Response to Comments Document: Land Disposal Restrictions Phase IV Final Rule Promulgating Treatment Standards for Metal Wastes; Mineral Processing Secondary Materials and Bevill Exclusion Issues; Treatment Standards for Hazardous Soils; and Exclusion of Recycled Wood Preserving Wastewaters; Volume 11: Comments Related to Newly Identified Mineral Processing Waste Treatment Standards, Grab vs. Composite Sampling, Radioactive Mixed TC-Metal Wastes, and Sulfide Waste Issues Printed


United States      Solid Waste and      EPA530-R-99-020k
Environmental Protection   Emergency Response     NTIS: PB99-155 921
Agency         (5305W)	April 1998	
Response to Comments
Document:  Land Disposal
Restrictions-Phase IV Final
Rule Promulgating Treatment
Standards for Metal Wastes;
Mineral Processing Secondary
Materials and Bevill Exclusion
Issues; Treatment Standards
for Hazardous Soils; and
Exclusion of Recycled Wood
Preserving Wastewaters;
Volume 11: Comments Related to
Newly Identified Mineral Processing
Waste Treatment Standards, Grab vs.
Composite Sampling, Radioactive Mixed
TC-Metal Wastes, and Sulfide Waste
Issues
      Printed on paper that contains at least 30 percent postconsumer fiber

-------
REPORT DOCUMENTATION | 1. Report No. | 2.
PAGE | |
| EPA530-R-99-020k |
1 1
4. Title and Subtitle
Response to Comments Document: Land Disposal Restrictions-Phase IV: Final Rule Promulgating Treatment
Standards for Metal Wastes and Mineral Processing Wastes; Mineral Processing Secondary Materials and Bevill
Exclusion Issues; Treatment Standards for Hazardous Soils, and Exclusion of Recycled Wood Preserving Wastes;
Volume 1 1 : Comments Related to Newly Identified Mineral Processing Waste Treatment Standards, Grab Versus
Composite Sampling, Radioactive Mixed TC-Metal Wastes, and Sulfide Waste Issues
7. Authors)
9. Performing Organization Name and Address
| 3. Recipient's Accession No.
I
| PB99-155921
I
| 5. Report Date
I ApriM998
1 6.
I
I
I
| 8. Performing Organization Rept No.
I
I
I
| 10. Project/Task/Work Unit No.
1
U.S. EPA |11. Contract © or Grant (G) No.
OFFICE OF SOLID WASTE |©
401 M STREET, SW |
WASHINGTON, DC 20460 (G)

I
12. Sponsoring Organization Name and Address 13. Type of Report & Period Covered

Response to Public Comment
L
| 14.
I
J
15. Supplementary Notes
16. Abstract (Limit: 200 words)

Responds to public comments relevant to appropriate sampling methods, treatment standards for newly identified mineral processing wastes,
radioactive waste mixed with toxicity characteristic metal wastes, sulfide waste issues.
17. Document Analysis a. Descriptors
b. Identifiers/Open-Ended Terms

-------
                   List of Phase IV Land Disposal Restriction Codes,
                     Description, and Location by Page Number

GRAB:  Grab Versus Composite Sampling  	1

MSTD:  Treatment Standards for Newly Identified Mineral Processing Wastes	59

RADl:  Radioactive Mixed TC Waste Issues	69

SULF:  Sulfide Waste Issues	107

-------
              Index of Commenters and Location of Comment, By Issue

 American Iron and Steel Institute (AISI)
      GRAB	40
 Department of Energy
      RADI	71, 79, 81
 Eastman Chemical Company
      MSTD	"	67
 Environmental Technology Council
      GRAB •	57
 FMC Corporation
      MSTD	67
 Horsehead Resource Development Company, Inc
      GRAB	51
 International Metals Reclamation Company, Inc. (INMETCO) and
      GRAB	"	30
 U.S. Department of Energy (DOE)
      RADI	101
 United States Department of Defense (DoD)
      GRAB	56
 INMETCO
      GRAB  	22
ASTSWMO
      RADI .'	99
Battery Council International (BCI) and Association of Battery R
      GRAB	32
CF Industries, Inc.
      GRAB	37
Chemical Manufacturers Association
      GRAB	28
Chemical Waste Management
      GRAB	1
Coalition on West Valley
      RADI	69
Cyprus Amax Minerals Company
      GRAB	42, 44
Department of Energy
      RADI	70, 76
FMC Corporation
      SULF	-.	-. ... 112
Hazardous Waste Management
      GRAB	27
Horsehead/Zinc Corp.

