Projected MACT Limits for Hazardous Waste Burning Boilers and Industrial Furnaces


Melissa L. Douglas

Gerald J. Drake

Compliance Strategies & Solutions, Inc.

1301 Regents Park Drive, Suite 203

Houston, Texas 77058

 

 

ABSTRACT


The United States Environmental Protection Agency (EPA) issued a notice of data availability (NODA) titled “NESHAPS: Standards for Hazardous Air Pollutants for Hazardous Waste Boilers and Industrial Furnaces” in the Federal Register on June 27, 2000 (hereafter database)1.  Comments on the accuracy and completeness of this database were due to the EPA by August 28, 2000.  The EPA intends to use this database as the basis for developing the maximum achievable control technology (MACT) emission limits for hazardous waste burning boilers and industrial furnaces (the Phase II Boiler MACT) not covered under the previously promulgated hazardous waste combustor (HWC) MACT.  A total of 115 sources were listed in the June 2000 database.

 

This paper presents an evaluation of the database utilizing the methodology Congress directed the EPA to follow when establishing the floor level emission limits for sources.  The emission parameters to be addressed in this paper include the following:


·         Dioxins/Furans;

·         Particulate Matter;

·         Mercury;

·         Semivolatile Metals;

·         Low Volatile Metals; and

·         Hydrochloric Acid/Chlorine Gas.

 

The floor level emission limits are calculated by averaging the emissions of the best performing 12 percent of sources for categories with 30 or more sources, or the best performing 5 sources when there are fewer than 30 sources.  The projections presented in this paper provide an indication of a possible floor level for each of the above parameters for the database as a whole, as well as individually for the combustion categories identified under this NODA.  The projected floor levels will also be compared to the HWC MACT standards actually promulgated in the Federal Register on September 30, 1999.  The pertinent air pollution control devices (APCD) used for these units will be presented.

 

INTRODUCTION


This paper presents a review of the data included in the EPA’s Phase II HWC MACT Data Base Report, dated June 2000.  The 115 units included in this database include 83 boilers, 11 heaters, 15 halogen acid furnaces (HAFs), 5 sulfuric acid recovery furnaces (SARFs), and one unit whose type was not identified.  The unidentified source was not included in this evaluation. This data was collected from formal regulatory compliance testing programs including certification of compliance (CoC), trial burn, and/or risk burn testing.  Multiple test conditions for the same source were frequently included in the database.  The EPA requested that interested parties provide comments on this database by August 28, 2000.  Accordingly, 57 facilities provided specific comments on their individual units and 17 corporate offices provided company-wide comments.  Approximately 1/3 of the facilities (41) did not provide any comments on the database.

 

It is important to note that the information included herein is based on the June 2000 database.  Prior to evaluating this data, a request was made to obtain an updated database based on the EPA’s request for comments.  This request was made to Mr. H. Scott Rauenzahn who was identified as the EPA contact for this database2.  Mr. Rauenzahn indicated that the database had been updated to include the new information but that it was not available for distribution.  He added that there were no immediate plans to include the updated database on the Internet.  He also indicated that the data was available in the docket and estimated that it would take an individual approximately one month to extract the data from the various commenters.  He went on to state that the additional data did not significantly impact the results from the June 2000 database.

 

Although the database includes a plethora of information regarding each individual unit, this paper is limited to addressing the stack gas emissions and the associated APCDs in use for these units.  The stack gas emissions for particulate matter (PM), total chlorine (TCl), dioxin/furan toxic equivalents (D/F TEQ), mercury (Hg), semivolatile metals (SVM) including cadmium and lead, and the low volatile metals (LVM) including arsenic, beryllium, and chromium have been evaluated to estimate the floor values for establishing the Phase II Boiler MACT emission standards.  Carbon monoxide (CO), hydrocarbons (HC), and destruction and removal efficiency (DRE) criteria are expected to remain unchanged and are not included in this evaluation.

