Nitric Acid New Source Performance Standards – EPA’s Responses to
the

Petition for Reconsideration

Background 

The initial Nitric Acid Plants New Source Performance Standards (NSPS)
were published in the Federal Register (FR) on December 23, 1971 (36 FR
24881) and codified at 40 CFR part 60, subpart G, pursuant to section
111 of the Clean Air Act (CAA). Pursuant to section 111(b)(1)(B) of the
CAA, we have reviewed the NSPS three times. The first review was
completed on June 19, 1979 (44 FR 35265). The second review was
completed on April 5, 1984 (49 FR 13654). No changes were made to the
NSPS as a result of these two reviews. However, based on the results of
the third review, which we completed in August 2012, we determined it
was appropriate to revise the NSPS. The revisions to the NSPS based on
the third review were published on August 14, 2012 (77 FR 48433). The
revisions included a change in the nitrogen oxides (NOx) emission limit,
from 3.0 pounds of NOx per ton of nitric acid production (3.0 lb/T) on a
3-hr basis to 0.5 lb/T on a 30-day average basis, and additional testing
and monitoring requirements.

Throughout the rulemaking process for this third review, we received
comments, data and information that supported these revisions. This
information is contained in the docket which is available at
http://www.regulations.gov. The revisions were proposed on October 14,
2011 (76 FR 63878). We received additional data and comments during the
comment period. These data and comments were considered and analyzed
and, where appropriate, revisions to the NSPS were made. The revised
NSPS (40 CFR part 60, subpart Ga) were finalized on August 14, 2012 (77
FR 48433). The NSPS apply to new, modified or reconstructed nitric acid
production units (NAPUs) that commence construction, modification or
reconstruction after October 14, 2011. 

Section 307(d)(7)(B) Requirements

Section 307(d)(7)(B) of the CAA states that “[o]nly an objection to a
rule or procedure which was raised with reasonable specificity during
the period for public comment (including any public hearing) may be
raised during judicial review. If the person raising an objection can
demonstrate to the Administrator that it was impractical to raise such
objection within such time or if the grounds for such objection arose
after the period for public comment (but within the time specified for
judicial review) and if such objection is of central relevance to the
outcome of the rule, the Administrator shall convene a proceeding for
reconsideration of the rule and provide the same procedural rights as
would have been afforded had the information been available at the time
the rule was proposed.” 

As to the first procedural criterion for reconsideration, a petitioner
must show why the issue could not have been presented during the comment
period, either because it was impractical to raise the issue during that
time or because the grounds for the issue arose after the period for
public comment (but within 60 days of publication of the final action).
In the EPA’s view, an objection is of central relevance to the outcome
of the rule only if it provides substantial support for the argument
that the promulgated regulation should be revised. See, e.g., the
EPA’s “Denial of the Petitions to Reconsider the Endangerment and
Cause or Contribute Findings for Greenhouse Gases under Section 202 of
the Clean Air Act,” 75 FR 49556, 49561 (August 13, 2010). See also, 75
FR 49556, 49560 – 49563 (August 13, 2010) and 76 FR 4780, 4786 –
4788 (January 26, 2011) for additional discussion of the standard for
reconsideration under CAA section 307(d)(7)(B).

On October 10, 2012, Dyno Nobel Inc. (DNI) submitted a Petition for
Reconsideration and judicial review of the final rule for Nitric Acid
Plants (77 FR 48433). The EPA has carefully reviewed the points raised
in the Petition and is denying the Petition for the reasons described
below. 

A.  Subcategory for Modified or Reconstructed Plants that Use NSCR to
Control NOx 

Claim: The petitioner states that “as EPA long ago recognized,”
nonselective catalytic reduction (NSCR) is not an end of pipe control
technology, as some would describe selective catalytic reduction (SCR).
The petitioner claims that there are significant and fundamental
differences between NAPUs using SCR and NSCR. NSCR is designed
differently from an SCR because an NSCR is built into, and is an
integral part of the operation of a NAPU, particularly in terms of heat
balance and energy generation for other parts of the NAPU. NAPUs with an
NSCR must keep temperatures elevated in order to create the reductive
environment within the unit that eliminates NOx. To ensure temperatures
and pressures are adequate for this reductive environment, the NSCR is
placed upstream of the tail gas expander. Because of these differences,
the petitioner believes the EPA should have established a subcategory
for NAPUs with NSCR with regard to emissions limits for modified or
reconstruction of existing sources.  