-------
      GRAB	4. 10. 13. 14
      SULF	107
INMETCO
      GRAB	15, 19, 24
Kennecott
      GRAB	46
Macalloy Corporation
      GRAB	55
Mineral Policy Center
      SULF	Ill
Mining Impact Coalition of Wisconsin Inc.
      SULF	113
Molten Metal Technology
      RADI	82
National Mining Association
      GRAB	44
National Mining Association
      MSTD	59
New York DEC
      RADI	70
New York State Department of Environmental Conservation
      MSTD	65
Nexsen, Pruet, Jacobs & Pollard, LLP (Counsel to Macalloy
      GRAB	49
The Ferroalloys Association (TFA)
      GRAB	32
The Fertilizer Institute
      GRAB	39
Westinghouse
      RADI	 85-87
Westinghouse Electric Corporation
      RADI	101
                                        in

-------
DCN   PH4P048
COMMENTER Chemical Waste Management
RESPONDER AC
SUBJECT  GRAB
SUBJNUM  048

COMMENT

With regard to the grab vs.  composite sampling issue CWM believes
that the Agency needs to reevaluate its position that wastewaters
require composite and nonwastewaters require grab samples.
Especially in light of the Agency's preamble discussion (See 60
Fed. Reg. at 43,683) which states:

"While UTS nonwastewater limits for metals specify a grab sample,
the data used to develop standards  included both grab and composite
samples."

The fact that UTS levels were not all  established using grab
indicates that the Agency recognizes that composite samples for
nonwastewaters can be  appropriate. CWM commented on the use of grab
versus composite samples during the Third Third rulemaking. CWM's
position on this issue has not changed and is presented below.

CWM believes that in contrast to grab sampling, composite sampling
is useful when the cost of analyzing multiple grab samples is
prohibitive if the waste is reasonably homogeneous. The composite
sample will represent the average concentration of the potential
grab samples. It is important to note that the total amount of a
constituent present in the composite sample is exult or greater
than any single potential grab sample making up the composite;
hence, the constituent may be more easily detected.

From a statistical perspective, when the number of potential
observations is limited,  composite sampling is clearly the method
of choice for providing  a "representative" index of a
constituent level. It can also be easily shown that when compared
to a treatment standard, a single composite sample will minimize
both false positive and false negative  rates relative to a single
sample.

To illustrate this point,  CWM conducted the following simulation
study during development of comments for the Third Third. First,

-------
ten background samples were drawn from a 109 normal distribution
that was generated by taking the antilogy of data drawn from a
normal distribution with mean zero and variance one. Second, a
treatment standard was then generated using the EPA method, i.e.,
EXP [ Y + 2.33 Sy], where Y is the exact mean of the natural log
transformed  values and Sy is the corresponding standard deviation.
Third, from the same distribution, a new"grab sample" was selected
and if this value now exceeded the limit a failure was
recorded.  Fourth, from the same distribution, four new samples were
selected and then arranged to form a single composite sample, and
if this composite sample exceeded  the first, a failure was
recorded.  Fifth, steps 1-4 were then repeated 10,000 times, and the
percent of failures for each of the two methods represent their
false positive rates. Inspection of figure 1 reveals that for a
difference  of zero SD units, the composite sample attains the
intended 1% false positive rates, but for the grab sample approach,
the false positive rate is on the order of 4% (i.e., 96  out of 100
measurements incompliance due to change alone).  Sixth, steps 1-5
were then  repeated for differences  of .2, .4, .6,..., 4.8, 5.0 SD
units between the data used to construct the treatment standard
and the simulated waste stream test data.  In this way, we can
compare grab samples and composite samples in terms of false
negative rate as well.  Inspection of figure 2 reveals  that, in
order to achieve a 5% false negative rate, the mean construct
level must be 4.2 SD units above the mean of the data used to
construct the treatment standard if we use grab samples, but only
3.4 SD units if we use a composite sample.  Recall that in all of
these cases, grab samples were used to construct the treatment
standard; therefore, this simulation directly reflects the actual
case that exists in the BOAT treatment standard problem.

The results of the simulation  reveals that by using composite
samples, not only do we achieve the intended false positive rate
(i.e., 99 out of 100 measurements are now in compliance due
to change  alone), but we also produce a test with a substantially
lower false negative rate.  Clearly,  the use of composite samples
is justified here, and if anything, it  should be required.
There is a reasonably straightforward explanation for these
results. The variability of an individual grab sample is o2;
however, the variability of a composite at 4 samples, has
variability of o2 or a standard deviation of 02.  This result
indicates that the composite samples will have one half of the
variability of the individual grab sample. By removing both high

-------
and low outlying values, the false positive rate is clearly
minimized, and a far more consistent result is obtained
when contamination is actually present; therefore, the false
negative rate is also minimized. If decisions are to be made based
on the result of a single analysis, composite sampling is clearly
the method of choice, regardless of whether the treatment standards
were derived from grab or composite samples.

As a final note, it is interesting (but not surprising) that when
composite samples are used, the elevated false positive rate of the
EPA treatment standard disappears. The reason for this again stems
from the difference between the distribution of an individual
value and the mean at four values. To the extent that a waste
stream is reasonably well mixed, the composite sample reflects the
average orate sample value.

Since the distribution of average value is narrower than that of
individual values, the same limit value (i.e., treatment standard)
will provide increased coverage of the distribution. This is
the reason why 96 of 100 grab samples will be in compliance but 99
out of 100 composite samples will be in compliance for a well
designed and properly operated facility.

As such, CWM encourages the Agency to allow use of both sampling
methods for nonwastewaters, and that for enforcement purposes the
Agency utilize the same method employed by the generator or TSDF,
or totally shift its reliance from grab samples to composite
samples.