 

The Clean Air Act (CAA), as amended by the CAA of 1990, directs the EPA to establish minimum emission standards for existing sources as follows3:


“Emission standards promulgated under this subsection for existing sources in a category or subcategory may be less stringent than standards for new sources in the same category or subcategory but shall not be less stringent, and may be more stringent than—

(A) the average emission limitation achieved by the best performing 12 percent of the existing sources (for which the Administrator has emissions information), excluding those sources that have, within 18 months before the emission standard is proposed or within 30 months before such standard is promulgated, whichever is later, first achieved a level of emission rate or emission reduction which complies, or would comply if the source is not subject to such standard, with the lowest achievable emission rate (as defined by section 7501 of this title) applicable to the source category and prevailing at the time, in the category or subcategory for categories or subcategories with 30 or more sources, or

(B) the average emission limitation achieved by the best performing 5 sources (for which the Administrator has or could reasonably obtain emissions information) in the category or subcategory for categories or subcategories with fewer than 30 sources.”

 

 

APPROACH


Using the CAA methodology, the projected MACT floor levels were derived using two separate approaches.  The first approach used the entire database as a source category.  The second approach separated the sources into subcategories (i.e., boilers, HAFs, SARFs, and heaters).  The projected results from these approaches are then compared to the results promulgated for incinerators, cement kilns, and light-weight aggregate kilns.

 

RESULTS

 

The first approach reviewed the database as a whole, hence the 115 sources as a single category.  The 14 best performers (top 12 percent) were evaluated for each parameter. Table 1 presents the projected floor levels (average values) using the first approach by identifying the top 14 performers for each parameter without regard for sub-categorization.  Table 1 also includes the maximum, minimum, and standard deviation values for the top 14 performers. Each unit was included only once in the top 14 performers, even if multiple test conditions for that unit would be identified as a top 14 performer.  The best performing test condition was included with the remaining conditions being ignored.

 

Table 1. Top 14 Performers (all categories combined)4.

 

 

PM (gr/dscf)

TCl (ppmv)

Hg (µg/dscm)

SVM (µg/dscm)

LVM (µg/dscm)

D/F TEQ

(ng /dscm)

Average

0.00067

0.05

0.09

1.08

3.19

0.0025

Maximum

0.00120

0.11

0.32

2.19

7.16

0.0043

Minimum

0.00010

0.00

0.00

0.08

0.14

0.0000

Std. Dev.

0.00036

0.038

0.084

0.771

2.252

0.00114

The data in this table is corrected to 7% oxygen.

 

The subcategories of the top 14 performers for each parameter are identified as follows:

 

·         PM – 8 Boilers, 2 HAFs, 2 SARFs, and 2 Heaters

·         TCl – 12 Boilers, 1 HAF, and 1 SARF

·         Hg – 5 Boilers and 9 HAFs

·         SVM – 5 Boilers, 8 HAFs, and 1 SARF

·         LVM – 5 Boilers, 8 HAFs, and 1 SARF

·         D/F TEQ – 14 Boilers

 

Table 2 presents projected floor levels (average values) using the second approach by identifying the top performers for each subcategory.  Table 2 also includes the maximum, minimum, and standard deviation values for the units evaluated.  Because there were 83 boilers (more than 30) in this subcategory, the top 12 percent of performers were evaluated (i.e., the best 10 boilers).  None of the other subcategories had more than 30 sources.  Therefore, the top 5 sources in each of the remaining subcategories were evaluated.

 

Table 2. Top Performers (subcategories)4.

 

Boilers

 

PM

(gr/dscf)

TCl

(ppmv)

Hg

(µg/dscm)

SVM

(µg/dscm)

LVM

(µg/dscm)

D/F TEQ

(ng/dscm)

Average

0.00080

0.03

0.63

4.07

6.52

0.0020

Maximum

0.00133

0.07

1.57

10.61

10.06

0.0032

Minimum

0.00010

0.01

0.00

0.42

2.91

0.0000

Std. Dev.