EPA’s Response: Section 111(b)(2) of the CAA states that “The
Administrator may distinguish among classes, types, and sizes within
categories of new sources for the purpose of establishing such
standards.” Pursuant to this discretionary authority, the general EPA
policy is that if one or more segments of an industry have significant
differences in class, type or size of production processes or designs,
which result in significant differences in the potential emissions, then
we consider developing a subcategory or subcategories for these
different segments. However, we believe it is inappropriate to establish
subcategories based on differences in control technologies, and decline
to exercise our discretion to do so here. 

As we explained in the preambles to the proposed and final rules, the
EPA clearly considers NSCR, as well as SCR, to be emissions control
technologies. Therefore, for this reason alone, it would not be
appropriate to establish a subcategory for units with NSCR. We are
unaware of the basis of DNI’s claim that the EPA “recognized” NSCR
as not being an emissions control technology. This is incorrect. None of
the 14 NAPUs in the U.S. using NSCR has additional NOx control after
NSCR. NSCR is the emissions control technology for these 14 NAPUs. There
are about 50 NAPUs in the U.S. The petitioner owns and operates seven of
the 14 NSCR units.

A possible justification for developing a subcategory (or a different
emissions limit) for existing NAPUs that become subject to this NSPS
through modification or reconstruction would be a determination that it
is technically impossible or cost prohibitive for a number of existing
NAPUs with NSCR to achieve the NSPS emission limit if these units are
modified or reconstructed. As described below in section C, there are
examples where some existing NAPUs with NSCR already achieve the NSPS
emission limit and, therefore, would have no additional control costs if
they were modified or reconstructed. Furthermore, the costs for most
other existing NSCR units to comply with the NSPS, if such units were
modified or reconstructed, are not expected to be excessive (as
described further in section D below). If it were the case that the
costs to comply with the NSPS for one specific NAPU with NSCR were
excessive, that would not by itself justify developing a subcategory for
one, or for a very limited number, of the NSPS affected facilities.
Thus, there is no basis for subcategorizing units using NSCR from units
using SCR. 

B.  Petitioner Believes EPA had Insufficient Data

Claim: The petitioner claims that the EPA cannot meet the legal
requirement that data relied on for setting an NSPS must be sufficiently
“representative of potential industry–wide performance” because
none of the plants that were tested for the NSPS review (i.e., test
plants) use NSCR. Also, the test plants in the EPA’s analysis do not
face the conditions, limitations in the ability to modify plant
operations, or the resulting impacts on NOx emissions that NAPUs with
NSCR face. The petitioner claims the EPA cannot meet the mandates of
National Lime Association v. EPA (DC Circuit 1981) and Sierra Club v.
Costle (DC Circuit 1981) that data used in setting an NSPS must be
sufficiently “representative of potential industry-wide
performance.”

EPA’s Response: As required under section 111(a) of the CAA, 1990
Amendments, NSPS emission limits are based on the emission levels
achieved by the best systems of emission reduction (BSER) while
considering the costs and any nonair quality health, environmental
impact and energy requirements of these systems. As noted in the
preamble to the final rule (77 FR 48436), we concluded that SCR is BSER
for NAPUs. As discussed in detail in the preambles to the proposed and
final rule as well as the Response to Comment document, we determined we
had ample emission test data to support SCR as BSER and that we had
sufficient data to determine the appropriate revised emissions limit for
NOx. However, we also explained that the NSPS does not require the use
and installation of a specific control device as long as the affected
facility complies with the NSPS. For example, as described in more
detail in section C below, available data indicate that some NAPUs with
NSCR can meet the revised limit without the need for any additional
controls. 