RESPONSE

The Agency's authority to establish treatment standards based on grab sampling has
already been upheld. Chemical Waste Management. 976 F. 2d at 34
B) The same opinion upheld the principle that it is inherently reasonable to require
compliance with treatment standards based upon the sampling methodology used to establish the
treatment standard. Id.
C) That is precisely what the Agency has done here: all treatment standards for
nonwastewaters are based exclusively upon data gathered through grab sampling.
D) There is no contradiction with wastewater treatment standards using composite
sampling to evaluate compliance. These standards were developed using composite sampling data
exclusively.

EPA thus does not view the issue of whether standards should be based on grab sampling

-------
to be presented here. As just noted, that is a settled issue. The only issue in this proceeding is
whether the compliance regime is based upon the sampling method used to establish the treatment
standard, and the answer to that question is "yes." The Agency has developed data for
nonwastewater treatment standards which is based exclusively upon grab sampling.
DCN PH4P067
COMMENTER Horsehead/Zinc Corp.
RESPONDER AC
SUBJECT GRAB
SUBJNUM 067

COMMENT

Horsehead Resource Development Company, Inc. ("HRD") and Zinc
Corporation of America ("ZC A") jointly submit these comments in
response to the United States Environmental Protection Agency's
("EPA" or the "Agency") proposed "Phase IV" land disposal
restrictions rulemaking. 60 Fed. Reg. § 43654 (August 22, 1995)
("Phase IV Proposal"). The Phase IV Proposal includes three
critical errors that require correction in the final rule. First,
EPA mistakenly asserts that the treatment standards for metal
constituents specify grab rather than composite sampling for
compliance purposes.

PH4P067
K061/GRAB

1. First Third LDR Rulemaking (August 17, 1988)
On August 17, 1988, EPA established a treatment standard for "high-zinc" K061
(15 percent or greater zinc) nonwastewaters as "no land disposal," rather than
setting concentration-based toxicity characteristic leaching procedure ("TCLP")
limits, because the Agency found that: 1) high temperature metals recovery
("HTMR") is the Best Demonstrated Available Technology ("BOAT") for K061;
and 2) K061 is "indigenous" to HTMR devices such as those operated by FIRD. 53
Fed. Reg. § 31138 (Aug. 17, 1988). Although EPA collected composite samples
from HRD's facility in 1987 in support of this rulemaking, it did not rely on the
data at that time because of the Agency's decision to establish the "no land
disposal" treatment standard.

2. Revised Treatment Standards for High-Zinc K061 (August 19, 1991).
EPA developed and promulgated numerical treatment standards for zinc and 12
other BOAT list metals for high-zinc K061 in 1991. 56 Fed. Reg. § 41164 (Aug.

-------
19,  1991) (EPA reserved the treatment standard for vanadium.) EPA used data
from four HTMR processes to set the TCLP limits:  1) International Metals
Reclamation Company ("INMETCO"); 2) HRD (both 1987 and  1991 data); 3)
International Mill Service ("IMS"); and 4) SKF:
1. All of the INMETCO data (which were based on processing of K061, K062 and
F006 wastes) reflected composite sampling.
2. All of the HRD data (including the 1987 EPA-collected data and the 1991 data
submitted by HRD) reflected composite sampling.
3. The IMS data sets were a mix of composite and grab samples.
4. Only a single SKF data set was used and the nature of the sample is unknown.
Most significantly, HRD's and INMETCO's composite data were used to establish
the  majority of the treatment standards.

Consistent with the sources of the data, in 1991 EPA established compliance with
the  K061 treatment standards based on the maximum for  any single composite
sample. See 56 Fed. Reg. at 41169 (table of BOAT treatment standards for K061)
(attached to these comments as Exhibit 1). In fact, the use of composite sampling
to determine compliance with the treatment standards was essential to implement
EPA's stated intent that the final BOAT standards be "achievable by all of the
major HTMR technologies." 56 Fed. Reg. at 41169 (emphasis added). Since all of
the  data that EPA used from HRD and INMETCO reflected composite samples,
EPA would have had no basis for stating was that the treatment standards were
"achievable by  all the major HTMR technologies" if compliance had been
measured on the basis of grab samples. This is equally true now, especially since
the  HRD, INMETCO, and IMS technologies still account for the preponderance
of HTMR processing of K061.

Given this information, EPA had no alternative in 1991 but  to require compliance
with the BDAT treatment standards based on the maximum for any single
composite sample. EPA neglected, however, to include that condition when it
transferred (in the  same 1991 rulemaking) the revised K061 treatment standards to
the  table at 40  C.F.R. § 268.41, which covered hundreds  of other wastes. Since the
treatment standards for other nonwastewaters generally were based on grab
samples, this error created the false impression that the K061 treatment standards
were also based on grab sampling, even though they actually were based on
composite sampling.

3.  Phase I LDR Regulations (August 18, 1992).

In August 1992, EPA simplified the treatment standards for K061  by merging the
standards for high-zinc and low-zinc K061 into a single set  of standards reflecting
the August 1991 high-zinc TCLP concentration limits, which were based on
HTMR. 57 Fed. Reg. 37194 (Aug.  18, 1992). EPA did not consider any additional

-------
data sets; it relied entirely on the INMETCO\HKD\IMS\SKF data, which consisted
principally of composite samples, from the 1991 rulemaking.

