es
ual Protection
Solid Waste arid
Emergency Response
(5305W) -
";-••-:'• PB90-190-715
:'v. February H 9 96
BiATBac
Document for
Spent Potliners from
Primary Aluminum :*
Reduction-K088
J'| «S (;
>#?<*]
» A*
*,*. ',\
*t .* i .
;f; {d'4
:, j1-. ,r Reproduced on>ap«HtiajlContair(t at teajt ^g
U J. 0«B«nm..n ol Co/r>n»rct
-------
I
I
I
50272-101
REPORT DOCUMENTATION
PAGE
1. REPORT NO.
EPA/530-R-96-015
4. Title and Subtitle | 5. Report Date
BOAT [BEST DEMONSTRATED AVAILABLE TECHNOLOGY] BACKGROUND DOCUMENT FOR SPENT | FEBRUARY 1996
POTLINERS FROH PRIMARY ALUMINUM REDUCTION - KOSfl I 6.
7. Author(s)
8. Performing Organization Rcpt. No
I
I
i
I
I
I
I
I
I
I
I
I
I
I
I
9. Performing Organization Name and Address
U.S. EPA
OFFICE OF SOLID UASTE
401 M STREET, SU
WASHINGTON. DC 20460
10. Project/Task/Work Unit No.
11. Contract(C) or Grant(C) No.
(C)
12. Sponsoring Organization Name and Address
13. Type of Report & Period Covered
TECHNICAL
U.
15. Supplementary Notes
16. Abstrsct
-------
I
I
I
I
I
I
I
I
FINAL
BEST DEMONSTRATED AVAILABLE TECHNOLOGY (BOAT)
BACKGROUND DOCUMENT
FOR
SPENT POTUNERS FROM PRIMARY
ALUMINUM REDUCTION -
K088
I
I Michael Petruska
Chief, Waste Treatment Branch
I
I Mary Cunningham
* Project Manager
I
— U. S. Environmental Protection Agency
I Office of Solid Waste
2800 Crystal Drive
_ Arlington, VA 22202
I
| February 29, 1996
I
I
-------
I
I
I
I
I
I
I
I
I
I
I
i
i
I
I
I
i
I
I
TABLE OF CONTENTS
EXECUTIVE SUMMARY ......................................... ES-1
1.0 INTRODUCTION ............................................. [-1
1 . 1 Regulator/ Background ............................ ." ........ i . i
1.2 Summary .............................................. 1-5
1.3 Contents of This Document .................................. 1-6
2.0 LAND DISPOSAL RESTRICTIONS FOR K088 ........................ 2-1
2.1 Summary of Basis for Listing of Spent Potliners ................... 2-1
2.2 Kev Points of Spent Potliner Standards and How Thev Reflect LDR GnaJs . 2-2
3.0 DETAILED DESCRIPTION Or SPENT POTLINERS .................... 3-1
3.1 Description of the Aluminum Production Industry ................... 3-1
3.1.1 Description of Aluminum Reduction Facilities ............... 3-1
3.1.2 Size and Geographical Distribution of Facilities .............. 3-9
3.1.3 Raw Materials .................................... 3-11
3.1.4 Aluminum Products and Their Uses ..................... 3-11
3.1.5 Aluminum Reduction Capacity ........................ 3-11
3.2 Waste Stream Characteristics ................................ 3-12
3.2.1 Waste Stream Status Under Other Regulations .............. 3-12
3.2.2 Waste Stream Descriptions ........................... 3-13
3.2.3 Amenability of Wastes to Chemical Analysis ............... 3-14
•3.2.3.1 SW-846 Method Applicability ................... 3-14
3.2.3.2 Sample Preparation Issues ..................... 3-17
3.3 Current Spent Potliner Waste Management Practices ................ 3-17
3.3.1 Waste Management Practices for K088 ................... 3-17
3.3.1.1 Description of K088 Waste Management Practices ..... 3-18
3.3.1.2 Waste Minimization and Recycling Activities . .' ...... 3-19
4.0 BOAT TREATMENT STANDARDS FOR SPENT POTLINERS ............. 4-1
4.1 Selection of Constituents for Regulation ......................... 4-1
4.1.1 BOAT List Constituents Present in K088 .................. 4-1
4.1.2 Other Constituents Present in K088 ...................... 4-3
4.1.3 Constituents Selected for Regulation in K088 ............... 4-3
4.2 Identification of Technologies for the Treatment of Spent Potliners • ....... 4-6
4.2.1 Nonwastewaters ................................... 4-9
4.2.1.1 Management Methods and Treatment Technologies ..... 4-9
4.2.1.1.1 Source Reduction ..................... 4-10
4.2.1.1.2 Total Recycle ....................... 4-11
4.2.1.1.3 Treatment Processes ................... 4-12
4.2.2 Wastewaters .................................... 4-36
-------
4.2.2.1 Applicable Treatment Technologies 4-37
4.2.2.2 Demonstrated Technologies 4-39
4.2.2.3 Identification of BOAT 4-40
4.3 Identification of BOAT Treatment Standards 4-40
4.3.1 Nonwastewaters 4-40
4.3.2 Wastewaters 4-43
5.0 REFERENCES 5-1
6.0 ACKNOWLEDGEMENTS 6-1
Appendix A Treatment Performance Database and Methodology for Identifying Universal
Treatment Standards for Constituents in Nonwastewater Forms of K.088
u
-------
LIST OF TABLES
Paue
ES-l BOAT Treatment Standards for Nonwastewater Forms of K088 for Organic
Constituents and Cyanide ES-5
ES-2 BOAT Treatment Standards for Nonwastewater Forms of K088 for Metals and
Fluoride • ES-6
ES-3 BOAT Treatment Standards for Wastewater Forms of K088 ES-7
l-l BOAT Treatment Standards for Nonwastewater Forms of K088 for Organic
Constituents and Cyanide 1-8
1-2 BOAT Treatment Standards for Nonwastewater Forms of K.088 for Metals and
Fluoride 1-9
1-3 BOAT Treatment Standards for Wastewater Forms of K088 1-10
3-1 Facilities Generating Spent Potliner 3-20
3-2 Waste Characterization Data for K088 3-21
3-3 Characterization of Wastewater Form of K088 3-27
3-4 Analytical Methods for Semivolatile Organics 3-29
3-5 Analytical Methods for Metals 3-30
3-6 Analytical Methods for Non-metallic Inorganics 3-31
3-7 Analytical Methods - Preparative (Digestion, Extraction, Cleanup) 3-32
3-8 Analytical Methods Instrumentation 3-33
4-1 BDAT Treatment Standards for Nonwastewater Forms of K088 for Organic
Constituents and Cyanide 4-45
4-2 BDAT Treatment Standards for Nonwastewater Forms of K.088 for Metals and
Fluoride 4-46
4-3 BDAT Treatment Standards for Wastewater Forms of K088 4-47
4-4 Scale and Description of Technologies Used to Treat K088 4-49
4-5 Operational Requirements 4-51
4-6 Fate of Components from K088 4-52
4-7 Treatment Performance Database for Organic Constituents (Nonwastewaters) •. . . 4-53
4-8 Treatment Performance Database for Metal Constituents (Nonwastewaters) 4-55
4-9 Treatment Performance Database for Chromium (Nonwastewaters) 4-59
4-10 Comparison of Non-Wastewater K.088 Treatment Standards to Treatment Data
For Organic Constituents 4-60
4-11 Comparison of Non-Wastewater K.088 Treatment Standards to Treatment Data
For Metals, Fluoride 4-61
4-12 Calculation of Universal Treatment Standards for Organic Constituents
(Wastewaters) 4-62
4-13 Calculation of Universal Treatment Standards for Metal Constituents
(Wastewaters) 4-63
4-14 Comparison of Wastewater K088 Treatment Standards to Treatment Data For
Organic Constituents 4-64
4-15 Comparison of Wastewater K088 Treatment Standards to Treatment Data For
Metals, Cyanide and Fluoride 4-65
in
-------
LIST OF FIGURES
3-1 Schematic of the Aluminum Reduction Process 3-2
3-2 Simplified Diagram of a Typical Pot 3-4
3-3 Simplified Diagram of VSS Pot 3-7
3-4 Simplified Diagram of HSS Pot 3-8
3-5 Locations of Aluminum Reduction Facilities 3-10
4-1 Schematic of the Alcan Caustic Leaching Process 4-14
4-2 Process Flow Diagram for the Barnard Process 4-19
4-3 Schematic of the Comalco Comtor Process 4-22
4-4 Diagram of Torbed Calciner 4-24
4-5 Schematic of the ESI Process 4-27
4-6 Schematic of the ORMET CMS System 4-31
4-7 Schematic of the Reynolds Treatment Process 4-35
IV
-------
I
I
I
1
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
EXECUTIVE SUMMARY
The U.S. Environmental Protection Agency (EPA or the Agency) is establishing Best
Demonstrated Available Technology (BDAT) treatment standards for the regulation of listed
hazardous waste identified in Title 40, Code of Federal Regulations. Section 261.32 (40 CFR
261.32) as K088. These BDAT treatment standards are being established in accordance with
the amendments to the Resource Conservation and Recovery Act (RCRA) of 1976 enacted by
the Hazardous and Solid Waste Amendments (HSWA) of November 8, 1984. Compliance
with the BDAT treatment standards would be a prerequisite for land disposal of restricted
wastes, as defined in 40 CFR 268.
Hazardous Waste Number K088 is generated from the primary reduction of aluminum.
This hazardous waste is defined as follows:
• Spent potliners from primary aluminum reduction
This background document provides the Agency's rationale and technical support for
developing BDAT treatment standards for K088 under the Land Disposd Restrictions (LDR)
program. This document also provides waste characterization data that may serve as the basis
for determining whether a variance from the applicable treatment standards is warranted for
•spent potliners.
The Agency's legal authority and the petition process necessary for requesting a
variance from the treatment standards are summarized in EPA's Final BDAT Background
Document for Quality Assurance. Quality Control Procedures and Methodologies (1). The
methodologies used for establishing the nonwastewater treatment standards for the constituents
selected for regulation in K088 are summarized in Appendix A of this document.
The EPA is developing a concept for K088 (and other wastes) that uses a hierarchy of
options for evaluating treatment or recycling technologies. For example, the Agency's
ES-1
-------
ultimate goal for hazardous waste is source reduction, that is, less or no production of
hazardous waste. The next preferred is total recycle or reuse. This would require that all the
waste generated be used as feedstock in the same process or another process. Next lower in
the hierarchy would be treatment technologies that can recover materials from the waste for
reuse. This option would most likely result in some residuals still needing to. be land
disposed, but preferred techniques would also significantly reduce the quantity and toxicity of
any waste destined for land disposal. Further down in the hierarchy would be technologies
that reduce quantity and toxicity without the recovery of materials for reuse. The next level
down would be technologies that lower toxicity alone and may increase volume of materials
for land disposal For example, while stabilization can sometimes render a metal-bearing
waste less hazardous, it often results in a great increase in volume which then has to be
landfilled At the base of this hierarchy are numerical treatment standards. Numerical
treatment standards are required for treatment or recovery technologies that result in residuals
requiring land disposal. Treatment residuals must comply with applicable treatment standards
prior to land disposal.
The Agency selected the constituents of concern for regulation in K088 based on the
1988 rule listing this waste as hazardous, information submitted by generators of spent
potliners in delisting petitions, sampling and analysis of spent potliner wastes performed by
the Agency and a significant quantity of d^ta that was available in the public record. The
Agency is regulating the land disposal of both nonwastewater and wastewater forms of K088
by establishing BOAT treatment standards numerically equivalent to universal treatment
standards (universal standards) with the addition of fluoride in nonwastewater forms of K088.
A universal standard is a single concentration limit established for a specific constituent
regardless of the waste matrix in which it is present, i.e., the same treatment standard applies
to a particular constituent in each waste code in which it is regulated. The Agency is
establishing two different sets of universal standards: one for nonwastewater forms of waste
and one for wastewater forms of waste. These two sets differ in the population of regulated
constituents and the individual universal standards. A more detailed discussion concerning the
determination of these treatment standards is provided in EPA's Final Best Demonstrated
ES-2
-------
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
Available Technology (BDAT) Background Document for Universal Standards. Volume A:
Universal Standards for Nonwastewater Forms of Listed Hazardous Wastes (2) and EP.Vs
Final Best Demonstrated Available Technology (BOAT) Background Document for Universal
Standards. Volume B: Universal Standards for Wastewater Forms of Listed Hazardous
Wastes (3).
Universal standards are based on a variety of technologies. Organic constituent
universal standards are based primarily on incineration for nonwastewaters and biological
treatment or carbon adsorption for wastewaters. Inorganic constituent universal standards are
based primarily on high temperature metal recovery for nonwastewaters and chemical
precipitation for wastewaters. These treatment standards were developed by examining
essentially all the BDAT treatment data the Agency had at the time. The universal standards
for wastewater forms of these wastes are based on treatment performance data from several
sources, including the BDAT database, the NPDES database, the WERL database, EPA
collected WAO/PACT* data, the HAD database, industry submitted leachate treatment.
performance data, data in the literature that were not already part of the WERL database, and
data in literature submitted by industry on the WAO and PACT* treatment processes. The
treatment standards for cyanide are based on alkaline chlorination. The treatment standard for
fluoride in nonwastewaters is a leachable concentration which was determined by the Agency
when granting a delisting for certain K.088 wastes.
Tables ES-1 and ES-2 present the BDAT treatment standards for nonwastewater forms
of K.G88. Table ES-3 presents the BDAT treatment standards for wastewater forms of these
wastes. The standards shown on the tables are numerically equivalent to the universal
standards for those constituents with the exception of fluoride in nonwastewaters. There was
no universal treatment standard promulgated for fluoride in nonwastewaters. Fluoride in the
nonwastawater form was not regulated in the multi-source leachate, based on the EPA's
determination that cyanide compounds, which are in the same treatability group, would
control the disposal/treatment of fluoride and that the fluorides were not expected to be
present at high concentrations. For K.088, however, fluoride is typically present at high
ES-3
-------
concentrations and can pose a threat to human health and the environment. The Agency thus
feels it is appropriate to regulate Piuoride in both wastewater and nonwastewater forms of
K.088. The nonwastewater standard is numerically equivalent to the leachable concentration
identified in a delisting decision for residues from the treatment of spent potlincrs. The
wastewater treatment standard for rluoride is from the Universal Treatment Standards.
ES-4
-------
I
I
I
1
I
1
I
1
1
I
I
I
I
f
I
I
I
I
I
Table ES-1 BOAT Treatment Standards for Nonwastewater Forms of K088 for
Organic Constituents and Cyanide
Regulated Constituent
Cyanide
Cyanide (amenable)
Acenaphthene
Anthracene
Benz(a)anthracene
Benzo(a)pyrene
Benzo(b and k)fluoranthene*
Benzo(g,h,i)perylene
Chrysene
Dibenz(a,h)anthracene
Fluoranthene
Indeno( 1 ,2,3-cd)pyrene
Phenanthrene '
Pyrene
Total Composition Concentration
(mg/kg)
Maximum for any Grab Sample
590
30
3.4
3.4
3.4
3.4
6.8
1.8
3.4
8.2
3.4
3.4
5.6
8.2
* The treatment standard for these constituents is expressed as a sum of their concentrations to
account for analytical concerns in distinguishing between the two compounds.
Ref.: 2) U.S. Environmental Protection Agency, Office of Solid Waste. Final Best
Demonstrated Available Technology (BOAT) Background Document for Universal Standards:
Volume A. Universal Standards for Nonwastewater Forms of Wastes. U.S. Environmental
Protection Agency, Washington, DC, July 1994
ES-5
-------
Table ES-2 BOAT Treatment Standards for Nonwastcwater Forms of K088 for
Metals and Fluoride
Regulated Constituent
Fluoride
Antimony
Arsenic
Barium
Beryllium
Cadmium
Chromium (total)
Lead
Mercury
Nickel
Selenium
Silver
Maximum for any Grab Sample (mg/L)
Using TCLP
48'
2.1
5.0
7.6
0.014
0.19
0.86
0.37
0.025
5.0
0.16
0.30
Ref.: 2) U.S. Environmental Protection Agency, Office of Solid Waste. Final Best
Demonstrated Available Technology (BDAT) Background Document for Universal Standards:
Volume A. Universal Standards for Nonwastewater Forms of Wastes. U.S. Environmental
Protection Agency, Washington, DC, July 1994
A Universal Treatment Standard was not promulgated for fluoride. The concentration presented is the
teachable concentration identified in the a delisting decision.
ES-6
-------
Table ES-3 BOAT Treatment Standards for Wastewater Forms of K088
Regulated Constituent
Acenaphthene
Anthracene
Benz(a)anthracene
Benzo(a)pyrene
Benzo(b and k)fluoranthene'
Benzo(g,h,i)perylene
Chrysene
Dibenz(a,h)anthracene
Fluoranthene
Indeno( 1 ,2,3-cd)pyrene
Phenanthrene
Pyrene
Cyanide (total)
Cyanide (amenable)
Fluoride
Antimony
Arsenic
Barium
Beryllium
Cadmium
Chromium (total)
Lead
Mercury
Nickel
Total Composition Concentration (mg/L)
Maximum for any 24 hour Composite
0.059
0.059
0.059
0.061
0.11
0.0055
0.059
0.055
0.068
0.0055
0.059
0.067
1.2
0.86
35
1.9
1.4
1.2
0.82
0.69
2.77
0.69
0.15
3.98
ES-7
-------
Table ES-3 BOAT Treatment Standards for Wastewater Forms of K088
ReguJated Constituent
Selenium
Silver
Total Composition Concentration (mg/L)
Maximum for any 24 hour Composite
0.82
0.43
' The treatment standard for these constituents is expressed as a sum of their concentrations to
account for analytical concerns in distinguishing between the two compounds.
Ref.: 3) U.S. Environmental Protection Agency, Office of Solid Waste. Final Best
Demonstrated Available Technology (BOAT) Background Document for Universal Standards:
Volume B. Universal Standards for Wastewater Forms of Wastes. U.S. Environmental
Protection Agency, Washington, DC, July 1994
ES-8
-------
I
I
I
I
I
I
I
I
1.0 INTRODUCTION
The U.S. Environmental Protection Agency (EPA or the Agency) is establishing Best
Demonstrated Available Technology (BOAT) treatment standards for the regulation of listed
hazardous waste identified in Title 40, Code of Federal Regulations. Section 261.32 (40 CFR
261.32) as K088. These BDAT treatment standards are being established in accordance with
the amendments to the Resource Conservation and Recovery Act (RCRA) of 1976 enacted by
the Hazardous and Solid Waste Amendments (HSWA) of November 8, 1984. Compliance
with the BDAT treatment standards would be a prerequisite for land disposal of restricted
wastes, as defined in 40 CFR 268. In 40 CFR 268.44, EPA supplies provisions, that if met,
may justify granting waste- and site-specific waivers for applicable treatment standards in 40
CFR 268.41-268.43.
Hazardous Waste Number K.088 is generated from the primary reduction of aluminum.
This hazardous waste is defined as follows:
• Spent potliners from primary aluminum reduction
This background document provides the Agency's rationale and technical support for
developing BDAT treatment standards for K088 under the Land Disposal Restrictions (LDR)
program. The Agency's legal authority and the petition process necessary for requesting a
variance from the treatment standards are summarized in EPA's Final BDAT Background
Document for Quality Assurance. Quality Control Procedures and Methodologies (1). The
methodologies used for establishing the nonwastewater treatment standards for the 'constituents
selected for regulation in KO88 are summarized in Appendix A of this document
1.1 Regulatory Background
On December 18, 1978 (43 FR 58946), EPA proposed its initial regulations for
hazardous waste management under Subtitle C of RCRA. These proposed regulations, among
1-1
-------
other things, identified a universe of "special wastes" that arc generated in large volumes,
were thought to pose less of a hazard than other hazardous wastes, and were thought not to be
amenable to all of the control techniques proposed for other types of RCRA hazardous wastes.
EPA identified waste materials from the "extraction, beneficiation, and processing of ores and
minerals," i.e., mining waste, as one such "special waste" under the proposed regulations.
On May 19, 1980, EPA promulgated the final hazardous waste management
regulations. In promulgating these regulations, the Agency did not finalize the "special waste"
category. The Agency listed as hazardous (as an interim final rule) eight wastes that are
generated from primary metal smelters (45 FR 33112, May 19, 1980, and 45 FR 47832, July,
16, 1980), including spent potliners from primary aluminum reduction.
In October of 1980, the Resource Conservation and Recovery Act (RCRA) was
amended by adding Section 3001(b)(3)(A)(ii) to exclude "solid waste from the extraction,
beneficiation, and processing of ores and minerals" from regulation as hazardous waste under
Subtitle C of RCRA, pending completion of a study and a report to Congress. EPA modified
its hazardous waste regulations in November 1980 to reflect this "mining waste exclusion,"
and issued preliminary and quite broad interpretation of the scope of its coverage. In
particular, EPA interpreted the exclusion to include "solid waste from the exploration, mining,
milling, smelting and reiining of ores and minerals" (45 FR 76618, November 19, 1980). To
be consistent with its interpretation of the scope of the exclusion expressed in the November
19, 1980 notice, the Agency suspended the listings for five smelter wastes which it
promulgated as final on November 12, 1980 (see 45 FR 76618). In addition, on January 16,
1981, the Agency suspended the other wastes that were promulgated as interim final on July
16, 1980. In suspending all of these listings, the Agency made it clear that although these
wastes met the criteria for listing in 40 CFR 261.11, they appeared to come within the scope
of the "mining waste exclusion." The K088 listing was included in the suspended listings.
In 1984, EPA was sued for failing to submit the required Report to Congress and
make the required regulatory determination by the statutory deadline (Concerned Citizens of
1-2
-------
I
I
I
I
1
I
I
Adamstown v.'EPA No. 84-3041, D.D.C., August 21, 1985). As a result of the lawsuit the
Agency essentially split the wastes that might be eligible for exclusion into two groups:
mining (mineral extraction and beneficiation) wastes and mineral processing wastes. On
October 2, 1985, under the court order in Adamstown. EPA proposed to narrow the scope of
the mining waste exclusion (50 FR 40292). Under this proposed ^interpretation, the
suspension of the six smelting waste listings would be removed since they would no longer be
considered "special wastes." Therefore, the notice proposed to relist the six smelter wastes,
including spent potliners.
Subsequently, on October 9, 1986, the Agency announced that it was withdrawing its
proposed reinterpretation (51 FR 36233). The Agency explained that it was withdrawing the
reinterpretation because the terms "high volume" and "low hazard" had pot been quantified in
the proposal and, therefore, the Agency was unable to determine the status of additional
wastes nominated by commenters as "special wastes" (51 FR 36234). While it did not view
the "high volume^ low hazard" standard as inherently unsound, EPA pointed to various
definitional problems it faced in determining how to group and classify these wastes. The
Agency concluded that its proposal had to be withdrawn because it failed to set on
"practically applicable criteria for distinguishing processing from non-processing wastes" and
because there was insufficient time to repropose a rule in light of the Adamstown deadline.
The withdrawal of the proposed reinterpretation effectively continued the suspension of the six
smelter waste listings.
Subsequently, two suits were filed against EPA challenging the Agency's decision to
withdraw its -proposed reinterpretation of the mining waste exclusion. The cases,
Environmental Defense Fund CEDF) v. EPA. No. 86-1584 (D.C. Cir.) and Hazardous Waste
Treatment Council v. EPA. No. 86-1691 (D.C. Cir.) were decided on July 29, 1988.