0.00043

0.020

0.599

3.811

2.505

0.00091

HAFs

 

PM

(gr/dscf)

TCl

(ppmv)

Hg

µg/dscm)

SVM

(µg/dscm)

LVM

(µg/dscm)

D/F TEQ

(ng/dscm)

Average

0.00147

0.79

0.04

0.39

0.78

0.1572

Maximum

0.00280

1.61

0.07

0.98

1.52

0.5090

Minimum

0.00044

0.00

0.00

0.08

0.14

0.0205

Std. Dev.

0.00096

0.637

0.024

0.373

0.500

0.20219

SARFs

 

PM

(gr/dscf)

TCl

(ppmv)

Hg

(µg/dscm)

SVM

(µg/dscm)

LVM

(µg/dscm)

D/F TEQ (ng/dscm)

Average

0.00105

2.13

1.44

7.20

9.31

0.0336

Maximum

0.00185

5.73

1.98

12.40

11.48

0.0462

Minimum

0.00027

0.08

0.91

2.00

7.14

0.0209

Std. Dev

0.00066

2.50

0.76

7.356

3.069

0.01788

Heaters

 

PM

(gr/dscf)

TCl

(ppmv)

Hg

(µg/dscm)

SVM

(µg/dscm)

LVM

(µg/dscm)

D/F TEQ

(ng/dscm)

Average

0.00248

35.92

N/A

36.10

N/A

N/A

Maximum

0.00490

N/A

N/A

N/A

N/A

N/A

Minimum

0.00097

N/A

N/A

N/A

N/A

N/A

Std. Dev.

0.00169

N/A

N/A

N/A

N/A

N/A

N/A – Not Available

The data in this table is corrected to 7% oxygen.

 

The data presented in Table 2 was obtained from the database sources for each subcategory as follows:

 

·         Boilers – Represented by the top 12% (10 units) for each parameter.

·         HAFs – Represented by the top 5 sources for each parameter.

·         SARFs – Represented by all 5 sources for PM, 4 sources for TCl, and 2 sources for Hg, SVM, LVM, and D/F TEQ.

·         Heaters – Represented by the top 5 sources for PM, 1 source for TCl and SVM, and no sources for Hg, LVM, or D/F TEQ.

 

Table 3 presents the data from all 115 Phase II Boiler MACT sources including multiple data from a single source.  This data is presented to provide a median range of where these units are operating today as a whole.  Table 3 also includes the maximum, minimum, and standard deviation values for this data set.

 

Table 3. All Phase II Sources and All Conditions4.

 

 

PM

TCl

Hg

SVM

LVM

D/F TEQ

Average

0.01690

38.20

1.45

34.83

63.25

0.3141

Maximum

0.09950

916.76

12.07

526.46

900.17

6.7736

Minimum

0.00010

0.00

0.00

0.08

0.14

0.0000

Std. Dev.

0.01827

115.88

2.519

88.872

135.22

0.98923

The data in this table is corrected to 7% oxygen.

 

Table 4 summarizes the information provided in the first three tables as well as the promulgated HWC MACT (i.e., incinerators, cement kilns, and light-weight aggregate kilns) values for each of these parameters for comparison purposes.

 

Table 4. Comparison of Phase II Boiler MACT Projected Results And HWC MACT Promulgated Results4,5.

 

 

PM

(gr/dscf)

TCl

(ppmv)

Hg

(µg/dscm)

SVM

(µg/dscm)

LVM

(µg/dscm)

D/F TEQ

(ng/dscm)

Top 14

Single Category

0.00067

0.05

0.09

1.08

3.19

0.0025

Boiler Subcategory

0.00080

0.03

0.63

4.07

6.52

0.0020

HAFs Subcategory

0.00147

0.79

0.04

0.39

0.78

0.1572

SARFs Subcategory

0.00105

2.13

1.44

7.20

9.31

0.0336

Heaters Subcategory

0.00248

35.92

N/A

36.10

N/A

N/A

Complete Data Set

0.01690

38.20

1.45

34.83

63.25

0.3141

Incinerators

0.015

77

130

240

97

0.2 or 0.4a

Cement Kilns

b

 

130

120

240

56

0.2 or 0.4a

Light-Weight Aggregate Kilns

0.025

230

47

250

110

0.2 or 0.4a

 

a0.4 ng TEQ/dscm allowed if the inlet temperature to the initial PM control device is less than or equal to 400oF.

b0.15 kg/Mg dry feed and 20% opacity.  If there is an alkali by-pass stack associated with the kiln or in-line kiln raw mill, the combined PM emissions from the kiln or in-line kiln raw mill and the alkali by-pass must be less than the PM emissions standard.