As in the case of the August 1991 rulemaking, EPA's preamble discussion of the
merger of the standards clearly states that compliance with the TCLP limits was
based on the maximum for any single composite sample. See 57 Fed. Reg. at
37209 (table of final treatment standards for K061, which indicates that the
treatment standards are based on composite sampling) (attached to these
Comments as Exhibit 2). Despite the specific and unambiguous preamble  statement
that compliance with the K061 treatment standards was based on composite
sampling, EPA again administratively erred by not transferring the composite
sampling requirement to the consolidated treatment standards table at 40  C.F.R. §
268.41, thus leaving the mistaken impression that compliance with the treatment
standards for K061  could be based on grab sampling.

As in the case of the August 1991 rulemaking, EPA's preamble discussion of the
merger of the standards clearly states that compliance with the TCLP limits was
based on the maximum for any single composite sample. See 57 Fed. Reg. at
37209 (table of final treatment standards for K061, which indicates that the
treatment standards are based on composite sampling) (attached to these
Comments as Exhibit 2). Despite the specific and unambiguous preamble  statement
that compliance with the K061 treatment standards was based on composite
sampling, EPA again administratively erred by not transferring the composite
sampling requirement to the consolidated treatment standards table at 40  C.F.R. §
268.41, thus leaving the mistaken impression that compliance with the treatment
standards for K061  could be based on grab sampling.
As confirmation of EPA's actual historical intent to require composite samples, the
Agency established  the TCLP limits for the generic exclusion from the
derived-from rule for HTMR residues from K061, K062 and F006 in terms of
composite samples.  See 57 Fed. Reg. 37209 (Table) (see Exhibit 2). Since the
HTMR generic exclusion was not integrated into a single table with hundreds of
other wastes, its composite sampling requirement was correctly transferred to 40
C.F.R. § 261.3(c)(2)(ii)(C)(l), which refers to the maximum for any single
composite sample.
C. The UTS Levels for Metal Nonwastewaters are Based on Composite Sampling.
1. Proposed Phase II LDR Regulations (September 14, 1993).
In 1993, in the proposed Phase II LDR regulations, EPA proposed UTS limits for
metal non-wastewaters based on the existing K061 treatment standard, which, as
explained above, were based on HTMR. Accurately reflecting the K061 treatment
standard sampling methodology, EPA expressed the UTS metal limits as the
maximum TCLP concentration in any single composite sample. 58 Fed. Reg. §
48092, 48103 (table of proposed UTS metal standards) (attached to these
Comments at Exhibit 3). As with the previous rules, however, EPA neglected to

-------
incorporate the composite sampling requirement in the consolidated table of all
UTS limits in the proposed regulations. See id. at 48200.

HRD, INMETCO and the Business Recycling Coalition informed EPA of this
problem in the comment period in the proposed rule. HRD and tr 2 other
commenters pointed out that EPA had established composite sampling as the basis
for compliance with the K061 treatment standards in 1991, and requested that
EPA conform the sampling requirement to the data used to establish the standards.

2. Final Phase II LDR Regulation. (September 19, 1994).

Inexplicably, despite EPA's clear precedent of establishing treatment standards for
K061 based on composite sampling, and given that the UTS metal treatment
standards were based largely on composite samples, EPA established compliance
with the metal UTS limits (and for the individual waste codes for the  regulated
limits) based on the maximum for any single grab sample. 59 Fed. Reg. 47980,
47999. EPA took this action even though, with three exceptions not relevant here,
the metals treatment standards are based on HTMR of K061, for which EPA had
already required compliance based on composite sampling, and for which nearly all
of the data were generated from composite samples.

Significantly, and contrary to EPA's stated policy that treatment standards must be
"achievable by all of the major HTMR technologies," EPA did not demonstrate
(and, indeed, did not have the data to demonstrate) in the Phase II rule that the
major HTMR technologies could achieve the specified UTS limits if compliance
were based on grab samples rather than composite samples.  Since EPA previously
established composite sampling as the basis for compliance with the K061 and
UTS for metal nonwastewaters, EPA had no legal basis to establish the grab
sampling requirement, and its decision to do so was arbitrary and capricious.

D. EPA Intended to Ensure that All Major HTMR Technologies Achieve the OTB.

As explained above, a primary consideration of EPA in setting the BOAT
standards for metals (including the UTS) was that they should be "achievable by all
the major HTMR technologies."  56 Fed. Reg. at 41169. EPA explained the BOAT
methodology for the K061 treatment standards in the Phase II LDR rulemaking:

First, treatment standards were calculated individually for each HTMR process.
Next, the four sets of standards were compared to one another. Based on this
comparison, the Agency selected the highest standard for each metal  from each of
the four processes to allow for process variabilityand detection limit  difficulties. . .
. The Agency believed that this approach would derive limits achievable by all the
major HTMR technologies. . . .