The U.S. Court of Appeals for the D.C. Circuit ruled in EOF v. EPA that EPA's
decision to withdraw the proposed reinterpretation and failure to relist the six smelting and
refining wastes was arbitrary and capricious. The Court found that EPA's inclusion of all
1-3
-------
smelting and refining wastes in the "mining waste exclusion" for ore processing wastes was
"impermissibly overbroad" and contrary to Congressional intent While the court conceded
that the statutory term "processing" is ambiguous, the Court nonetheless found EPA's
interpretation to be unreasonable in light of "clear" legislative history that suggested that
Congress had intended the Bevill Amendment to be limited to those ore processing wastes
which meet EPA's 1978 "special waste" concept, i.e., those solid wastes that are high volume
and low hazard,
In its order for relief, the Court directed EPA to relist the six smelter wastes by
August 31, 1988. The Court noted that, regardless of the status of any additional processing
wastes, the six smelter wastes clearly would not fit any definition of "high volume, low
hazard." In summary, the Court found that the six wastes cannot, as a matter of law, be
excluded from regulation under the Bevill amendment and must be regulated under Subtitle C
if they meet the listing or identification criteria for hazardous wastes under 40 CFR 261.10
and 261.11. On September 13, 1988, the Agency published the final rule promulgating the
hazardous listing for K088 waste generated fiom the primary reduction of aluminum (53 FR
35412). •
In 1990, in American Mining Congress v. EPA No. 88-1835 (D.C. Cir.) six petitioners
sued the Agency over the final listing decisions on the six smelting wastes. On July 10, 1990
the courts remanded five of the six listings, however, the court upheld the final listing for
spent potliners from primary aluminum reduction.
The hazardous waste listing program and the LDR program define "wastewater"
quantitatively to mean forms of hazardous wastes with less than one percent total organic
carbon (TOC) and less than one percent total suspended solids (TSS). Although K088 wastes
meet the definition of nonwastewaters as generated, EPA establishes treatment standards for
both wastewater and nonwastewater forms of listed wastes to ensure that any waste streams
that meet the definition of wastewater are also treated to meet appropriate treatment standards
prior to land.disposal. Streams generated from the treatment of K088 containing less than one
1-4
-------
As discussed in the Advance Notice of Proposed Rulemaking (ANPRM) published in
October 24, 1991 (56 FR 55180-55182), EPA was seeking development of concentration-
based standards for K088 so as to allow the use of any appropriate technology chat can
achieve the numerical values.
1.2 Summary
The Agency is regulating the land disposal of both nonwastewater and wastewater
forms of Hazardous Waste Number K088 by establishing BOAT treatment standards
numerically equivalent to universal treatment standards (universal standards) as published in.
the final Phase II land disposal restrictions rulemaking (59 FR 47982, September 19, 1994).
A universal standard is a single concentration limit established for a specific constituent
regardless of the waste matrix in which it is present, i.e., the same treatment standard applies
to a particular constituent in each waste code in which it is regulated The Agency is
establishing two different sets of universal standards: one for nonwastewater forms of waste
and one for wastewater forms of waste. These two sets differ in the population of regulated
constituents and the individual universal standards. A more detailed discussion concerning the
determination of these treatment standards is provided in EPA's Final Best Demonstrated
Available Technology (BDAT> Background Document for Universal Standards^Volume A:
Universal Standards for Nonwastewater Forms of Listed Hazardous Wastes (2) and EPA's
Final Best Demonstrated Available Technology (BOAT) Background Document for Universal
Standards. Volume B: Universal Standards for Wastewater Forms of listed Hazardous
Wastes (3).
Universal standards are based on a variety of technologies. Organic constituent
universal standards are based primarily on incineration for nonwastewaters and biological
treatment or carbon absorption for wastewaters. Inorganic constituent universal standards are
1-5
-------
based primarily on high temperature metal recovery for nonwastewaters and chemical
precipitation for wastewaters. These treatment standards were developed by examining
essentially all the BOAT treatment data the Agency had at the time. The universal standards
for wasiewater forms of these wastes are based on treatment performance data from several
sources, including the BOAT database, the NPDES database, the WERL database, EPA
collected WAO/PACT* data, the EAD database, industry submitted leachate treatment
performance data, data in the literature that were not already part of the WERL database, and
data in literature submitted by industry on the WAO and PACT* treatment processes. The
treatment standards for cyanide are based on alkaline chiorination. The treatment standards
for fluoride nonwastewaters is a leachatc concentration which was determined by the Agency
when granting a delisting for certain KG Si 8 wastes.
Tables 1-1 and 1-2 presents the BOAT treatment standards for nonwastewater forms of
KQ88. Table 1-3 presents the BOAT treatment standards for wastewater forma of these
wastes. The standards shown are numerically equivalent to the universal standards for those
constituents with the exception of fluoride in nonwastewaters. There was no Universal
Treatment Standard promulgated for fluoride. The nonwastewater standard is numerically
equivalent to the leachable concentration required in a delisting decision for residues from the
treatment of spent potliners. The wastewater treatment standard for fluoride is from the
Universal Treatment Standards.
1.3 Contents of This Document
Section 2.0 of this document summarizes the BOAT treatment standards, the basis for
listing spent potliners as hazardous, and how BOAT treatment standards reflect the goals of
the Land Disposal Restrictions program. Section 3.0 describes the industry and processes
generating K088 and presents data characterizing these wastes. Existing management
practices for spent potliners are also described in Section 3.0. Section 4.0 explains the
methodology and rationale for the selection of the regulated constituents, discusses treatment
1-6
-------
I
I
• technologies for the* wastes, and presents the determination of the BOAT treatment standards
for these wastes. In addition, potential reuse and recycling, source reduction, pollution
| prevention, and waste minimization alternatives are discussed in Section 4.0. References are
listed in Section 5.0 and are cited numerically within the document in parentheses (e.g., (1)).
I
I
I
1
I
I
I
I
I
Acknowledgements are provided in Section 6.0. Tables are located at the end of each section.
1-7
-------
Table 1-1 'BOAT Treatment Standards for Nonwastewater Forms of K088 for
Organic Constituents and Cyanide
Regulated Constituent
Cyanide
Cyanide (amenable)
Acenaphthene
Anthracene
Benz(a)anthracene
Benzo(a)pyrene
Benzo(b and k)fluoranthene*
Benzo(g,h,i)perylene
Chrysene
Dibenz(a,h) anthracene
Fluoranthene
Indeno( 1 ,2,3-cd)pyrene
Phenanthrene
Pyrene
Total Composition Concentration
(rag/kg)
Maximum for any Grab Sample
590
30
3.4
3.4
3.4
3.4
6.8
1.8
3.4
8.2
3.4
3.4
5.6
8.2
* The treatment standard for these constituents is expressed as a sum of their concentrations to
account for analytical concerns in distinguishing between the two compounds.
Ref.: 2) U.S. Environmental Protection Agency, Office of Solid Waste. Final Best
Demonstrated Available Technology (BOAT) Background Document for Universal Standards:
Volume A. Universal Standards for Nonwastewater Forms of Wastes. U.S. Environmental
Protection Agency, Washington, DC, July 1994
1-8
-------
I
I
I
I
I
I
I
1
I
I
I
I
I
I
I
I
I
I
I
Table 1-2 BOAT Treatment Standards for Nonwastewater Forms of K.088 for
Metals and Fluoride
Regulated Constituent
Fluoride
Antimony
Arsenic
Barium
Beryllium
Cadmium
Chromium (total)
Lead
Mercury
Nickel
Selenium
Silver
Maximum for any Grab Sample (mg/L)
Using TCLP
48
2.1
5.0
7.6
0.014
0.19
0.86
0.37
O.G25
5.0
0.16
0.30
Ref.: 2) U.S. Environmental Protection Agency, Office of Solid Waste. Final Best
Demonstrated Available Technology fBDAT) Background Document fo- Tniversal Standards:
Volume A. Universal Standards for Nonwastewater Forms of Waste?. U.S. Environmental
Protection Agency, Washington, DC, July 1994
1-9
-------
Table 1-3 BOAT Treatment Standards for Wastewater Forms of K.088
Regulated Constituent
Acenaphthene
Anthracene
Benz(a)anthracene
Benzo(a)pyrene
Benzo(b and k)fluoranthene*
Benzo(g,h.i)pcrylene
Chrysene
Dibenz(a,h) anthracene
Fluoranthene
Indeno( 1 ,2,3-cd)pyrene
Phenanthrene
Pyrene
Cyanide (total)
Cyanide (amenable)
Fluoride
Antimony
Arsenic
Barium
Beryllium
Cadmium
Chromium (total)
Lead
Mercury
Nickel
Total Composition Concentration (mg/L)
Maximum for any 24 Hour Composite
0.059
0.059
0.059
0.061
0.11
0.0055
0.059
0.055
0.068 ,
0.0055
0.059
0.067
1
1.2 |
0.86 1
35 |
1.9 1
1.4 |
1.2 |
0.82 1
0.69 |
2.77
0.69 |
0.15 I
3.98 1
1-10
-------
Table 1-3 BOAT Treatment Standards for Wastewater Forms of K088
Regulated Constituent
Selenium
Silver
Total Composition Concentration (rag/L)
Maximum for any 24 Hour Composite
0.82
0.43
' The treatment standard for these constituents is expressed as a sura of their concentrations to
account for analytical concerns in distinguishing between the two compounds.
Ref.: 3) U.S. Environmental Protection Agency, Office of Solid Waste. Final Best
Demonstrated Available Technology (BDAD Background Document for Universal Standards:
Volume B. Universal Standards for Wastewater Forms of Wastes. U.S. Environmental
Protection Agency, Washington, DC, July 1994
1-11
-------
2.0 LAND DISPOSAL RESTRICTIONS FOR K088
2.1 Summary of Basis for Listing of Spent Potliners
As presented in the Listing Background Document (5), the Agency determined that
spent potliners are a solid waste that may pose a substantial present or potential hazard to
human health or the environment when improperly transported, treated, stored, disposed of, or
otherwise managed. The Agency determined that spent potliners contain toxic constituents
that are mobile and/or persistent in the environment and therefore, are capable of reaching
receptors in harmful concentrations. The information that supports these findings is presented
in the Listing Background Document, and in the RCRA Docket supporting the listing of spent
potliners. Specifically, spent potliners are listed as a hazardous waste because:
• Spent potliners contain significant amounts of iron cyanide complexes and free
cyanide. EPA has detected both iron cyanide complexes and free cyanide in
spent potliners in significant concentrations.
• Free cyanide is extremely toxic to both humans and aquatic life if ingested.
• Available data indicate that significant amounts of free cyanide and iron
cyanide will leach from potliners if spent potliners are stored or disposed in
unprotected piles outdoors and are exposed to rainwater.
• Damage incidents have been reported that are attributable to improper disposal
of spent potliners, demonstrating migration, mobility, and persistence of waste
constituents and demonstrating that substantial hazard can result from improper
management of this waste.
• Generation of large quantities of the waste increases the potential for hazard if
mismanagement should occur.
Spent potliners from primary aluminum reduction listed as K088 (which are more fully
described in Section 3.0) include the carbon portion of the materials contained inside the
electrolytic reduction cell and does not include other material contained in the pot such as the
collector ban, steel shell, or thermal insulation composed of insulating brick, or alumina.
2-1 .
-------
I
I
I
2.2 K*v points of Spent Potliner Standards and How They Reflect LDR Goals
The LDP. program is designed to protect human health or the environment by
• prohibiting the land disposal of RCRA hazardous wastes unless specific treatment standards
are met.
I
I
I
I
I
I
I
I
I
I
I
I
I
In RCRA Section 3004(m), Congress directed the Agency to:
"...promulgate...levels or methods of treatment...which substantially diminish the
toxicity of the waste or...the likelihood of migration of hazardous constituents ...so that
short-term and long-term threats to human health and the environment are minimized."
Key provisions of the LDR program require that: (1) treatment standKds are met prior
?fi ^B
to land disposal, (2) treatment is not evaded by long-terra storage, (5) acnu»eatmeni occurs
r%t
rather than dilution, (4) recordkeeping and tracking follow a waste from "cradje to grave"
(i.e., generation to disposal), and (5) certification verifies that the specified treatment
standards have been met.
*•• '
The Agency is establishing treatment standards for both non-wastewater and
lit W
wastewater forms of this waste as concentrations numerically equivalenFto the universal
9f
treatment standards for the constituents selected for regulation in these wastes! As discussed
in the Advance Notice of Proposed Rulemaking (ANPRM) publishedjn October 24, 1991 (56
FR 55180-55182), EPA was seeking development of concentration-based standards for K088
so as to allow the use of any technology that can achieve the numerical values. The EPA is
developing a concept for K.088 (and other wastes) that uses a hierarchy of options for
evaluating treatment or recycling technologies. For example, the Agency's ultimate goal for
hazardous waste is source reduction, that is, less or no production of hazardous waste.
Another environmentally acceptable option would be total recycle or reuse. This would
require that all the waste generated be used as feedstock in the same process or another
process. Because these options are unlikely to be feasible for all wastes, the Agency would
look further down the hierarchy to find an acceptable option. Next lower in the hierarchy
2-2
-------
would be treatment technologies that can recover materials from the waste for reuse. This
option would most likely result in some residuals still needing to be land disposed but
preferred techniques would also significantly reduce the quantity and toxicity of any waste
destined for land disposal. Further down in the hierarchy would be technologies that reduce
quantity and toxicity without the recovery of materials for reuse. The next ievel down would
be technologies that lower toxicity alone and may increase volume of materials for land
disposal. For example, while stabilization can sometimes render a metal-bearing waste less
hazardous, it often results in a great increase in volume which then has to be landfilled. At
the base of this hierarchy pyramid are numerical treatment standards. Numerical treatment
standards are required because most treatment or recovery technologies result in some
residuals requiring land disposal. When this occurs, the residuals must comply with the
treatment standard before being landfilled.
The Agency believes that establishing treatment standards for the constituents in spent
potliners as equivalent to the corresponding universal standards meets it goal of minimizing
threats to human health and the environment from land disposal. EPA has identified a broad
range of treatment, recycling, reclamation, and reuse practices as applicable technologies to
K.088. Any technology that actually recovers some of the value in K088 is the Agency's
preference for treatment Because the Agency does not have adequate data on the products or
residues created by these recycling technologies, numerical treatment standards are being set.
The universal standards for nonwastewater and wastewater forms of the waste were developed
based on treatment performance data used to promulgate previous BOAT treatment standards,
and therefore, have already.been determined to meet the Agency's requirements of BOAT.
2-3
-------
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
3.0 DETAILED DESCRIPTION OF SPENT POTLINERS
This section describes the industry generating Hazardous Waste Number K088, the
facilities generating these wastes, the processes generating the waste, the physical and
chemical characteristics of this waste, and waste management practices of facilities generating
these wastes.
3.1 Description of the Aluminum Production Industry
Hazardous Waste Number K088 is generated by the aluminum manufacturing-industry.
This section includes a description of the aluminum industry, the size and geographic
distribution of aluminum reduction facilities, other manufacturing operations performed at
aluminum reduction facilities, raw materials used at aluminum reduction facilities, and
aluminum reduction end products and their uses.
3.1.1 Description of Aluminum Reduction Facilities
Aluminum reduction facilities are classified by the U.S. Office of Management and
Budget under Standard Industrial Classification (SIC) code 3334, which is under major
heading 33, primary metal industries (6).
The production of aluminum occurs in four distinct steps: (1) the mining of bauxite
ores, (2) the refininj of bauxite to produce alumina (AljO3), (3) the reduction of alumina to
aluminum metal, and (4) casting of the molten aluminum. Since this report focuses on the
generation of spent potliner from primary aluminum reduction, only that portion of the
process will be described. Figure 3-1 provides a simplified process flow diagram of the
aluminum reduction process. The diagram shows the location of the generation of the spent
potliner in the process.
3-1
-------
c
o
(ft
tt
0)
O
O
oZ
c
o
3
•o
c
E
3
<
0)
E
o
o
CO
I
CO
o
3
3
il
9
4)
1*
•o c
o c
(/) 4
"*
w
i
£
1°
>,
a
&
w
_«
°w
u
V
m
;
OJ
S~™
U
i k
-?
s!
f!
c o
< X
t ^
Crushing and
Classifying
t
nil
3C <"",
i
Si
«- _
2 1
u E
f *
u
a
•o
52
3|
£<
a
1
w
a
j k.
?
3
- k
o»
. "*
«
£
M
* * W
1^1
U.-|
• *5
« 3 OT
(90
t
1
u
^ '«.
^ 1
> D
t \t
s
2
^
u
E
9
Tj
2
s.
it
c
<
j
s
0
«•
o
1
i
•0
09
o
*:
" * e
« e
1= 'i
12 • 1
»
1 - c
c?9 ?-: S
•? — £ 5.^ ti
E «s ^ w E
5 * * B C ^ . m
•* <• «• a ~ ST ^
->
c.
^
1
e
<
c
c
~
'
(
'
4
!
!
(
L
C
-!
i_
c
&
o
u
u
ra
J^
3
V3
U
~
e
C^l
c
£
<
SJ
'w
C_
1)
u.
t£
•y>
w
«>
.1
; ;/;
> u
. u
•> c
) (3
i E
?£
* L.
: ^
3 —
- U
^ Si
3 I
"• ry
§ e
3Z
= T
3 C
5 *
s <*
> S
2 'i
. T
O •:
2 0
u
u
o
cl
"2
u
1
a
_w
c
U
V5
>
C
u
/?
O
u
a
/*
V2
C
£
_c
u
u
c
c
ra
S
in
R
U
^r
•£.
S
c
k»
c^
c
<*»
Z
*—
D
i
<—
minum Senmcnl
j^
= 2
50
C
o
,o
-------
I
I
I
I
I
I
I
I
I
I
I
All primary aluminum produced in the United States is manufactured by the Hall-
Heroult Process. Aluminum is refined by dissolving alumina (aluminum oxide) in a molten
cryolite (Na3 6) bath. An electric current is then introduced reducing the alumina to
aluminum. This technique for reducing aluminum from its oxide was discovered in 1886
simultaneously in the United States and France. The reduction process requires high purity
aluminum oxide, carbon, electrical power, and an electrolytic cell. The reduction takes place
in carbon-lined, steel electrolytic hall cells or pots. Pots consist of a steel container lined with
refractory brick with an inner lining of carbon. The carbon liner is usually up to 15 inches
thick and serves as the cathode in the electrolytic circuit collecting aluminum ions from the
molten bath. The size of a pot ranges from 6x18 to 14x42 feet (7). Figure 3-2 shows a
general sketch of a typical pot. These pots are connected in series to form a potline. Potlines
may contain 100 to 250 cells electrically connected in series. Incoming alternating current is
transformed directly to DC at high voltages and is fed to a line of pots connected in series.
In this way, the operation is essentially at constant current but the individual voltages can be
varied on each pot The electrical supply is direct current on the order of several hundred
volts and 60,000 to 100,000 amps. The carbon liner of the pot is electrically active and
constitutes the cathode of the cell when covered with molten aluminum. The electrolysis
takes place in a molten bath composed principally of cryolite. The electrolyte consists of a
mixture of the following materials:
Cryolite ' 80 to 85%
Calcium fluoride 5 to 7%
Aluminum fluoride 5 to 7%
Alumina 2 to 8%
The essential ingredient of the electrolyte is cryolite which is the best flux for alumina.
Various additions to the cryolite modify its physical and chemical properties and thus improve
cell performance. Aluminum fluoride and calcium fluoride lower the freezing point of the
electrolyte. The function of the electrolyte is to enable physical separation between the
3-3
-------
O
a.
o
'5.
o
E
2
(5
o
T3
.2
"5.
E
GO
CO
o
3
O)
iZ
c
o
1)1
it:
c/:
4J
a
E
U '
fl.
O «
5
e -
_0
u
u is
o c
4J
2
c .
-------
I
I
•
• cathodically produced aluminum and the anodically evolved oxides of carbon while also
enabling electrolytic decomposition of the alumina,
I
I
I
I
I
I
I
I
I
The composition of the bath varies as electrolysis proceeds. Electrolyte is absorbed by
the lining, which becomes saturated in the first 80 to 85 days of operation, the electrolytic
bath normally operates at approximately 950 °C and alumina is added periodically to the bath
to maintain a relatively constant concentration of alumina in the molten bath. The aluminum
reduction reaction results in reducing aluminum in the trivalent state to liquid metal at the
cathode. Oxygen appears at the anode and reacts with the anode to form a mixture of 75
percent carbon dioxide and 25 percent carbon monoxide and consumes the carbon anode (8).
The main electrochemical reaction occurring is represented by the equation
2A1203 (dissolved) + 3C(s) ** 4A1(1) + 3CO:(g)
• with the aluminum being deposited at the bottom of the cell.
• There are basically two types of cells that are used for the production of aluminum.
• The primary difference between the cells is the manner in which the anode is baked and
consumed. Two methods of replacing the anodes are practiced.' These are referred to as
• prebaked anode (intermittent replacement) and the Soderberg anode (continuous replacement).
™ For either system, the anode preparation begins in the anode paste plant, where petroleum
• coke and pitch are hot blended. For prebaked anodes, the anode paste is pressed in molds,
™ and the anodes are baked in an anode bake plant The baked anodes are used to replace
• consumed anodes. In the Soderberg anode system, the anode paste is not baked initially, but
is fed continuously in the form of briquettes through the shell of the pot As the anode is
consumed in the pot it must be continually lowered to maintain a constant depth of anode
immersed within the electrolyte. Additional paste is added to the top of the anode to replace
the consumed anode. As the paste approaches the hot bath, the paste is baked in place to
form the anode. Soderberg anodes are supported in the pot by one of two methods: vertical
stud Soderberg (VSS) or by horizontal stud Soderberg (HSS) (7). Figures 3-3 and 3-4 show
3-5
-------
sketches of the VSS and HSS type of Soderberg cells. While all electrolytic pots operate on
the same principles, the pots produced by each aluminum company mayvary in design.
The cathode of the aluminum reduction cell is a carbon liner on which the pool of
molten aluminum rests. Alumina is added to the bath intermittently to maintain the
concentration of dissolved alumina within the desired range. Typically cells have an
automatic or semiautomatic feeding system. At predetermined intervals the overlaying crust is
broken into the bath to replenish the alumina content of the bath and a predetermined amount
of alumina is dumped from hopper to reform the crust. Molten aluminum is withdrawn
intermittently from the bottom of the bath at a rate of approximately 1 ton per day. The
molten aluminum is collected from each pot by siphoning a measured aliquot from the cell
into a transportable vessel. The aluminum is then taken directly to the casting process to be
cast into ingots or pigs as the final product in a separate casthouse facility or it is taken to a
holding furnace.
It is essential for purity of the product aluminum and the structural integrity of the cell
that the molten aluminum be isolated from the steel shell. Over the life of the cathode, the
carbon materials become impregnated with the cryolite electrolytic solution. As the cryolite is
absorbed into the cathode, the integrity of the lining can be reduced andcracks or heaving of
the lining can occur. A.service life of four to seven years for a potliner is common. In
several cases, it has been reported that a pot can remain in service for up to ten years (25).
Aluminum companies are working to extend the life of potliners, however, longer use of a
potliner may affect the quality of the aluminum produced Some of the factors that may
impact the pot life include strength of the pot shell, cell preheat procedures, quality of cathode
blocks and sidewall blocks, and type of ramming paste. A pot "fails" when iron is detected in
the molten aluminum, when cell voltage increases, or when the shell leaks molten metal or
electrolyte. The iron contamination can be caused by the development of cracks or by erosion
in the carbon lining, which allow electrolyte to come in contact with the steel collector bars or
steel shell. Upon failure of a liner, the cell is emptied and cooled, The pot is then removed
from the cellroom to a working area or dismantled in place. By mechanical drilling and/or
3-6
-------
0
o
CL
CO
CO
o
E
2
Q
T3
gO>
"5.
