris Beekman, Air Quality Modeling Specialist at Ohio

EPA on July 23, 2018 (2015-2016 ozone and NO2 data) and July 25, 2018 (2017 ozone and NO2 data),

and mail correspondence with Chris Beekman, Air Quality Modeling Specialist at Ohio EPA on

September 12, 2019.

20 American Meteorological Society/Environmental Protection Agency Regulatory Model (AERMOD).

AERMOD Implementation Guide, AERMOD Implementation Workgroup, U.S. Environmental

Protection Agency, Office of Air Quality Planning and Standards, Air Quality Assessment Division,

Research Triangle Park, North Carolina, EPA-454/B-18-003. Last Revised April, 2018.

21 BEEST Version 12.01 was used, which implements the EPA AERMOD algorithm version 19191 and

BPIP-Prime. The BEEST model has been certified for use in regulatory applications by U.S. EPA and

is also the model used by Ohio EPA.

https://www.pca.state.mn.us/sites/default/files/06100067-004-aqpermit.pdf

http://www.adem.state.al.us/newsEvents/notices/may17/pdfs/5outokumpu.pdf

http://www.earth.google.com/




- 117 -

22 GeoTIFF is a public domain metadata standard which allows georeferencing information to be

embedded within a TIFF file. United States Geological Survey (USGS) website:

https://www.usgs.gov/.

23 Applicability of Appendix W Modeling Guidance for the 1-hour NO2 National Ambient Air Quality

Standard, Tyler Fox, Leader Air Quality Modeling Group, U.S. EPA, Research Triangle Park, June 28,

2010.

24 Sensitivity Analysis of PVMRM and OLM in AERMOD, MACTEC, September 2004.

25 Email correspondence from Chris Beekman Air Quality Modeling Specialist at Ohio EPA on April 20,

2018.

26 Email correspondence from Nicole Patella, Administrative Professional, Northeast District Office,

Ohio EPA on May 23, 2018.

27 Memorandum, Guidance on the Development of Modeled Emission Rates for Precursors (MERPs) as

a Tier 1 Demonstration Tool for Ozone and PM2.5 under the PSD Permitting Program, Richard A.

Wayland, Director Air Quality Assessment Division, U.S. EPA, Research Triangle Park, April 30,

2019.

28 https://mesonet.agron.iastate.edu/sites/locate.php

29 In a March 2011 permitting guidance, the U.S.EPA observed that “…it is not necessary for applicants

or permitting authorities to assess impacts from GHGs in the context of the additional impacts analysis

or Class I area provisions of the PSD regulations… EPA believes that the most practical way to address

the considerations reflected in the Class I area and additional impacts analysis is to focus on reducing

GHG emissions to the maximum extent.” PSD and Title V Permitting Guidance for Greenhouse Gases,

EPA-457/B-11-001, March 2011, at pages 47-48.

30 Air Quality Criteria for Oxides of Nitrogen – Summary of Vegetation Impacts, U.S. EPA, Region 7.

31 Workbook for Visual Impact Screening and Analysis (Revised), EPA OAQPS, October 1992, EPA-

454/R-92-023.

32 Email correspondence from Chris Beekman Air Quality Modeling Specialist at Ohio EPA on

September 5, 2018.

33 Federal Land Managers’ Air Quality Related Values Work Group (FLAG), Phase I Report—Revised

(2010), Natural Resource Report NPS/NRPC/NRR—2010/232, P. 18.

https://www.usgs.gov/

https://mesonet.agron.iastate.edu/sites/locate.php

LEGEND

Petmin

Ashtabula, OH

Process Flow DiagramDate Revised: 12/1/2019

Screening

Iron

Process Gas

Iron Ore PelletsIron Ore PelletsPellet Coating

Iron Ore PelletsIron Ore PelletsHYL DRI Reactor

Coated PelletsCoated Pellets Closed Electric Arc

ReactorPig Caster

Liquid IronLiquid Iron Pig IronPig Iron

Process Gas Heat

Recovery, Venturi,

Quenching System

Dirty

Process

Gas

CO2 & H2S Removal

Natural GasNatural Gas

Reducing

Natural Gas

Independent

Slag

Processing

Company

Slag

EAF

Baghouse

EAF

Baghouse

Stack A4Stack A1

Process

Gas Heater

Dust

Collector

Dust

Collector

Thermal Oxidizer

Flare (A3)Flare (A3)

Stack A7Stack A7

Air Emissions

Dirty

Process

Gas

Intermittent

Baghouse/ Dust

Collector

SO2

Scrubber

Process

Gas

CO2 &

H2S

Natural GasNatural Gas

CO2 &

SO2

DRI PelletsDRI Pellets

Packed Tower

Scrubber

Slag

Facility Profile ReportFacility Name: Petmin USA Incorporated

ID: 0204012023

-

-

-

-

-

Page 1 Facility Profile Report (0204012023): Petmin USA Incorporated

Facility : 0204012023 Apr 9 2020, 15:56:31

Facility Information

Facility ID: 0204012023

FacilityName: Petmin USA Incorporated

Facility Description: Merchant Pig Iron Production

Address1: 1003 Bridge Street

Address2:

City: Ashtabula State: Ohio

Zip Code: 44004

Portable:

Operating Status: Not Installed

Permitting Classification: TV PER Due Date: None

Transitional Status: None

Title V Permit Status: None Title V Certification Report Due Date:

Emissions Reporting Category for2019:

TV Status: Reminder Sent

Anticipated Emissions ReportingCategory for 2020:

TV

Core Place ID: 539756

Latitude: 41.9014

Longtitude: -80.78492

Yearly Emissions Reporting Category

Year Category Enabled Status

2019 TVx

Reminder Sent

SIC Codes

3312 Blast Furnaces And Steel Mills

NAICS Codes

331110 Iron and Steel Mills and Ferroalloy Manufacturing

Contacts

Contact Type Contact Person Phone Number Email Start Date End Date

Owner Incorporated,Petmin USA

(216)479-6876 05/23/2018

Primary Farinha,Palmira

(216)479-6876 [email protected]

02/22/2018

Billing Farinha,Palmira

(216)479-6876 [email protected]

02/22/2018

Contact Detail For : Incorporated, Petmin USA

Prefix: First Name: Petmin USA

Middle Name: Last Name: Incorporated

Suffix:

Company Title: Operating Company Name: Petmin USA Incorporated

-


Address 1: 1003 (Upper) Bridge Street

Address 2:

City: Ashtabula Zip Code: 44004

State: Ohio

Work Phone No: (216)479-6876 Secondary Phone No.:

Address 2: Secondary Ext. No.:

Mobile Phone No.: Pager No.:

Fax No: Pager PIN No.:

Email:

Email Pager Address:

Contact Detail For : Farinha, Palmira

Prefix: First Name: Palmira

Middle Name: Last Name: Farinha

Suffix:

Company Title: Operating Company Name: Petmin USA Incorporated

Address 1: P.O.Box 14280

Address 2:

City: Cleveland Zip Code: 44114

State: Ohio

Work Phone No: (216)479-6876 Secondary Phone No.:

Address 2: Secondary Ext. No.:

Mobile Phone No.: (216)532-4461 Pager No.:

Fax No: Pager PIN No.:

Email: [email protected]

Email Pager Address:

Federal Rules Applicability

Subject to MACT: Subject to PSD:

Subject to NESHAPS: Subject to non-attainment NSR:

Subject to NSPS: Subject to 112(r):

Subject to Title IV:

-

--


Emission Unit : B001 Apr 9 2020, 15:56:31

Emission Unit Information

DAPC Emissions Unit ID: B001

DAPC Description: 15.17 MMBtu/hr, natural gas-fired, startup boiler

Company Equipment ID: Startup boiler

Company Description: 15.165 mmbh natural gas fired boiler used at rated capacity only for start up(maximum anticipated 200 hr/yr). For the other 8560 hr/yr, the boiler will operateon low fire at 3.033 mmbh. The weighted average input is 3.31 mmbh.