-------
Memorandum from Anita Cummings to the Administrative Record for UTS (July
29,  1995) at 12 (CS2F-S0024) (emphasis added). Since all of the data EPA relied
upon from HRD and INMETCO reflected composite samples, the Agency would
not  have been able to state that the treatment standards were "achievable by all the
major HTMR technologies" if compliance were measured on the basis of grab
samples. EPA's grab sampling requirement therefore is contrary to the Agency's
BOAT methodology, which is grounded in common sense.

E. In. This Phase IV Proposal,  a Composite Sampling Requirement Is Consistent
With the Basis for the UTS and the Agency's Intent.
EPA should affirm that the treatment standards for UTS metals are K061  are based
on composite samples for three reasons. First, EPA established the composite
sampling requirement for K061 in 1991. Second, virtually all of the UTS metals for
which HTMR is BOAT are based on composite samples. (HTMR is the BOAT for
K061). Third, EPA intended that the BOAT metal standards would be achievable
by all HTMR technologies, and HTMR is the basis for most of the standards.

Affirming in this rulemaking that composite sampling is the appropriate basis for
determining compliance with the HTMR-based limits is particularly important
because HRD and INMETCO are responsible for the preponderance of HTMR
processing of K061. Since EPA does not have any data to show that HRD,
INMETCO, or any other HTMR operator can achieve the BOAT and UTS
standards on a grab sampling basis,  and the rulemaking precedent is based on
composite sampling, the Agency cannot state that grab  sample-based standards are
"achievable by all of the major HTMR technologies." Accordingly, to make the
UTS metal  standards and the K061  standards consistent with the applicable
rulemaking precedent, and the Agency's stated objectives, composite sampling
must be the basis for determining compliance for these standards.
HRD believes that,  given the regulatory development history of the K061 and UTS
metal treatment standards, the failure to specify that compliance  with the UTS
limits for metals (and with the nonwastewater TCLP limits for metals in listed
wastes) was an inadvertent oversight, and that EPA's statement in the proposed
Phase IV rule preamble concerning  the grab sample requirement is incorrect.
Accordingly, we suggest the following revisions to the appropriate regulations to
codify the composite sampling  requirement:

The second sentence of 40 C.F.R. § 268.40(b) should be revised to read as
follows:
"For all nonwastewaters, compliance with concentration level standards for
constituents other than metals is based on grab sampling; for metals, compliance
with concentration level standards is based on composite sampling.
The last sentence of footnote 3 to 40 C.F.R. § 268.48
Table UTS should be revised to read as follows:

-------
       "All concentration standards for nonwastewaters other than those for metals are
       based on analysis of grab samples; in the case of metals, concentration standards
       for nonwastewaters are based on analysis of composite samples."

       These corrections to the existing regulations will conform the sampling
       requirements for UTS metals to reflect the data and rationale EPA used to
       establish the BOAT treatment standards for K061 in 1991. and also confirm that
       composite sampling is the basis for compliance with the K061 treatment standards.
RESPONSE

       EPA does not view the commenter's recitation of past history to be germane, except
insofar as it reiterates the predicate for the reason for revising the current standards so that both
the means of developing the standard and the means of compying with it are based upon the same
sampling methodology: grab sampling.  The Agency reiterates that treatment standards for metal
constituents are to be based on grab samples for compliance purposes. The Agency prefers that
compliance with LDR standards for nonwastewaters be based on grab samples (a one-time sample
taken from any part of the waste), rather than composite samples (a combination of samples
collected at various locations for a given waste, or samples collected over time from that waste).
The Agency believes that grab samples generally reflect maximum process variability, thus would
reasonably characterize the range of treatment system performance.  (See discussion on this issue
at 54 FR 26605, June 23, 1989 and  55 FR 22539, June 1,  1990.) The grab sampling also meets
the ultimate objective of the LDR program, that all of the hazardous waste  to be land disposed be
treated in a way that minimizes the threats that land disposal could pose, not just the average
portion of the waste to be so treated (a possible result if composite sampling were to be used). In
addition, since grab sampling is based on individual sampling events, it facilitates enforcement
proceedings for EPA and authorized states.

-------
DCN PH4P067
COMMENTER Horsehead/Zinc Corp.
RESPONDER AC
SUBJECT GRAB
SUBJNUM 067

COMMENT

Consistent with previous treatment standards established over
several years and in multiple rulemakings, and to conform to the
Agency's intent, EPA should confirm that compliance for
the nonwastewater metal constituents in the UTS and listed
hazardous waste K061 is based on composite rather than grab
sampling.

RESPONSE

EPA has amended the rule to avoid this result. The treatment standards now reflect
exclusively the performance of technologies based upon grab sampling, and compliance will
likewise be measured exclusively by means of grab sampling. The Agency reiterates that
treatment standards for metal constituents are to be based on grab samples for compliance
purposes. The temporary exception of K061, K062, and F006 wastes managed at two HTMR
facilities, Horse head Resource Development Company Inc., and International Metals
Reclamation Company Inc. is no longer in force now that the standards are revised to be based
upon grab sampling data exclusively.