E
CO
CO
I
CO
0)
3
O)
iZ
u
s/1
c
c3
.^ir
.§
—
c
u
c
u:
w
,0
**••
c
U
2
U
o
Q
»^
c
E
^c
T
>
^
—
UMB
O
e:
c
u
15
o
Js
3
C/3
••i
c
"l
E
00
E
—
C
' ?-
c
<;
&
03
C
W
h.
CL
U
U i.
!<=
*3
• <— c
O a
« "c
,H «
(C —
<- 00
O 0
. u
>> c
U (Q
5 §
ca o
<^
c -^
o —
'S U
0 U
O i
2£
CL ^
"«3 j
gz
o w
_w
'>
c .
UJ -
VJ •:
=5 L
O
**•
5
3
s
3
^
^
H
V5
ZJ
>
t_
u
0
«
U
u
i!
2
c
CO
c
c
a.
u
c
u
C
t(Z
C
nj
s
V)
a
13
V)
1
.u
c
**
2
u
^
o
c
E
c
V
£
•
•
<
T
r*.
as
_«
~
<3
*
CJ
Q
K
c
0
si
c
'•i
5
^
>^
0
c
-------
•a
. c
u
§.£
'2 U
2 o
S-2
c .=
u
c
CO
o
c
u
cfl
u
=0
r~>
<—.
u
-------
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
soaking in water, the steel shell is stripped of the carbon lining. There are two portions of
spent potliner. These are designated as first cut and second cut potliner. First cut potlinsr
consists of the upper portions of the carbon from the bottom block and side walls. Second
cut material is the thermal insulation composed of carbon insulating brick or alumina. The
first cut carbon lining is the subject of the K088 listing.
3.1.2 Size and Geographical Distribution of Facilities
At the time of the original listing, May 19, 1980 (45 F_£ 33112), EPA had identified
30 primary aluminum reduction plants generating spent potliners (5). Since that time several
have closed or become inactive. As of 1993, there were 23 aluminum reduction facilities and
generators of spent potlinen operated by 13 companies or consortia in the United States. One
of these facilities was temporarily closed in 1993. These companies produced an estimated
3,700,000 metric tons of aluminum in 1993 and generate between 100,000 and 120,000 metric
tons/year of spent potliner (32, 33). Primary aluminum smelters are located in the following
states:
Indiana Ohio
Kentucky (2) Oregon (2)
Maryland South Carolina
Missouri Tennessee
Montana Texas
New York (2) Washington (7)
North Carolina West Virginia
Table 3-1 (located at the end of this section) provides a list of facilities generating
spent potliners in 1991. This list includes the location of the facility, quantity of spent
potliner generated and the management practice. Figure 3-5 shows a map of the U.S. with the
approximate location of each facility. The overall aluminum reduction capacity in the U.S. is
discussed in Section 3.1.5 below.
3-9
-------
a:
in
z
Zj
h-
O
CL
LU
a.
o
£
LU
OJ
o
CO
LU
O
<
LL
LU
O
CO
o
3
in
•
CO
o
f- CO O> O »- CN| CO
T- •»- »- CN CM
-------
3.1.3 Raw Materials
The primary raw materials used in the manufacture of aluminum are alumina, cryolite,
aluminum fluoride, calcium fluoride, carbon, petroleum coke, and pitch. The process also
uses large quantities of electrical energy.
3.1.4 Aluminum Products and Their Uses
Aluminum and its alloys have properties that make it one of the most widely used
metals in the world. The best known property of aluminum is its light weight. Its specific
gravity is 2.7 and is approximately one-third as dense as iron, copper or zinc. Despite its
light weight, it can be made strong enough to replace heavier and more costly metals in many
applications. Aluminum and its alloys are highly resistant to corrosion making them very
useful in coating applications. Its high electrical conductivity and comparative low density
make aluminum ideal for. many electrical transmission and distribution uses. Because
aluminum is an excellent conductor of heat, it is widely used in heat exchange applications
such as radiators and cooling coils. In addition, aluminum is an excellent reflector of all
forms of radiated energy, which results in wide use in roofing materials and building
insulation. Because aluminum is effective at keeping-heat in or out it is also widely used as
food wraps.
3.1.5 Aluminum Reduction Capacity
Domestic primary aluminum production decreased significantly in 1993. In response
to rising world inventories and falling prices, more than half of the domestic primary
aluminum smelters announced temporary shutdowns in production capacity during the year.
By the end of 1993, domestic smelters had closed 769,000 tons of annual capacity and were
operating at approximately 80 percent of engineered or rated capacity. The estimated
aluminum reduction capacity in 1993 was approximately 4,163,000 metric tons per year.
3-11
-------
3.2 Waste Stream Characteristics
3.2.1 Waste Stream Status Under Other Regulations
Under the Clean Water Act, the discharge of pollutants into surface waters and
Publicly-Owned Treatment Works (POTWs) from primary aluminum smelting facilities is
regulated under the Aluminum Segment of the Non-ferrous Metals Manufacturing Point
Source Category (40 CFR Part 421 Subpart B). This subpart includes effluent limitations and
standards for cyanide, fluoride, antimony, nickel, aluminum, benzo(a)pyrene, oil and grease,
TSS, and pH for wastewaters discharged from certain aluminum reduction processes. This
includes wastewaters from the anode and cathode paste plants, anode bake plant, cathode
reprocessing, potline and potroora air pollution control, aluminum degassing, pot repair and
soaking, and aluminum casting.
Of the 26 constituents selected for regulation for spent potliners under the LDR
program anthracene, antimony, arsenic, barium, beryllium, cadmium, chromium, lead,
mercury, nickel, selenium, and silver are regulated under the Emergency Plannkngand
Community Right-to-Know Act (EPCRA) Section 313. Under Section 313, facilities that
manufacture, process, or otherwise use these chemicals, and that meet; certain other criteria,
must report the releases and transfers or these chemicals.
Under the Clean Air Act, the Agency has set Standards of Performance for New
S'ationary Sources that requires any facility that commences construction or modification after
October 23, .1974 to meet emissions standards for fluoride and opacity. In addition, under the
Clean Air Act (CAA), Section 112, National Emission Standards for Hazardous Air Pollutants
CNESHAP) program, Primary Aluminum Production facilities are listed as a source category
scheduled for rulemaking by November 15, 1997 (see 58 FR 63941, December 3, 1993).
These NESHAP rules, once promulgated, are intended to limit the emissions of hazardous air
pollutants from certain facilities.
3-12
-------
Waste characterization data for nonwastewater forms of K088 were obtained during
EPA-conducted sampling at three facilities during 1990 and data submitted with two delisting
petitions (8, 9, and 10). While the Agency realizes that there is an abundance of
characterization data on spent potliners, most of it is limited to a few constituents (e.g.,
cyanide) and generally lacks the rigorous QA/QC requirements of the land disposal program.
Thus, the Agency has limited its summary of characterization data to samples of spent
potliner from three studies. This data includes information from spent potliners from eight
aluminum reduction facilities. Table 3-2 presents the waste characterization data. The BOAT
List constituents generally found in this waste included polynuclear aromatic hydrocarbons,
metals, and cyanide. One non-BDAT List constituent, fluoride, was also detected at high
concentrations in these wastes. Typical concentration of constituents in potliners are as
follows: 50% carbon (block), 10-15% sodium, 10-15% fluoride, 0.5% cyanide, <0.1%
polynuclear aromatics, <0.1% phenols and various concentrations of hazardous metals.
Available data shows that levels of cyanides appear to vary from pot to pot and within
a pot. Within a pot, levels can vary between the bottom of the pot liners and its side or end
walls. Cyanide is generally found in higher concentrations at the side wall where the bottom
block carbon is exposed to air. Total cyanide may vary by two orders of magnitude within a
single pot. In contrast to cyanide, fluoride is generally found in the bottom block carbon
since it is in direct contact with the molten fluoride salt. EPA expects that waste variabilities
with respect to cyanides (total) could be minimized during size reduction activities prior to
thermal treatment or recovery.
For wastewater forms of K088 there. limited data available. The Agency has two
sets of data on the generation of wastewater during the treatment of nonwastewater forms of
K088. The first is a quench water from the treatment of K088 using the Ormet process (8)
and the second is a scrubber water blowdown from the pilot-scale incineration of K088 (11).
Table 3-3 presents these waste characterization data.
3-13
-------
3.2.3 Amenability of Wastes to Chemical Analysis -
3.2.3.1 SW-846 Method Applicability
The following provides details of the analytical methodologies which-can be used for
determining constituent concentrations in spent potliners for comparison with the BOAT
treatment standards. It is assumed that the waste samples which will be analyzed are
representative of the waste stream and that appropriate preservation techniques are used, if
necessary, prior to shipment to the laboratory. In addition to the following information, it is
assumed that all analyses will be performed according to the guidance provided in the Best
Demonstrated Available Technology (BOAT) Background Document for Quality
Assurance/Oualitv Control Procedures and Methodology (I). This document provides
guidance regarding preparation of analysis plans, QA/QC, calibration, and procedures for data
reduction, validation, and reporting.
EPA-approved methods for the analysis of BDAT List constituents in nonwastewater
and wastewater forms of wastes are presented in the Agency's Test Methods for Evaluating
Solid Waste, Physical/Chemical Methods (SW-846), Third Edition (USEPA, 1986X12). Each
BDAT List constituent selected for regulation in spent potliners, except fluoride, is listed as a
target analyte by at least one SW-846 method. Fluoride does not have an SW-846 method.
An American Society of Testing and Materials (ASTM) method is recommended for the
analysis of fluoride. Table 3-4 through 3-6 lists the SW-846 and ASTM methods applicable
to the analysis of each constituent selected for regulation in spent potiiner. These tables
provide examples of extraction, cleanup and analytical methods used for the regulated
constituents.
Semivolatiles
All semivolatile organic compounds listed in the treatment standards set for spent
potliners (acenaphthene, anthracene, benz(a)anthracene, benzo(a)pyrene, benzo(b and
3-14
-------
k)fluoranthene-, benzo(g,h,i)perylene, chrysene, dibenz(a,h)anthracene, fluoranthene,
indeno(l,2,3-cd)pyrene, phenanthrene, and pyrene) may be analyzed according to Method
8270A or B of SW-846. The method uses gas chromatography/mass spectrometry (GC/MS)
and a capillary column.
As noted in Step 7.1 of Method 8270A, all waste samples must be prepared by either
SW-846 Method 3540A, Method 3550, or Method 3580A prior to GC/MS analysis. All
liquid samples must be prepared by either SW-846 Method 3510A or Method 3520A. Direct
injection is not appropriate for these extracts. The choice of a preparation method may be
made by each individual facility, but must be specified in the sampling and analysis plan
(S \P) prepared prior to the treatment test.
As noted in Step 7.2 of Method 8270A, the extracts from the preparation methods may
be cleaned up by any of a variety of SW-846 methods prior to GC/MS analysis. Applicable
methods which may be needed include Methods 3610A, 3630A and 3640. The SAP should
provide the flexibility for the laboratory to perform an SW-846 cleanup method, if required
on individual samples. These constituents may also be analyzed by 8250, 8100, and 8310.
(12)
The Agency is regulating benzo(b)fluoranthene and benzc(k)fluoranthene as a sum in
K088 wastes. The Agency recommends the use of SW-846 Method 8270A or B, which
requires the use of a GC/MS for measurement of the concentration of these compounds.
When analyzing for these compounds using this method, these two stereoisomers co-elute.
Since the two constituents may not be accurately quantified separately, the Agency is
regulating these constituents as a sum in both wastewater and nonwastewater forms of waste.
Cyanide
Cyanide analyses may be performed according to SW-846 Method 9010A or 9012.
The methods must be followed, as written, for liquid samples. However, because these
3-15
-------
methods do not specify a sample size for solid samples, the same guidance given for F006
and F019 samples is recommended for K088 samples (see 55 £& 22578 and 55 F_E 3870).
Specifically, cyanide analyses should be performed with 10 g of solid and the solid should be
distilled for one hour and fifteen minutes. This time limit does not include the time needed
for cooling the distillation flask, which should be at least an additional 15 minutes.
Spectrophotometric, titration, or automated procedures may be used to quantify the cyanide
following the distillation procedure.
Leachable Constituent Concentrations
All samples to be analyzed for leachable quantities of metals and fluoride are based on
the Toxicity Characteristic Leaching Procedure (TCLP) extraction. The analytes may all be
analyzed from the same extract, which must be generated according to the March 29, 1990
version of the TCLP, as amended on June 29, 1990, August 10, 1990, and November 24,
1992. Caution should be' used when extracting these samples, due to the potential production
of toxic gas. The extraction vessels should be carefully monitored for at least the first 8
hours of the extraction and vented into a hood.
Fluoride
Extracts or wastewaters may be analyzed for fluoride using an ion selective electrode
(ISE) method. SW-846 does not include this method, but it is available from the ASTM as D
1179-30, Method B (13). The ASTM method describes a manual ISE method and includes
precision data. No preparation of the sample extract should be necessary.
Metals
All metals except arsenic, selenium, and mercury may be analyzed according to
SW-846 Method 6010A. The extract should be digested according to SW-846 Method 3050A
prior to ICP-AES analysis. Arsenic and selenium may be analyzed according to SW-846
3-16
-------
I
• methods 7060 and 7740, respectively. Preparation techniques for the TCLP extract are listed
in the individual atomic absorption (AA) methods. Mercury analysis of the TCLP extract
must be performed according to SW-846 Method 7470.
I
Additional guidance on which method is appropriate for a specific sample is found in
• the appropriate SW-846 method (12). Table 3-7 lists prepatory method. Table 3-8 lists the
instrumentation required for each method,
I
3.2.3.2 Sample Preparation Issues
I
I
I
I
I
I
I
I
I
Sample preparation issues for cyanide were discussed in the above section.
3.3 Current Spent Potliner Waste Management Practices
This section describes waste management practices at aluminum reduction facilities
including routing of waste, treatment units and practices, disposal of treatment residues, and
waste minimization, pollution prevention and recycling practices.
The principle information source available for waste generation rates and waste
| management practices are 1991 Biennial Reporting System National Oversight Database
'(BRS) information, 1988 requests for information on the generation and management of spent
• potliners, as well as delisting petitions submitted to the Agency (8, 10).
I 3.3.1 Waste Management Practices for K088
I
Spent potliners from primary aluminum reduction, listed as Hazardous Waste Number
K.088, include the carbon portion of the materials contained inside the electrolytic reduction
cell and do not include other material contained in the pot such as the collector bars, steel
shell, or thermal insulation composed of insulating brick or alumina. These materials can be
segregated but have in the past been co-disposed.
3-17
-------
Upon failure of a liner in a pot, the cell is emptied, cooled, and the lining is removed.
This operation can be performed in place or the pot may be removed from the cell room to a
working area (depending on space limitation). The steel shell is stripped of the podiner by
mechanical drilling and/or soaking in water. The potliner removed from the cell is generally
stored in rail cars, dumpsters or in covered piles prior to treatment or disposal.
Information on the generation and management of spent potliners was collected from
the 1991 Biennial Reporting System National Oversight Database (BRS database). The BUS
database was searched for information on the 23 domestic aluminum reduction facilities. The
search was limited to the non-CBI information available in the BRS database. Of the 23
facilities, four were not found in the BRS database. Information for the remaining four
facilities was collected from delisting petitions and other publicly available sources. One
reason that these facilities were not in the BRS database is that the facilities may have
claimed that their 1991 National Biennial Reports contain Confidential Business Information
(CBl). Reports containing CBI data are not kept in the public BRS database. Table 3-1
presents the 1991 data on K.088 generation and management reported in the Biennial
Reporting System for generators of K088.
In 1991, spent potlinen were managed primarily in RCRA Subtitle C landfills. Of the
approximately 100,000 tons generated in 1990 over 80 percent were landfilled. This practice
is changing as new treatment technologies become available. In late 1993, Reynolds Metals
started its process for treating spent potliner and began treating all of its spent potliner by this
process (see Section 4.2.1.1.3 for a description of the Reynolds process). The bulk of the
residues from this process are no longer subject to the Subtitle C regulations because of a
delisting granted to Reynolds Metals for the kiln residues. Currently all of the kiln residues
from the Reynolds process are landfilled in a Subtitle D landfill. Reynolds has also begun
accepting wastes from several other facilities including the six Alcoa facilities (Alcoa, TN,
Rockdale, TX, Wenatchee, WA, Massena, NY, Warrick, IN, and Badin, NC). It is anticipated
3-18
-------
I
I that Alcoa will continue to send its spent potliner to the Reynolds facility for treatment. (27,
28, 29, 30, 31) If all of the Alcoa and Reynolds spent potliner was treated at the Reynolds
I facility, over 50 percent of the spent potliner generated in the U.S would be treated at this
facility.
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
3.3.1.2 Waste Minimization and Recycling Activities
Based on the 1991 BRS database, no pollution prevention or waste prevention
activities were occurring at the aluminum reduction facilities. While limited information
regarding waste minimization activities at aluminum reduction facilities is available, primary
aluminum reducers are actively working to .increase the life of pots. By increasing pot life,
generation of spent potliner is decreased, as pots are changed out less frequently. Several
factors impacting the life of a cell have been studied. Research has concluded that the factors
affecting pot life include:
• Use of stronger steel shells that limit the deformation of the cathode
Cell preheat procedures,
• Use of high quality cathode blocks
• Use of sidewall blocks with higher thermal conductivity
Use of ramming pastes with better physical and mechanical characteristics
• Balancing of the magnetic field in the cell.(36)
One potential additional method to reduce the generation of spent potliners is to
increase the recovery and reuse of post-industrial and post-consumer aluminum. Increased use
of secondary aluminum could replace some demand for virgin aluminum. If the increased
demand for aluminum, is in part met by secondary aluminum sources, additional primary
aluminum reduction capacity may not be required thus potentially reducing the quantity of
spent potliners generated. In 1993, the quantity of aluminum recovered from purchased scrap
in the U.S. was approximately 1.6 million tons, of which approximately 40 percent was
derived from pre-consumer materials and 60 percent from post-consumer material. (32, 33)
3-19
-------
Table 3-1 Facilities Generating Spent Potliner
Company
A lean
Alcoa
Alcoa
Alcoa
Alcoa
Alcoa
Alcoa
Alumax
Alumax/Eastalco
Alumax/Intalco
Columbia Aluminum
Corp.
Columbia Falls
Aluminum Co.
Kaiser Aluminum
Kaiser Aluminum
National Southwire
Noranda Aluminum
Northwest Aluminum
Ormet Corporation
Ravenswood
Aluminum Corp.
Reynolds
Reynolds
Reynolds
Vanalco
Location
Sebree. KY
Evansville. FN
Badin, NC
Alcoa, TN
Rockdale, TX
Wenatchee, WA
Massena, NY
ML Holly, SC
Frederick. MD
Femdale. WA
GoldendaJe, WA
Columbit Falls, MT
Tacoma. WA
Spokane. WA
Hawesville. KY
New Madrid, MO
The Dalles, OR
Hannibal. OH
Ravenswood, WV
Massena, NY
Longview, WA
Troutdale, OR
Vancouver, WA
1992 Reduction
Capacity (Thousand
Metric Tons)
180
300
115
210
315
220
125
184
170
275
168
168
73
200
186
215
82
245
161
123
204
121
116
1991 Spent
Potliner
Generation
Rate (tons)
2.927
9.500(1989)
3,097
3.923
16,068
4.657
2,974
1,727
4.000(1989)
W97
1.65
4.129
5.084
•3,200(1989)
2,762
5.542
5,236
6.407
5584
3.720 (1989)
6,728
2J25
3,920
Management/Treatment
Techniques
Landfill
Not Available
Landfill, stabilization, or
chemical precipitation
Landfill
Landfill
Landfill
Landfill
Landfill
Not Available
Not Available
Landfill
Landfill
Landfill
Not Available
Landfill
Landfill or Stabilized
Landfill
Landfill
Landfill
Not Available
Laadfill
Landfill
Landfill
References: (8, 10, 26, and 32)
3-20
-------
I
I
I
I
I
I
I
I
00
3O
o
2
13
Q
c
_o
U
(N
ea
b
w
s
a
Z
'5
ei
U.
00
r~
NO
m
C^4
"™
U
!
ntration
i
o
'5
8.
s
o
U
V
a
Conslid
Z
Z
z
z
z
8
s
^O
§
5
E
aluminu
3
v"
v
o
V
oo
6
q
w-i
q
i
•ON
rr
V
?
a
antimon
r-
i
o
o
o
0
-
V
f*;
Y
9
arsenic
o
i
o
oo
o
oo
1
o
o
r-l
i
ON
O
O
o
§
ON
rr
3
rr
1
0
1
•••
NO
^*
cn
ON
(N
rn
O
m
m
(N
en yi
322
u u u
e-
o
t/i
"° 2 12 w
"o o o r;
§,^§,6
U o jj L.
10
oo
§30
-5 sj -^
•gZ ^
>« C • O
£ u. S J
rf e- § rf
fill
5 >,|| <^
i aril I
zslllS-
^ G < < Z ac
Z '^ — fN rn rr
-------
<
z
<
z
so
(N
oo
<
Z
<
Z
oo
oo
O
a
a
c
o
I
u
u
I
U
e
3
Z
fN
04)
*
"9
ON
so
OO
<
z
(N
o
V
<
Z
<
Z
<
Z
ri
o
CM
<
Z
n
<
Z
<
o
V
(^
OS
Z
<
Z
<
Z
so
SO
O
V
OS
so
<
Z
oo
SO
s
\
~o
J3
g
JU
"o
so
o
*I'11
l|s
-1
2.ill! I
^ -3 < < Z as
Z U. — (N CO TT
-------
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
30
(30
O
c
_o
j
u
u
JS
E
3
Z
'3
•2
=°
(^
*0
m
c.
—
'M
^
M
£
0
entrati
u
e
o
mpositiol
0
U
9t
O
H
'c
-
Z
Z
Z
^
Z
Z
S
m
in
00
E
r~
o
d
V
i
m
O4
O
V
m
d
V
o
in
d
V
0
in
d
V
o
in
d
(N
>n
V
OS
•g
~
X
m
O
V
o
in
OO
o
m
O
o
o
r-
o
150-180
V
";
0
.S
o
m
V
00
(N
m
I
CM
OO
I
m
r-i
vO
1
OO
O
r-
n
O
«n
^r
o
OS
0
o
0
00
^^
3
c
u
T3
C
U
Z
1
^»
Q
0
fN
fS
o
o
«n
^f
Z
Z
Z
u
J3
(0
c
u
u
•o
1
u
i!
P Z
o|
* p^
3 n
•k
«n
"™
0
• L. - O
2 "1 I J
"S • 3 3 •£ -s
O C —" C ^
N « 2 ^ U
•^ 1 S3 S
•< « S ^3 « -^
s^ a "t3 —
- * «r s c o
o ^ o c 2 c
z S ^ 5 o ^
^ -3 « Z a:
Z U, — fN rn ^r
-------
SO
ri
v
o
V
V
I
in
O
V
O
vn
O
V
in
O
0
vn
O
o
vn
0
O
O
V
o
p
o'
V
vn
o
vn
O
O
V
O
vn
O
8 °°
SS
in
V
O
vn
O
NO
?
30
3O
o
v2
2
Q
c
1
u
u
A
E
a
Z
«
u.
in
e
_o
75
w
e
o
o
G5
V
o
o
V
o
s
O
vn
O
o
vn
O
VI
O
I
o
U
O
O
\J
o
w-i
0
o
V
vrv
o
vn
O
o
vn
0
0
u-i
0
vn
O
m
r
V
vn
o
V
o
vn
O
Ci
V
m
.