C.  Achievability of Complying with Standard for Existing Plants Using
NSCR 

Claim: The petitioner claims that there are insufficient data showing
that existing plants using NSCR can meet the emission limit. The
petitioner notes that there was no notice in the proposed rule that the
EPA would use continuous emissions monitoring system (CEMS) data from
DNI’s Oregon plant as the primary basis for supporting the rule for
NSCR. The petitioner believes that this plant is not representative of
NAPU using NSCR in the U.S. for at least three reasons. First, the plant
is very small, only 70 tons per day (TPD) of nitric acid is produced.
DNI’s largest NAPU produces 520 TPD. Second, the plant is not a
typical NAPU as it was built for the sole function of scrubbing
emissions from an on-site urea plant. This plant functions without
maximizing acid purity or production rate as do other NAPUs. Thus, the
relative vessel sizes and flow rates that are found in an NSCR used at a
commercial NAPU are much different than DNI’s Oregon plant. Third, the
fuel used for the plant’s NSCR (purge gas from an on-site ammonia
plant – 65-percent hydrogen) is unusual. Four of DNI’s seven NAPUs
with NSCR use methane which, according to the petitioner, is the common
fuel for NSCRs at NAPUs. The petitioner also claims that the use of
hydrogen in fuel results in significantly lower NOx emissions at the
Oregon plant. Furthermore, the petitioner states that the only other
data cited by the EPA to support that the NSPS emissions limit is
achievable for NAPUs with NSCR are monthly averages from the JR Simplot
plant, which, according to the EPA, indicate a “strong probability”
that this NAPU with NSCR would comply with this NSPS.

 

EPA’s Response: As noted in the preambles to the proposed and final
rules, the EPA has determined that SCR is BSER, so there is no
requirement that the EPA conclude that NAPUs with NSCR can meet the
revised limit without additional controls. Nevertheless, we still
evaluated the potential for NAPUs with NSCR to achieve the revised
limit. As explained further below, we disagree that there are
insufficient data showing that some existing plants with NSCR can
achieve the revised limit.  

As noted in the preamble to the proposed rule, we had long term CEMS
data from two NAPUs with NSCR which indicated difficulty in meeting the
NSPS emission limit. However, we also noted we had monthly block
averages (often 30-day periods, which are the same time period of the
NSPS emission limit) from two other facilities using NSCR, including
DNI’s Oregon plant (also known as the Deer Island plant or St. Helens
plant) and the JR Simplot-Helm plant, that indicated these plants are
achieving the NSPS limit. We noted that the monthly averages for both
plants ranged from 0.15 to 0.36 lb NOx/T of 100-percent acid; well below
the emission standard of 0.50 lb/T. We also noted that since these data
were not rolling 30-day averages like the NSPS emission limit, but
rather monthly averages, it was not exactly a direct comparison with
CEMS data. However, we stated that these monthly averages indicate that
at least some NAPUs with NSCR would be able to achieve the emission
limit without additional controls (76 FR 63882). 

In the preamble to the final rule (77 FR 48437), we provided additional
information and analyses that indicate the same two NAPUs with NSCR
described in the proposal (i.e., DNI’s Oregon Plant and JR
Simplot-Helm plant) can meet the NSPS emission limit. Regarding the DNI
Oregon plant, we received CEMS data for this plant after proposal that
allowed us to calculate 30-day rolling averages. Based on these data, we
determined that the maximum rolling 30-day emission rate was 0.21 lb/T
acid for the DNI Oregon plant, which is well below the NSPS limit of
0.50 lb/T.  Regarding the JR Simplot-Helm plant, we explained again
that, based on available data, this plant had monthly averages that
ranged from 0.15 to 0.36 lb/T, which indicates this unit can also
achieve the limit of 0.50 lb/T. Therefore, similar to the proposal, we
concluded again in the preamble to the final rule that the NOx emission
limit was achievable for some NAPUs with NSCR.  

Since the final rule, we have re-analyzed the JR Simplot data, which
includes CEMS data. This re-analysis of JR Simplot CEMS data showed that
it is possible to determine 30-day rolling averages for this unit. For
the period, January 1, 2011, thru August 1, 2011, the 30-day rolling
averages range from 0.063 to 0.19 lb/T. The average 30-day average was
0.11 lb/T. Thus, this unit, which is controlled with NSCR, easily meets
the NSPS emission limit.  

