Primary Reformer Bact 8-9-19
Primary Reformer Bact 8-9-19
Primary Reformer Bact 8-9-19
Analysis Update
Agrium Kenai, Alaska
Kenai Nitrogen Operation
5 August 2019
Project No.: 0497868
Version 1.0
Document history
www.erm.com Version: 1.0 Project No.: 0497868 Client: Agrium Kenai Nitrogen Operation 5 August 2019
P:\Projects\0497868 Nutrien Kenai Air Permit Rev.DG\Agency Request\Response Documents\Primary Reformer BACT 8.9.19.docxP:\Projects\0497868 Nutrien Kenai Air
Permit Rev.DG\Agency Request\Attachment C BACT jrk 8.2.19 rev.docx
Signature Page
5 August 2019
Jason Krawczyk
Project Manager
© Copyright 2019 by ERM Worldwide Group Ltd and / or its affiliates (“ERM”).
All rights reserved. No part of this work may be reproduced or transmitted in any form,
or by any means, without the prior written permission of ERM.
www.erm.com Version: 1.0 Project No.: 0497868 Client: Agrium Kenai Nitrogen Operation 5 August 2019
P:\Projects\0497868 Nutrien Kenai Air Permit Rev.DG\Agency Request\Response Documents\Primary Reformer BACT 8.9.19.docxP:\Projects\0497868 Nutrien Kenai Air
Permit Rev.DG\Agency Request\Attachment C BACT jrk 8.2.19 rev.docx
PRIMARY REFORMER BACT ANALYSIS UPDATE CONTENTS
Kenai Nitrogen Operation
CONTENTS
1. BEST AVAILABLE CONTROL TECHNOLOGY (BACT) BACKGROUND ................................ 2
1.1 Introduction .................................................................................................................................... 2
1.2 Regulatory Basis for BACT Analysis.............................................................................................. 2
1.3 Five-Step Top-Down BACT Process ........................................................................................... 33
www.erm.com Version: 1.0 Project No.: 0497868 Client: Agrium Kenai Nitrogen Operation 5 August 2019 Page i
P:\Projects\0497868 Nutrien Kenai Air Permit Rev.DG\Agency Request\Response Documents\Primary Reformer BACT
8.9.19.docx\\usinddc01\Data\Indianapolis\Projects\0497868 Nutrien Kenai Air Permit Rev.DG\Application Documents\Attachment C BACT Documents\Attachment C
BACT 5.16.19.au.docx
PRIMARY REFORMER BACT ANALYSIS UPDATE BEST AVAILABLE CONTROL TECHNOLOGY (BACT) BACKGROUND
Kenai Nitrogen Operation
1.1 Introduction
Agrium U.S. Inc. (Agrium) was issued Air Quality Control Construction Permit AQ0083CPT06 on
6 January 2015 for the proposed restart of a portion of it fertilizer production facility (Facility) at the
Kenai Nitrogen Operation in Kenai, Alaska. In a letter dated 4 March 2016, the Alaska Department of
Environmental Conservation (ADEC) extended the deadline by which construction must commence
by eighteen (18) months until 6 January 2018. In a second letter dated 3 October 2017, the ADEC
extended the deadline by which construction must commence by an additional eighteen (18) months
until 6 July 2019.
Since the issuance of the ADEC letter dated 3 October 2017, Agrium has decided to replace the five (5)
existing 37.6 MMBtu/hr Solar Turbines identified as Units 55, 56, 57, 58, and 59. The replacement
Solar Turbines will each have a maximum rated heat input capacity of 55.443 MMBtu/hr. The new
Solar Turbines will utilize the existing Waste Heat Boilers (Units 50, 51, 52, 53, and 54) for heat
recovery. Due to the increase in heat input capacities of the new Solar Turbines, the required
supplemental heat input capacity of the 50.0 MMBtu/hr Waste Heat Boilers have decreased. The
Waste Heat Boilers once integrated with the new Solar Turbines, will now only have heat input
capacities of 46.729 MMBtu/hr, each. Since the heat input capacities of the Waste Heat Boilers are
changing, as are the potential emissions, Agrium is providing updated top-down Best Available Control
Technology (BACT) analyses for these affected units, in addition to the top-down BACT analyses for
the new Solar Turbines.
In addition, Agrium is proposing to install Selective Catalytic Reduction (SCR) for Nitrogen Oxides
(NOx) control on the Package Boilers (Units 44, 48, and 49). These emission units went through
Prevention of Significant Deterioration (PSD) BACT as part of the permitting for AQ0083CPT06.