U
CO t/> CO
"° 2 2
"o o o
c c c
o u u ^
oi o: oi O
vn -o
oo
u
u
u
u
-------
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
oo
oo
o
o
I
5
1
6
u
la
fN
U
V
E
a
Z
£»
'3
91
Lfa
30
-
*
in
T
fN
—
'S
£
B
e
15
V
o
U
o
'S
Q.
e
e
U
H
u
2
r
z
o
V
q
v"
o
v"
o
V
o
vn
fN
o
V
5
o
?
(•J
1 .1
q
"v
i
in
d
o
V
o
q
o
in
O
O
V
o
o
o
V
§
Q
i
o'
m
O
Z
o
V
in
fN
in
fN
d
V
in
d
v
0
o
o
0
9
0
vn
o
9
i
u
E
c
onofluor
E
a
'u.
-O
O
V
o
v
q
V
oo
fN
0
7
i
^
o
oo
vO
?
O
o
i
J3
•«*
c
">»
X
J*
>»
•£
fN
ri
r-i
C3
O
vO
O
V
0
V
1
oo
o
V
o
\7
1
Q
0
oo
?
1
?
c
u
2
X
c
a
d
d
vO
o
V
o
V
1
OO
^y
i
\r\
^^
m
6
V
1
«
o
oo
?
i
?
hracene
c
A
a
1
Jg
§
i
o
V
o
oo
fN
Ov
fN
fN
fN
5
1
«
o
oo
o
V
o
?
^
o.
a
1
J*
6 £
O ,
d d
v
o
V
o
o
oo
(^
VO
(N
i
vn
fN
1
Q
0
00
V
1
v
o
c
uoranthe
G
1
o
d
v
q
v
o
o
oo
r-
VO
fN
"?
fN
1
O
V
0
oo
o
V
vO
?
4>
C
uoranthe
c
2"
1
1
o
d
v
q
V
o
^
T
^m
0
v"
1
O
V
0
00
?
1
?
U
"B
^
00
JS
fN
§
O
v
q
v
o
o
fN
O
oo
oo
1
CO
vn
T
"
1
<^j
o
oo
VO
?
O
v
X
0
o
d
v
q
v
oo
T
fN
,,
O
V
o
v"
1
o
V
o
oo
VO
o
V
1
?
u
c
o
i
•fi.
' ^
1
'"5
U
>: So
Z u
O O O —
c c = i
>%>%>> c
u u u *-
Cti Ci 2i O
in
co
^•^5
CJ *^ (J
«j U >
"3 z =^
^« t. ^_- O
z - i -.
•H g S | -3
>> § 2 I «
c >» J u < ^
< " -^ ^< « 2
- ^ «" g c o
- >> o c 2 =
z = ^ ^ o ?T
< ^ « z -
Z — —;
-------
oo
oo
o
^
J
(Q
73
G
o
1
^
u
o
to
%
Jg
u
u
1
fS
1
r**>
U
J3
0}
f-
u
V
^
E
3
Z
*
'3
e>
U.
00
r~-
o
w-i
^
rn
r\i
'o*
t
£
_e
««
2
"S
V
(J
e
o
r \
W
tg
1
o
U
"m
4rt
A
H
v
3
««
^«
i/i
f
f
r«
CO
m
O
q
v"
o
v
o
v"
o
7
2
c^
o
v
¥
o '
oo
VO
c>
v
o
u§
^^?
9'
v
2
M
1
Q.
">»
s
r-
vri
vO
(N
O
q
v"
o
rr
(N
o
r^
oo
r~
1
rr
m
fN
6
. v"
9'
v
o
oo
vo
Q
\7
v
o
S
^CJ
9
V
u
u
!r
c.:
*
^
3
O
V
q
V*
o
fS
3
r-
ri
i
(N
O
v"
9
v
o
00
^o
?'
o
vS
^C7
9
V
S
2
X
^D.
^
CJ
i
f*\
r{
r
m
vO
Od
"
m
3
v
9
v
o
oo
VO
?'
o
s?
^•3
9
V
S
u
tz
<
2
<
2
<
2
<
2
<
2
o
ON
9
V
o
oo
vO
o
^7
v
O
«n
fl
O
w
J2
CO
I
1
£
^Z,
<
^«
<
2
<
2
1
<
2
•<
2
00
en
O'
(/i
2
•§
I '-5
s
_o
CJ
fl
X
>,
CJ
C
U
<
,2
«^
b
2
Q«
1
C
U
5
2
>
e
U
CO
b
^ ^
i
s
'i
.0
3
CO
U
"S
!
I
CO
E
2
vt
i
1
a.
T3
U
C
*
**^
L
^c;
c
o
v
M
^
t
l_
£
t
c
ec
b
c!
+•
1.
t
c
s*
Ol
.
c
e
_o
«^
CJ
T3
U
f^
OC,
>
a
c
W
0.
4J
*
3 -a
' U
2 S
3 S
i J3
> 3
i
N
!J ^'
^ il
CJ
§
a
S o
u
2 « 3 •=
I '3 ° 5
«i 03 r- '2
J2~ J2~ ^2* S
2 2! 2 ~
o ^o 13
"5 -5 "5 Q.
o o o —
c c c H
X >> X C
u.
at al on O
>/^ \o r~- od
O
1
^_
^^
"«3
5
o
Z
2
>»
u
U
>>
'G
CO
U.
>
2
S
S
ca"
o
CJ
<
—
•t
j*
o
' k«
U
-o
u
Uu
o"
JJ
S3
•^
max/Eas
3
<
r^i
2
*
u
Z
*
£
3
C
'i
3
<
co
T3
c
r3
L«
O
Z
r«i
5"
u
'>
so
o
1
e/i"
2
u
2
V)
2
o
^
•j
*s
T
-------
Table 3-3 Characterization of Wastewater Form of K088
Constituent
Acenaphthene
Anthracene
Benz(a)anthracene
Benzo(a)pyrene
Benzo(b)fluoranthene and
Benzo(k)fluoranthene
Benzo(g,h,i)perylenc
Bis (2-ethylhexyl) phthalate
Chrysene
Dibenz(a,h) anthracene
Di-n-butyl phthalate
Di-n-octyl phthalate
Fluoranthene
Indeno( 1 ,2,3-cd)pyrene
Phenanthrene
Pyrene
Cyanide (total)
Fluoride
Antimony •
Arsenic
Barium
Beryllium
Cadmium
Chromium (total)
Process Data (mg/L)
Ormet Process
<0.01
<0.01
<0.01
<0.01
<0.02
<0.01
<0.01
<0.0l
<0.0l
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01-0.018 '
4.9-150
<0.02
<0.004-0.014
<0.01-0.08
<0.01
<0.01
<0.02-0.55
EPA Test Bum
<0.02
<0.02
<0.02
<0.02
<0.04
<0.02
<0.02
<0.02
0:02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.005
0.3
<0.05
<0.03
17.8
<0.03
-------
Table 3-3 Characterization of Wastewater Form of K088
Constituent
Lead
Mercury
Nickel
Selenium
Silver
Thallium
Vanadium
Zinc
Process Data (mg/L)
Ormet Process
<0.005-0.007
<0.0002
<0.04-0.1
<0.004
<0.01
<0.005
<0.01-0.04
<0.01-0.55
EPA Test Burn
-
<0.002
<0.2
^0.5
<0.05
<0.5
<0.2
5.9
Ref.: 8) Ormet Corporation, Petition for Exclusion for Vitrified Product from Spent Potliner.
Submitted to U.S. Environmental Protection Agency. Ormet Corporation, Hannibal, OH,
April 1994
11) U.S. Environmental Protection Agency, Office of Research and Development Pilot-scale
Incineration Tests of Spent Potliners from the Primary Reduction of Aluminum (TCOSS) U.S.
Environmental Protection Agency, Cincinnati, OH., 1991.
3-28
I
I
I
-------
Table 3-4 Analytical Methods for Semivolatile Organics
^BBBGBBEEaKKSE^S-^^SE^B^SS^SS^E
Contaminant
Anthracene
Benz(a)anthracene
Benzo(a)pyrene
Benzo(b)fluoramhene
Benzo(k)fluoranthene
Benzo(g,h,i)perylene
Chrysene
Dibenz(a,h)anthracene
Fluoranthene
Indeno(l,2,3-cd)pyrene
Phenanthrene
Pyrene
SW-846 Method Number
Extraction Method
Water/Wastewater
3510A
3520A
Solids
3 540 A
3550
3580A1
Clean-up Method
3610A
3630A
3640
Analysis Method
.
8250
8270A
8100
8310
1 - Method only applicable to wastes that are soluble in the extraction solvent.
Ref.: 12) U.S. Environmental Protection Agency, Office of Solid Waste, SW^846 Test
Methods for Evaluating Solid Waste Physical/Chemical 'Methods. Third Edition.
Washington, D.C. November, 1986.
3-29
-------
Table 3-5 Analytical Methods for Metals
Contaminant
Antimony
Arsenic
Barium
Beryllium
Cadmium
Chromium
Lead
Mercury
Nickel
Selenium
SW-846 Method Number
Digestion Method
WATER1 SOLID2
3005A
3005A
3005A
3005A
7060
7061 A
3005A
3005A
3005A *
3005A
3005A
3005A *
3005A
3005A
3005A *
3005A
3005A
3005A*
3005A
3005A
3005A'
•7470
3005A
3005A
3005A
7740
7741
3050A *
3050A *
3050A •
3050A
3050A
7061 A
3050A
3050A
3050A *
3050A
3050A
3050A
3050A
3050A
3050A
3050A
3050A
3050A
3050A
3050A
3050A
7471
3050A
3050A
3050A
3050A
7741
Analysis Method
6010 (ICP)
7040 (AA Direct Aspiration)
7041 (AA Furnace)
6010 (ICP)
7060 (AA Furnace)
7061 A (AA Gaseous Hydride)
6010 (ICP)
7080 (AA Direct Aspiration)
7081 (AA Furnace)
6010 (ICP) I
7090 (AA Direct Aspiration)
7091 (AA Furnace)
6010 (ICP) I
7130 (AA Direct Aspiration)
7131 (AA Furnace)
6010 (ICP)
7190 (AA Direct Aspiration)
7191 (AA Furnace)
6010 (ICP') I
7420 (AA Direct Aspiration;
7421 (AA Furnace)
7470 (CVAA) water !
7471 (CVAA) solid
6010 (TCP)
7520 (AA Direct Aspiration)
6010 (ICP)
7740 (AA Furnace)
7741 (AA Gaseous Hydride)
3-30
-------
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
Table 3-5 Analytical Methods for MetaJs
Contaminant
Silver
SW-846 Method Number
Digestion Method
WATER1 SOLID'
3005A
7760A
7761
3050A
3050A *
3050A *
Analysis Method
6010 (ICP)
7760A (AA Direct Aspiration)
7761 (AA Furnace)
* - Digestion is typically performed by the method cited in the table, although the
metal is not specifically referred to in the SW-846 Method provided.
1 - As an alternative, all metals listed in this table may be digested in aqueous
samples by Method 3015, SW-846 Proposed Update No. 2.
2 - As an alternative, all metals listed in this table may be digested in solid matrix
samples by Method 3051, SW-846 Proposed Update No. 2.
Ref: 12) U.S. Environmental Protection Agency, Office of Solid Waste, SW-846 Test
Methods for Evaluating Solid Waste Physical/Chemical Methods. Third Edition.
Washington, D.C. November, 1986.
Table 3-6 Analytical Methods for Non-metallic Inorganics
Contaminant
Cyanide
Fluoride
Analysis Method
SW-846 • 9010A or 9012
ASTM - D 1179-80, Method B
Ref.: 12) U.S. Environmental Protection Agency, Office of Solid Waste, SW*846 Test
Methods f?r Evaluating Solid Waste Physical/Chemical Methods. Third Edition.
Washington, D.C. November, 1986.
13) American Society of Testing and Materials (ASTM) D 1179-80, Method B
3-31
-------
Table "3-7 Analytical Methods • Preparative (Digestion, Extraction, Cleanup)
Method Number
3510A
3520A
3540A
3550
3580A
3610
3630
3640
3005A
3050A
3015 *
3051 *
Method Title
Separatory Funnel Liquid-Liquid Extraction
Continuous Liquid-Liquid Extraction
Soxhlet Extraction
Ultrasonic (Sonication) Extraction
Waste Dilution
Alumina Column Cleanup
Silica Gel Cleanup
Gel-Permeation Cleanup
Acid Digestion of Waters for Total Recoverable or Dissolved
Metals for Analysis by FLAA (Direct Aspiration) or ICP
Spectroscopy
Acid Digestion of Sediments, Sludges, and soils by ICP, FLAA
(Direct Aspiration), or GFAA (Graphite Furnace) Spectroscopy
Microwave Assisted Acid Digestion of Aqueous Samples and
Extracts
Microwave Assisted Acid Digestion of Sediments, Sludges, Soils
&0ils
* • Proposed method included in SW-846, Third Edition, Update No. 2; promulgation
pending.
Ref.: 12) U.S. Environmental Protection Agency, Office of Solid Waste. SW-846 Test
Methods for Evaluating Solid Waste Physical/Chemical Methods. Third Edition.
Washington, D.C. November, 1986.
3-32
-------
Table 3-8 Analytical Methods Instrumentation
MethodrNumfaer-
6010
7040
7041
7060
7061
7080
7081
7090
7091
7130
7131
7190
7191
7420
7421
7470
7471
7520
7740
7741
7760A
7761
8100
8250 •
MWf fin • ...«*—
Inductively Coupled Plasma-Atomic Emission Spectroscopy
Antimony by Atomic Absorption, Direct Aspiration
Antimony by Atomic Absorption, Furnace Technique
Arsenic by Atomic Absorption, Furnace Technique
Arsenic by Atomic Absorption, Gaseous Hydride
Barium by Atomic Absorption, Direct Aspiration
Barium by Atomic Absorption, Furnace Technique
Beryllium by Atomic Absorption, Direct Aspiration
Beryllium by Atomic Absorption, Furnace Technique
Cadmium by Atomic Absorption, Direct Aspiration
Cadmium by Atomic Absorption, Furnace Technique
Chromium by Atomic Absorption, Direct Aspiration
Chromium by Atomic Absorption, Furnace Technique
Lead by Atomic Absorption, Direct Aspiration
Lead by Atomic Absorption, Furnace Technique
Mercury in Liquid Waste, Manual Cold- Vapor Technique
Mercury in Solid or Semisolid Waste, Manual Cold-Vapor
Technique
Nickel by Atomic Absorption, Direct Aspiration
Selenium by Atomic Absorption, Furnace Technique
Selenium by Atomic Absorption, Gaseous Hydride
Silver by Atomic Absorption, Direct Aspiration
Silver by Atomic Absorption, Furnace Technique
Polynuclear Aromatic Hydrocarbons by GC/FID
•Gas Chromatography/Mass Spectrometry (packed GC column;
ll
3-33
-------
Table 3-8 Analytical Methods Instrumentation
Method Number
8270A
8310
901 OA
• 9012
Method Instrumentation
Gas Chromatography/Mass Spectrometry (capillary GC column)
Polynuclear Aromatic Hydrocarbons by HPLC
Total and Amenable Cyanide (Colorimetric, Manual Technique)
Total and Amenable Cyanide (Colorimetric, Automated UV
Technique)
Ref.: 12) U.S. Environmental Protection Agency, Office of Solid Waste, SW-846 Test
Methods for Evaluating Solid Waste Physical/Chemical Methods. Third Edition.
Washington, D.C. November, 1986.
3-34
-------
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
4.0 BOAT TREATMENT STANDARDS FOR SPENT POTLINERS
4.1 Selection of Constituent*; for Regulation
This section presents the methodology and rationale for selecting constituents for
regulation in nonwastewater and wastewater forms of K088.
4.1.1 BOAT List Constituents Present in K088
The constituents identified in K088 wastes are presented in the waste characterization
data tables presented at the end of Section 3. Samples of K088 were analyzed for numerous
organic constituents including those amenable to analysis by SW-846 methods 8240 and 8270.
In addition, samples were analyzed for cyanide, fluoride and BOAT List metals. One sample
of K088 collected by the Agency was analyzed for all the BOAT List constituents with the
exception of the following constituents which were not amenable to chemical analysis.
benzal chloride 1,2-diphenylhydrazine
benzenethiol 2-cthoxyethanol
p-benzoquinone ethylene oxide
3-chloropropionitrile hexachlorophene
dibenzo(a,e)pyrene methanol
dibenzo(a,i)pyrene phthalic anhydride
3,3'-dimethoxybenzidene resorcinol
diphenylaminc tris(2,3-dibromopropyl)phosphate
Waste characterized as K088 contained the BDAT List constituents presented below.
4-1
-------
Organics
Inorganics
antimony
arsenic
barium
beryllium
cadmium
chromium
copper
lead
mercury
nickel
selenium
silver
thallium
vanadium
zinc
acetone
acetonitrile
acrolein
anthracene
benz(a)anthracene
benzene
benzo(a)pyTene
benzo(b)fluoranthene
benzo(g,h,i)p«ylene
benzo(k)fluoranthene
bis(2-ethylhexyl)phthalate
butyl benzyl phthalate
carbon disulfide
chloroform
chrysene
di-n-octyl phthalate
dibenz(a,h)anthracene
dichlorodifluoromethane
ethyl cyanide
fluoranthene
hexachlorodibenzofurans
indeno(l ,2,3-cd)pyrene
methyl ethyl ketone
methylene chloride
phenanthrene
pyrene
pyridine
toluene
trichloromonofluoromethane
cyanide
sulflde
phosphorous
4-2
I
I
I
-------
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
4.1.2 Other Constituents Present in K088
Wastes characterized as K088 were also analyzed for fluoride, a non-BDAT List
constituent. Fluoride was found in significant quantities in all samples. This is expected
because a fluoride-based electrolyte (cryolite (sodium aluminum fluoride)) is used in the
aluminum reduction cell. In addition, several other constituents including aluminum, calcium,
cobalt, iron, iithium, magnesium, manganese, molybdenum, potassium, sodium, strontium, tin,
and phosphorous were also found in K088 waste samples.
4.1.3 Constituents Selected for Regulation in K.088
i
Lists of the constituents selected for regulation in K.088 are presented in Tables 4-1
through 4-3. These constituents were selected based on waste composition data collected by
the Agency. All of the constituents selected for regulation were found in concentrations of
regulatory concern (i.e., under plausible improper management scenarios, the constituent
concentrations likely to be present in groundwaters are expected to be significantly higher
than their health-based levels of concern). Several of the constituents including cyanide and
fluoride were contaminants found in groundwater and soils of five aluminum reduction sites
listed on EPA's National Priorities List (NPL). These sites all had groundwater or soil
contamination resulting from the disposal of spent potliners and other aluminum reduction
wastes. (50, 51, 52, 53, 54)
Fluoride is present in K088 wastes at high concentrations, often at concentrations of
greater than 10 percent Untreated concentrations of this magnitude can cause significant
adverse effects to human health or the environment Fluoride, although not an Appendix VIII
constituent or BOAT List constituent, is regulated in the F039 (multi-source leachate) and
included in the K.088 delisting decision granted to Reynolds Metals. Damage to the
environment from fluoride from spent potliners has also been documented. The Agency has
identified one Record of Decisions (RODs) (50) from primary aluminum smelters and :v.o
4-3
-------
additional NPL sites (51, 52) that show groundwater contaminated with fluoride. In addition,
the Agency has set a maximum contaminant limit (MCL) of 4.0 mg/1 for fluoride developed
under the Safe Drinking Water Act to protect drinking water from excessive levels of
fluoride. The Agency believes that threat to human health or the environment from potential
migration of fluoride from spent potliner would not be minimized without including it in the
treatment standards for K088.
It should also be noted that including fluoride is consistent with other actions. For
example, under the Clean Water Act, the discharge of pollutants into surface waters and
Publicly-Owned Treatment Works (POTWs) from primary aluminum smelting facilities is
regulated under the Aluminum Segment of the Non-ferrous Metals Manufacturing Point
Source Category (40 CFR Part 421 Subpart B). This subpart includes effluent limitations and
standards for, among other things, fluoride for wastewaters discharged from certain aluminum
reduction processes. This includes wastewaters from the anode and cathode paste plants,
anode bake plant, cathode reprocessing, potline and potroom air pollution control, aluminum
degassing, pot repair and soaking, and aluminum casting. In addition, under the Clean Air
Act, the Agency has set Standards of Performance for New Stationary Sources that requires
any facility that commences construction or modification after October 23, 1974, to meet
emissions standards for fluoride and opacity. Thus, the risk from the uncontrolled emission of
fluoride to the environment has been recognized under other programs.
Studies indicate that human health effects associated with acutely toxic doses of
fluoride can include gastroenteritis, muscular weakness, chronic convulsions followed by
depression, pulmonary congestion, and respiratory and cardiac failure. Prolonged ingesticn of
fluorides can result in the development of exostatic lesions and, in some cases, a general
thickening of some bones, stiffness, lameness, appetite impairment, and poor reproduction.
(Fluorides as Environmental Contaminants J.R. Bodnar, Institute of Environmental Sciences
and Engineering, University of Toronto, November 1972) (37).
4-4
-------
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
The animals most effected by fluoride appear to be cattle, followed by sheep, swine,
horses, turkeys, and chickens. The main problems associated with chronic fluoride intake are
striking dental changes, including brown discoloration, pitting, and fast wearing away of the
teeth. Intermittent stiffness and lameness appear as the bone changes become more marked.
Anorexia, inanition, and emaciation may occur. Decreased milk production and rough hair
coats may then be expected. (Fluorine from Veterinary Toxicology by R. D. Radeleff, Lea
and Febiger, Phila. 1964 pp 145-149.) (38) Additional human health and environmental
effects of fluoride exposure are presented in the Summary of Articles on the Health and
Environmental Effects of Fluoride (39) in the docket for the final rule. As a result, fluoride is
regulated in both wastewater and nonwastewater forms of K088.
Phthalates were not included in the list of regulated constituents because the Agency
believes that phthalates are not present in the spent potlmer at significant concentration. The
Agency believes that phthalates are not present in the raw materials used in the manufacture
of potliners (e.g., coke, pitch, etc.) and that the aluminum reduction process would not
generate phthalates since the starting materials are not present (i.e., phthalic acid and an
alcohol source). In addition, no ester bond (such as in a phthalate ester) would survive the
temperatures employed in either coke production or aluminum reduction.- The Agency
believes that the analytical data showing the presence of phthalates is likely to have been
caused by sampling contamination or contamination in the laboratory. Phthalates are a
common laboratory contaminant that can be introduced during laboratory extraction and
analysis procedures from such items as PVC gloves, PVC piping and tubing (34). Thus, the
Agency is not regulating phthalates in the Land Disposal Restrictions for spent potliners.
Another document prepared in this rulemaking concerning constituents in K088 wastes
contained in the docket is "Analysis of Proposed Regulated Constituents for K088" (34).
With regard to other organic constituents, the Agency believes that the ^olynuclear
aromatic hydrocarbon (PAH) constituents should be included in the treatment standards for
K.088 wastes. "The Agency has data that shows that each of the organic constituents selected
4-5
-------
for regulation were found in spent potliner above the UTS concentrations. The PAHs are
very toxic and present at concentrations greater than levels at which toxic effects can occur.
One NPL site The Agency considered monitoring PAHs with benzo(a)pyrene (BAP) as an
indicator parameter for other PAHs. The Agency, however believes that the concentration of
one constituent does not always reflect the concentration of similar constituents in a waste.
Surrogate analyses would assume that ail PAHs are present at similar concentrations. Because
of the variability of concentrations found in K088 wastes, BAP may not be present while
other PAHs may be present. The Agency has data characterizing spent potliner that does not
contain detected concentrations of BAP but has other PAHs at concentrations exceeding the
treatment standards. Analyzing for BAP alone does not ensure that other constituents are not
present above the treatment standards.