With this re-analysis of the JR Simplot data, we have CEMS data from two
NAPUs with NSCR that clearly achieve the NSPS limit. Parameters for the
JR Simplot plant refute DNI’s contention that the DNI plant in Oregon
is only able to meet the emission limit because it is non-representative
of NAPUs with NSCR for the three points that DNI mentions. The JR
Simplot plant is larger (230 TPD) than DNI’s Oregon plant (41 TPD).
The sole plant function at JR Simplot is to produce nitric acid. It also
uses a fuel for NSCR (butane) that is similar to methane used in four of
the seven NSCR controls at DNI plants. 

While reviewing DNI’s reconsideration petition, we requested CEMS data
for DNI’s other plants with NSCR. DNI submitted data for two of their
other units with NSCR, including the only unit with NSCR that DNI
identified as being considered for modification (i.e., Unit 3 at DNI’s
Cheyenne, Wyoming plant). These data indicate that for many 30-day
periods, Unit 3 had emissions below the NSPS limit. We acknowledge that
this existing NAPU has not achieved emissions below the NSPS limit for
all periods in the past. However, that does not mean that they would not
be able to comply with the NSPS emission limit in the future if they
decide to modify this unit, especially if they take reasonable measures
to improve process operations and/or add emissions controls during the
modification or reconstruction process.

It is uncertain what equipment and/or controls DNI would have to install
in order to be in continuous compliance with the NSPS if they decided to
modify or reconstruct this unit. Nevertheless, we expect that by adding
a pre-heater, they could significantly reduce their NOx emissions during
startups and shutdowns. Furthermore, by changing the catalyst more
frequently, they could decrease their NOx emissions. We believe by
applying these two relatively inexpensive measures, it is likely that
this DNI unit could be in continuous compliance with the revised NSPS
limit. Otherwise, the other option would be to install an SCR control
which would be more expensive, but still reasonable. Costs for
retrofitting an SCR are described below in section D.

 

D.  Costs of Complying for Modified or Reconstructed NAPUs Currently
Using NSCR 

Claim: The petitioner claims that it would not be possible to meet the
NOx standard without cost-prohibitive retrofits for existing NAPUs with
NSCR. The petitioner stated that it is considering modifying a plant
that currently uses NSCR and claims that retrofit costs required for an
NSCR plant to install SCR and meet the NSPS were not considered in the
rulemaking.

The petitioner also claims the EPA stated in its proposed rule that it
was “uncertain” whether NAPUs with NSCR could meet the NOx emission
limit as the EPA had not reviewed any CEMS data for these units that met
this limit. The petitioner states that the EPA did not evaluate the
additional steps that are required for these units to meet the emission
limit. Also, according to the petitioner, the EPA did not evaluate the
impacts these steps would have on operations and the costs for these
steps.  

In December 2012, after DNI submitted its petition, in response to
questions from the EPA, DNI supplied information on estimated costs for
its Donora, PA plant to retrofit SCR to an existing NAPU with NSCR. With
these estimates, DNI asserts that the total annual cost to retrofit the
Donora unit to install an SCR would be an estimated $1,565,000, and that
the cost effectiveness of this control would be $11,000 per ton of NOx
removed.

EPA’s Response: As noted in the preamble to the final rule (77 FR
48439), we evaluated the costs of controls for an actual NAPU plant that
was modified. The plant installed SCR at a capital cost of $2,700,000.
The cost effectiveness was $1,360/T of NOx removed. We concluded that
these costs and cost effectiveness were reasonable.

The EPA concludes that retrofitting a NAPU with an NSCR to add SCR (or
add other control measures) to meet the NSPS would not, in the vast
majority of circumstances, incur substantially more costs than
retrofitting any other NAPU. With regards to the Donora unit, it has a
unique design, is the only remaining NAPU of its kind and is equipped
with a 1960s era NSCR device that is considerably less effective than
more modern NSCR devices. For this reason, the Donora NAPU emits
considerably more NOx than most NSCR controlled NAPUs, and retrofitting
it with an SCR may present engineering challenges and potential
additional costs that would not be present for retrofitting the vast
majority of other NAPUs. Therefore, the EPA believes these estimated
costs provided by DNI are not representative of a typical existing NAPU
with NSCR that may need to retrofit controls to meet the NSPS.