Completion of Initial InstallationDate:

Begin Installation/Modification Date:

Commence Operation AfterInstallation or Latest Modification

Date:

Title V EU Classification: Significant Exemption Status: NA

Boiler/Turbine/Generator DesignCapacity:

Boiler/Heater ORIS Boiler ID:

Processes

Emission Process Information

Process ID: Startup Boiler

Company Process Description: Startup Boiler

Source Classification Code (SCC): 1-02-006-02

Egress points(s) directly associated with this process

Aux Boiler

-

--


Emission Unit : F001 Apr 9 2020, 15:56:31


DAPC Emissions Unit ID: F001

DAPC Description: Unpaved plant roadways and parking areas

Company Equipment ID: Plant Roadways

Company Description: Plant roadways and parking areas





Date:

Title V EU Classification: Insignificant Exemption Status: NA


Not Applicable Design Capacity Units:

ORIS Boiler ID:

Processes


Process ID: Plant Roadways

Company Process Description: Plant Roadways


Control equipment(s) directly associated with this process

Dust Control

-

--


Emission Unit : P001 Apr 9 2020, 15:56:31


DAPC Emissions Unit ID: P001

DAPC Description: 218.9 MMBtu/hr, natural gas and process gas-fired, process gas heater, with top-gastreatment with a thermal oxidizer and scrubber.

Company Equipment ID: Process gas heater

Company Description: Process gas heater





Date:



Boiler/Heater ORIS Boiler ID:

Processes


Process ID: Process gas heater

Company Process Description: Process gas heater



CTO

-

--





DAPC Description: 15.00 MMBtu/hr, natural gas-fired ladle dryer / preheater, vented to the EAFbaghouse.

Company Equipment ID: Ladle preheater

Company Description: Ladle preheater: 15.00 mmbh, natural gas fired





Date:




ORIS Boiler ID:

Processes


Process ID: Ladle preheat

Company Process Description: Ladle preheat and dry



EAF/Cast BH

-

--






Company Equipment ID: Ladle preheat (backup)

Company Description: Ladle preheat: 15.00 mmbh, natural gas fired





Date:

Title V EU Classification: Not Applicable Exemption Status: NA



ORIS Boiler ID:

Processes


Process ID: Ladle preheat backup

Company Process Description: Ladle preheat backup



EAF/Cast BH

-

--






Company Equipment ID: Ladle drying station

Company Description: Ladle drying station: 15.00 mmbh, natural gas fired





Date:

Title V EU Classification: Not Applicable Exemption Status: NA



ORIS Boiler ID:

Processes


Process ID: Ladle dry station

Company Process Description: Ladle dry station



EAF/Cast BH

-

--





DAPC Description: 3,131 HP, diesel fuel-fired, emergency generator

Company Equipment ID: Emergency Generator #1

Company Description: Emergency generator - not yet ordered but preliminary specifications are for a 3131HP, diesel-fired, Tier 4 certified RICE, operating a maximum of 100 hr/yr





Date:




ORIS Boiler ID:

Processes


Process ID: Emer. Generator #1

Company Process Description: Emergency Generator #1



Emrg Gen Stk

-

--





DAPC Description: 3,131 HP, diesel fuel-fired, emergency generator

Company Equipment ID: Emergency Generator #2

Company Description: Emergency generator - not yet ordered but preliminary specifications are for a 3131HP, diesel-fired, Tier 4 certified RICE, operating a maximum of 100 hr/yr





Date:




ORIS Boiler ID:

Processes


Process ID: Emer. Generator #2

Company Process Description: Emergency Generator #2



Emrg Gen Stk

-

--





DAPC Description: 158 HP, diesel fuel-fired, black start generator

Company Equipment ID: Black Start Generator

Company Description: Black start generator - not yet ordered but preliminary specifications are for a 158HP, diesel-fired, Tier 4 certified RICE, operating a maximum of 100 hr/yr





Date:




ORIS Boiler ID:

Processes


Process ID: BlackStart Generator

Company Process Description: Black Start Generator



Blk-Strt Gen

-

--





DAPC Description: Quenching & pre-wastewater treatment, equipped with a flare.

Company Equipment ID: Quenching & pre-wastewater treatment

Company Description: Process gas conditioning: quenching & pre-wastewater treatment





Date:




ORIS Boiler ID:

Processes


Process ID: Quench

Company Process Description: Quenching & wastewater treatment



NH3 Scrubber

Flare

-

--





DAPC Description: Emergency fire pump (high pressure), diesel engine

Company Equipment ID: High Pressure Emergency Diesel Engine

Company Description: High pressure emergency diesel engine for fire fighting pump.





Date:




ORIS Boiler ID:

Processes


Process ID: HP Emg D Eng

Company Process Description: High pressure emergency diesel engine for fire fighting pump.



HP Emg Eng S

-

--





DAPC Description: Emergency fire pump (low Pressure), diesel engine.

Company Equipment ID: Low Pressure Emergency Diesel Engine

Company Description: Low pressure emergency diesel engine for fire fighting pump.





Date:




ORIS Boiler ID:

Processes


Process ID: LP Emg D Eng

Company Process Description: Low pressure emergency diesel engine for fire fighting pump.



LP Emg Eng S

-

--





DAPC Description: Electric Arc Furnace, including DRI loading, smelting, tapping, pouring, andcasting.

Company Equipment ID: EAF

Company Description: Electric Arc Furnace, including DRI loading, smelting, tapping, pouring, and casting





Date:




ORIS Boiler ID:

Processes


Process ID: EAF and Casting

Company Process Description: EAF and Casting



EAF/Cast BH

-

--





DAPC Description: Raw materials handling, including screening and transfer via conveyor system.

Company Equipment ID: Material Handling

Company Description: Conveying, transferring and screening of iron ore pellets





Date:




ORIS Boiler ID:

Processes


Process ID: Material Handling

Company Process Description: Pellet Handling



Screen bldg

-

--


Emission Unit : TMP208268 Apr 9 2020, 15:56:31


DAPC Emissions Unit ID: TMP208268

DAPC Description:

Company Equipment ID: Cooling Tower

Company Description: Cooling Tower





Date:

Title V EU Classification: Trivial Exemption Status: De minimis



ORIS Boiler ID:

Processes


Process ID: Cooling Tower

Company Process Description: Cooling Tower


-

--





DAPC Description:

Company Equipment ID: Fines Handling

Company Description: Transfer of iron ore fines from screening to truck where it is transported offsite.





Date:

Title V EU Classification: Not Applicable Exemption Status: De minimis



ORIS Boiler ID:

Processes


Process ID: Fines Handling

Company Process Description: Transfer of iron ore fines from screening to truck where it is transportedoffsite.


-

--





DAPC Description:

Company Equipment ID: Bulk Flux Silos

Company Description: Silos for storing bulk flux.





Date:




ORIS Boiler ID:

Processes


Process ID: Bulk Flux Silo

Company Process Description: Bulk silo's for storing flux


-

--





DAPC Description:

Company Equipment ID: Remet Storage Piles

Company Description: Storage piles for storing remet.





Date:




ORIS Boiler ID:

Processes


Process ID: Remet Storage Piles

Company Process Description: Storage piles for storing remet.