The Agency is regulating benzo(b)fluoranthene and benzo(k)fluoranthene as a sum in
K088 wastes. The Agency recommends the use of SW-846 Method 8270, which requires the
use of a GC/MS for measurement of the concentration of these compounds. When analyzing
for these compounds using this method, these two stereoisomers co-elute. Since the two
constituents may not be accurately quantified separately, the Agency is regulating these
constituents as a sum in both wastewater and nonwastewater forms of waste.
4.2 Identification of Technologies for the Treatment of Spent Potliners
Presented in the sections below is the Agency's discussion of technologies for
treatment of nonwastewaterand wastewater forms of spent potliner. However any treatment
technology which reduces the concentration of regulated constituents to the level of the
treatment standards, and is not considered impermissible dilution, is also acceptable.
Typically, in order to establish BOAT, the Agency first identifies which technologies
are "applicable" for treatment of the constituents of interest An applicable technology is one
which, in theory, can treat the waste in question or a waste similar to the waste in question in
terms of parameters that affect treatment selection. Detailed descriptions of the technologies
4-6
•KK
-------
identified as applicable for the treatment of listed hazardous wastes are provided in EPA's
Final Treatment Technology Background Document (14). The identification of treatment
technologies as applicable for treating BDAT List constituents is based on evaluation of
current waste management practices, current literature sources, field testing, data submitted by
equipment manufacturers and industrial concerns, plus engineering judgment-of EPA technical
staff personnel.
The Agency next determines which of the applicable technologies are "demonstrated"
for treatmc'- of the wastes. To be designated as demonstrated, a technology must be used in
a full-scale operation for treatment of the waste of interest or a similar waste. Technologies
that are available only at pilot- or bench-scale operations are not considered demonstrated
technologies. The Agency determines which of the demonstrated technologies is best, based
on a thorough review of all performance data available on treatment of the waste of concern
or wastes judged similar, and determines whether this 'best" demonstrated technology is also
commercially "available." If the "best" demonstrated technology is "available," then the
technology is determined to represent BDAT.
For K088, however, EPA has identified a broad range of treatment, recycling,
reclamation, and reuse practices as applicable technologies to K088. Any technology that
actually recovers some of the value in K088 is the Agency's preference for treatment. For
spent potliners, the Agency does not have adequate data on the products or residues created
by these recycling technologies. Therefore, numerical treatment standards are being set. In
examining treatment and recycling technologies for K088, the Agency discovered a number of
technologies that appear very promising for actually recovering some of the value in K088
while still destroying the hazardous constituents. Some of these technologies also claim to
process K088 into marketable products. These technologies vary from being pilot-scale
projects to full-size plants that are ready to treat K088 wastes. For example, some industrial
furnaces and calciners enable the use of fuel values and the reuse of valuable materials such
as aluminum sodium fluoride salts and un-bumed carbon. It is believed that while the carbon
in K088 can be used as a reducing agent for metals processed in iron and steel furnaces, the
4-7
-------
sodium aluminum fluoride salts can also serve as a substitute for fluor-spar (calcium fluoride)
when K088 is pretreated with limestone. The fluor-spar serves as a furnace fluxing agent for
industrial furnaces that make iron and steel. Thermal technologies such as incineration
(regulated under §264), industrial furnaces (exempted from or regulated under §266), or
calcines (usually exempted from §266) ensure the destruction of cyanides and other organic
constituents of concern in K088 (if properly designed and operated).
Available data show that when these thermal technologies are used, the total K088
mass content of the materials fed to these thermal furnaces ranges from 5% to 20% total feed.
In addition, available information appears to indicate that K088 has little or no impact on the
quality of the products manufactured by these thermal processes. K088 wastes can have value
not only for their carbon value used in place of coke, but also for the reducing value of the
carbon. In addition, the fluoride can be used as a fluxing agent A more detailed discussion
of these technologies is presented in a later section of the report
The EPA is developing a concept for K088 (and other wastes) that uses a hierarchy of
options for evaluating treatment or recycling technologies. For example, the Agency's
ultimate goal for hazardous waste is source reduction, that is, less or no production of
hazardous waste. The next preferred option is total recycle or reuse. That would require that
all the waste generated be used as feedstock in the same process or another process. Next
lower in the hierarchy would be treatment technologies that can recover materials from the
waste for reuse. This option would most likely result in some residuals still needing to be
land disposed but preferred techniques would also significantly reduce the quantity and
toxicity of.any waste destined for land disposal. Further down in the hierarchy would be
technologies that reduce quantity and toxicity without the recovery of materials for reuse.
Last would be technologies that lower toxicity alone and may increase volume of materials for
land disposal. For example, while stabilization can sometimes render a metal-bearing waste
less hazardous, it often results in a great increase in volume which then has to be landfilled.
At the base of this hierarchy are numerical treatment standards. Numerical treatment
standards are required for treatment or recovery technologies that result in residuals rc
4-8
-------
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
land disposal. Treatment residuals must comply with applicable treatment standards prior to
land disposal.
An analysis of residues from K088 treatment technologies leads EPA to believe that
the concentration-based universal treatment standards can be routinely achieved. As discussed
in the Advance Notice of Proposed Rulemaking (ANPRM) published in October 24, 1991,
EPA was seeking development of concentration-based standards for K088 so as to allow the
use of any technology that can achieve the numerical values. The Agency notes, however,
that when it establishes concentration-based treatment standards, the regulated community may
use any non-prohibited technology to treat the waste to meet the treatment standards.
Compliance with concentration-based treatment standards requires only that the effluent
concentration be achieved; once achieved, the waste may be land disposed. The waste need
not be treated by the technology identified as BDAT; in fact, concentration-based treatment
standards provide flexibility in the choice of treatment technology. Any treatment, including
recycling or any combination of treatment technologies, unless prohibited (e.g., impermissible
dilution) or unless defined as land disposal (e.g., land treatment), can be used to achieve these
standards.
4.2.1 No n waste waters
This section presents the Agency's discussion of management methods and treatment
technologies for nonwastewater forma of spent potliner.
4.2.1.1 Management Methods and Treatment Technologies
EPA's progress in improving environmental quality through its media-specific
pollution programs has been substantial. Over the past two decades standard industrial
practice for pollution control concentrated to a large extent on "end of pipe" treatment and/or
disposal of hazardous and nonhazardous wastes. However, EPA realizes that there are limits
to the degree of environmental improvement that can be achieved under these programs by
4-9
-------
emphasizing management after pollutants have been generated. EPA believes that eliminating
or reducing discharges and/or emissions to the environment through implementation of cost
effective source reduction and environmentally sound recycling practices can provide
additional environmental improvements.
Because nonwastewater forms of the K088 wastes contain organic constituents as well
as cyanide, fluoride and metals, applicable technologies include those tt -n destroy or reduce
the total amount of various organic compounds as well as recover or fixate the metals and
fluorides present. There are numerous technologies either available or being developed that
recycle or recover the value (carbon, fluoride, etc.) in K088. The information presented in
the following sections is not intended to be a comprehensive discussion, but rather is intended
to give the public some idea of the available options that are known to the Agency. The
Agency would like to discuss first those methods that are the top of the envisioned hierarchy
including source reduction, recycling and reuse of spent potliner. The following sections
present information provided to the Agency for technologies that can be used for the
treatment of spent potliners and recovery of materials for reuse. A comparison of these
processes is presented in Tables 4-4, 4-5, and 4-6.
4.2.1.1.1 Source Reduction
Source reduction opportunities for aluminum reduction facilities are somewhat limited.
The industry has not found a commercially feasible alternative to the Hall-Heroult aluminum
reduction process (see Section 3.1 for a description of the aluminum reduction process),
therefore this process is likely to be used for the foreseeable future. Primary aluminum
reducers are actively working to increase the life of pots. By increasing pot life, generation
of spent potliner is decreased as pots are changed out less frequently. Several factors
impacting the impact life of a cell have been studied Research has concluded that the factors
affecting pot life include:
Use of stronger steel shells that limit the deformation of the cathode
Cell preheat procedures,
4-10
-------
I
I
I
I
I
• Use of high quality cathode blocks
• Use of sidewall blocks with higher thermal conductivity
• Use of ramming pastes with better physical and mechanical characteristics
Balancing of the magnetic field in the cell.(36)
One potential method to reduce the generation of spent potliner is to increase the
recovery and reuse of post-industrial and post-consumer aluminum. Increased use of
secondary aluminum could replace some demand for virgin aluminum. If the increased
demand for aluminum is in part met by secondary aluminum sources, additional primary
aluminum reduction capacity may not be required thus potentially reducing the quantity of
spent potliners generated.
4.2.1.1.2 Total Recycle
Total recycle or reuse of a waste material within the same process or an external
process eliminates the generation of a waste for treatment and disposal and subsequently
generates no treatment residuals. The Agency has information that indicates that 5 to 10
percent of the spent potliner can be reused in the manufacture of new potliners or cell anodes.
Thus, while total reuse or recycle within the same process may not be feasible, recycling up
to 10 percent of spent potliners would reduce treatment and disposal requirements for some of
the material.
Recycling to Carbon Liners
The carbon portion of the spent potliner can be used in the manufacture of new
potliners. The high density blocks can be made of anthracite coal and tar pitch with some
spent potliner mixed in. Manufacturers have experimented with the use of 10 to 50 percent of
the blocks being made from spent potliner. Successful use of spent potliners in this manner is
dependent on the affect on the failure rate of the pot. Data have indicated that 10 percent
4-11
-------
substitution causes no statistical variation in the failure rate of a pot but a 50 percent content
increases the failure rate of pots. (15) Alcoa researchers have reported that limited plant trials
confirmed that untreated spent potliner could be used in cathode blocks and seams without
adversely affecting cell operating parameters. They also stated that a number of issues
including industrial hygiene, baking furnace refractory deterioration, and the ability to produce
consistent blocks from a variable material would need to be addressed before
commercialization. (35)
Recycling to Anodes
This technology involves the use of the carbon portion of the spent potliner to replace
portions of coke aggregate used in the production of Soderberg anodes used in the reduction
cell. Using the spent potliner reduces raw material costs, and very little preparation is needed.
Impurities are destroyed in the reduction cell or removed in emission control systems.
However, carbon consumption in the anodes is increased and metal quality may be degraded
when spent potliner is used in the anodes. (15) One facility has tried the technology and is
discontinuing it due to significant contamination of product. One other facility is
experimenting with the concept (16)
4.2.1.1.3 Treatment Processes
The following descriptions present several technologies in use or in development to
treat spent potliners and recover and reuse the chemical value in spent potliner. The process
descriptions are presented- in alphabetic order by firm or technology with no implied
recommendation or preference.
Alcan International Limited - Low Caustic Leaching and Liming
Alcan has developed what it believes to be a cost-effective hydrometallurgical route to
destroy cyanide in spent potliner (SPL) and recover chemical value of fluoride, sodium,
4-12
-------
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
carbon and aluminum, and provide a zero-discharge process (17). Alcan has piloted a Low
Caustic Leaching and Liming (LCL&L) process to destroy cyanide and convert fluoride to
acid grade fluorspar. Sodium and aluminum are recovered as a sodium aluminate caustic feed
to the Bayer plant operations. Alcan believes that the remaining brick and carbon are
valuable as a chemical reducing agent Alcan has tested this process in a one metric ton/day
unit and has plans to scale up to a 60,000 metric tons/year plant. Figure 4-1 presents a
general schematic of the process flow for the Alcan process. Alcan's process involves five
major process steps:
Spent potliner dismantling and crushing
Grinding and classification
Extraction and leaching of spent potliner
Cyanide destruction and crystallization of sodium fluoride from leachate
Causticization of the sodium fluoride liquor
1. Spent potliner dismantling and crushing
Alcan dismantles its pots removing excess alumina bath and aluminum metal. The
spent potliner is then segregated into first cut and second cut fractions. These fractions are
crushed and homogenized separately so that the process can be examined using varying spent
potliner combinations.
2. Grinding and Classification
The spent potliner is crushed using an impact crusher mill to an average size of 20
mm and separately ground using a Willow hammer mill reducing the aggregate size to an
average of 0.5 mm. A screening operation produced particle size fractions ranging from
minus 84 mesh to minus 28 mesh.
3. Extraction and Leaching
The finely ground spent potliner is digested in hot, dilute caustic solutions. Multiple
agitated cascade reactors digest the slurry, extracting fluorides, sodium, alumina, silica, and
free and complexed cyanides into the leach liquor. The cryolite present reacts with caustic
and decomposes to sodium fluoride and sodium aluminate:
4-13
-------
Na3AlF6 + 4NaOH -> 6NaF + NaA102 + 2H20
Any intercalated sodium and any remaining aluminum metal as well as other metals
will dissolve with the evolution of hydrogen:
2Na + 2H20 -» 2NaOH
and
2A1 + 2NaOH + 2H20 -+ 2NaA102 + 3H2
Aluminum nitrides and carbides will generate ammonia and methane gas respectively:
A1N + NaOH + 2H20 -> NaA102 + NH4OH
Al4Cj + 4NaOH + 4H2O -> 4NaAlO2 +3CH4
The solubility of sodium fluoride and sodium aluminate depends on the caustic
concentration, as well as on temperature and silica content. Therefore the caustic
concentration of the leachate will dictate the SPL/leachate ratio. A lower caustic
concentration will limit the driving force to decompose cryolite while higher concentration
will limit fluoride solubility and require a more dilute slurry.
The solid residues remaining are separated by filtration and washed with .water. Alcan
considers these residues a high ash industrial fuel. Organic constituents are likely to be
present in this carbon material.
4. Cyanide Destruction and Sodium Crystallization from Leachate
The destruction of cyanide in the spent potliner leachate is performed using an alkaline
hydrolysis. The caustic leachate from the digestion unit is enriched with caustic to reach a 60
g/L NaOH concentration. The liquor is then reacted in a stainless steel plug flow reactor.
The initial decomplexing of ferrocyanides occurs in the absence of oxygen:
4-15
-------
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
2[Fe(CN)J + 40H -» 2FeO + 12 CN + 2H20
This is followed by the hydrolysis of the cyanide ion:
CN -i- 3H20 -* NH4OH +• HCOO
The liquor is flashed and any iron complexes formed are filtered out The caustic
leachate is then fed to an evaporator/crystallizer which raises the caustic concentration to
approximately 225 g/L NaOH causing the NaF to precipitate out of solution.
5. Causticization of Fluoride Liquor
This operation transforms the highly soluble sodium fluoride into insoluble calcium
fluoride. The sodium fluoride precipitated in step 4 is redissolved and the coprecipitated
insoluble impurities are filtered out The sodium fluoride is then neutralized with additions of
milk of lime in two cascading precipitators. The calcium fluoride generated may be converted
for use as a feedstock to produce aluminum fluoride and the caustic liquor produced is
returned to the extraction step of the process.
Ausmelt Technology Corporation
Ausmelt Technology Corporation, a U.S, subsidiary of Ausmelt Ltd, Melbourne,
Australia (Ausmelt) develops technologies for the recovery and/or treatment' of metal-bearing
wastes. Ausmelt has developed a submerged lance technology that can be used for the
reclamation and treatment of spent aluminum potliners. The technology is a pyrometallurgical
system originally developed for processing metallurgical ores and concentrates. The process
utilizes a bath smelting technology with applications in many waste treatment and recovery
fields at low capital and operating costs. Heavy metals recovered from wastes by Ausmelt's
technology may be recycled to mainstream uses. The slags and other residues generated by
the process are stable and suitable for use as aggregate, shotblasting grit, or other apph.ations.
4-16
-------
Ausmelt's technology centers on a simple lance through which process air and fuel are
delivered beneath the surface of a liquid slag bath. The steel lance is protected from the
furnace contents by a coating of frozen slag which is maintained by the cooling effect of the
combustion gases passing down the lance. The tip of the lance is normally well below the
static slag level of the furnace contents, which results in the process gases being injected
deeply into the slag, creating very rurbuJent conditions in the bath. These conditions promote
high mass and energy transfer rates and ensure the smelting capacity is high per unit volume
of furnace capacity. The Ausmelt furnace is characterized by flexibility, ease of operation,
low capital and operating costs, and significant environmental advantages. Fugitive off-gases
and dust carry-over are minimal and furnace conditions can be controlled to generate slag
products that pass the Toxicity Characteristic Leaching Procedure (TCLP).
Ausmelt's technology has been applied successfully to many processes, including zinc
fuming from slag, recovery of cobalt and platinum group metals fiorn slags, recovery of zinc
from residues, slags and dusts, smelting and processing complex ores and concentrates, sieel
plant dust processing, zinc (each residue smelting, iron making, and nickel laterite smelting.
The process also has been applied for the recovery of copper, nickel, lead, zinc, tin, and
precious metals. It also can recover metallurgical wastes such as leach residues, blast furnace
slags, emission control dusts, and, of particular relevance here, spent aluminum potliners.
Ausmelt believes that the process is not solely a thermal destruction/treatment process;
it can recover materials from spent potliners for further use. Ausmelt has been working with
Alcoa of Australia on joint trials utilizing Ausmelt's top submerged lance technology to
process spent potliners. In addition to destroying the toxic organic constituents in the waste,
the process achieves two important recovery goals. First, the process recovers fluorides,
which may be reused in the aluminum smelting process. Second, Ausmelt's technology
recovers metals from the potliners that then can be incorporated into other products. These
products include metal alloys, metal oxides, and slag-based products, such as mineral wool
fiber.
4-17
-------
Barnard Environmental, Inc.
Barnard Environmental, Inc. (BEI) plans to recycle spent potliner by using it to
manufacture industrial glass and ceramic products using the patented Terra-Vit process
licensed to BEI by Battelle Memorial Laboratory. BEI is in the design stages of building a
recycling facility in Richland, Washington. The melter will process approximately 30,000
tons of spent potliner (SPL) per year with capability of processing 54,000 tons per year. It
will produce approximately 32,500 tons of glass per year. BEI also plans to build a second
plant in the Eastern U.S. These facilities will offer recycling as an alternative to treatment or
land disposal for spent potliner. The Terra-Vit process transforms, oxidizes, and vitrifies
spent potliner in a high temperature, joule-heated, refractory-lined melter into glass. The
process incorporates and immobilizes a broad range of elemental oxides into a chemically
durable glass that can be formed to useful products. Figure 4-2 presents a process flow
diagram for the Barnard process.
The facility consists of several components. Deliveries of incoming materials will
arrive to meet glass production requirements minimizing on-site storage. Raw materials are
fed using a hopper system. SPL from the hoppers is processed in a crushing circuit
Although the melter can accommodate large size feedstock, energy conservation and glass
consistency are best served by reducing the size of the SPL to minus 1/2 met As needed,
crushed SPL is fed to the melter using a ram feeder and the charge falls by gravity into the
melter tower. The large molten reservoir serves as a mixer for feed materials. With a three
to five days of residence time in the molten mass, incoming feed materials of varying
chemical characteristics are homogenized. The discharged glass is a moving average of the
material being fed. The SPL gradually settles toward the molten glass surface. Glass in the
melter is kept in a molten state by resistance hating produced by an electrical current that
passes between electrodes submerged beneath the glass surface. Glass-making materials form
a path for the electrical current which melts the material; there is no open flame. Air is
introduced just above the molten glass surface. During vitr.ration, carbon from ihc spent
potliner primarily acts as a reducing agent and volatilizes in the vitrification process This
4-18
-------
>
C
-------
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
reaction heats the air. As the oxygen depleted gas is drawn into the off-gas system, it heats
and drys the SPL charge above. In the melter, extremely high temperatures are obtained to
melt and incorporate many constituents into glass. Two products of this process are glass and
off-gases. Molten glass flows out of the melt chamber via an underflow as it is displaced by
additives to the chamber. The molten glass is then formed as desired into products. For
example, the glass can be formed into material suited for roofing material or blasting grit. It
can also be used to form glass fiber for insulation, synthetic materials, ceiling tiles, or large
castings. Glass conditioning agents can be added at this point to improve or modify glass
characteristics. The process includes an off-gas treatment which can recycle the recovered
particulates and liquids into the rnelter. Another potential wastes stream from the melter is
process residuals such as reduced metals (e.g., iron and aluminum) from the bottom of the
melter.
Comalco COMTOR Process
Comalco Aluminum Ltd., an Australian company, has developed the COMTOR
process, which is designed for total recycle (19, 20, and 21). The Comtor process is a full-
scale calcination process claimed to thermally destroy the cyanide in K088 and produce useful
products from the fluoride and carbon values in K088. The liquid product is called Bayer
Grade Caustic (BGC). BGC is a caustic soda that is widely used in various manufacturing
processes. The solid product of the third stage can be used as kiln grade spar (KGS), a
feedstock material for cement kilns. The Comalco Comtor Process is a three phase process
that is designed to reduce the size of the spent potliner particles, utilize a hot air catalyst to
destroy cyanide and produce a porous calcine, and recover available fluoride contained in the
spent pot'iner.
Comdco has completed pilot studies on all three stages of the COMTOR Process and
has been operating the first two stages in a full-scale demonstration plant for over one year at
its Boyr.i smelter facility in Australia. Construction of the third stage at Boyne has been
4-20
-------
completed and ComaJco expects to begin production of BGS and KGS for saJe in local
markets in July 1995.
Comalco has also recently licensed the operation of the first two stages of the
COMTOR Process at a smelter in New Zealand. Comalco and the New Zealand smelter are
in the process of reconfiguring the third stage to accommodate local markets for the BGC and
KGS products.
The COMTOR Process is composed of three distinct phases 1) Feed Preparation, 2)
Calcining, and 3) Fluoride recovery. Figure 4-3 shows a general process flow for the Comtor
process.
1) Feed Preparation
When spent potliner is removed from the reduction pot, the spent potliner particle size
ranges from an ultra fine dust to pieces 1-2 feet long. To treat this material effectively it is
necessary to perform size reduction to obtain a relatively uniform particle size. The
COMTOR process uses a semi-autogenous grinding mill to grind raw SPL into less-than-one
millimeter particles. It is necessary to grind the SPL this finely in order to meet the
specifications for the TORBED unit Tramp aluminum and steel are segregated from the SPL
during this stage for reuse in the smelting process. Problems in reducing the size of spent
potliner include high dusting rates, high noise rates, and wear on size reduction equipment.
Reducing the size of the spent potliner prior to processing, however, improves the rate and
quality of treatment during both the calcining stage and the fluoride recovery phase of the
Comtor process. The small particle size required by the Comtor process also allows the use
of pneumatic material transfer systems in the process.
2) Calcination
In the second stage of the COMTOR Process, the ground SPL is fed into a processing
chamber called the TORBED unit. Within the TORBED, small amounts of SPL are rapidly
spun around a circular chamber with pre-heated air in the temperature range of 600°C to
4-21
-------
I
I
I
I
I
I
I
I
I
1
I
I
I
I
I
I
I
I
I
Figure 4-3 Schematic of the COMTOR Process
FAILED REDUCTION CELL CONTAINING
RAW SPENT POTUNINO MATERIAL
J
Cathode Bars
(Recycled)
Controlled Pot
Breakout Procedure
1 »t CUT
2nd CUT
, Tramp Materiel
(Refuse)
Off Qas
(To FIKer)
Comtor Stag* On*
SPL COMMINUTION
Reduces SPL to
Usable Feed Size
Comtor Stag* Two
TORSO CALCINATION
Resource
Recovery
Ume
Water
Destroys Laachable
Cyanides and PAH's
Comtor Stage Three
SPL ASH TREATMENT
Cauttk
Leaching
Stabilize
LMcKabke Fluorides
Dry Mb
KGS
Caustic
Liquor
To Cament Kiln To Baver Process
Lime Gypsum
Inert Material
Ref.: 20)
"The Comtor Process for the Treatment of Spent Potlining" Brochure from
Comaico and AISCO, January 31, 1994
-------
750°C. The Torbed is essentially a gas/solid contacting device. In the Torbed unit, preheated
process gas is forced up through a distributor, which consists of a radial tract of horizontal
deflectors, into the process chamber. Spent potliner feed material is gravity fed from the top
of the process chamber. Impingement of the gas on the spent potliner particulates is such that
a toroidal suspension is formed in the reaction chamber. The particle bed is of fixed capacity,
so that every unit of additional feed displaces a unit of material from the bed. The dynamic
inventory in the process area is approximately 8-10 kg. The air in the TORBED is heated by
a device that is ancillary to the TORBED unit. The ground SPL never comes in contact with
the heating device or any flame, nor is it consumed as a fuel. Rather, the hot air is produced
using natural gas and acts as a catalyst only for the purpose of breaking down the cyanides.