The data in this table is corrected to 7% oxygen.

 

APCD

 

The type of APCD used by the evaluated units varies significantly.  In fact, many of the units were not equipped with any type of APCD.  This section presents the APCD used, or lack thereof, for each of the various scenarios presented above.  A key to the abbreviations for the APCD equipment is provided in Table 5.

 

Table 5. Abbreviations for APCD Equipment4.

 

Abbreviation

Description

3STGHCLABS

3 Stage HCl Absorbers

ABS

Absorber

CC

Contact Cooler

CLWS

Chlorine Wet Scrubber

CONV

Converter

CWS

Caustic Wet Scrubber

DM

Demister

DQ

Dry Quench

DT

Drying Tower

ESP

Electrostatic Precipitator

FF

Fabric Filter

GC

Gas Cooler

HCLABS

HCl Absorber

HEPA

High Efficiency Particulate Air

ME

Mist Eliminator

MGCLREC

Magnesium Chlorine Recovery

PBS

Packed Bed Scrubber

RH

Reheater

Q

Quench

QT

Quench Tower

S

Scrubber

SDA

Spray Dryer Absorber

SD

Spray Dryer

SEP

Separator

SO3ABS

S03 Absorber

SO3CON

S03 Converter

VE

Ejector Venturi

VS

Venturi Scrubber

WESP

Wet Electrostatic Precipitator

WHB

Waste Heat Boiler

WQ

Wet Quench

WS

Wet Scrubber

WS-WS

2 Stage Absorber Wet Scrubbers

 

The first approach used the 115 sources as a single source category.  The APCD used for the 14 best performers (top 12%) for each emission parameter are identified as follows:

 

Table 6. APCD For The Top 14 Performers (Best 12% of all sources)4.

 

APCD

PM

TCl

Hg

SVM

LVM

DF TEQ

None

9

10

2

2

3

9

WHB/QT/GC/WESP/SO3CON/WS-WS

1

 

 

1

2

 

Q/HCLABS/WS

1

1

 

1

1

 

WHB/QT/CC/WESP/DT/CONV/SO3ABS/ME

1

 

 

 

 

 

QT/ABS/WS

1

 

 

 

 

 

Q/HCLABS/CWS

1

 

 

 

 

 

Unknown

 

1

 

 

 

3

HCLABS/CWS

 

1

 

 

 

 

WHB/QT/DT/SO3CON/ABS/WS/DM

 

1

 

 

 

 

DQ/HCLABS/CLWS

 

 

1

 

 

 

VS/WS

 

 

1

1

 

 

WHB/HCLABS/WS

 

 

2

2

1

 

WHB/Q/HCLABS/VE/CLWS

 

 

1

1

1

 

WHB/VS/Q/HCLABS/VS/CLWS

 

 

1

1

1

 

WHB/HCLABS/CWS

 

 

1

1

1

 

WHB/Q/HCLABS/VS/WS

 

 

1

 

1

 

WQ/3STGHCLABS/S/CWS

 

 

1

1

 

 

ESP

 

 

1

2

 

1

MGCLREC/VS/SEP/DM

 

 

1

1

1

 

FF

 

 

1

 

1

1

SD/FF/PBS/RH/HEPA

 

 

 

 

1

 

DQ/HCLABS/VS/CLWS

 

 

 

 

1

 

 

 

Tables 7, 8, 9, and 10 present the APCD types for the top performers in each subcategory using the second approach.  Table 7 presents the top 12% of boilers.  Table 8 presents the top 5 sources for HAFs.  Table 9 presents the top 5 sources for SARFs.  Table 10 presents the top 5 sources for heaters.

Table 7. APCD For Boilers (top 12%)4.