-

-

-

-


Control Equipment : CTO Apr 9 2020, 15:56:31

Control Equipment Information

Equipment Type: Thermal Incinerator

DAPC Description:

Company ID: CTO

Company Description: Catalytic thermal oxidizer converting H2S into SO2

Operating Status: Operating Initial Installation Date:

Manufacturer: TBD Model: TBD

Specific Equipment Type information

Equipment Description: Catalytic thermal oxidizer used for converting H2S from dirty process gas into SO2

Min Operating Temp: 338

Temperature Sensor Location: TBD

Combustion Chamber ResidenceTime:

0.8

Inlet Gas Flow Rate:

Outlet Gas Flow Rate: 761278

Inlet Gas Temp:

Outlet Gas Temp: 554

Pollutants Controlled

Pollutant Design ControlEfficiency(%)

OperatingControlEfficiency(%)

CaptureEfficiency(%)

Total CaptureControl(%)

Hydrogen Sulfide(seeModification)

99 99 100 99

Associated Control Equipments And Egress Points

Control equipment(s) directly associated with this control equipment

Wet Scrubber

-

-

-

-


Control Equipment : Dust Control Apr 9 2020, 15:56:31


Equipment Type: Fugitive Dust Suppression

DAPC Description:

Company ID: Dust Control

Company Description: Plant Roadways dust control

Operating Status: Not Operating Initial Installation Date:



Suppressant Agent Type: Water

Equipment Description: Water truck or sprayer

Method of Application: Water truck or sprayer

Application Rate - specify units: As needed

Application Frequency: As needed






PE (Filt) - Primary PM,Filterable Portion Only

99.9 95 100 95

PM10 (Filt) - Primary PM10,Filterable Portion Only

99.9 95 100 95


Egress points(s) directly associated with this control equipment

Roads

-

-

-

-


Control Equipment : EAF/Cast BH Apr 9 2020, 15:56:31


Equipment Type: Filter/Baghouse

DAPC Description:

Company ID: EAF/Cast BH

Company Description: EAF and Casting baghouse




Filter/Baghouse Type: Pulse Jet

Equipment Description: EAF and Casting baghouse

Pressure type: negative

Fabric Cleaning Mechanism: Pulse jet

Operating Pressure Drop Range: 5-9

Lime Injection/fabric Coating Agent: No

Lime Injection/Fabric Coating AgentType:

Lime Injection/Fabric Coating FeedRate:

Bag Leak Detection System: No

Inlet Gas Flow Rate: 505229


Inlet Gas Temp: 205.7

Outlet Gas Temp: 205.7







99.9 99.9 100 99.9


99.9 99.9 100 99.9



A-4

-

-

-

-


Control Equipment : Flare Apr 9 2020, 15:56:31


Equipment Type: Flare

DAPC Description:

Company ID: Flare

Company Description: Flare




Flare Type: Elevated - Open

Elevated Flare Type: Non-assisted

Equipment Description: Flare

Ignition Device: Yes

Flame Presence Sensor: Yes



Inlet Gas Temp: 85

Outlet Gas Temp: 3310






CO - Carbon Monoxide 98 98 100 98



A-3

-

-

-

-


Control Equipment : NH3 Scrubber Apr 9 2020, 15:56:31


Equipment Type: Wet Scrubber

DAPC Description:

Company ID: NH3 Scrubber

Company Description: Ammonia scrubber




Wet Scrubber Type: Packed Bed

Equipment Description: Ammonia scrubber

Operating Pressure Drop Range: <15

pH Range for Scrubbing Liquid: >9.5

Scrubber Liquid Recirculated: No

Scrubber Liquid Flow Rate: 44

Scrubber Liquid Supply Pressure: TBD



Inlet Gas Temp: 70

Outlet Gas Temp: 70






Ammonia 100 100 100 100



A-9

-

-

-

-


Control Equipment : Screen bldg Apr 9 2020, 15:56:31


Equipment Type: Filter/Baghouse

DAPC Description:

Company ID: Screen bldg

Company Description: Screen building




Filter/Baghouse Type: Reverse Air

Equipment Description: Screen building baghouse

Pressure type: negative

Fabric Cleaning Mechanism: Reverse air

Operating Pressure Drop Range: <15

Lime Injection/fabric Coating Agent: No

Lime Injection/Fabric Coating AgentType:

Lime Injection/Fabric Coating FeedRate:

Bag Leak Detection System: No



Inlet Gas Temp: 70

Outlet Gas Temp: 70







99.9 99.9 100 99.9


99.9 99.9 100 99.9



A-7

-

-

-

-


Control Equipment : Wet Scrubber Apr 9 2020, 15:56:31


Equipment Type: Wet Scrubber

DAPC Description:

Company ID: Wet Scrubber

Company Description: SO2 wet scrubber

Operating Status: Operating Initial Installation Date:



Wet Scrubber Type: Packed Bed

Equipment Description: Packed bed wet scrubber for controlling SO2 emissions

Operating Pressure Drop Range: TBD

pH Range for Scrubbing Liquid: TBD

Scrubber Liquid Recirculated: TBD

Scrubber Liquid Flow Rate: TBD

Scrubber Liquid Supply Pressure: TBD

Inlet Gas Flow Rate:

Outlet Gas Flow Rate:

Inlet Gas Temp:

Outlet Gas Temp:






SO2 - Sulfur Dioxide 95 95 100 95



A-1

-

-

-

-

-

-


Egress Point : A-1 Apr 9 2020, 15:56:31

Egress Point Information

Release Type: Stack-Vertical

DAPC Description:

Company ID: A-1

Company Description: A-1 Process Gas Heater

Operating Status: Not Operating

Base Elevation (ft): 575.0 Fenceline Distance (ft): 1200.0

Release Height (ft): 390.0

Building Dimension

Length (ft) 660.0 Width (ft): 380.0

Height (ft): 40.0

Egress Latitude and Longitude

Latitude: 41.9053 Longitude: -80.79177

Stack Details

Shape: Round Cross Sectional Area (square ft): 8.04

Diameter (ft): 8.04

Temp At Max. Oper (F): 383.0 Flow At Max. Oper (acfm): 87780.0

Temp At Avg. Oper (F): 383.0 Flow At Avg. Oper (acfm): 87780.0

EIS Information

Horizontal Collection Method: Global Positioning Method, with unspecified parameters

Horizontal Accuracy Measure: 100 Meter Accuracy

Reference Point: Point where a substance is released

Horizontal Reference Datum: World Geodetic System of 1984

Coordinate Data Source Code: An Organization or individual that contracts to perform work

CEM Data

Description H2S SO2 NOX CO THC HCL HFL O TRS CO2 FLOW OPACITY PM

-

-

-

-

-

-





DAPC Description:

Company ID: A-3

Company Description: A-3 Flare




Building Dimension


Height (ft): 40.0



Stack Details


Diameter (ft): 1.67



EIS Information






CEM Data


-

-

-

-

-

-





DAPC Description:

Company ID: A-4

Company Description: A-4 EAF baghouse




Building Dimension


Height (ft): 40.0



Stack Details

Shape: Round Cross Sectional Area (square ft):

Diameter (ft): 15.09



EIS Information






CEM Data


-

-

-

-

-

-





DAPC Description:

Company ID: A-7

Company Description: A-7 Material Handling baghouse




Building Dimension


Height (ft): 40.0



Stack Details

Shape: Square Cross Sectional Area (square ft): 3.14

Diameter (ft): 2



EIS Information






CEM Data


-

-

-

-

-

-





DAPC Description:

Company ID: A-9

Company Description: A-9 Ammonia scrubber




Building Dimension


Height (ft): 40.0



Stack Details


Diameter (ft): 2



EIS Information






CEM Data


-

-

-

-

-

-


Egress Point : Aux Boiler Apr 9 2020, 15:56:31



DAPC Description:

Company ID: Aux Boiler

Company Description: Startup boiler stack




Building Dimension


Height (ft): 40.0



Stack Details


Diameter (ft): 2



EIS Information






CEM Data


-

-

-

-

-

-


Egress Point : Blk-Strt Gen Apr 9 2020, 15:56:31



DAPC Description:

Company ID: Blk-Strt Gen

Company Description: Black-Start Generator Stack

Operating Status: Operating

Base Elevation (ft): 575.0 Fenceline Distance (ft):


Building Dimension


Height (ft): 40.0



Stack Details


Diameter (ft): 0.25



EIS Information






CEM Data


-

-

-

-

-

-


Egress Point : Emrg Gen Stk Apr 9 2020, 15:56:31



DAPC Description:

Company ID: Emrg Gen Stk

Company Description: Emergency Generator Stack




Building Dimension


Height (ft): 40.0



Stack Details

Shape: Other Cross Sectional Area (square ft):

Diameter (ft): 1



EIS Information






CEM Data


-

-

-

-

-

-


Egress Point : HP Emg Eng S Apr 9 2020, 15:56:31



DAPC Description:

Company ID: HP Emg Eng S

Company Description: Stack for high pressure emergency diesel engine for fire fighting pump.




Building Dimension


Height (ft): 40.0



Stack Details


Diameter (ft):

Temp At Max. Oper (F): Flow At Max. Oper (acfm):

Temp At Avg. Oper (F): Flow At Avg. Oper (acfm):

EIS Information






CEM Data


-

-

-

-

-

-


Egress Point : LP Emg Eng S Apr 9 2020, 15:56:31



DAPC Description:

Company ID: LP Emg Eng S

Company Description: Stack for low pressure emergency diesel engine for fire fighting pump.




Building Dimension


Height (ft): 40.0



Stack Details


Diameter (ft):

Temp At Max. Oper (F): Flow At Max. Oper (acfm):

Temp At Avg. Oper (F): Flow At Avg. Oper (acfm):

EIS Information






CEM Data


-

-

-

-

-


Egress Point : Roads Apr 9 2020, 15:56:31


Release Type: Fugitive-Area

DAPC Description:

Company ID: Roads

Company Description: Roads



Building Dimension


Height (ft): 40.0



Area Source Dimensions

Length (ft): 4050.0 Width (ft): 30.0


EIS Information






Plume Temp (F): 60.0

Summary Petmin USA Inc.

Source of Emissions Stack OEPA ID NOx CO SO2 PMTotal PMCondensable PMFilterable PM10 PM2.5 VOC HAP NH3 CO2

Process gas heater A-1 P001 82.71 48.92 3.46 7.14 5.36 1.79 7.14 7.14387 5.17 - - 307,490.06

Startup boiler A-2 B001 2.78 5.47 0.04 0.49 0.37 0.12 0.49 0.49 0.36 - - 7,814.43

Flare (Quench & ww treatment) A-3 P008 1.97 8.97 0.01 0.22 0.16 0.05 0.22 0.22 4.05 - 0.01 3,405.29

Ladle preheat* A-4 P002 9.29 2.26 0.04 - - - - - 0.35 - - 7,729.41

Ladle preheat (backup)* A-4 P003 9.29 2.26 0.04 - - - - - 0.35 - - 7,729.41

Ladle dry station* A-4 P004 9.29 2.26 0.04 - - - - - 0.35 - - 7,729.41

EAF and Casting A-4 P901 368.72 474.06 0.00 54.44 1.11 53.33 54.44 43.55 6.06 0.11 - 49,094.69

Emergency Generator #1 A-10 P005 0.17 0.90 0.00 0.01 0.01 - 0.01 0.01 0.05 - - 181.75

Emergency Generator #2 A-10 P006 0.17 0.90 0.00 0.01 0.01 - 0.01 0.01 0.05 - - 181.75

Black Start Generator P007 0.01 0.06 0.00 0.00 0.00 - 0.00 0.00 0.00 - - 9.09

High Pressure Emergency Pump 0.10 0.09 0.00 0.01 0.01 - 0.01 0.01 0.00 - - 17.90

Low Pressure Emergency Pump 0.07 0.07 0.00 0.00 0.00 - 0.00 0.00 0.00 - - 13.64

Material Handling A-7,Fug. P902 - - - 1.29 - 1.29 1.29 0.89 - - - -

Roadways/Parking Fugitive F001 - - - 0.22 - 0.22 0.22 0.02 - - - -

Remet Storage Piles (de minimis) Fugitive - - - - 0.05 - 0.05 0.05 0.05 - - - -

Fines Handling (de minimis) Fugitive - - - - 0.13 - 0.13 0.06 0.01 - - - -

Bulk Flux Silos (de minimis) N/A - - - - 0.02 - 0.02 0.02 - - - - -

Cooling Tower (de minimis) N/A - - - - 0.02 - 0.02 0.02 0.00 - - - -

484.57 546.22 3.63 64.03 7.02 57.01 63.97 52.40 16.80 0.11 0.01 391,397

100 100 100 25 15 10 100 25 75000

PSD "Major Source" ? YES YES NO YES YES YES NO NO YES

* PM emissions included in EAF and Casting emissions

TOTAL PROJECT-WIDE POTENTIAL TO EMIT

PROJECT POTENTIAL TO EMIT BY EMISSIONS UNIT (tons/yr)

Prepared By: Glen Greenwood and Jake Leitnaker 11/14/2019

Reviewed By: Joe Hollowell 11/18/2019 Page 1 of 15

Combustion Petmin USA Inc.

ObjectiveCalculate annual potential to emit from various, natural-gas combustion units at the facility.

Method

Use AP-42 emission factors, except as noted.

Assume 8,760 hours of operation per year at rated capacity.

Emission Factors

Natural gas BTU content is assumed to be: 1,020 BTU per scf. Reference AP-42 Appendix A.

Natural Gas Combustion Factors

CO 84 lb/MMscf 8.24E-02 lbs/MMBTU NOx 6.80E-02 lbs/MMBTU

SO2 0.6 lb/MMscf 5.88E-04 lbs/MMBTU CO 3.10E-01 lbs/MMBTU

PMTotal 7.6 lb/MMscf 7.45E-03 lbs/MMBTU SO2 2.37E-04 lbs/MMBTU

PMCondensable 5.7 lb/MMscf 5.59E-03 lbs/MMBTU VOC 1.40E-01 lbs/MMBTU

PMFilterable 1.9 lb/MMscf 1.86E-03 lbs/MMBTU

VOC 5.5 lb/MMscf 5.39E-03 lbs/MMBTU Process Gas Heater

CO2equiv 120000 lb/MMscf 1.176E+02 lbs/MMBTU Emission Factors - Tenova

NOx 6.40E-02 lbs/MMBTU

Additional Specific Emission Factors SO2 4.56E-03 lbs/MMBTU

Low-NOx burner guarantee: Startup Boiler CO 5.00E-02 lbs/MMBTU

NOx 18 g/GJ 4.18E-02 lbs/MMBTU

Ladle Preheaters/Dryers

Emission Factors - CEBA (burner manufacturer)


CO 3.44E-02 lbs/MMBTU

Average Flare Input Capacity

Average

Usage

(MMBTU/hr) Time (hr/yr)

Annual

Usage

(MMBTU/yr)