Precise temperature control eliminates any need for direct contact between the heat source and
the SPL as well as any need for the addition of anti-agglomeration agents into the process.
Solids are transported out of the calciner using a pneumatic transfer system to a
product holding bin. Exhaust gases contain material elutriated from the calciner. The exhaust
gases are sent through a baghouse and then the particle free gas is sent through the smelter
dry scrubbing system. A general arrangement of the Torbed calciner is shown in Figure 4-4.
Through this process, the cyanides in the SPL are broken down. The break-down of the
cyanides in turn creates porosity in the SPL particles to enable effective digestion in the third
stage. The solid particulates are then separated from the air stream and conveyed to the feed
hoppers of the third stage.
The calcine material that exits the TORBED unit consists mainly of carbon and
fluoride compounds, and inert residues of the refractory component of the SPL. At this point,
the calcine may either be fed into the third stage of the COMTOR Process for recycling into
commercial products, or fixed and stabilized with a lime/gypsum fixing agent to meet the
LDR requirements for landfilling.
Spent potliner tends to have a wide variation in the calorific value and the presence of
low melting point salts can make calcination a difficult operation without material
4-23
-------
Figure 4-4 Diagram of the Torbed Calciner
STATIONARY
DEFLECTORS
PROCESS
GAS
INLET
Ref.: 19) "On-site Engineering Report • Comtor Process for Spent Potlining Treatment",
Submitted to US.Environmental Protection Agency RCRA Docket F-91-CSP-
FFFF, January 9, 1992.
-------
agglomeration. Comalco states that Torbcd calciner allows temperatures to be controlled in a
very narrow range, thus, allowing destruction of cyanide without allowing particle
agglomeration and without the use of anti-agglomeratives used in other thermal treatment
processes.
3) Fluoride Recovery
In the third stage of the COMTOR Process, the beneficiation stage, the calcine is
mixed with a lime slurry and a caustic solution and then filtered. Both a liquid product and a
solid product are produced. The liquid product is called Bayer Grade Caustic (BGC). BGC
is a caustic soda that is widely used in various manufacturing processes, including the refining
of alumina, the pulping and bleaching of paper, and for desulfurization and scrubbing in the
petroleum industry.
The solid product of the third stage can either be used as kiln grade spar ("KGS" - a
feedstock material for cement kilns) or landfilled. The potential land applications of KGS
would not be problematic as KGS is expected to meet the UTS standards.
The physical structure of the COMTOR Process is comprised of modular units. The
grinding (or comminution) unit is one modular unit, the TORBED is another modular unit and
the beneficiation equipment forms another modular unit These modular units can be installed
in various configurations depending on the needs or limitations of an aluminum smelter.
Elkem
Elkem Technology is a Norwegian company which has done bench scale testing
consisting of smelting K088 along with iron ore to produce pig iron and a slag. Elkem plans
to submit a delisting petition for the slag. The process uses the carbon in K088 to act as a
reducing agent and destroys the cyanides and other toxic organics, while rendering all other
constituents immobile in a glassified, inert slag. For each ton of K088, they produce 0.85 ton
4-25
-------
of iron. Elkem plans a demonstration plant in the U.S. next year. They also plan to pilot a
process to recover fluoride from the molten slag. )
Elkem has piloted its process at a scale of 200-300 kg/hr of spent potliner and plans to
install a 1 ton/hr demonstration plant Elkem feeds a mixture of spent potliner and iron ore to
a smelter. Quartz is added to form a glassy slag that ties up the fluoride, sodium and other
impurities for land disposal. Elkem has also piloted a process to recover fluoride from the
molten slag. This process uses steam to drive off the fluoride as hydrogen fluoride which can
be reacted with hydrated alumina to make alumina fluoride for recycling to the aluminum
process.
Envirosciencc
Enviroscience, Inc. (ESI) has completed a pilot plant demonstration sponsored by
Kaiser Aluminum, Vanalco and Columbia Falls Aluminum Company (23). Their process uses
K088 and K.061 (electric arc furnace dust (EAF dust)) to produce zinc oxide, mineral wool
fiber and pig iron. The K088 is formed into pellets then heated to approximately 3000 °F in
a furnace, with lime and silica being added to attain an -optimal acid:base ratio for proper fiber
formation. The carbon and the cyanide from the potliner are used-to reduce the metals in the
K.061. The non-reducible metal oxides are spun into a mineral wool from the molten slag.
The ESI technology employs extractive metallurgy followed by vitrification of non-reducible
metal oxides. In this process, volatile and reducible metals are used as their respective oxides
and elements. Non-reducible metal oxides are converted into a man-made vitreous fiber
identical in chemical content to a mineral wool fiber known as rock or slag wool. This
process involves recycling of K088 (and K061 (EAF dust)) into commercial products. The
spent potliner provides certain secondary fluxes to modify the viscosity of the primary fluxes,
and clean the coke and the furnace walls. The sodium is converted into sodium oxide much
as soda ash does in other furnaces. The fluoride is converted into calcium fluoride
(fluorspar), eliminating the need for fresh fluorspar and soda ash in the furnace. Figure 4-5
presents a schematic of the ESI process.
4-26
-------
M k M
•ft*
II *»§«
So •= o.s
9 J> w wu.
i
(/)
w
0)
o
o
Q.
55
LU
0)
'o
(O
m
E
Q)
r»
Am
O
en
U)
<±
g>
3
.2*
i^ S
"• i
o
3
^
01
2
— *
—
•
01
— c
• m
55
2
o
•
•
^ «
O
«*
0
u
1 •"
u.
3 A
A
I
&
o
t k
1
u
> k
»
9
o
X.
*^T* *
r
; »
!
O)
3
£
>k
?
„ 1
a
o
u
TJ
O
o
•/I
i/s
a
O
JO
u
Qi
SA
'^
ID
CJ
C
i
.1 8
> c
c u
c
0 .t=
" I
E L~
3
"o 2
0=1 .2
ry 'G
rt L.
O O
•
u o
H U
-------
I
The amount of spent potliner material used in the reduction process will vary greatly.
The reducing agents in SPL vary between each producer as well as each pot. The amount of
metals to be reduced from K061 vary by waste generator and type of waste. The type of
metal to be reduced determines the amount of oxygen bound to the metal as well as the most
effective reducing agent. The key to the process is to grind the SPL into a powder, blend the
powder and test the powder. Carbon is the reducing agent that covers the widest range of
metals as well as imparting higher value as carbide in some metals produced. Compositions
of spent potliner have varied from a low of 13% to a high of 60% carbon content. Using
electric arc furnace dust (K061) as a standard for the reducible metals (iron, nickel, zinc, and
chrome), the average amount of SPL required for reduction becomes less than 20% by weight.
Because the reduction process is endothermic, no heat value can be derived from the carbon.
The type of furnace or heat source makes little difference to the process. The shaft furnace
has worked well because the pellets formed by the waste can be slowly brought up to the
reduction temperature and held there for the 15 minutes required for total reduction. This is
done continuously as charge moves down the shaft. Any other furnace that allows 15 minutes
retention time at the reduction temperature before melt of the pellet will work.
The best use of this material is to directly convert it into fiber. This is done by
putting the material into a high temperature furnace. Electric submerged arc or plasma arc
furnaces work well for this application. The brick is melted and spun into a high-quality,
high-temperature acid refractory insulation. This insulation would have a melting point much
higher than mineral wool and only slightly lower than "Certa Wool".
Another way to use this material is as the source of both Si02 and A12O3 in standard
mineral wool production. The furnace used in the process is unlined. The water-cooled shell
freezes slag in a thin layer on the furnace walls. Because the furnace has no lining, the walls
do not erode and change the slag components as in other furnaces. The high A1203
concentration in the brick must be reduced in the slag with the addition of SiOj and CaO in
the charge. This can be done with the standard furnace and process when very little brick
need to be processed.
4-28
-------
ESI recently completed a pilot plant demonstration project in Vancouver, Washington
sponsored by three primary aluminum companies. Carbon and carbon compounds (including
cyanide) in the potliner were used as the reductant to extract zinc, cadmium, lead and iron
contained in EAF dust. Mineral wool fiber and slags suitable for other commercial products
were produced from the non-reducible metals contained in the potliner, EAF dust and the
surrogates.
An 18" I.D. countercurrent mineral wool type shaft furnace was used in the project.
The processing steps were:
A. Blending of the waste materials
B. Pelletizing with water and binder
C. Curing
D. Drying
E. Charging to the furnace with fuel
F. Collection of products.
EAF dust contains reducible oxides such as FeO, ZnO, PbP, CuO, NiO, and Cr2O,
along with non-reducible oxides of CaO and A1203. Potliner contains carbon and cyanide as
reductants; A1203 and SiO2 for fiber production and fluoride and sodium for secondary fluxing
agents. Experiments were conducted with EAF dust/potliner ratios of 40:30 to 60:30. Lime
and silica were added to obtain a proper acid:base ratio that assures proper fiber formation
from the molten slag.
Zinc, cadmium and lead oxides (in the presence of halogens) volatilize at the 2,800
degrees fahrenheit furnace operating temperature. These oxides are recovered in the bag
house. Recovery exceeded 95%. A silvery pig iron containing 98% iron was obtained from
the bottom of the shaft furnace. Aluminum, calcium, magnesium, fluoride, sodium and silica
remained in the slag as fiber forming compounds.
4-29
-------
I
I
I In a commercial mineral wool plant, rock and slag are melted in a shaft furnace to
produce molten slag. This molten slag is converted to fiber using a wheel spinning process.
I The wheel spinner converts 70%-80% of the molten slag into fiber. The remaining 20%-30%
is shot (unfiberized slag). At commercial facilities, shot is separated from the fiber and the
| shot is usually landfilled because commercial mineral wool plants normally do not have
pelletizing plants. In the ESI process, the shot would be pelletized and returned to the
0 furnace.
I
I
I
I
I
I
I
ESI believes that extractive metallurgy combined with fiber production (vitrification)
can be a cost effective recycle process compared to thermal treatment, secure landfill disposal
or standard HTMR processes. The process provides for the production of commercial
products from select inorganic liquid, semi-solid and solid hazardous wastes.
Ormet Primary Aluminum Corporation
• Ormet Corporation has used a pilot-scale combustion melting system (CMS)
* vitrification process to treat K088 wastes (8). The process involves the rapid suspension
• heating of the waste and other additives in a preheater prior to physical and chemical melting
which occur within a cyclone reactor. Ormet has submitted a petition to the EPA requesting a
• delisting of their residues from this process. They intend to scale-up this plant upon receiving
a delisting of their waste. They claim the process produces a nonhazardous reusable product
with the qualities of industrial glasses. Figure 4-6 presents a schematic of the CMS system.
• Ormet has conducted pilot-scale spent potliner vitrification process studies. The intent
of the pilot-scale vitrification process studies, was to render the spent potliner nonhazardous
• and produce a nonleachable glass product that could be beneficially reused. The vitrification
system is an extension of an advanced multifuel capable CMS being developed for high
• temperature process hearing applications. The primary components of the basic CMS are a
feedstock preheater and a cyclone reactor (melter).
4-30
-------
U)
to
CO
CO
o
0)
o
0)
0)
o
CO
0)
3
O)
-------
I
I
1) Vitrification Feedstock Preparation
— FO' Jie pilot study, representative samples of spent potliner were reduced to a nominal
• minus one-half inch. The spent potliner was further ground to a minus 40 mesh (0.425
I millimeter) in an on-site vibrating ball mill. This size reduction was required to ensure
complete carbon oxidation and feedstock melting in the CMS.
Based on results of pilot-scale surrogate testing, it was determined that various other
glass-forming ingredients should be added to the spent potliner to improve the melting
characteristics of spent potliner and to improve the physical/chemical properties of the
vitrified product from spent potliner.
2) Vitrification Process
I
I
I
I
I The CMS consists- of two primary assemblies; an in-flight suspension combustor, and a
cyclone separation and melting chamber. Fuel combustion and in-flight suspension
I preheating of the vitrification process feedstock (spent potliner and other glass-forming
ingredients) take place in a high combustion intensity combustor.
I
A scaling analysis was performed on the CMS process to determine whether the test
• results of the pilot-scale system utilized for the vitrification of the spent potUner may be used
in the development of a full-scale system. They have performed several studies on the scale-
• up of the CMS process using b
-------
CMS have been identified ind designs for a proposed full-scale CMS system with a
processing throughput capability of SO tons per day has been completed.
The primary concern for the CMS process scale-up for vitrification of the spent
potliner is to ensure that there is sufficient combustor residence time to achieve complete
oxidation of all the organic constituents. During the pilot-scale testing, all of the carbon in
the spent potliner was effectively oxidized. The scaling criteria used in the design of the
proposed full-scale system provides a greater residence within the full-scale system than was
available in the pilot-scale system, and therefore, will allow more than sufficient residence
time for complete carbon conversion.
The second major concern for the CMS process scale-up is to ensure that the fluid
dynamic of the larger system provides for stable combustion of the organic constituents in the
spent potliner. Several simulations were conducted at various length to diameter ratios to
determine the optimum configuration of the full-scale system. Results of the modeling has
indicated that as the combustor size increases, the mixing ability of the combustor increases
providing additional flame and combustion stability.
Economic feasibility of this process will be dependant upon the demand for the
vitrified product as recycle/reuse, as new potliner make-up material and reuse as an industrial
grade glass. If the demand is low or non-existent the vitrified product will be disposed of in
a landfill
Pyrosolfolysis
Pyrosulfolysis is a two step process being investigated by Martin Marietta and owned
by Commonwealth Aluminum Corporation. It is claimed that this technology enables the
destructioo of cyanides and allows the recovery of fluorides as hydrofluoric acid by burning
K088 in the presence of a sulfur source. The first step is prctreatment of K088 in order to
reduce its particle size to 1/S-inch. The pretreated K088 is then placed In a fluidized bed
4-}3
-------
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
combustion (FBC) unit with suitor dioxide added stoichiometricaJly. Hydrofluoric acid can be
recovered by dry scrubbing the off-gases from the FBC and any sulftir dioxide can also be
scrubbed ou»~
Reynolds Metals
Reynolds Metals has developed a process to thermally treat spent potliner in a rotary
kiln (10). The thermal treatment destroys the cyanide and reduces the amount of teachable
fluoride by converting soluble forms of fluoride salts to insoluble calcium fluoride. Reynolds
Sfeuls petitioned the Agency through the delisting process to exclude its kiln residue
generated from the treatment of spent potliner wastes by its rotary kiln process. Reynolds
received an exclusion for kiln residues from its process. As a result, spent potliner treated in
the Reynolds facility and meeting the delisting conditions set out in its delisting decision are
excluded from the lists of hazardous wastes contained in 40 CFR $261.31 and $261.32.
Figure 4-7 presents a general schematic of the Reynolds treatment process.
In the rotary kiln treatment process, spent potliner is first crushed and milled to a 3/8-
inch particle size. Brown sand and limestone are also ground to 3/S-inch particles. Brown
sand is an alkaline mud generated in the two-stage process of extracting alumina from
bauxite2. In the past, Reynolds stored this waste oo-site at the Bauxite facility and has built
up a significant stockpile of this miffrial, Reynolds mines the brown sand from the dry lake
beds at their Hurricane Creek facility and crushes it for ust in their treatment process. The
crushed spent potliner, brown sand, and limestone can then be blended in varying ratios
depending on the results of initial spent potliner characterization (i.e.. the greater the cyanide
and fluoride levels in the spent potliner, the more brown sand and limestone is added). In
general, the spent potliner can contribute between 30 to 45 percent of the influent to the
treatment process generating th« petitioned waste. Reynolds adds the brown sand to help
' Reynolds reports dui brown uod tut approxbMUty SH aluminum oxide, 25H lilkoo dioxide, 3H
sodium ctrbottBUL 4IH calcium oxide. I OH fork oxide, and 4H tiooiun dioxide.
-------
o
o
o
0)
"5
o>
o
>»
o
DC
O
5
-------
I
I
I
I
I
I
I
I
I
I
prevent the mixture being treated from agglomerating in the kiln. Limestone reacts with the
soluble fluoride salts (sodium fluoride and cryolite) in spent potiiner to form stable, relatively
insoluble calcium fluoride, thereby reducing the leaching potential of fluorides in the kiln
residue.
The rotary kiln is approximately 250 feet in length and 9.5 feet in diameter and
operates counter-currently. Natural gas is used to heat the kiln to the 1,200 *F operating
temperature at the burner end. The flue gas is sent through cyclones and an electrostatic
precipitator (ESP) to remove solids. Reynolds recycles the solids from the cyclones to the
kiln: solids generated from the ESP art handled as hazardous waste. The kiln residue is
cooled by contact spraying with lake water and stored in waste piles. Reynolds disposes of
the kiln residue at non-hazardous waste landfill
4.2.2 Wastewaters
Although K088 wastes meet the definition of nonwastewaters as generated, EPA
establishes treatment standards for both wastewater and noowastewater forms of listed wastes
•to ensure that any waste streams that meet the definition of wastewater are also treated to
—meet appropriate treatment standards prior to land disposal Streams generated from the
•treatment of K088 containing less than one percent total organic carbon (TOC) and less than
I one percent total suspended solids (TSS) are defined as wastewater forms of these wastes to
which the wastewater treatment standards in this rule apply. As • result, this section presents
the Agency's determination-of:
I
I
Applicable technologies
Demonstrated technologies
The Best Demonstrated Available Technology (BOAT)
for treatment of wastewater forms of K08S.
I
I
4-36
-------
4.2.2. t Applicable Treatment Technologies
Applicable treatment technologies for organic*, cyanide, fluoride and metals in
waste water forms of K088 include those that destroy or reduce the total amount of various
organic compounds and cyanide in the waste, and recover or fix the fluoride and meuUs in the
waste. The Agency has identified the following technologies as being applicable for treatment
of wastewatcr forms of these spent potliner wastes.
• Biological treatment (including aerobic fixed film, aerobic lagoon, activated
sludge, filtration, anaerobic fixed film, rotating biological contactor, sequential
batch reactor, and trickling filter technologies);
• Carbon adsorption (including activated carbon and granular activated carbon
technologies);
• Chemical oxidation (including alkaline chJorination);
Chemically assisted clarification (including chemical precipitation technology);
• PACT* treatment (including powdered activated carbon addition to activated
sludge and biological granular activated carbon technologies);
• Reverse osmosis; and
• Total recycle or revise.
The concentration and types of waste constituents determine which technology is most
applicable. A brief discussion of each of the technologies identified as applicable for the
treatment of constituents in wastewater forms of L *S is given below (3)
Biological Treatment
Biological treatment includes aerobic fixed film, aerobic lagoons, activated sludge,
anaerobic fixed film, routing biological contactor, sequential batch reactor, and melding filter
technologies. Biological treatment is a destruction technology in which organic constituents in
-------
I
I
•woste^aiers art biodegraded. This technology generates two treatment residuals: i treated
effluent and a waste biosludge. Waste biosludge may be land disposed without further
•treatment if the concentrations of its regulated constituents art equal to or below their BOAT
treatment standards.
I
Carbon Adsorption
I
Carbon adsorption is a separation technology in which hazardous organic constituents
•m wastewaters are selectively adsorbed onto activated carbon. This technology generates two
treatment residuals: a treated effluent and spent activated carbon. Spent activated carbon is
•Dficn reactivated, recycled, or incinerated.
Chemical Oxidation
Chemical oxidation is a destruction technology in which some dissolved organic
ompounds art chemically oxidized to yield carbon dioxide, water, salts, simple organic acids.
d sulfates. Alkaline chlorination is a chemical oxidation technology that enables the
cstruction of cyanide in aqueous solutions or sludges. This technology generates one
eatment residual: treated effluent
I
I
BUT
i
• Chemically Assisted Clarification
• Chemically assisted clarification, including chemical precipitation, is a separation
Vxhnology in which the addition of chemicals during treatment results in the formation of
recipitates from the organic and inorganic constituents in the wastewater. The solids formed
•ire then separated from the wastewater by sealing, clarification, tod/or polishing filtration
i"his technology generates two treatment residuals: treated wastewater effluent and separated
•solid precipitate. The solid precipitate may be Und disposed without further treatment if it
meets the applicable BOAT treatment standards for regulated constituents.
I
-------
PACTS Treatment
PACTS treatment is a combination of carbon adsorption and biological treatment. It
is a destruction technology in which hazardous organic constituents arc biodegraded and
selectively adsorbed onto powdered activated cirbon. This technology generates two
treatment residuals: a treated effluent and spent carbon/biosludge. Spent carbon b often
regenerated and recycled to the process or incinerated.
Reverse Osmosis
Reverse osmosis is a separation technology in which dissolved organic* (usually salts)
art removed from a wastewater by filtering the wasttwater through a semi permeable
membrane at a pressure greater than the osmotic pressure caused by the dissolved organic
constituents. This technology generates rwo treatment residuals: the treated effluent and the
concentrated organic materials that do not pass through the membrane.
Total Recycle or Reuse
Total recycle or reuse of a waste material within the same process or an external
process eliminates the generation of a waste for treatment and disposal and subsequently
genenies no treatment residuals.
4.2.2.2 Demonstrated Technologic*
The Agency docs not have any information that indicates that any of the above
technologies is used for the treatment of wastewater forms of K088. The Agency is not
aware of any wastewater forms of K088 being currently generated.
4-39
-------
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
4.2 13 Identification of BOAT
The Agency believes thai alkaline chJorinatioo of JC088 wastewaters to destroy cyanide
followed by chemical precipitation is likely to be an appropriate treatment Alkaline
chlorination is demonstrated for the destruction of cyanide and chemically assisted
clarification is demonstrated for metals and fluoride.
4.3 Identification of BDAT Treatment Standards
The Agency is transferring universal treatment standards to the constituents selected
for regulation in nonwastewater and wastewater forms of K088. A universal standard is a
single concentration limit established for a specific constituent regardless of the waste matrix
in which it was present Universal treatment standards are intended to be used to replace
treatment standards in previously promulgated waste codes and as the treatment standards for
listed hazardous waste codes in the future.
This section presents the universal standards that were transferred to the constituents
selected for regulation in nonwastewater and wastewater forms of K088 and the specific data
used to determine these treatment standards.
4.3.1 Nonwastewatrrs
The Agency is transferring universal standards to the constituents selected for
regulation in nonwastewater forms of K088. Tables 4-7. 4-8 and 4-9 present specific
treatment performance data used to determine the universal standards for the constituents
regulated in the K08C wastes.