The Agency prefers that compliance with LDR standards for nonwastewaters be based on grab
samples (a one-time sample taken from any part of the waste), rather than composite samples (a
combination of samples collected at various locations for a given waste, or samples collected over
time from that waste). The Agency believes that grab samples generally reflect maximum process
variability, and thus would reasonably characterize the range of treatment system performance.
(See discussion on this issue at 54 FR 26605, June 23, 1989 and 55 FR 22539, June 1, 1990).
The grab sampling also meets the ultimate objective of the LDR program, that all of the
hazardous waste to be land disposed be treated in a way that minimizes the threats that land
disposal could pose, not just the average portion of the waste to be so treated (a possible result of
using composite sampling). In addition, since grab sampling is based on individual sampling
events, it facilitates enforcement proceedings for EPA and authorized states.
10

-------
DCN   PH4P067
COMMENTER Horsehead/Zinc Corp.
RESPONDER AC
SUBJECT GRAB
SUBJNUM 067

COMMENT

Despite years of rulemaking to the contrary, in late 1994 EPA
revised its regulations without explanation to provide that
compliance with the treatment standards for nonwastewater
hazardous wastes, including treatment standards for metal
constituents, is based on grab samples. 59 Fed.Reg. § 47980, 48046
(Sept. 19, 1994); codified at 40 C.F.R.§ 268.40(b). EPA explained
its interpretation of this requirement in the preamble to the Phase
IV Proposal:

The issue of grab versus composite sampling has been raised as
needing clarification. As previously promulgated, these metal
treatment standards specify grab samples. . . .  The use of grab
versus composite samples does not mean more frequent sampling is
necessary. Grab samples normally reflect maximum process
variability, and thus will reasonably characterize the range
of treatment system performance. The sampling  analysis for both
wastewater and nonwastewater is composite and grab respectively (40
CFR 268.41 and 268.43).
60 Fed. Reg.  at 43683/2.

With respect to hazardous waste K.061 (emission control dust/sludge
from the primary production of steel in electric arc furnaces) and
the UTS for metals, EPA's position directly conflicts with previous
LDR rules in which EPA required compliance with treatment
standards for metal constituents and K061 based on composite
sampling. These rulemakings established compliance based on
composite sampling as early as 1991. Moreover, the data EPA used
to develop treatment standards for K061 and the UTS for metals were
based largely on composite samples from several recyclers of K061.
Perhaps most  relevant, the two principal commercial off-site recyclers
of K061 submitted composite sampling data to EPA
during the development of the treatment standards for K061. EPA's
statement in the Phase IV proposal therefore represents an unlawful
change in position by the Agency. Therefore, as we explain below
in more detail, and based on the history of the K061  treatment
                                           11

-------
standards, EPA should confirm that compliance with the metal
treatment standards in general and the K061 standards in
particular is based on composite sampling (since EPA used the K061
treatment standards as the basis for nearly all of the metal
standards).

By way of background, EPA established compliance with the
treatment standards for K061 based on composite samples in 1991. 56
Fed. Reg. § 41164, 41169 (Aug. 19, 1991). Then, in the 1994 Phase
II LDR rule, EPA used the K061 treatment standards as the basis
for 11 of the HUTS for metal nonwastewaters. See 59 Fed. Reg. at
47998. (In other words, the Phase II LDR metal treatment standards
also were based on composite sampling.) Notwithstanding the
source of the treatment standard data, EPA inexplicably changed the
basis for compliance with the metal standards in the Phase II final
rule from composite to grab. Id. at 48046 (new 40 C.F.R.
§268.40(b)). I/ Given the inconsistency between the basis for the
metal treatment standards(composite sampling), which was developed
over many years and with a significant regulatory history, and the
purported compliance standard (grab sampling), EPA should correct
the error(first made in the Phase II final rule) and confirm that
the standard for compliance with the UTS metal nonwastewaters and
K061 is composite sampling.

RESPONSE

Again, the earlier rules obviously did not establish an immutable principle that the
treatment standard for K061 or any other prohibited waste, has to be based on composite
sampling. Since the earlier data bases reflected some composite sampling data, EPA has indicated
that until the standards for these wastecodes was amended, compliance could reflect a composite
sampling regime. However, EPA has developed a new data base for metals treatment based
exclusively upon grab sampling data. Consequently, compliance will be based on that same
sampling regime. (Indeed , to do otherwise would provide an unwarranted regulatory break for
facilities. This is because the new treatment standards reflect maximum variability, and therefore
are higher values than standards which would have reflected composite sampling. Obviously,
facilities should not now get the benefit of the higher numerical treatment standards, but the more
lenient sampling regime.)

The Agency thus reiterates that, the treatment standards for metal constituents are to be
based on grab samples for compliance purposes, with the temporary exception of K061, K062,
and F006 wastes managed at two HTMR facilities, Horse head Resource Development Company
Inc. and International Metals Reclamation Company Inc. ,The Agency prefers that compliance
with LDR standards for nonwastewaters be based on grab samples (a one-time sample taken from
12

-------
any part of the waste), rather than composite samples (a combination of samples collected at
various locations for a given waste, or samples collected over time from that waste). The Agency
believes that grab samples generally reflect maximum process variability, and thus would
reasonably characterize the range of treatment system performance. (See discussion on this issue
at 54 FR 26605, June 23, 1989 and 55 FR 22539, June 1, 1990). The grab sampling also meets
the ultimate objective of the LDR program, that all of the hazardous waste to be land disposed be
treated in a way that minimizes the threats that land disposal could pose, not just the average
portion of the waste to be so treated (a possible use of composite sampling). In addition, since
grab sampling is based on individual sampling events, it facilitates enforcement proceedings for
EPA and authorized states.