Normal Conditions 2.24 8183.00 18329.92

264 100 26400

Shutdown with Venting 40 25 1000

Operational Contingencies 25 442 11050

Emergency Venting 111 10 1110

Total: 57,889.92 MMBTU/yr

Total Average: 6.61 MMBTU/hr

Emission Calculations

Total PTE - Includes Combustion Only

Input Capacity

MMBTU/hr NOx CO SO2 PMTotal PMCond. PMFilt. VOC CO2equiv

Startup boiler (Aux. Boiler) 15.17 2.78 5.47 0.04 0.49 0.37 0.12 0.36 7,814

Flare 6.61 1.97 8.97 0.007 0.22 0.16 0.05 4.05 3,405

Ladle preheat 15.00 9.29 2.26 3.86E-02 0.49 0.37 0.12 0.35 7,729

Ladle preheat (backup) 15.00 9.29 2.26 3.86E-02 0.49 0.37 0.12 0.35 7,729

Ladle drying station 15.00 9.29 2.26 3.86E-02 0.49 0.37 0.12 0.35 7,729

TOTALS 32.62 21.22 0.16 2.18 1.64 0.53 5.47 34,408

Potential to Emit - Tons Per Year

Emission Factors - AP-42 Section 1.4, July 98 &

40 CFR Part 98, Subpart C

Equipment Description

Case

Start-up

Flare Emission Factors - AP-42

Section 13.5, dated 12/16, except SO2

from Tenova

An annual weighted-average MMBTU/hr for the flare was calculated by multiplying the MMBTU/hr for each type of

venting event by the estimated maximum hours per year for each event, summing up the MMBTU/yr for all of the

events, and then dividing by 8,760 hours/year. The estimated hours are conservatively high, "worst-case"

estimates'



Process Gas Heater Petmin USA Inc.

Objective

Calculate annual potential to emit from Process Gas Heater and CO2 removal.

Method

Use emission factors supplied by Tenova and AP-42, as noted.


Emission Factors

Natural gas BTU content is assumed to be: 1,020 BTU per scf. Reference AP-42 Appendix A.

Natural Gas Combustion Factors

CO 84 lb/MMscf 8.24E-02 lbs/MMBTU

SO2 0.6 lb/MMscf 5.88E-04 lbs/MMBTU

PM10 7.6 lb/MMscf 7.45E-03 lbs/MMBTU

PMCondensable 5.7 lb/MMscf 5.59E-03 lbs/MMBTU

PMFilterable 1.9 lb/MMscf 1.86E-03 lbs/MMBTU

VOC 5.5 lb/MMscf 5.39E-03 lbs/MMBTU

CO2equiv 120,000 lb/MMscf 1.176E+02 lbs/MMBTU

Process Gas Heater

Emission Factors - Tenova


SO2 4.56E-03 lbs/MMBTU

CO 5.00E-02 lbs/MMBTU

Calculations

lb/hr ton/yr

NOx 42.2 46.01 5.04E-06 3,745,240.1 18.88 82.71

CO 41 28.01 2.98E-06 3,745,240.1 11.17 48.92

CO2 164021 44.01 1.87E-02 3,745,240.1 70,203.21 307,490.06

VOC and PM10 Emissions

lb/hr ton/yr

VOC 5.39E-03 219 1.18 5.17PM10 7.45E-03 219 1.63 7.14

PMConensable 5.59E-03 219 1.22 5.36PMFilterable 1.86E-03 219 0.41 1.79

SO2 Emissions

lb/hr ton/yr

SO2 12 64.06 2.00E-06 396,244.5 0.79 3.46

1) Emission factors obtained from Tenova.

2) Conversion factor of: ppmV x MW/385.1x106 = lb/scf, was obtained from AP-42 Appendix A.

3) Flow rate converted from Nm3/h

35.3147 ft3

(20+273.1) K 37.9 SCF

1 m3

(0+273.1) K 1 Nm3

4) Emission Factors - AP-42 Section 1.4, July 98

Emission Factors - AP-42 Section 1.4, July 98 &

40 CFR Part 98, Subpart C

Potential Emissions

(8,760 hr/yr)

Gas

Flowrate

(scf/h)3

x =

Pollutant

Emission

Factors

(ppmV)1

Molecular

Weight

(g/mol)

Emission

Rate (lb/scf)2

Gas

Flowrate

(scf/h)3

Potential Emission

(8760 hr/yr)

Pollutant

Emissions

Factor4

(lb/MMBTU)

Input

Capacity

(MMBTU/hr)

Potential Emissions

(8,760 hr/yr)

Emission

Rate (lb/scf)2Pollutant

Emission

Factors

(ppmV)1

Molecular

Weight

(g/mol)



Bulk Flux Silos Petmin USA Inc.

Bulk Flux Storage Silos (assume cement is worst case)

Variable Data Units Reference

A 0.47 lb PM10/ton throughput AP-42 Table 11.12-2 (6/06)

B 4,239 CF silo capacity 120 m3 - largest silo

C 94 lb/CF bulk density of cement

D 199.23 tons silo capacity: B x C ÷ 2000 lb/ton

E 93.64 lb PM10/day dust generated: max 1 silo fill/day: A x D

F 99.9% filter efficiency Specification -for product conservation

G 0.094 lb PM10/day max daily emission rate: E x (1 - F)

0.0171 ton PM10/year max daily emission rate in TPY: G ÷ 2,000 lb/ton x 8,760 hr/yr

Bulk silos for flux (lime, limestone, dololime, or bauxite) and cement can hold up to 120 cubic meters each

and are equipped with filters to prevent product loss during pneumatic filling. According to Engineering

Guide #37, the filter should not be considered as control equipment. Assume that a silo could be

completely filled in one day.



EAF and Casting Petmin USA Inc.

Objective

Calculate annual potential to emit from the electric arc furnace (EAF).

Method

Use AP-42, Section 12.5.1 emission factors, except as noted.


EAF Calculations

Value Units Value Units lb/hr ton/yr

NOx 1.4 lb/ton 60.13 ton/hr 84.18 368.72

CO 1.8 lb/ton 60.13 ton/hr 108.23 474.06

VOC 0.023 lb/ton 60.13 ton/hr 1.38 6.06

CO2 186.41 lb/ton 60.13 ton/hr 11,209 49,095

Conversion

Factor

Value Units gr/lb lb/hr ton/yr

PM10 0.0025 gr/scf 34,801,675 7,000 12.43 54.44

PM2.5 = 80% of PM10 = 9.94 43.55

HAP Calculations

Pollutant

% of Raw

Material

Total PM10

ton/yr

Total HAP

ton/yr

Mn 0.18% 54.44 0.098

Cr 0.02% 54.44 0.009

TOTAL HAP: 0.107

PM10 emissions are based on the minimum, guaranteed absolute efficiency of the baghouse, and the gas flow rate.

The maximum fraction of PM2.5 was provided by the EAF design engineer team.

Calculations, except for PM2.5 (see footnote), are based on AP-42 emission factors for minimills, which typically have less

consistent and controlled inputs. These factors are expected to be conservative. AP-42 emission factors for sulfur, lead and

specific organic compounds were not used because they are not expected to be present in process inputs.

A complete cycle of the EAF processes 153 tons of DRI and flux over ~132 minutes, consisting of ~112 minutes of smelting

and ~20 minutes for tapping to the ladle, yielding ~132.3 tons of metal and ~20.7 tons of slag.

The fumes generated during the smelting are extracted from the roof of the furnace via a flexible suction pipe cooled with

water leading to a sedimentation chamber ( “drop-out box”). The configuration of the sedimentation chamber makes it

possible to separate coarse particles from the gas and the addition of air to the hot gas (800 °C) combusts most of CO.

The final emission factor was provided by the EAF design engineers.