 

APCD

PM

TCl

Hg

SVM

LVM

DF TEQ

None

8

9

5

4

6

6

Q/HCLABS/CWS

1

 

 

 

 

 

HCLABS/CWS

1

1

 

 

 

 

VS/WS

 

 

1

1

 

 

ESP

 

 

1

3

1

1

FF

 

 

1

1

2

 

SDA/ESP

 

 

1

 

 

 

Q/WS

 

 

1

 

 

 

SD/FF/PBS/RH/HEPA

 

 

 

1

1

 

Unknown

 

 

 

 

 

3

 

 

Table 8. APCD For HAFs (top 5 sources)4.

 

APCD

PM

TCl

Hg

SVM

LVM

DF TEQ

Q/HCLABS/WS

1

1

 

1

 

1

QT/ABS/WS

1

 

 

 

 

1

WHB/Q/HCLABS/VS/WS

1

 

 

 

1

1

WHB/QT/WS

1

 

 

 

 

 

WHB/HCLABS/CWS

1

 

1

1

1

 

WHB/Q/HCLABS/VE/CLWS

 

1

1

 

1

 

WHB/VS/Q/HCLABS/VS/CLWS

 

1

1

1

1

 

Q/HCLABS/CWS

 

1

 

 

 

 

DQ/HCLABS/VS/CLWS

 

1

1

 

 

 

WHB/HCLABS/WS

 

 

1

1

1

 

MGCLREC/VS/SEP/DM

 

 

 

1

 

1

WQ/3STGHCLABS/S/CWS

 

 

 

 

 

1

 

 

Table 9. APCD For SARFs (top 5 sources)4.

 

APCD

PM

TCl

Hg

SVM

LVM

DF TEQ

WESP

2

1

 

 

 

 

WHB/QT/GC/WESP/SO3CON/WS-WS

1

1

1

1

1

1

WHB/QT/CC/WESP/DT/CONV/SO3ABS/ME

1

1

 

 

 

1

WHB/DT/QT/SO3CON/ABS/WS/DM

1

1

1

1

1

 

 

 

Table 10. APCD For Heaters (top 5 sources)4.

 

APCD

PM

TCl

Hg

SVM

LVM

DF TEQ

None

5

1

 

1

 

 

 

 

CONCLUSIONS

 

The results of the top performers from either a single category approach or a subcategory approach indicate projected floor levels that are approximately 0.5 to 3 orders of magnitude less than the limits promulgated for incinerators, cement kilns, and light-weight aggregate kilns.  One reason for this disparity could be that some of the test conditions included in the database were not designed to demonstrate worst-case operating conditions.  For example, during risk burn testing chlorine and metals are generally not spiked.  Using emissions data from these conditions to establish the MACT limits will result in emission limits much more stringent than evaluating conditions designed to set the metals and chlorine emission limits for a unit.  Another possible reason could be the burning of clean fuels in units that do not have APCDs.  In this case, the facilities burning low ash and chlorine streams could cause the projected results to be lower.

 

More than half of the affected facilities provided comments on the June 2000 database.  However, Mr. Rauenzahn indicated that the new data had little impact on the overall database results.  Hence, it is likely that the additional information provided by the commentors did not eliminate the top performing test conditions used in these projections. 

 

It is interesting to note that the data set that most closely approximates the emission limits established for the HWC MACT are those that include all the data from all sources and all test conditions as presented in Table 3.  In this case, except for mercury and SVM, the floor values would be essentially equivalent to the HWC MACT standards.

 

REFERENCES

 

1.      Federal Register; June 27, 2000; pp 39581 – 39584.

 

2.      Personal conversation with Mr. H. Scott Rauenzahn, Work Assignment Manager for the June 2000 Phase II HWC MACT Data Base Report, US Environmental Protection Agency, in January 2001.

 

3.      42 United States Code 7412(d)(3) as amended by the Clean Air Act of 1990.

 

4.      EPA’s HWC MACT Page.  http://www.epa.gov/hwcmact (accessed January 2001).

 

5.      Federal Register; September 30, 1999; pp 52860, 52875, and 52891.