The Agency also notes that the revised UTS numbers promulgated in the final Phase IV
LDR rule for all the TC metals is based on treatment performance data from grab samples.
Therefore, the Agency believes that the UTS for the TC metals could be achieved through grab
samples from commercially available HTMR and/or stabilization technologies.

DCN PH4P067
COMMENTER Horsehead/Zinc Corp.
RESPONDER AC
SUBJECT GRAB
SUBJNUM 067

COMMENT

Given this information, EPA had no alternative in 1991 but to require compliance with the BOAT
treatment standards based on the maximum for any single composite sample. EPA neglected,
however, to include that condition when it transferred (in the same 1991 rulemaking) the revised
K061 treatment standards to the table at 40C.F.R. § 268.41, which covered hundreds of other
wastes. Since the treatment standards for other nonwastewaters generally were based on grab
'samples, this error created the false impression that the K061 treatment standards were also based
on grab sampling, even though they actually were based on composite sampling.

RESPONSE

This comment again is not germane to the final rule, except in establishing the reasons the
Agency needed to amend the rule if grab sampling were to be used as the exclusive means of
compliance sampling, and explaining the interim sampling regime EPA adopted for these wastes
until the rule was amended. In the second supplemental Phase IV proposed rule (62 FR 26047,
May 12, 1997), EPA noted that the current treatment standards for hazardous waste K061, K062,
and F006 were based partially on the use of composite rather than grab sampling. The Agency
also noted that the BOAT technology for K061 was HTMR and data used to develop parts of the
standard came from composite samples. Based on concerns raised by HTMR facilities regarding
the treatability of TC metals using grab samples, the Agency in the second supplemental proposal,
13

-------
temporarily allowed two HTMR facilities. Horse head Resource Development Company Inc. and
International Metals Reclamation Company Inc.. to comply with the current treatment standards
for K061, K062. and F006 through the use of composite samples. However, EPA's ultimate
intent is to require compliance with UTS on a grab sampling basis for all facilities.

The Agency notes that the revised UTS numbers promulgated in the final Phase IV LDR
rule for all the TC metals is based on treatment performance data from grab samples. Therefore,
the Agency believes that the UTS for the TC metals could be achieved through grab samples from
commercially available HTMR technologies.

DCN PH4P067
COMMENTER Horsehead/Zinc Corp.
RESPONDER AC
SUBJECT GRAB
SUBJNUM 067

COMMENT

As in the case of the August 1991 rulemaking, EPA's preamble
discussion of the merger of the standards clearly states that
compliance with the TCLP limits was based on the maximum for
any single composite sample. See 57 Fed. Reg. at 37209 (table of
final treatment standards for K061,which indicates that the
treatment standards are based on composite sampling) (attached to
these comments as Exhibit 2). Despite the specific and unambiguous
preamble statement that compliance with the K061 treatment
standards was based on composite sampling, EPA
again administratively erred by not transferring the composite
sampling requirement to the consolidated treatment standards table
at 40 C.F.R. § 268.41, thus leaving the mistaken impression
that compliance with the treatment standards for K061 could be
based on grab sampling.

RESPONSE

This comment again is not germane to the final rule, except in establishing the reasons the
Agency needed to amend the rule if grab sampling were to be used as the exclusive means of
compliance sampling and explaining the interim sampling regime EPA adapted for these wastes
until the rule was amended. [In the second supplemental Phase IV proposed rule (62 FR 26047,
May 12, 1997), EPA noted that the current treatment standards for hazardous waste K061, K062,
and F006 were based partially on the use of composite rather than grab sampling. The Agency
also noted that the BOAT technology for K061 was HTMR and data used to develop parts of the
standard came from composite samples. Based on concerns raised by HTMR facilities regarding
14

-------
the treatability of TC metals using grab samples, the Agency in the second supplemental proposal,
allowed two HTMR facilities, Horse head Resource Development Company Inc. and International
Metals Reclamation Company Inc., to comply with the current treatment standards for K061,
K062, and F006 through the use of composite samples. However, EPA's ultimate intent is to
require compliance with UTS on a grab sampling basis for all facilities.