Pollutant

AP-42 Emission Factor

Section 12.5.1 (04/09) Production Rate

Potential Emissions

(8,760 hr/yr)

Peak Flow

Rate

Duration of

Flow Rate (%

of time)

Potential Emissions

(8,760 hr/yr)

Calculations are based on the supplier's analysis of the iron ore pellets, which are the only input materials. The highest

listed HAP content in any of the proposed iron ore pellets is shown as 0.18% Mn and 0.016% Cr. It is assumed that these

percentages would be maintained in the PM10 emissions from the EAF.

Operation Pollutant

Outlet ConcentrationExhaust

Flow Rate

(scf/h)

Prepared By: Jake Leitnaker 12/9/2019

Reviewed By: RJ Schratz 12/9/2019 Page 5 of 15

Material Handling Petmin USA Inc.

Description

Plant Throughput: 96.93 TPH

850,000 tons/yr @ 8,760 hr/yr

Outdoor Conveyor Transfer Point (AP-42 Section 11.19.2, 8/04)

EF = 0.0011 lb/ton (uncontrolled PM10 emission factor)

0.00031087 lb/ton (uncontrolled PM2.5 emission factor)

Total Throughput = 850,000 tons per year iron ore pellets

Transfer Point Throughput (TPY) PM10 PM2.5 PM10 PM2.5

Loader A to Receiving Bin A 425,000 0.0011 0.00031 0.23375 0.06606

Loader B to Receiving Bin B 425,000 0.0011 0.00031 0.23375 0.06606

Receiving Bin A to Conveyor

1A 425,000 0.0011 0.00031 0.23375 0.06606

Receiving Bin B to Short

Conveyor 1B 425,000 0.0011 0.00031 0.23375 0.06606

Conveyor 1A to Conveyor 2

(Main Conveyor) 425,000 0.0011 0.00031 0.23375 0.06606

Conveyor 1B to Conveyor 2

(Main Conveyor) 425,000 0.0011 0.00031 0.23375 0.06606

Conveyor 2 to Feed Bin 850,000 0.0011 0.00031 0.4675 0.13212

Feed Bin to Conveyor 3 850,000 0.0011 0.00031 0.4675 0.13212

Conveyor 3 to Screening Bin 850,000 0.0011 0.00031 0.4675 0.13212

2.805 0.792717

80 80

0.56 0.16

Indoor Primary Screening and Conveyor Transfer

0.0025 gr/dscf control efficiency of baghouse

7,800 scfm flowrate of baghouse

60 hours per minute

7,000 grains per pound

2,000 pounds per ton

8,760 hours per year

Controlled PM10/2.5 = 0.17 lb/hr

Controlled PM10/2.5 = 0.73 TPY

Controlled Emissions PM10 PM2.5

Outdoor Material Handling0.56 0.16

Indoor Material Handling 0.73 0.73

Total PE (TPY): 1.29 0.89

Total Uncontrolled (TPY):

Total Controlled (TPY):

% control efficiency for water (RACM, p.

2-48)

Iron ore pellets are stored and loaded to Petmin's conveying system by Kinder Morgan. Pellets are

screened to remove undersized material. Other materials, i.e., slag and ingots are solidified metal from the

EAF and are not included here because they are not expected to contain fines.

Emission factors for PM2.5 from screening and transfer points are shown in AP-42 Section 11.19.2 as "ND",

or "No Data". The controlled emission factors for the same processes show values for both PM10 and

PM2.5. The ratio of the controlled emission factors for the two pollutants was applied to the uncontrolled

PM10 emission factor to calculate a PM2.5 emission factor.

Indoor transfer points and the screening operation are in buildings which are kept under negative pressure

by the applicable baghouses. Capture efficiency is presumed to be 100%. Baghouse control efficiency is

used to calculate these emissions

Uncontrolled

Emission Factor

(lb/ton)

Uncontrolled Potential

Emissions (TPY)



Roadways Petmin USA Inc.

Data

This site is not located in an area identified in Appendix A of OAC 3745-17-08.

EF = ((k*(s/12)^a*(W/3)^b))((365-p)/365)

PM10 PM2.5

k = 1.5 0.15

a = 0.9 0.9

b = 0.45 0.45

s = 6 % = mean silt content of unpaved roads (AP-42 Table 13.2.2-1 Iron/Steel Plants)

p = 150 days/yr with > 0.01 inch precipitation, from Figure 13.2.2-1.

95% total control efficiency

Vehicles

Tare

tons

Load

tons

W

Average

tons Material

Throughput

ton/yr

Road

Length

ft.

VMT

mi/yr

PM10

EF

lb/VMT

Unc.

PM10

ton/yr

Control

PM10

ton/yr

PM

EF

lb/VMT

Unc.

PM2.5

ton/yr

Control

PM2.5

ton/yr

Slag Pot Carrier 118.5 50.5 143.75 Slag 82,516 350 108 2.70 0.15 0.01 0.27 0.01 0.00

Slag Truck 35 20 45 Slag 82,516 1,900 1,485 1.60 1.19 0.06 0.16 0.12 0.01

Fines Truck 35 20 45 Fines 16,977 1,800 289 1.60 0.23 0.01 0.16 0.02 0.00

Maintenance 2 0 2 None 0 3,000 4,148 0.39 0.82 0.04 0.04 0.08 0.00

CO2 Processing 35 20 45 CO2 127,750 2,000 2,420 1.60 1.94 0.10 0.16 0.19 0.01

TOTAL: 8,450 4.32 0.22 0.43 0.02

Calculation of vehicle miles traveled (VMT) is based on the total annual potential throughput being hauled over the round-trip distance from the

Petmin property line to the storage piles. An estimated average of ten round-trips per day from the office to the farthest point in the plant is used

for maintenance work VMT.



Remet Storage Piles Petmin USA Inc.

Objective

Calculate annual potential to emit from storage piles.

Method


EF = k*0.0032*(U/5)^1.3/(M/2)^1.4 EF = 1.7*(s/1.5)*((365-p)/235)*(f/15) where: where:

k = 0.35 particle size multiplier for PM10 s = 0.5 silt content of stored material, wt. %

U = 8.86 mean wind speed, miles per hour p = 150 days/yr with > 0.01 inch precipitation

M = 1.5 % moisture content of material f = 30 % of time wind speed exceeds 12 mph

A = 1.0 total surface area of piles, acres

EF = 0.003525 pounds/ton

EF = 1.04 pounds/day/acre

2 load-in

2 load-out Control eff: 80% control efficiency for water

RACM, p. 2-47

Throughput = 8,000 metric tons / year theoretical throughput

8,818 short tons / year theoretical throughput

LOAD IN/LOAD OUT WIND EROSION

UNCONTROLLED TSP EMISSIONS = 0.06 tons per year UNCONTROLLED TSP EMISSIONS = 0.19 tons per year

80% control efficiency for water

RACM, p. 2-48

LOAD IN/LOAD OUT WIND EROSION

CONTROLLED TSP EMISSIONS = 0.012 tons per year CONTROLLED TSP EMISSIONS = 0.04 tons per year

PM

Uncontrolled PM Controlled

0.06 0.01 TPY

0.19 0.04 TPY

0.25 0.05 TPY

Mean wind speed from NNDC Global Climate Online: 1982 - 2011 data for Youngstown, OH

Iron ore pellets are received, stored, and loaded into Petmin's pellet handling system by Kinder-Morgan (K-M). Petmin has no pellet storage piles. Iron

ingots are stored by Petmin briefly before transfer to K-M or a third party. Remet is stored in two separate piles onsite. These temporary storage piles

are addressed below.