Under the Air Quality Control Construction Permit, BACT for NOx was identified as use of ultra low
NOx burners. SCR is considered to provide the same, if not a higher, control efficiency than the use of
ultra low NOx burners.
The BACT reviews for the affected units identified above at the Facility i were included as Attachment
C to the 2019 PSD permit application.
In an email dated 26 July 2019, Mr. Dave Jones, Environmental Engineering Assistant I, ADEC – Air
Quality – Juneau, requested an updated BACT for the Primary Reformer. The ADEC identified a
stationary source in the RACT/BACT/LAER Clearinghouse (RBLC) (Emberclear Gas to Liquids, RBLC
ID No. MS-0092) with a steam methane reformer using an oxidation catalyst to control carbon
monoxide (CO) emissions down to 5 ppmv at 3% oxygen. RBLC ID No. MS-0092 had not been
entered in to the RBLC at the time of the initial permitting for AQ0083CPT06. During the initial PSD
permitting, oxidation catalysts were determined to be not technically feasible. Since an oxidation
catalyst can also be used as control for volatile organic compound (VOC) emissions, and RBLC ID
No. MS-0092 included a limit based on the use of catalytic oxidation for this pollutant, a VOC BACT
analysis has been included in this document. Section 3.0 of this document contains the CO and VOC
BACT analyses for the Primary Reformer (Unit 12). Section 4.0 contains an evaluation of new RBLC
results for PM, PM10, PM2.5, NOx, and CO2e control requirements associated with permits issued
since the original PSD permit was issued in January 2015.
“An emission limitation based on the maximum degree of reduction of each pollutant subject
to regulation under [the CAA] emitted from or which results from any major emitting facility,
which the permitting authority, on a case-by-case basis, taking into account energy,
environmental, and economic impacts and other costs, determines is achievable for such
facility through application of production processes and available methods, systems, and
www.erm.com Version: 1.0 Project No.: 0497868 Client: Agrium Kenai Nitrogen Operation 5 August 2019
PRIMARY REFORMER BACT ANALYSIS UPDATE BEST AVAILABLE CONTROL TECHNOLOGY (BACT) BACKGROUND
Kenai Nitrogen Operation
techniques, including fuel cleaning, clean fuels, or treatment or innovative fuel combustion
techniques for control of each such pollutant.”
Based on projected potential emission rates, BACT is required for the following criteria pollutants:
Nitrogen Oxides (NO X)
Carbon Monoxide (CO)
Volatile Organic Compounds (VOC)
Particulate Matter (PM)
Particulate Matter ≤ 10 microns in aerodynamic diameter (PM10)
Particulate Matter ≤ 2.5 microns in aerodynamic diameter (PM2.5)
In addition, the proposed project is subject to a BACT review for the greenhouse gas (GHG)
pollutants under EPA’s Tailoring Rule. The regulated GHGs include the following:
Carbon Dioxide (CO2)
Methane (CH4)
Nitrous Oxide (N2O)
Carbon Dioxide Equivalent (CO2e)
Where CO2e represents the CO2 equivalence of the emissions. CO2e emissions are calculated as
the sum of the mass emissions of each individual GHGs adjusted for its respective global warming
potential (GWP). The GWP values are included in Table A-1 of the Greenhouse Gas Mandatory
Reporting Rule found in 40 CFR 98, Subpart A.
www.erm.com Version: 1.0 Project No.: 0497868 Client: Agrium Kenai Nitrogen Operation 5 August 2019
PRIMARY REFORMER BACT ANALYSIS UPDATE BEST AVAILABLE CONTROL TECHNOLOGY (BACT) BACKGROUND
Kenai Nitrogen Operation
The third step is to rank the technologies not eliminated in Step 2 in order of descending control
effectiveness for each pollutant of concern. If the highest ranked technology is proposed as BACT, it
is not necessary to perform technical or economic evaluation of the selected or less effective control
technologies identified as outlined in Step 4. Potential adverse impacts, however, must still be
identified and evaluated.
The fourth step entails an evaluation of energy, environmental, and economic impacts for determining
a final level of control. The evaluation begins with the most stringent control option and continues until
a technology under consideration cannot be eliminated based on adverse energy, environmental, or
economic impacts. The economic or “cost-effectiveness” analysis is conducted in a manner consistent
with EPA’s OAQPS Control Cost Manual, Sixth Edition1 and subsequent revisions.
Cost effectiveness is expressed in terms of dollars per ton of pollutant removed ($/ton). The costs in
the numerator of that expression are determined by adding the annualized capital cost and the annual
operation and maintenance costs of a given control device under evaluation. Annualized costs are
determined by the following equation:
Where:
PV = Present value of the equipment;
i = Interest rate (cost of money); and
n = Number of years of the life of the equipment.