Universal standards for organic constituents regulated in K088 were based on
incineration treatment performance data. These data represent the BDAT for wastes Included
in previous rulemakinfs and therefore have been judged to meet the Agency's requirement of
4-40
-------
BOAT. The treatment standard for fluoride oo&wastewtten is a leachate concentration uhich
was determined by the Agency when granting t delisting for certain KOSS wastes. The
treatment standard for fluoride was based on the Reynolds delisting petition. This
concentration is based on the MCL of 4.0 mg/1 for fluoride developed under the Safe
Drinking Water Act and the dilution attenuation factor developed from the EPACML model
used during the development of the verification testing limits set for the Reynolds delisting
decision. The Agency selected this level after consideration of a variety potential treatment
standards. The Agency first calculated a technology-based treatment standard using the
BOAT methodology. This calculation was performed using the available treatment data from
the Reynolds process, Ormet process and the Test Burn performed by the EPA. Based on this
data, a treatment standard of approximately 29 mg/1 was calculated. The Agency considered
•
this concentration to be overly stringent and not likely to be readily achievable by the
available technologies. The Agency also believed that this standard did not help the Agency
meet object of the safe disposal of the KOSS waste. Thus, the Agency selected the 48 mg/1 is
a readily achievable standard for KOSS wastes. The treatment standard for cyanide is based
on alkaline chlorination (transferred from F006). While most of the technologies described
above are use a thermal destruction process to destroy cyanide, the Agency is transferring the
Universal Treatment Standard based on alkaline chlorination. The Agency would like to note
that one of the technologies described (Alcan) did use alkaline hydrolysis for cyanide
destruction. The Agency, however, does not want to force the use of a treatment technology
so any treatment, including recycling or any combination of treatment technologies, unless
prohibited (e.g.. impermissible dilution) or unless defined as land disposal (e.g., land
treatment), can be used to achieve these standards. However, the Agency has developed a
preliminary treatment standard for cyanide based on KOSS treatment data submitted to the
Agency, as shown in •Development of BDAT Number for Cyanide in K088" (46). Treatment
standards for metals are based on HTMR technology.
Tables 4-10 and 4-11 pruent the treatment standards for nonwastewater forms of KOSS
as well as a summary of data available to the Agency on treated nonwastewater forms of
KOSS. EPA'» performance data supporting the concentration levels and list of regulated
4-41
-------
I
I
I constituents are from the Reynolds Metal Company's thermal treatment process and EPA's
. rotary kiln incineration of K088. An analysis of residues from these two technologies as well
• ax cuu submitted on Comalco's Comtor fluidized bed calcination process and Ormet's
• vitrification process leads EPA to believe that the concentration-based universal treatment
™ standards can be routinely achieved by a variety of technologies.
Each one of the treatment studies relied oo the collection of composite samples.
• Discrete grab samples of untreated and treated K088 were collected and then were
consolidated per run in order to characterize the wastes. However, the frequency and
protocols of how composites were collected varied somewhat. This information is
documented in the respective testability studies or onsite engineering reports. While the
treatment studies relied on composite samples for characterization, the Agency is requiring
that discrete grab samples meet the standards. Toe Agency believes that the extensive
crushing and grinding as well as the additives needed for some technologies will tend to
reduce the variability of constituent concentration in treated spent potliner. The Agency
believes that grab sampling is appropriate for testing compliance with the treatment standards.
I
I
I
I
I Each study shows that thermal destruction is the best technology to treat the complexes
of iron cyanide present in K088. As discussed in all the various technologies above, it is
| possible to destroy the cyanide. EPA believes that treatment standards for cyanides must be
based on residues from the destruction of cyanides prior to any stabilization or ultimate
I disposal. The legislative history of 3004
-------
K088 treatment technologies leads EPA to believe that the concentration-based universal
treatment standards can be routinely achieved. As discussed in the Advance Notice of
Proposed Rulemalcing (ANPRM) published in October 24. 1991, EPA was seeking
development of concentration-based standards for K08S so as to allow (he use of any
technology that can achieve the numerical vlues. The Agency notes, however, that when it
establishes concentration-based treatment standards, the regulated community may use any
non-prohibited technology to treat the waste to meet the treatment standards. Compliance
with concentration-based treatment standards requires only that the effluent concentration be
achieved; once achieved, the waste may be land disposed. The waste need not be treated by
the technology identified as BOAT; in fact, concentration-based treatment standards provide
flexibility in the choice of treatment technology. Any treatment, including recycling or any
combination of treatment technologies, unless prohibited (e.g.. impermissible dilution) or
unless defined as land disposal (e.g.. land treatment), can be used to achieve these standards.
Details regarding the methodology used to develop the nonwastewater universal
treatment standards and the treatment standard database are presented in Appendix A of this
document A more detailed discussion is provided in EPA's Final Best Demonstrated
Available Technology fBDAT> Background Document for Universal Standards; Volume A.
Universal S***^ards for Nonwastewater Forms of Wastes (2)
4.3.2 Wastewaten
The Agency is transferring universal standards to the constituents selected for
regulation in wastewcter forms of K088. Tables 4-12 and -M 3 present specific treatment
performance data used to determine the universal standards for the constituents regulated in
the K088 wastes.
The universal standards for wastewater forms of these wastes are based on treatment
performance data from several sources, including the BOAT database, the NPDES database.
(he WERL database, EPA collected WAOPACT* data, the EAD database, industry submitted
4-43
-------
leachate eeatment performance data, data in the literature that were not already part of the
WERL database, and data in literature submitted by industry on the WAO and PACT*
treatment processes. The treatment standard for fluoride wastewaters is based on lime
conditioning followed by sedimentation as developed for the F039 • multi-source
t
standards.
Tables 4-14 and 4-15 present the treatment standards for wastewater forms of K088 i
well as the available data on wastewater forms of K.088. An analysis of this data from K08!
treatment technologies leads EPA to believe that the concentration-based universal treatment
standards can be routinely achieved. Details regarding the methodology used to develop the
nonwastewater universal treatment stzndards and the treatment standard database are present*
in Appendix A of this document A more detailed discussion is provided in EPA's Final Be
Demonstrated Available Technology (BOAT! Background Document for Universal
-------
Table 4-1 BOAT Treatment Standards for Nonwastewater Fonnj of K088 for
Organic Constituents tod Cyanide
Regulated Constituent
Cyanide
Cyanide (amenable)
Acenaphthene
Anthracene
Benz(a)anthracene
Benzo(a)pyrene
Benzo(b and k)fluoranthene*
Benzo(gJU)perylene
Chrysene
uiDenziajijeninracene
Fluoranthene
Indeno(1.2JTene
Total Composition Conceutrtdon
(»|/ki)
MAiimnm for any Single Grab
590
30
3.4
3.4
3.4
3.4
6.8
1.8
3.4
8.2
3.4
3.4
5.6
8.2
* The treatment standard for these constituents Is cxpc eased as a sum of their concentrations to
account for analytical concerns In diffingnithing between the two compounds.
Ref.: 2) U.S. Environmental Protection Agency, Office of Solid Waste. Final Best
ology (BPAT^ Background Document for Universal Standards!
Volume A. UnivenaJ Standards for Nonwaatewater Forms of Wastes. U.S. Environmental
Protection Agency, Washington, DC, July 1994
4-45
-------
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
TabU 4-2 BOAT Treatment Standards for Nonwastcwater Fonni of K08S
for Metals and Fluoride
RefuUted Constituent
Fluoride
Antimony
Arsenic
Barium
Beryllium
Cadmium
Chromium (total)
Lead
Mercury
Nickel
Selenium
Silver
Leaeaabk CoMeatrado* («f/L)
UdflfTCLP
4S»
2.1
5.0
7.6
0.014
0.19
0.74
0.37
0.025
5.0
0.16
OJO
Ref: 2) U.S. Environmental Protection Agency, Office of Solid Waste. Final Best
Demonstrated Available Technology fBDAT) Background Document for Universal Standards-
Protection Agency. Washington. DC, July 1994
U.S. Environmental
1 A Univtml Trancac Standard wti aof pronulpud tot fluoride Tb« cooccntmioa pracoted it the
l«*dubU uxxmamKm idaitiA«d la the fUyoolds dtiimnf
4-46
-------
Tiblt 4-3 BOAT Treatment Standards for Wastewater Forms of K083
Regulated Constitueot
Acenaphthene
Anthracene
Benz(a)amhracene
Benzo(a)pyrene
B«nzo(b and k)fluoranthene*
Benzo(g,!U)perylene
Chrysene
Dibenz(a,h)anthracene
Fluoranthene
Indeno( 1 ,2,3-cd)pyrene
Phenanthrene
Pyrcne
Cyanide (total)
Cyanide (amenable)
Fluoride
Antimony
Arsenic
Barium
Beryllium
Cadmium
Chromium (total)
Lead
Stecury
Nickel
Total Composition Concentration (mt/L)
Maximum for any Static Grab
0.059
0.059
0.059
0.061
0.11
0.0055
0.059
0.055
0.063
0.0055
0.059
0.067
1.2
0.86
35
1.9
1.4
1.2
0.82
0.69
2.77
0.69
0.15
3.98
4-47
-------
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
Table 4-3 BOAT Treatment Standards for Wutewatir Forms of KOS8
Regulated Constituent
Selenium
Silver
Total Composition Concentration (m|/L)
Maximum for any Single Grab
0.82
0.43
* The treatment standard for these constituents is expressed as a sum of their concentrations to
account for analytical concerns in distinguishing between the two compounds.
Ret: 3) U.S. Environmental Protection Agency, Office of Solid Waste. Final Best
Demonstrated Available Technology fBDATt Background Document for Upiversal Standards:
Protection Agency, Washington, DC July 1994
4-4S
-------
00
o
I
2
.8
e «
liii
liliJl
-------
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
S
.5
>**
i
z
o
at
a.
i
(JJ
Z
flu
I
U!ji !
I jf
*""" x 9 * 'I • *•
ISl'h-*11
il!iH?H
ii
i
f
1
Jsj
£ ««2
§
1
ill1
H" J»
!i
Hfl"
i]
V.
I
4
«»1
§
ills
•4 -"^
Jl
»
I
j
S-a
5bi!
*^
!
•
II
Jl
o
I
tt**l
i
^^
i
I
i
*HI
iJ*!li*
*= e-rfi "*
Hi
Jj
1
i
3
i
1
c
I
^1
-
i
-------
-------
1
I
a
j
£
5
i
£
I
1
jt
12
i
II
I
2
j
l\
I
J
1
s
*
,2
* i c*
ifli
4
j
1
r
8
1
i
ii
i
1
2
I
\
h
i
2
i
1
M
1
j
J
I
1
S
s
JJ
-------
« ^"
I
Accmcy
S
J
ill
i
8
8
S
I
1
I
1
1
I
i
1
«•%
i
8
!
j
v-»
2
-------
i!
8
l!f
!)
liji
hi
i** r
}
8
£
*
8
2
.
3
1
1
8
1
1
S
j
i
I
i
2
i
1
I
J
I
r
*
J
J
i
j
f
n
*j
j
• i
!l 11!
i if ,s
I I
1 |
1
Ij
ii 5! M1
j jj M
i '{
^ * •
t
s
K
i
1
j
II
I
I
'i 5 li si M
spi H-S I
fi fj jl fj j;
j| Jl (!c jl fj
-------
h- '
;;.,»;
P
N
I
w
I
8
ii
(^
i
1
11
II!
i
s
r
I
06
1
222EX
III!5
Lovdfof
8 P
I
I
"*8
£5
1
KMI
M
i \*\'b
-------
1
1
1
1
1
i !
V
4
1 I
Z
1 i
1 1
1 1
1
A
1 1
i
• o-
•s
1 1
H
1 :
1 :
1
1
1
3
£
I
e
f
5
6
^
\\
*
\
&
|||
]1
III
*
Hi
* 1*1
II1
°
^l
^fl 1
1
II;!|
Jto 3
N]^
slli
»
I
2
I
il!i!
|JH
1
t
M
IS
|
f»
n
e
;
r»
rf?
-------
^•N
§
§
£
;
1
3
3
3
e
3
3
3
'
t
i
il
3
l
Uvdfcr
1
H
oo
9
6
I
3
i
-------
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
S.
6
b
l
I
»
ii
'!!
if
n
w%
K06
2
i
i!
3
p
£
*
K06I-SKF
§
I
i
5
gcfs
2 £ *•
i
E
I
I
if
I
1:
I
p
* »
-------
Ill
I
!
H
Sllj
I J JH I
I
iii
>,,
-------
I
I
I
I §
I
I I
I
I
I
I
I
I
I
I
I
I
I
I
oe
3
•t
c.
o
I
I
<•«
s
I
0.
a a
i
3
I
1
I
c
i
I
2
I
.!
]
*
I
If
*
Iff
I
"*S<5
13
' 3
3*
3
s
li 1
g
oT
I
-------
2
a
V)
M
3
*
•c
J
O
i
i
§
I
i
a
j
j.
"3iJ
i
i!'
I'
~ P
I
I
I
ta
B.
!
I
I
a
I
o
.
1
-------
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
• I
Table 4*12 Calculation of Universal Treatment Standards for Organic Constituents
[Wastewaten)
Owtltitot StltttH
For RMvlcttoA
AothnctM
B*Q*»«ttfe*C«»
ILLJ. WL \niuumrtmu
BauoOOfnuHiiuncM
Suaofbauo(b)fiiionfUboM
BauoCaJLOporyteM
• r i
fenirtfcyraM
ChrywM
nttiMrf* hWrfLxmrjnj
lsiu*itHm_Hmmim m mm
fn/tMtVI t T_. fLiLT JJLJ
uuauH I j^~ca)pyTtBt
PhfnvnhriM
Pyrcn*
Avwflt KflbMojl
04)10
04)10
04)10
04)10
04)10
•
O.OOIO
04)10
04)10
04)10
0.012
O.OOIO
0.010
04)11
VwUbflity
5.9
5.9
5.9
5J
5.9
•
5J
5.9
5.9
3.3
5.9
5J
5.9
5.9
Aeanqr
Comtdoa
.
*
.
.
-
•
.
.
•
.
•
•
.
-
Ultra*
Tlwtec
Studar
<«l/t)
0.059
0.059
0.059
0.055*
• 0.059*
0.11*
0.0055
0.061
0.059
0.055
0.061
0.0055
0.059
0.067
R*f: 3) U3. Fn*krt'tfnf"*il PtotocH*** A*«WV rwviM */ CHILI a/>^. cu«i n^* rv»^^..ti»«^! Au.n.M>
rechnclocv fBDAT) Badccreand C
Wutewttcr Forms of Wnt*^ If A.
ecumcot for UnfvtmJ {fta^njli VohaM B. Unlvcnal $ttn
4^2
dgds for
94
-------
Table 4-13 Calculation of Universal Treatment Standards for Metal Constituents
(Wastewaters)
For RtftbitfM
• A ntifflooy
Annie
Barium
Beryllium
Cadmium
ChromfamCtotal)
•
Mtrcury
Nlckd
S«l«nium
Silvtr
AmtfalffiMa*
Coactttrvtioa)
(•it)
0.47
0.34
021
020
0.13
OJ7
0^0
0.036
0.94
0.20
0.096
VadabttUy
Factor
4.1
4.1
4.1
4.1
SJ
4.9
3.3
4.1
42
4.1
4J
Aeoiracjr
ComctidB
Factor
*
.
.
.
*
.
.
.
.
-
Utlraul
TfMUiMt
Slaadard
1.9
1.4
1.2
O.R
'0.69
2.77
0.69
O.IS
3.91
O.S2
0.43
4-63
-------
1
J
I
2
§
ui
£
•o
I
$
!
I
i
J
I
?
w->
8
s
-------
I
H
oo
•q
*J
21
I!
8
j
{
9
-nro
9
9
9
9
D
i
5
3
I
3
3
S «.i
ill
S
I
1 flfl-j
=
2
-------
5.0 REFERENCES
1.
3.
4.
5.
6.
7.
9.
10.
U.S. Environmental Protection Agency, Office of Solid Waste. Final tea
Demonstrated Available Technology (BOAT) Background Document for
Assufance.OuaHtv Control Procedures, and Methodologies. U.S. Environmental
Protection Agency, Washington. D.C. October 23. 1991.
U.S. Environmental Protection Agency, Office of Solid Waste. Final B«t
Demonstrated Available Technology (BOAT) Background Document for Universal
Environmental Protection Agency, Washington, DC, July 1994
U.S. Environmental Protection Agency, Office of Solid Waste. Final Best
Demonstrated Available Technology (BDAT> Background Document for Universal
Standards; Volume B. Universal Standards for Wastewater Forms of Wastes. U.S.
Environmental Protection Agency, Washington, DC, July 1994
U.S. Environmental Protection Agency, Office of Solid Waste. Final BOAT
Background Document for U and P Wastes and Mold-source Lcachate; Volume A/
May 1990.
U.S. EPA. Background Document. Resource Conservation and Recovery Act. Subtitle
C • Identification and Listing of Hazardous Waste, May 2, 1980.
U.S. Office of Management and Budget. Standard Industrial Classification Manual
Washington. D.C. 1987.
U.S. Environmental Protection Agency, Office of Water, Development Document for
Effluent Limitations Guidelines md New Source Performance Standards for the
Primary Aluminum Smelting Sub<**t^gofV of the Aluminum Segment of tha
Notifcrrous MctaJi Manufacturing Point Source Category. U.S. Environmental
Protection Agency, Washington, DC, March 1974
Ormct Corporation, Petition for Exclusion for Vitrified Product from Spent Potliner.
Submitted to U.S. Environmental Protection Agency. Ormct Corporation, Hannibal,
OH. April 1994
U.S. Environmental Protection Agency, Office of Research and Development
ChanctcriradQfl for Soent Potllngn from the Primary Reduction of Aluminum. U.S.
Environmental Protection Agency, Cincinnati, OH, April 16, 1991
Reynolds Metals Company, Petition for Exclusion for Spent PotHner Generated at
Reynolds Mcttli Cotroanv Hurricane Creek Facility. Hurricane Creek- AK. Submitted
5-1
-------
to U.S. Environmental Protection Agency, Reynolds Metals Company, Richmond,
VA. August 10. 1989
11. U.S. Environmental Protection Agency, Office of Research and Development pj|oj-
scale Incineration Tests of Spent Potlinen from the Primary Reduction of Aluminum
(K088) U.S. Environmental Protection Agency, Cincinnati. OH., 1991.
11 U.S. Environmental Protection Agency, Office of Solid Waste, SW.846 Test Methods
for Evaluating Solid Waste Physical/Chemical Methods. Third Edition. Washington.
D.C November. 1986.
13. American Society of Testing and Materials (ASTM) D 1179-80, Method B
14 U.S. Environmental Protection Agency, Office of Solid Waste. Final Treatment
Technology Background Document. Washington, D.C., May, 1990. '
•
15 U.S. Environmental Protection Agency, Office of Solid Waste. Technical Feasibility
Assessment of Spent Potliner Management Potions in the Aluminum Smelting
iodusjQc, Washington, D.C, September, 1986.
16. Managing Spent Potlining: A Report to the Washington State Department of Ecology,
Submitted by Alcoa. Columbia Aluminum, Intalco, Reynolds Metals, and Vanalco.
February. 1989.
17. "Chemical Recovery from Spent Potlining", Proceedings of an International
Symposium sponsored by the Extraction tn4 Processing Division of The Minerals.
Metals & Materials Society, February 27 • March 3,1994.
19. "On-site Engineering Report • Comtor Process for Spent Potlining Treatment*.
Submitted to US.Environmental Protection Agency RCRA Docket F-91-CSP-FFFF,
January 9,1992.
20. The Comtor Process for the Treatment of Spent Potlining* Brochure from Comalco
and A1SCO. January 31,1994
21. Further Development of the Comtor Process for SPL Treatment, Light Metals, 1993
22. "New Technology Takes Aim at Aluminum Processing Waste", Chemical Engineering
Magazine, May 1994, pp. 47-52
23 Technical Report Volume I on Envirosdence, Inc. Process Using Recyclable Inorganic
Wastes to Produce Commercial Products. Enviroscience. 1/1/94.
5-2
-------
24. Report to Congress on Special Wastes from Mineral Processing, EPA/530-SW-90-
070C, July 1990.
25. Regulatory Impediments to the Reclamation and Reuse of Spent Potliner from Primary
Aluminum Reduction, Jack Goldman, Jeffrey S. Holik, From Environmental
challenges of the 1990's Proceedings, EPA/600/9-90/039, September. 1990
•
26. U.S. Environmental Protection Agency, 1991 Hazardous Waste Report, 1991 Biennial
Reporting System National Oversight Database.
27. Reynolds Metals Company, Report on Twenty Dav Verification of Kiln 2. Report
Submitted to U.S. Environmental Protection Agency, Office of Solid Waste, January
25.1994.
28. Reynolds Metals Company, Report of Four Dav Verification of Spent Potliner from
the Aluminum Company of America Wenatchee Works. Wenatchee. Washington-
March. 1994. Report Submitted to U.S. Environmental Protection Agency, Office of
Solid Waste, April 7,1994.
29. Reynolds Metals Company, Report of Four Dav Verification of Spent Potltner from
1994. Report Submitted to U.S. Environmental Protection Agency, Office of Solid
Waste, April 7.1994.
30. Reynolds Metals Company, Report of Four Dav Verification of Spent PotHner from
1994, Report Submitted to U.S. Environmental Protection Agency, Office of Solid
Waste, April 19,1994.
•
•
31. Reynolds Metals Company. Report of Four Dav Verification of Spent Pot liner from
1994, Report Submitted to U.S. Environmental Protection Agency, Office of Solid
Waste. April 19. 1994.
32. U.S. Department of the Interior. Bureau of Mines, 1992 Minerals Yearbook: Volume •
1. Metals and Minerals.
33.
U.S. Department of the Interior, Bureau of Mines, 1994 Mineral Commodity
Summaries*
34. Analysis of Proposed Regulated Constituents for K088, Working Paper under Quick
Response Task 2, Work Assignment 116, Radian Corporation and SAJC, August 19,
1994.
5-3
-------
35. Beach-scale Tests on Revse of Spent Potliniog in Cathodes. Paper in Light Metals.
1992. ed. by E.R. CutshalL
36. The Importance of Pot Shell Design on the Life of Electrolytic Cells, G. Concas. F.
Gregu. and G. Soletta, Light Metals. 1989.
37. Fluorides as Environmental Contaminants J.R. Bodnar, Institute of Environmental
Sciences and Engineering, University of Toronto, November 1972
38. "Fluorine" from Veterinary Toxicology, R. D. Radelef& Lea and Fe^iger, Phila. 196*4
pp 145-149
39. Summary of Articles on the Health and Environmental Effects of Fluoride, October
1994.
40. Fax transmittal from D.D. Macauley, Reynolds Metals Company to Mary Cunningham,
US. EPA, regarding Updated Information. December 22, 1994.
41. Fax transmittal from Bob Sims, ESI to Mary Cunningham. U.S. EPA, regarding
Revisions to Draft Tables. December 22, 1994. .
41 Correspondence from 'Serena Wiltshire, Freedman. Levy, Kroll A Simonds to Mary
Cunningham. U.S. EPA, regarding the COMTOR Process. December 1, 1994.
43. Correspondence from EJL Bob, ORMET to Mary Cunningham, U.S. EPA, regarding
Proposed Rule for K088. November 30, 1994.
44. Fax transmittal from Alan Smith, Ausmelt Technology Corporation to Mary
Cunningham, U.S. EPA, regarding K088 Treatment Rules. November 29, 1994.
•
45. Fax transmittal from Michel Lalonde, Akaa Smelters and Chemicals Ltd. to Mary
Cunningham, U.S. EPA, regarding Treatment Technologies for K088. November 25.
1994.
46. Memorandum to Mary Cunningham, EPA/OSW, from Richard Weisman, Radian,
-Development of BDAT Number for Cyanide in K088," November 21. 1994.