Remet consists of partially-reduced iron pellets. Moisture content is set at 1.5%, based on company information. Fines are estimated by the equipment

manufacturer at 0.5%. Remet is only produced during shutdown of the process and Petmin estimates that the maximum throughput is 8,000 metric tons

per year.

The pig iron from the process is cast into small ingots and placed in a stockpile prior to loading on trucks. The ingots handling of is not included in these

calculations because they have negligible fines and handling is not expected to result in negligible emissions.

These calculations predict emission rates below the de minimis threshold (OAC rule 3745-15-05) from all storage piles combined. Therefore, these units are listed in

the application as "de minimis."

EMISSION SUMMARY

Load-in/load-out =

Wind erosion =

Total emissions =

EMISSION FACTOR CALCULATION - REMET

LOAD IN/LOAD OUT

(AP-42, Table 13.2.4 (11/06))

WIND EROSION

(USEPA: Control of Open Fugitive Dust Sources, Eq. 4-9 (9/88))



Fines Handling Petmin USA Inc.

Objective

Calculate annual potential to emit from iron ore fines loading and transportation.

Method

EF = k*0.0032*(U/5)^1.3/(M/2)^1.4 where:

k = 0.74 particle size multiplier for PM

0.35 particle size multiplier for PM10

0.053 particle size multiplier for PM2.5

U = 8.86 mean wind speed, miles per hour

M = 1.5 % moisture content of material

PM EF = 0.0074524 pounds/ton

PM10 EF = 0.0035248 pounds/ton

PM2.5 EF = 0.0005338 pounds/ton

2 transfer points

Throughput = 16,977

Potential Emissions

PM 0.13 tons per year

PM10 0.06 tons per year

PM2.5 0.01 tons per year

Mean wind speed from NNDC Global Climate Online: 1982 - 2011 data for Youngstown, OH

Iron ore fines are separated from iron ore in a vibrating screener. The fines are then transported

to a fines bin outside the building on a conveyor. The fines are then loaded into a truck and

shipped offsite.

Calculations for the outdoor portions of this loadout process are based off of AP-42 13.2.4

EMISSION FACTOR CALCULATION - Fines Handling

(AP-42 13.2.4 (11/06))

These calculations predict emission rates below the de minimis threshold (OAC rule 3745-15-

05). Therefore, these units are listed in the application as "de minimis."

tons / year theoretical throughput



Emergency Equipment Petmin USA Inc.

Objective

Calculate annual potential to emit from diesel-fired emergency engines and pumps at the facility.

Method

Use AP-42 emission factors, except as noted.

Diesel Fired Emergency Generators (2)

The diesel fuel used in the units will contain less than 0.0015% sulfur.

BTU capacity of each engine is: 3,131 HP

All PM is assumed to be </= 1 μm in size and therefore condensable, AP-42 Section 3.3-6, Table 3.3-1 Footnote b.

Emission CalculationsTier 4 Standards >560 kW (except CO2 from AP-42 Table 3.4-1, 10/96)

Emission Factor Power Output

g/bhp-hr HP lb/hr each ton/yr (100 hr)

NOx 0.5 3,131 3.45 0.17

CO 2.6 3,131 17.95 0.90

PM/PM10 0.022 3,131 0.15 0.01

VOC 0.14 3,131 0.97 0.05

CO2 526.6 3,131 3,635 181.75

Emission Factor Sulfur Content Heat Content Fuel Usage

lb/MMBTU % MMBTU/gal gal/yr ton/month ton/yr

SO2 1.01 0.0015 0.1375 5,800 5.0E-05 6.0E-04

SO2 emissions factor obtained from AP-42, section 3.4, Table 3.4-1.

Diesel Fired Black Start Generator

The diesel fuel used in the unit will contain less than 0.0015% sulfur.

BTU capacity of this engine is 158 HP

Emission CalculationsTier 4 Standards, 56 - 130 kW (except CO2 from AP-42 Table 3.3-1, 10/96)

Emission Factor Power Output

g/bhp-hr HP lb/hr ton/yr (100)

NOx 0.3 158 0.10 5.22E-03

CO 3.7 158 1.29 6.44E-02

PM/PM10 0.015 158 5.22E-03 2.61E-04

VOC 0.14 158 0.05 2.44E-03

CO2 522.1 158 181.86 9.09

The engines are new emergency CI RICE located at an area source of HAPs. MACT-ZZZZ, 40 CFR 63.6590(c)(1), states

that such engines shall comply with MACT-ZZZZ by complying with NSPS-IIII. 40 CFR 60.4211(f) limits non-emergency

engine operation, including readiness testing, to 100 hours per year. 40 CFR 60.4211(c) requires the engine to meet Tier

4 standards.

Assume 100 hours of operation per year at rated capacity for each of two planned emergency generators and the black

start generator (consistent with MACT-ZZZZ and NSPS-IIII requirements).

The engine is a new black start CI RICE located at an area source of HAPs. MACT-ZZZZ, 40 CFR 63.6590(c)(1), states

that such engines shall comply with MACT-ZZZZ by complying with NSPS-IIII. 40 CFR 60.4211(c) requires the engine to

meet Tier 4 standards.

PollutantPotential Emission

PollutantPotential Emissions







SO2 1.01 0.0015 0.1375 770 6.7E-06 8.0E-05

SO2 emissions factor obtained from AP-42, section 3.4, Table 3.4-1.

Emergency Fire Fighting Diesel Pumps

Two Emergency Fire Fighting Pumps will be at the facility. One will be for high pressure and one will be for low pressure.

Each engine is a new RICE engine that burns ultra low sulfur diesel fuel.

Hours of operation per year: 100

1

Emission Factors for Stationary Fire Pump Engines

NOx+VOC 4.0 g/KW-hr 3.0 g/HP-hr

CO 3.5 g/KW-hr 2.6 g/HP-hr

PM 0.20 g/KW-hr 0.15 g/HP-hr

2

Emission Factors From Manufacturer (Clarke Fire Pump Engines)

High Pressure Pump (311 HP)

NOx 3.8 g/KW-hr 2.83 g/HP-hr



VOC 0.05 g/KW-hr 0.037 g/HP-hr

Low Pressure Pump (237 HP)

NOx 3.79 g/KW-hr 2.83 g/HP-hr



VOC 0.12 g/KW-hr 0.089 g/HP-hr

3

Emission Factors for Large Stationary Diesel Engines

SO2 3.7 g/HP-hr 1.01 lb/MMBTU

CO2 522.10 g/HP-hr 165 lb/MMBTU

NPSP Subpart IIII, Table 3

Engine Power: 130-225 kW (100-175 HP) and 225-450 kW (300-600 HP)

AP-42, section 3.4, Table 3.4.1

The emission factors were obtained from NSPS Subpart IIII Table 3, AP-42, section 3.4, Table 3.4.1, and a representative

manufacturer.

The calculations below are based on representative make and model of the engines since the actual fire pumps have not

been selected at this time.

Potential EmissionPollutant




High Pressure Pump

HP: 311

Emission Factor Power Output Potential Emissions

g/HP-hr HP lb/hr ton/yr (100)

NOx 2.83 2 311 1.94 0.0971

CO 2.6 1 311 1.78 0.0891

PM/PM10 0.15 1 311 1.03E-01 0.0051

VOC 0.037 2 311 0.03 0.0013

CO2 522.10 3 311 357.97 17.8987



SO2 1.01 0.0015 0.1375 8,400 7.3E-05 8.7E-04

Low Pressure Pump

HP: 237

Emission Factor Power Output Potential Emissions

g/HP-hr HP lb/hr ton/yr (100)

NOx 2.83 2 237 1.48 0.0738

CO 2.6 1 237 1.36 0.0679

PM/PM10 0.15 1 237 7.84E-02 0.0039

VOC 0.089 2 237 0.05 0.0023

CO2 522.10 3 237 272.80 13.6398



SO2 1.01 0.0015 0.1375 6,300 5.5E-05 6.6E-04

PollutantEmission Factor

Source



PollutantEmission Factor

Source



Waste Water Petmin USA Inc.