The annual mass (ton) of pollutant removed is determined by multiplying the annual uncontrolled
emission rate by the expected control efficiency. The uncontrolled emission rate may, in some cases,
be the rate after some level of control. In addition, the annual emission rate may be the potential to
emit, or a level based on limited hours of operation.
The fifth and final step is to select as BACT the emission limit from application of the most effective of
the remaining technologies under consideration for each pollutant of concern.
1 USEPA, OAQPS Control Cost Manual, Sixth Edition (Research Triangle Park, NC, 2002)
www.erm.com Version: 1.0 Project No.: 0497868 Client: Agrium Kenai Nitrogen Operation 5 August 2019
PRIMARY REFORMER BACT ANALYSIS UPDATE SUMMARY OF AFFECTED EMISSION UNITS AND POLLUTANTS
Kenai Nitrogen Operation
www.erm.com Version: 1.0 Project No.: 0497868 Client: Agrium Kenai Nitrogen Operation 5 August 2019
PRIMARY REFORMER BACT ANALYSIS UPDATE CRITERIA POLLUTANT BEST AVAILABLE CONTROL TECHNOLOGY
Kenai Nitrogen Operation (BACT) ANALYSIS
The auxiliary section of the Primary Reformer is subject to NSPS Subpart D, under terms of 1998
Consent Decree. EPA determined that even though the primary function of the reformer is to reform
process gas, the auxiliary section is a discrete unit whose primary function is to produce steam.
Review of the RBLC database identified two control technologies for control of CO emissions from
reformers. The use of Good Combustion Practices (GCP) is cited as BACT for nearly every entry;
however, the use of an oxidation catalyst is cited in a single RBLC entry (Emberclear Gas to Liquids
(Emberclear), RBLC ID No. MS-0092). Emission limits range from 0.0194 lb/mmBtu to 0.06 lb/mmBtu
for natural gas combustion using GCP. The CO limit for the one unit equipped with an oxidation
catalyst was identified as 5 ppmv at 3% O2. Available control technologies for the control of CO
emissions include good combustion practices, oxidation catalyst, and thermal oxidation.
Oxidation Catalyst
Oxidation catalysts use a noble metal catalyst to reduce the activation energy of the oxidation
reaction:
2CO + O2 → 2CO2
Although oxidation catalysts are used to reduce CO emissions from natural gas-fired combustion
turbines, they have limited demonstration in reducing CO emissions from natural gas-fired boilers and
have not been demonstrated for natural gas reformers. The note in the RBLC for Emberclear’s steam
methane reformer’s oxidation catalyst BACT verification status states that demonstration of
www.erm.com Version: 1.0 Project No.: 0497868 Client: Agrium Kenai Nitrogen Operation 5 August 2019
PRIMARY REFORMER BACT ANALYSIS UPDATE CRITERIA POLLUTANT BEST AVAILABLE CONTROL TECHNOLOGY
Kenai Nitrogen Operation (BACT) ANALYSIS
compliance with the CO emission limitation has not been verified; however, this technology is carried
forward for control of CO emissions from the Primary Reformer.
Thermal Oxidation
Thermal oxidation has never been required nor used on a natural gas-fired reformer, and the
effectiveness of the technology in reducing CO emissions from natural gas-fired reformer is
questionable. Thermal oxidation would involve injecting additional air into the flue gas and heating the
oxygen enriched mixture to approximately 1,500 °F to oxidize CO to carbon dioxide. However, since
the combustion of the reheat fuel would itself result in CO emissions, there is no evidence that thermal
oxidation would result in overall reductions in CO emission.
Since thermal oxidation has never been demonstrated on a natural gas-fired reformer, and because
there is no evidence that it could reduce CO emissions, thermal oxidation is not a technically feasible
CO control technology for the Primary Reformer.
3. High temperatures and low oxygen levels in the primary combustion zone
4. High enough overall excess oxygen levels to complete combustion and maximize thermal
efficiency
5. Proper fuel gas supply system designed to minimize effects of contaminants or fluctuations in
pressure and flow on the fuel gas delivered
Combustion efficiency is dependent on the gas residence time, the combustion temperature, and the
amount of mixing in the combustion zone. Each of these parameters is incorporated into the design of
the burners and the combustion zones of the Primary Reformer to optimize combustion and minimize
fuel consumption. In addition to the above parameters the level of oxygen in the Primary Reformer is
important to GCP. Therefore, combustion control is accomplished primarily through reformer design
as it relates to time, temperature, mixing, and through reformer operation as it relates to excess
oxygen levels. Combustion design for modern reformers is intended to simultaneously minimize
formation of CO and NOx emissions. This is a difficult task, since emissions of NOx and emissions of
CO are inversely related. That is, measures used to reduce NOx emissions often lead to increases in
CO emissions. Therefore, the reformer design to minimize CO emissions is interrelated with the
reformer design to minimize NOx formation.