Most existing NAPUs use SCR, the control technology that we determined
is the BSER. There are no expected additional control costs for most (or
all) of these units if they are modified or reconstructed and become
subject to the NSPS, as we expect most (or all) of existing NAPUs with
SCR to be in compliance with the emission limit. For a few cases,
minimal changes may be needed for NAPUs with SCR and, if so, costs are
expected to be low.  

For the 14 NAPUs with NSCR, some already are complying with the NSPS
(e.g., the DNI Oregon and JR Simplot plants), so little or no control
costs would be incurred if these existing NAPUs with NSCR become subject
to this NSPS through modification or reconstruction. Control costs may
well be minimal or zero for several other of the 14 NSCR units. Some
NAPUs with NSCR could incur larger costs if they become subject to the
NSPS, but the costs are generally not expected to be more than $1,360/T.
Finally, even if there are one or a few NAPUs with NSCR that would not
be able to meet this NSPS in a cost-effective manner, courts have
recognized that an NSPS may be considered "adequately
demonstrated" even if not all existing sources would be able to meet
it. Portland Cement Assn. v. Ruckelhaus, 486 F.2d 375, 391 (D.C. Cir.
1973) ("We begin by rejecting the suggestion of the cement manufacturers
that the Act's requirement that emission limitations be ‘adequately
demonstrated' necessarily implies that any cement plant now in existence
be able to meet the proposed standards.").

It should be noted that rather than the Donora unit, the NAPU that DNI
is considering modifying is Cheyenne Unit 3. The additional air
pollution control/practices and associated costs needed for Cheyenne
Unit 3 are uncertain. However, as shown in section C above, CEMS data
for this unit show many periods of compliance. Furthermore, there are a
number of possible options under which this unit could meet the revised
NSPS for a reasonable cost. First, an SCR might not have to be
installed. For example, we believe Unit 3 could potentially comply by
fine tuning the NSCR, adding additional catalyst to the NSCR, adding a
preheater to lower NOx emissions during startups and replacing the
catalyst more frequently. If these measures did not achieve the NSPS
emission limit, Cheyenne’s Unit 3 could possibly comply with the NSPS
by installing a “mini” SCR with much less catalyst and using much
less ammonia than a full-size SCR. In any event, making retrofits or
other improvements necessary to meet the NSPS at Cheyenne Unit 3 would
be considerably less expensive than what DNI claims would be necessary
for the Donora unit. Since DNI’s cost estimates are entirely based on
the cost of retrofits to the Donora unit, the cost of retrofitting
Cheyenne Unit 3 would be much lower than estimated by DNI, and are
likely consistent with the EPA’s estimated cost-effectiveness value of
$1,360/T for a modified or reconstructed NAPU (77 FR 48442).
Nevertheless, as stated above in section A, if the costs for most plants
would be reasonable, but costs for an individual NAPU are considerably
higher, that itself would not justify developing a subcategory for one
or for a very limited number of NAPUs. 

Finally, an in-depth look at DNI’s Donora plant estimates suggests
that these costs could be overestimated. We note that the estimated
annual cost provided by DNI associated with retrofitting the Donora unit
is mostly energy costs ($916,000 per year or 59 percent of the total
annual cost) calculated based on steam losses associated with the
removal of an economizer that DNI asserts is necessary to accommodate
the installation of an SCR. We believe more cost-effective options
likely exist for replacing this lost steam, such as installing a
replacement economizer, that were not evaluated in DNI’s analysis.

E.  DNI Did Not Have a Chance to Comment on Analysis Presented in Final
Rule 

Claim: The petitioner states that section 307(d)(7)(B) of the CAA
requires the EPA to grant reconsideration when the petitioner:

	“[C]an demonstrate to the Administrator that is was impracticable to
raise [an] objection [during the period for public comment] or if the
grounds for such objection arose after the period for public
comment…and if such objection is of central relevance to the outcome
of the rule.”