47. BaniMZi1 Environmental Inc. Company Brochure on Terra-Vit Process and Process
.•low Diagram
48. Test Results: Demonstration Test Melt for Recycling Aluminum Spent Potliner.
Barnard Environmental Corporation. Volume 2, July 1995
5-4
-------
I
49. Request .for Determination of Legitimate Recycling of Aluminum Spent Potiiner.
_ Barnard Environmental, Inc. Volume 3. July 1995
50. Record of Decision for Martin Marietta Corporation. The Dalles, Oregon, ROD Date
• 09/29/88. NTIS Report Number EFA/RODft 10-88/017
51. CERCLIS Data for Alcoa, Vancouver. WA from EPA's National Priorities List Site
• List. CERCLIS Database
52. CERCLIS Data for Kaiser Aluminum Mead Works, Mead, WA from EPA's National
• Priorities List Site List CERCLIS Database
53. CERCLIS Data for National Soutmvire Aluminum Company, Hawesville, K.Y from
• EPA's National Priorities List Site List. CERCLIS Database
54. CERCLIS Data for Ormet Corporation. Hannibal, OH from EPA's National Priorities
• List Site List. CERCLIS Database
I
I
I
I
I
I '
I
I
I
I
I
-------
6.0 ACKNOWLEDGEMENTS
Radian Corporation and Science Applications International Corporation (SA1Q
provided technical support for the development of this document for the U.S. Environmental
Protection Agency. Office of Solid Waste under Contract Number 68-W3-0001. This
document was prepared under the direction of Michael Petrusca, Chief, Waste Treatment
Branch; Larry Rosengrant. Section Chiet Treatment Technology Section; David Levy, Project
Officer. Mary Cunningham served as the Project Manager. Steve Silverman served as EPA
legal advisor.
The following personnel from SAIC supported the development of this document:
Mary Wolfe. Project Director, and the SAIC technical soft Henry Huppert Scott Cullen.
Paul White, and Margo Jackisch.
The following personnel from Radian Corporation supported the development of this
document: Gayle Kline, Program Manager; Richard Weisman, Project Director.
6-1
-------
Appendix A
Treatment Perfbn
•Database tnd
Methodology for Identifying Universal Treatment Standards
for Constituent* in Nonwastewater Forms of
KOSS
A-l
-------
This appendix presents the development of the universal treatment standards for the
constituents selected for regulation in nonwastewater forms of K088. Section A.I presents the
methodology for determining nonwastewater universal standards and introduces the universal
standards database. Section A.2 presents a constituent-by-constituent discussion of the
determination of the universal standards for each constituent selected for regulation.
•
A. I Methodology for Detennlninf BOAT Untvenal Treatment Standard^
The performance data presented in Appendix A represent the universal standards
database for the constituents regulated in K088. These data consist of the treatment
performance data used to develop nonwastewater treatment standards in the First. Second, and
Third Thirds and Phase I Land Disposal Restrictions Program rulemaking efforts. In order to
determine the universal treatment standard, the Agency examined the treatment performance
data used in calculating each treatment standard applicable to a specific constituent
. i
The Agency chose which treatment standards to transfer as the universal standard on a
constituent-by constituent basis. Six factors were considered in selecting the "best" standard
from the available treatment standard perfonuaace da's:
(1) Where possible, the Agency preferred performance data (i.;.. the matrix spike
recovery data, detection limit, and variability factor for the same constituent.
(2) The matrix S{.ue recovery data were evaluated to determine whether acceptable
recoveries were obtained according to EPA's quality assurance/quality control
guide lii
(3) When performance data from the same constituent were unavailable, the
Agency used performance data from a constituent with similar composition and
functional groups.
(4) When evaluating the matrix spike recovery data, the Agency preferred to use a
matrix spike recovery for specific constituent instead of t value averaged over a
group of constituents (e.g., volatile organics).
A-2
-------
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
(5) The method detection limit was examined to determine if it could be met
routinely by industry.
(6) The treatment standard corresponding to the "best" data was compared to the
detection limits used to calculate other treatment standards to determine if the
constituent could be treated to similar levels in similar waste codes.
Determination of Treatment Sttndirdi for Nenwastewtter Forms of K088
Treatment standard data for the constituents regulated in nonwastewater forms of K088
ore presented in Table A- 1. A constituent-by-constituent discussion of the determination of
(he universal treatment standard for each of these constituents is given below. A more
detailed discussion of the determination of the universal treatment standards is provided in
EPA's Final Beat Demorts|r»ted Available Technology (BOAT) Background Document fo^
Universal Standards. Volume A! Universal Statvfa*ds for Nonwastewater Forma of Listed
Haraitlmn Wastes (14). ,
•
Attaaphthent
The universal standard for acenaphthene was determined to be 3.4 mg/kg, based upon
the K035 treatment standard data. The Agency chose to use the K035 treatment standard data
rather than the F039 rrrafmrnt standard data because the F039 treatment standard was
promulgated incorrectly as 4.0 mg/kg instead cf 0.8 mg/kg. The Agency believes that a
standard of 0.8 mg/kg may not be reasonably achieved based on detection limits reported for
in other incineration
AatfcnctM
The universal standard for anthracene was determined to be 3.4 mg/kg, based upon the
K015 and K035 treatment standard data. Tne universal standard for inthraccne was not based
upon the F039 treatment standard cUta because the F039 standard was promulgated incorrectly
A-3
-------
as 4.0 mg/kg instead of 0.8 mg/kg. The Agency believes that a standard of 0.8 mg/kg may
not be reasonably achieved based upon detection limits reported foir anthracene from other
incineration tests.
Benz(a)antbracent
The treatment standard for benz(a)anthncene was determined to be 3.4 mg/kg, based
upon the K035 treatment standard. The Agency chose to use the K035 treatment standard
data since these data represent the use of an accuracy correction factor and detection limit
from the same constituent as the constituent of concern. The Agency believes that a treatment
standard of 3.4 mg/kg may be reasonably achieved based on detection limits reported for
benz(a)anihracene in other waste codes.
Beazo(a)pyre0e
The treatment standard for benzo(a)pyrene was determined to be 3.4 mg/kg. based
upon the K035 and K.060 treatment standards. The Agency chose to use the K03S and K060
treatment standard data since these data represent the use of both an accuracy correction (actor
and detection limit from the same constituent as the constituent of concern. The Agency
believes that a treatment standard of 3.4 mg/kg may be reasonably achieved based on the
detection limits reported for benzo(a)pyrent in other waste codes.
*
Beozo(b)flaorutheae aad B*azo(k)fiBoriatheae
The treatment standard for the sum of benzo(b)fluoranthene and benzo(k,fluoranthene
was determined to be 6.S mg/kg. based upon the sum of the F039 treatment standards for
benzo(b)Quocanthene and benzo(k)fluora&theae. As explained in Section 3.2.3.1, these
• »
constituents are regulated as § sum to account for analytical problems In distinguishing
between the two compounds In noowastewater tnatrirfi
A-4
-------
B«nzo(g»n4)perylene
The universal standard for benzo(gju)perylene was determined to be 1.8 mg/kg, based
upon the F039 treatment standard, which represents the only concentration-based treatment
standard the Agency has promulgated to date for this constituent. The F039 treatment
standard was calculated from the incorrect accuracy correction factor. The treatment
performance dati transferred for the universal standard are correct
Cbryscne
The universal standard for chrysene was determined to be 3.4 mg/kg, based upon the
K087 and K035 treatment standard data. The Agency chose to use the K087 and K035
treatment standard data since these data represent the use of both an accuracy correction factor
and detection limit from the same constituent as the constituent of concern. The Agency
believes that a universal standard of 3.4 mg/kg may be reasonably achieved based upon
detection limits reported for chrysene in other waste codes.
I
I
I
I
I
DibeBz(a4i)aathrace0e
The treatment standard for.dibenz(a^)anthncene was determined to be 8.2 mg/kg,
based upon the F039 and U063 treatment standards. The Agency chose to use the F039 and
U063 treatment standard data since these data leptexut the use of an actual matrix spike
recovery. TK» Agency believes that a treatment standard of 8.2 mg/kg may be reasonably
achieved based on detection limits reported for dJbenz(aJi)anthraccne In other waste codes.
The universal standard for fluonnmeat was determined to be 3.4 mg/kg, based upon
the K035 and K087 treatment standard dan. The Agency chose to use these data because
A-5
-------
I
they represent the use o£ both the accuracy correction (actor and detection limit from the same
constituent as the constituent of concern.
The treatment standard for indeno(l,2J-cd)pyrenc was determined to be 3.4 rag/kg.
based upon the K035 and K087 treatment standards. The Agency chose to use the K035 and
K087 treatment standard data since these data represent the use of both on accuracy correction
factor and detection limit from the same constituent.
Phtaanthrtnt
The universal standard for phenanthrene was determined to be 5.6 mg/kg, based upon
the K019 treatment standard data. The Agency chose to use these data because they represent
the use of an accuracy correction factor and detection limit from the same constituent as the
constituent of concern. The Agency chose a universal standard of 5.6 mg/kg to account for
regulatory flexibility based on variations in treatment of this constituent
Pyrene
The universal standard for pyrene was determined to be 8.2 mg/kg, based upon the
K035 and F039 treatment standard data. The Agency chose to use these data because they
•
represent the us* of both aa accuracy correction factor and detection limit from the same
• •
constituent as the constituent of concern. The Agency believes that transfer of data from
KOOl and U051, with a treatment standard of 1J mg/kg, is not reasonable for a universal
standard based on detection limits from other incineration tests.
A-6
-------
AJ Identification of Universal Standards for Cyanide
The Universal Treatment Standard for cyanide was determined to be 590 mg/kg total
cyanide and 30 mg/kg amenable cyanide. These standards are based on the most difficult
cyanide wastes to treat including F006, F019, FOl 1. F012 and metal-cyanide P wastes. The
treatment standard for cyanide is based on incineration, alkaline chlorination and electrolytic
oxidation followed by alkaline chlorination. This standards also includes specification of a 10
grams sample size and a 75 minute distillation time during the analysis for the purposes of
complying with the universal standards.
A.4 Identification of Universal Stindirdi for Fluoridt
The treatment standards for fluoride in nonwastewaten is a teachable concentration
which was determined by the Agency when granting a delisting for certain K088 wastes. The
Agency granted a delisting for kiln residues generated from the treatment of spent potlinen at
the Reynolds Metals facility in Gum Springs, Arkansas. This exclusion set conditional testing
limits on the kiln residue for fluoride at a teachable concentration of 48 mg/L. Based on the
information collected from K08S treatment processes, this concentration is achievable by at
least to treatment technologies (i.e^ Reynolds and Onset). Thus the Agency, does not believe
that this concentration is technology forcing.
Identification of Untvmal Standirdi for Metal Conittromti
The Agency determined universal standards for metal constituents using the following
methodology:
(I) The Agency selected metal constituents for regulation as presented in Section }
4.0.
(2) For each metal constituent selected, the Agency listed BOAT treatment
performance data according to waste code; data included the concentration in
A-7
-------
.j-Htfima-f
the TCLP leachate or the detectioo limit of the constituent in the treated waste.
the accuracy correction (actor used (and its basis), and the variability factor.
(3) The Agency removed data corresponding to waste codes excluded from
consideration in universal standards;
(4) The Agency evaluated the data on a constituent by constituent basis to
determine the data most appropriate to establish a universal standard.
These steps are described in more detail below.
The development of universal standards for metal constituents began with the selection
of metal constituents proposed for regulation from the BOAT List of hazardous constituents.
t
Universal standards for metal constituents were determined utilizing treatment
performance data that had been used to develop noowastewater treatment standards in the
First Second, and Third Thirds and Phase I mlmnlring efforts.
To determine a universal standard for a particular constituent, it was necessary to
examine the data used in calculating each promulgated treatment standard for that constituent.
The data used to compute the treatment standard include the concentration of the constituent
in the treated waste, an accuracy correction factor, and a variability factor. When treatment
performance data were not available for treatment of a specific waste code, data were
•
transferred from treatment of a similar waste. These proposed universal standards were
chosen on a constitueni-by-constitucnt basis. Five factors were considered in selecting a
treatment standard value:
(1) Where possible, the Agency preferred to use treatment performance data from
the technology believed to be "best* for treatment of metal constituents in
universal standards wastes, KTMR.
(2) Where possible, the Agency preferred to use treatment performance data (i.e.,
the, concentration in the TCLP extract of the treated waste, matrix spike
recovery data, and variability factor for the constituent of concern.
A-S
-------
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
(3)
The Agency evaluated the matrix spike recovery data to determine whether the
recoveries were within the acceptable range of values as identified in EPA's
Generic Quality Assurance Project Plan for Land Disposal Restrictions Proaran]
(REF).
(4)
(5)
The Agency examined the concentration in the TCLP extract of the treated
waste to determine if it could be routinely met by industry.
The Agency compared the treatment standard corresponding to the "best" data
to the concentration in the TCLP extracts of the treated waste obtained for
other waste codes to determine if the constituent could be treated to similar
levels in similar waste codes.
The Agency preferred to use data from the performance of HTMR processes to
develop the proposed universal standards. Since metals cannot be destroyed, treatment
options are limited, and typically include technologies that can either recover the metal(s) or
incorporate the metals in t stable matrix resistant to leaching, The Agency, believes that the
"best" treatment for metal constituents is recovery, especially in cases of high waste metal
concentrations. HTMR appears to be the most matrix-independent of the applicable
technologies (i.e., it consistently achieves the same levels of treatment performance regardless
of Influent matrix compositions). HTMR also generally decreases the amount of material sect
for land disposal, and incorporates metals that are not recoverable into a stable slag matrix
The use of HTMR is consistent with the national policy, identified in the Hazardous
and Solid Waste Amendments (HSWA) to RCRA. to reduce the quantity of hazardous
constituents disposed (this is in contrast to non-recovery technologies, such as stabilization,
which are not intended to reduce the total metal concentration or metal volume in the waste .
and in fact, can increase volumes being sent to landfills). In addition, because metals are
recovered instead of land disposed, ore processing is reduced thus saving energy and pollution
of another source.
A-9
-------
The Agency reviewed other non-HTMR performance date (i.e.. stabilization) and
determined that the proposed universal standards for most metals could be achieved by
stabilization tor a wide variety of nonwastewater matrices.
A constituent-by*constituent discussion of the determination of the universal standards
is given below.
Antimony
The universal standard for antimony was determined to be 2.1 mg/L in the TCLP
extract based upon the K061-HTMR treatment standard data. The Agency chose to use these
data because they represent the treatment performance of a HTMR process. The universal
standard for antimony was not based upon the K021 and F039 treatment standard data
because these data represent the performance of incineration, which is not considered a
demonstrated technology for metal constituents in nonwastewater forms of universal standard
wastes.
Arsenic
The universal standard for arsenic was determined to be 5.0 mg/^ in the TCLP extract
based upon the F039 treatment standard. The F039 treatment standard was established as
equivalent to the toxiciry characteristic (TC) regulatory level for arsenic (D004).
The Agency established BOAT for arsenic as slag vitrification. The universal standard
was not based upon K061-HTMR data because the Agency believes that this technology is not
"best" for treatment of arsenic in universal standards wastes. The available slag vitrification
treatment standard dau (K031, K084. K101. K102, P010. P011, P036, P038. and U136) show
treatment to a leachate concentration of I.I mg/L (Using the Extraction Procedure (EP)
toxicity test). The universal standard based on* this value would yield a standard of S.6 mg/L
using the EP toxicity test Because the characteristic level for arsenic of S.O mg/L In the
A-10
-------
I
I
TCLP extract is limilar in magnitude to the standard calculated from slag vitrification, the
Agency believes that it is valid to default to the characteristic level for the universal standard
for arsenic.
Barium
The universal standard for barium was determined to be 7.6 rag/L in the TCLP extract
based upon the K061-HTMR treatment standard data. The Agency chose to use these data
because they represent the treatment performance of a HTMR process. The Agency believes
that an universal standard based upon K061-HTMR treatment standard data could be routinely
met by industry using HTMR because the applicability of the HTMR process is matrix*
independent (i.e.. the technology consistently achieves the same levels of treatment
performance regardless of influent matrix compositions). Additionally, the Agency reviewed
stabilization data and determined that the proposed universal standard for barium could be
achieved by stabilization for t wide variety of waste matrices. The Agency, therefore, does
not believe that the universal standard would be technology forcing.
Beryllium
The universal standard for beryllium was determined to be 0.014 mg/L in the TCLP
extract based upon the K061-HTMR treatment standard data. The Agency chose to use these
data because they represent the only concentration-baaed oonwutewater treatment standards
the Agency has promulgated to date for this constituent. Additionally, these data represent the
treatment performance of t HTMR process.
Cadmiaa
The universal standard for cadmium was determined to be 0.19 mg/L la the TCLP
extract based upon the K061-HTMR treatment standard data. The Agency chose to use these
data because they represent the treatment performance of a HTMR process. The Agency
A-ll
-------
believes that a Universal standard based upon K06I-HTMR treatment standard data could be
routinely met by industry because the applicability of the HTMR process is matrix*
independent (i.e.. the technology consistently achieves the same levels of treatment
performance regardless of influent matrix compositions). Additionally, the Agency reviewed
stabilization treatment standard data for cadmium and determined that the proposed universal
standard could be achieved by stabilization for a wide variety of waste matrices.
Lead
The universal standard for lead was determined to be 0.37 mg/L in the TCLP extract
based upon the K061-HTMR treatment standard data. The Agency chose to use these data
because they represent the treatment performance of a HTMR Process. The Agency believes
*
that an universal standard based upon K061-HTMR treatment standard data could be routinely
met by industry because the applicability of the HTMR process is matrix-independent (i.e., the
technology consistently achieves the same levels of treatment performance regardless of
influent matrix compositions). Additionally, the Agency reviewed the stabilization treatment
standard data for lead and determined that the proposed universal standard could be achieved
by stabilization for a wide variety of waste matrices.
Mercnry
The Agency is ruablithing two universal standards for mercury, 0.20 mg/L in the
TCLP extract for low-mercury subcategory RMERC residues and 0.025 mg/L in the TCLP
extract for low-mercury subcategory non-RMERC residues. Low-mercury subcategory wastes
are mercury wastes containing concentrations of mercury less than 260 mg/kg. RMERC is the
recovery of mercury by roasting/retorting.
The universal standard for mercury in low-mercury subcategory RMERC residues was
determined to be 0.20 mg/L in the TCLP extract This determination was based upon the
A-12
-------
I
I
I
_
m
I
I
I
I
I
I
K106. U15I. P065. and P092 treatment standards for low-mercury subcategory RMERC
residues. which were established as equivalent to the TC regulatory level for mercury (0009).
The universal standard for mercury in low-mercury subcategory non-RMERC residues
was determined to be 0.025 mg/L in the TCLP extract This determination was based i-poa
the K071. F039, K106. and U151 treatment standard data for low-mercury subcatcgory non-
RMERC residues. The Agency chose to use these data because they represent the treatment
performance of the technology selected as BOAT for mercury in low-mercury subcategory
wastes, acid leaching.
Nlcktl
• The universal standard for nickel was determined to be 5.0 mg/L in the TCLP extract
* based upon the K061-HTMR treatment standard data. The Agency chose to use these data
• because they represent the treatment performance of a HTMR process. The Agency believes
that an universal standard based upon K061-HTMR treatment standard data could be routinely
• met by industry because the applicability of the HTMR process is matrix-independent (Le^ the
technology consistently achieves the same levels of treatment performance regardless of
• influent matrix compositions). Additionally, the Agency reviewed stabilization treatment
standard data for nickel and determined that the proposed universal standard could be
• achieved by stabilization for a wide variety of wastt matrices.
I
Scleateai
The universal standard for selenium was determined to be 0.16 mg/L in the TCLP
extract based upon the K061-HTMR treatment standard data. The Agency chose to use these
data because they represent the treatment performance of a HTMR process. The Agency
believes that in universal standard based upon K06l>HTMR treatment standard data could be
routinely met by industry because jhc applicability of the HTMR process is matrix-
A-13
-------
independent (i.e.. the technology consistently achieves the same levels of treatment
performance regardless of influent matrix compositions). Additionally, the Agency reviewed
stabilization data and determined that the proposed universal standard for selenium could be
achieved by stabilization for a wide variety of waste matrices. The Agency, therefore, does
not believe that the universal standard would be technology forcing.
Silver
The universal standard for silver was determined to be 0.30 mg/L in the TCLP extract
based upon the K061-HTMR treatment standard data. The Agency chose to use these data
because they represent the treatment performance of .a HTMR process. The Agency believes
that an universal standard based upon K061-HTMR treatment standard data could be routinely
met by industry because the applicability of the HTMR process is matrix-independent (Le., the
technology consistently achieves the same levels of treatment performance regardless of
influent matrix compositions). Additionally, the Agency reviewed stabilization treatment
•
standard data for silver and determined that the proposed universal standard could be achieved
by stabilization for a wide variety of waste matric
Calculation of Universal Standard for Chromium
*
The Agency developed a universal standard for chromium based on the stabilization
treatment performance data. EPA evaluated waste characterization and treatment performance
data for chromium from several sources, including data on the performance of HTMR and
stabilization technologies for treating chromium, EPA selected the'stabilization data to
develop the universal standard for chromium because these data represent treatment of
chromium in difficult to treat wastes, including stripping liquids, plating and pelletizing
• .
operation wastes, and cleanout wastes from plating tanks. The Agency believes that these
data represent effluent values that can be routinely achieved by industry. The universal
standard for chromium was determined to be 0.86 mg/L in the TCLP extract based upon the
treatment standard developed from the stabilization treatment performance data.
#
A-14
-------
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
!
8
fa
B
8
1
3
9 9
tt
6
j
r
14
r\
§
fat
s
r
I
*•*
s
8
ti
(4
8
1
§
5
8
fc
I
I
ii
r
Jf
ih
I«
ai
s
-------
•
Ch
Jl 11
vr Ir
ti
•. <
r
*? I
li
2,8.
M
S f
*
I *•
'f
1
r
s
L
t
I
r^
§
8
I). KUS.
KOC7
8
I
3
8
f
\
8
s
8
»-^
§
3
O
8
I
^
I
*
f
I
4<* ^ «tf^ *« , «M • •* I** I*.
fi
?
=.?
r.
I
7
-------
I
I
I
I
I
I
I
I
I >
I
I
I
I
I
I
I
I
I
f
3
is
j
i
;
I
r
*
I
r
!
.1
-------
&>
)
a,
n
H
*.
S'j
F
r
#ft
1
-H
w • **
f
r
r
--
ii
JI p 5*
(fig
I .I'i
r
i
\\
a,
9
ifr
!?
IT
!
\
a.
m
m
ilir
ii.»i
si i
if
I?
a.
B
fr
a,
f
E
E
B'
El
I
8"
a,
r
t=
E
E
B
f
B
-------
I
I
<
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
;r
'l?
1
8
f
fl»
i
I
f
r
1
1
B
1
p«.
!
•• '
r;
5
r
5
S
P
§
f
a,
f
e
I:
**
J
'I
I
r
f
s
§
•
?
f
}w
>
I
i
c
3t
H
.
Lin
I
-------
- ~- \
i
II
8P
a
x
S
It
3
n
to U
3
- o
£ £* P.- .—
i
V
i
8
i
3
&
I
l??|
A
1
i
• r
!
^t
>
-------
I
I
I
I
I
I
I
I
I
I*
I
I
I
I
I
I
I
I
r
SS33358
*v<*~QSo&
." .-J" .*.-£.-
*
8
i
i
r
I
K
8
i
U03
e
*
55
ill
p
tr
i
i
n
8
-------
g
u
^
a
A
iff!
\W
I
I
1
E
1?
-------
I
I
»..-• • 'i-
i-
"•3
e
2
*
K06
f
L*
F024.K02I.
KOt
e
fe
5?
1
33
X.
5 2 3
i
i!
<
3
i'
-------
-------
-------
-------
-------
*)pii i«Bft i*'
1
til
I?
q
iU
mm
M **f
*Bi*|
I
B
fr
a,
f
E
8
r
i
» •
liil
tl't
jgf It
S v»
>
r
5
s
f
r
Ml
I! i
if
r
a,
r
?
------- | |