Objective

Calculate annual potential to emit from the ammonia scrubber located at the facility.

Method

Guaranteed scrubber performance provided in ppmv, provided by Tenova.


Emission Calculations

NH3

Emission Factor: 1 ppmv

Flowrate: 60,000 scf/hr

1 partsv 17.03 g 760 torr 1 mol K 1 1,000 mg 1,000 L 0.759 mg

1,000,000 partsv 1 mol 62.4 L torr 273.15 K 1 g 1 m3

1 Nm3

0.759 mg 2.20E-06 lb 1 Nm3

4.41E-08 lb

Nm3

1 mg 37.9 scf 1 scf

4.41E-08 lb 60,000 scf 2.64E-03 lb

1 scf 1 h 1 hr

2.64E-03 lb 1 ton 8760 hr 0.012 ton

1 hr 2000 lb 1 yr 1 yr

Ammonia is stripped out of facility's wastewater and processed

through a packed bed scrubber for neutralization and phase

x =

x x =

x x x x x

x =

x x =

Prepares by: Jake Leitnaker 11/21/2019

Reviewed by: Joe Hollowell 11/22/2019 13 of 15

Cooling Tower Calculator Petmin USA Inc.

Cooling Tower Particulate Air Emissions Calculator

COOLING TOWER SPECIFICATIONS:

Enter specifications into blue cells

Drift loss 0.0010%

Circulating water flow rate 10,510 gpm

Total dissolved solids 260 ppm

Density of TDS constituents 2.5 g/cc Average density of common salts (CaCO3, CaSO4, CaCl2, NaCl, Na2SO4, Na2CO3)

Volume of a sphere V = 4/3*π*r3

Annual drift 53 lb H2O/hr

ANNUAL EMISSIONS:

Total Particulate Emissions 0.019 ton/yr

PM10 Emissions 0.016 ton/yr

PM2.5 Emissions 0.005 ton/yr

Water Drop Size Distribution for Low Efficiency Drift Eliminators*

Based on a drift rate of 0.001%

Droplet H2O Droplet Solids Emissions

Dia. % mass Mass Vol. Dia. PM PM10 PM2.5

(micron) % mass smaller (g) (cc) (micron) (lb/hr) (lb/hr) (lb/hr)

22 0.43 0.43 5.6E-09 5.8E-13 1.0

29 1.49 1.92 1.3E-08 1.3E-12 1.4

44 3.76 5.68 4.5E-08 4.6E-12 2.1

58 2.09 7.77 1.0E-07 1.1E-11 2.7 7.8%

65 1.86 9.63 1.4E-07 1.5E-11 3.1

87 1.56 11.19 3.4E-07 3.6E-11 4.1

108 1.43 12.62 6.6E-07 6.9E-11 5.1

120 1.26 13.88 9.0E-07 9.4E-11 5.6

132 1.09 14.97 1.2E-06 1.3E-10 6.2

144 1.32 16.29 1.6E-06 1.6E-10 6.8

174 5.81 22.1 2.8E-06 2.9E-10 8.2

300 5.04 27.14 1.4E-05 1.5E-09 14.1 27.1%

450** 4.17 31.31 4.8E-05 5.0E-09 21.2 31.3%

600 4.01 35.32 1.1E-04 1.2E-08 28.2

750 4.00 39.32 2.2E-04 2.3E-08 35.3

900 4.03 43.35 3.8E-04 4.0E-08 42.3

1,050 4.57 47.92 6.1E-04 6.3E-08 49.4

1,200 5.46 53.38 9.0E-04 9.4E-08 56.4

1,350 6.80 60.18 1.3E-03 1.3E-07 63.5

2,250 17.99 78.17 6.0E-03 6.2E-07 105.8

2,400 21.83 100 7.2E-03 7.5E-07 112.9

* EPA. 1979. Effects of Pathogenic and Toxic Material Transport Via Cooling Device Drift - Vol. 1

Technical Report. EPA-600/7-79-251a. November 1979.

** Maximum droplet size governed by atmospheric dispersion. Larger droplets fall to the ground before

evaporating into a particle (EPA 1979).

Low Efficiency

Prepared by: Glen Greenwood

Reviewed by: Joe Hollowell 11/18/2019 Page 14 of 15

Cooling Tower Calculator Petmin USA Inc.

Water Drop Size Distribution for High Efficiency Drift Eliminators*

Based on a drift rate of 0.0003%

Droplet H2O Droplet Solids Emissions

Dia. % mass Mass Vol. Dia. PM PM10 PM2.5

(micron) smaller (g) (cc) (micron) (lb/hr) (lb/hr) (lb/hr)

10 0 5.2E-10 5.4E-14 0.5

20 0.196 4.2E-09 4.4E-13 0.9

30 0.226 1.4E-08 1.5E-12 1.4

40 0.514 3.4E-08 3.5E-12 1.9

50 1.816 6.5E-08 6.8E-12 2.4

60 5.702 1.1E-07 1.2E-11 2.8 5.7%

70 21.348 1.8E-07 1.9E-11 3.3

90 49.812 3.8E-07 4.0E-11 4.2

110 70.509 7.0E-07 7.2E-11 5.2

130 82.023 1.2E-06 1.2E-10 6.1

150 88.012 1.8E-06 1.8E-10 7.1

180 91.032 3.1E-06 3.2E-10 8.5

210 92.468 4.8E-06 5.0E-10 9.9

240 94.091 7.2E-06 7.5E-10 11.3 94.1%

270 94.689 1.0E-05 1.1E-09 12.7

300 96.288 1.4E-05 1.5E-09 14.1

350 97.011 2.2E-05 2.3E-09 16.5

400 98.34 3.4E-05 3.5E-09 18.8

450** 99.071 4.8E-05 5.0E-09 21.2 99.1%

500 99.071 6.5E-05 6.8E-09 23.5

600 100 1.1E-04 1.2E-08 28.2

* Reisman, J. and G. Frisbie. 2002. “Calculating Realistic PM10 Emissions from Cooling Towers.”

Environmental Progress & Sustainable Energy. American Institute of Chemical Engineers. Volume 21,

Issue 2, pp. 127-130. July 2002.

** Maximum droplet size governed by atmospheric dispersion. Larger droplets fall to the ground before

evaporating into a particle (EPA 1979).

EXAMPLE CALCULATIONS: Low Efficiency

Annual drift:

10,510 gal water 8.33 lb 60 min 0.001% (drift) = 53 lb water drift

1 min 1 gal water 1 hr hr

Total Particulate Emissions

53 lb water 260 lb PM 31.3% PM = 0.004 lb PM = 0.019 ton PM

hr 1E+6 lb water hr yr

PM10 Emissions

53 lb water 260 lb PM 27.1% PM10 = 0.004 lb PM10 = 0.016 ton PM10


PM2.5 Emissions

53 lb water 260 lb PM 7.8% PM2.5 = 0.001 lb PM2.5 = 0.005 ton PM2.5


Prepared by: Glen Greenwood

Reviewed by: Joe Hollowell 11/18/2019 Page 15 of 15

PTI/PTIO Application A0066143 Petmin USA Incorporated ... · 1. 2. 3.--Petmin USA Incorporated -...

Documents

Transcript of PTI/PTIO Application A0066143 Petmin USA Incorporated ... · 1. 2. 3.--Petmin USA Incorporated -...