GCPs are planned for the Primary Reformer at the facility and represent the baseline BACT for the
Primary Reformer; therefore, an oxidation catalyst represents the highest ranked level of control for
CO emissions from the Primary Reformer.
A cost evaluation for a catalytic oxidizer on the Primary Reformer is has been performed as part of
this BACT analysis. The estimate results in a cost per ton of CO removed of $16,200 per ton. This
level of cost is considered to be economically infeasible; therefore, catalytic oxidation is eliminated for
consideration as representing BACT for CO emissions.
www.erm.com Version: 1.0 Project No.: 0497868 Client: Agrium Kenai Nitrogen Operation 5 August 2019
PRIMARY REFORMER BACT ANALYSIS UPDATE CRITERIA POLLUTANT BEST AVAILABLE CONTROL TECHNOLOGY
Kenai Nitrogen Operation (BACT) ANALYSIS
Agrium proposes the use of Good Combustion Practices as the BACT for CO emissions from the
Primary Reformer. CO Emissions from the Primary Reformer will be limited to 43.45 lb/mmcf for a 3-
hour average. Initial compliance with the proposed emission limit will be demonstrated by conducting
a stack test.
3.1.2 BACT Evaluation for VOC Emissions from the Primary Reformer
Review of the RBLC database identified two control technologies for control of VOC emissions from
reformers. The use of Good Combustion Practices (GCP) is cited as BACT for nearly every entry;
however, the use of an oxidation catalyst is cited in a single RBLC entry (Emberclear Gas to Liquids
(Emberclear), RBLC ID No. MS-0092). Emission limits range from 0.0014 lb/mmBtu to 0.0055
lb/mmBtu for natural gas combustion using GCP. The VOC limit for the one unit equipped with an
oxidation catalyst was identified as 5 ppmv at 3% O2. Available control technologies for the control of
VOC emissions include GCP, oxidation catalyst, and thermal oxidation.
For the same reasons given for CO control from the Primary Reformer exhaust, thermal oxidation is
eliminated from further consideration.
The note in the RBLC for Emberclear’s steam methane reformer’s oxidation catalyst BACT verification
status states that demonstration of compliance with the VOC emission limitation is unknown; however,
this technology is carried forward for control of VOC emissions from the Primary Reformer.
GCPs are planned for the Primary Reformer at the facility and represent the baseline BACT for the
Primary Reformer; therefore, an oxidation catalyst represents the highest ranked level of control for
VOC emissions from the Primary Reformer.
A cost evaluation for a catalytic oxidizer on the Primary Reformer is has been performed as part of
this BACT analysis. The estimate results in a cost per ton of COVOC removed of $158,102 per ton.
This level of cost is considered to be economically infeasible; therefore, catalytic oxidation is
eliminated for consideration as representing BACT for VOC emissions.
www.erm.com Version: 1.0 Project No.: 0497868 Client: Agrium Kenai Nitrogen Operation 5 August 2019
PRIMARY REFORMER BACT ANALYSIS UPDATE BEST AVAILABLE CONTROL TECHNOLOGY (BACT) ANALYSIS
Kenai Nitrogen Operation UPDATES
This section of the analysis is provided as a supplement to the BACT analyses performed for the
original PSD Construction Permit application for KNO, submitted in October 2014. This section
provides an evaluation of RACT/BACT/LAER Clearinghouse (RBLC) results associated with permits
issued since the original PSD permit was issued in January 2015. Based on the information provided
below, KNO concludes that no new permits have been issued since the issuance of AQ0083COT06
that contain BACT limits for PM, PM10, PM2.5, NOx, and CO2e that are inconsistent with the BACT
determinations made for KNO as part of the original PSD Construction Permit.
Tables summarizing RBLC entries since the issuance of AQ0083COT06 were provided in Attachment
B to the 2019 PSD permit application. The results of all three analyses for emission units contained in
the KNO PSD permit are summarized below:
www.erm.com Version: 1.0 Project No.: 0497868 Client: Agrium Kenai Nitrogen Operation 5 August 2019
PRIMARY REFORMER BACT ANALYSIS UPDATE
Kenai Nitrogen Operation