The petitioner further states the standards for reconsideration are
clearly met. First, no interested party, including DNI, had any notice
in the proposed NSPS that the EPA would utilize CEMS data from DNI’s
Oregon plant as the primary basis for supporting the rule’s
application to NAPUs with NSCR and rejecting the petitioner’s request
for a separate emission limit for these units. This is clearly stated in
the preamble to the final rule as the EPA admits that these data were
evaluated “after proposal.” If the EPA had included this evaluation
in the proposed rule, the petitioner states that they would have
commented that DNI’s Oregon plant is not representative of overall
industry performance of NAPUs with NSCR.  

Also, the petitioner stated that, as noted in their subsequent comments
regarding the DNI Oregon plant, their objection is most certainly “of
central relevance to the outcome.” The petitioner states that DNI’s
Oregon plant is the only hourly CEMS data showing compliance with the
NSPS emission limit. Thus, the petitioner believes that the Nitric Acid
NSPS is arbitrary and capricious as it relates to NAPU with NSCR.

EPA’s Response: First, it should be noted that data for DNI’s Oregon
plant were described in the preamble to the proposed rule. At proposal,
we stated we had monthly block averages for this NAPU, as well as for
the JR Simplot plant (76 FR 63882). These averages were well below the
NSPS emission limit. These averages were not 30-day rolling averages,
however, these data still indicated these two NSCRs could likely meet
the NSPS emission limit of 0.50 lb/T on a 30-day rolling average.

The data that the EPA further evaluated after proposal and prior to
promulgation were hourly CEMS data for DNI’s Oregon plant that we
requested in the preamble to the proposed rule. These data were
requested to assist in evaluating whether NSCR can achieve the NSPS
emission limit. The rolling 30-day averages for this NAPU with NSCR meet
the NSPS emission limit. We also had data from another NAPU controlled
by NSCR, JR Simplot, as described above (in section C). In the proposal,
we noted that monthly averages from the JR Simplot plant were below the
revised limit. After the final rule was promulgated, we re-analyzed data
from the JR Simplot plant based on evaluation of hourly CEMS data. The
re-evaluation of the JR Simplot hourly data also show that JR
Simplot’s emissions are well below the NSPS emission limit and could
easily comply with the NSPS. The CEMS data for these two NAPUs with NSCR
show that the NSPS emission limit can be achieved by some NAPUs with
NSCR.

F.  Conclusion

In response to the reconsideration petition, we have reviewed the
data/analysis supplied by the petitioner. Also, we have reviewed and
analyzed previously available data to evaluate the claims made in the
reconsideration petition. Our responses to all issues raised by the
petitioner are explained in detail above in sections A through E, and
summarized below. 

Regarding the petitioner’s request for a subcategory for modified or
reconstructed nitric acid plants that use NSCR, the main points are as
follows:

First, in section A, we explained that we believe it is inappropriate to
establish subcategories based on differences in control technologies, so
it is inappropriate to establish a subcategory for plants using NSCR.
Also, we believe it is technically possible for existing NAPUs with NSCR
to achieve the NSPS emission limit if they were to be modified or
reconstructed. 

Second, as described in sections C – E above, we believe it is
feasible for some units with NSCR to comply with this NSPS without the
need for any additional controls as some existing units with NSCR are
already achieving the NSPS emission limit. 

Third, as described in section D above, we believe other units with NSCR
that are modified or reconstructed, could comply with the NSPS limit by
improving their controls at reasonable costs. 

Finally, regarding the petitioner’s argument that the EPA did not meet
the legal requirement of having representative emission data to set an
emission limit, as explained in section B above, we had ample test data
to support SCR as BSER and to establish a revised emission limit.

Therefore, based on our review and evaluation of all issues raised by
the petitioner and relevant available data and information, we have
concluded that reconsideration is not warranted.

	

  PAGE   \* MERGEFORMAT  8 

Enclosure

  PAGE   \* MERGEFORMAT  1 