DCN PH4P096
COMMENTER INMETCO
RESPONDER AC
SUBJECT GRAB
SUBJNUM 096

COMMENT

Introduction

These Comments are submitted by The International Metals Reclamation Company, Inc.
("INMETCO"), a wholly owned subsidiary of Inco United States, Inc., which in turn is a wholly
owned subsidiary of Inco Ltd., a Canadian corporation. INMETCO's Comments address a
narrow aspect of the proposed Phase IV Land Disposal Restrictions ("LDR") rule, the question of
whether the Best Demonstrated Available Technology ("BOAT") universal treatment standards
("UTS") for metals should specify grab or composite sampling. See 60 Fed. Reg. 43654, 43683
(August 22, 1995). Performance data from INMETCO's high temperature metals recovery
("HTMR") process were used as one of the critical data sets to establish the BD AT/UTS limits
for metals, and EPA intended that the resulting treatment standards would be consistently
achievable by INMETCO's process and other HTMR operations.
INMETCO believes that compliance with the nonwastewater UTS limits
for metals under the LDR program should be based on composite,
rather than grab, sampling. In these Comments, we explain why. In
Part I below, we describe INMETCO's HTMR process and outline
briefly the environmental and other benefits that an HTMR process
like INMETCO's provides. In Part II, we show that the data
underlying EPA's development of the nonwastewater BD AT/UTS limits
for metals do not support a requirement.that the limits be,met on
the basis of grab sampling. In Part III, we note that expressing the
nonwastewater UTS limits for metals in terms of composite sampling
15

-------
would be consistent with the approach that EPA has followed in
evaluating delisting petitions for metal-bearing wastes and in
setting generic exclusion levels from the "derived-from rule" for
HTMR slags. We also explain why measuring compliance on the basis
of composite sampling would not create any special difficulties
for the enforcement program.

I.  Description of INMETCO's HTMR Process and
Its Environmental Benefits

As far as we are aware, INMETCO, located
in Ellwood City,  Pennsylvania, currently operates the only
pyrometallurgical process in the United States designed to recover
nickel and chromium from metal-bearing secondary materials.  In
brief,  the process works as follows. Various hazardous and
nonhazardous nickel- and chromium-containing secondary materials
(including K061, K062, and F006) generated primarily in the
specialty steel and metal finishing industries are transported
(under hazardous waste manifests, where applicable) to INMETCO,
where they are temporarily stored in accordance with the company's
Part B hazardous waste storage permit before entering the HTMR
process. In recent years, Ni-Cd batteries have become an
increasingly important feedstock for INMETCO, which is the only
facility in the United States capable of thermally processing
Ni-Cd batteries.  The various metal-bearing materials are shredded,
where necessary, and blended together with carbon fines (which
promote the reduction of metal oxides) to produce a feed material
which is pelletized and fed into a Rotary Hearth Furnace ("RHF").
In the RHF, nickel oxides and some iron oxides are reduced to
the metallic state, and all chromium is reduced from the
hexavalent state to the trivalent form.  The heated, partially
reduced pellets are removed from the RHF, mixed with carbon
additives, dolomite and metal additives, and then fed into an
Electric Arc Furnace. INMETCO uses dolomite, rather than straight
lime, in order to ensure that its slag coproduct, once cooled, will
form into a hard solid matrix that resists abrasion, dusting, and
leaching. By carefully controlling the  slag chemistry, INMETCO is
able both to extend the life of the chrome refractory lining of its
electric furnace and to ensure that its slag has suitable and
uniform properties for use in road building and similar
construction applications when it is crushed and sized after
cooling. This allows it to compete in the marketplace with mined
aggregates and steel manufacturing slags. In the Electric Arc
                                           16

-------
Furnace ("EAF"), chromium oxides and the remaining iron oxides are
reduced to the metallic state, and a molten metal and slag
coproduct are produced. Periodically (typically from three to five
times per day) the molten metal is tapped from the EAF and cast
into remelt alloy pigs weighing approximately 14 kg through the use
of a twin strand pig caster.  The remelt alloy consists of
approximately 10 percent nickel, 14 percent chromium, 68 percent
iron, 1.5 percent manganese, 1.0 percent molybdenum, and the
balance carbon and other minor elements.  For the most part, the
remelt alloy is returned, under tolling agreements, to customers
who furnished various metal-containing feed materials and who use
the remelt alloy as a feedstock to produce specialty steel.  The
balance of the remelt alloy is sold to other specialty steel mills
in the United States and abroad.  The coproduct slag also is tapped
from the furnace periodically (typically five to seven times per
day) and is then cooled in a thin sheet, crushed, sized, and sold
as aggregate for road building and related applications, just as
comparable steelmaking  slags have been used for decades in the
United States.  In 1994, INMETCO recycled more than 58,000 tons
of secondary materials ~ including approximately 18,000 tons
of K061; 10,600 tons of hazardous and nonhazardous filter
cakes;8,900 tons of swarf; 8,600 tons of mill scale; 2,800 tons
of grindings; 2,200 tons of Ni-Cd batteries; and 1,700 tons
of nickel and chromium catalysts. From these materials, INMETCO
was able to recover for direct reuse (in the case of
nickel,  chromium, iron, manganese, and molybdenum) or for
further processing at an outside HTMR facility (in the case of
cadmium, lead, and zinc) approximately 98 percent of the nickel,
86percent of the chromium, 96 percent of the iron, 60 percent
of the manganese, 92 percent of the molybdenum, 97 percent of
the cadmium, 87 percent of the lead,  and more than 99 percent of
the zinc. By the second  quarter of 1996, INMETCO expects to
be operating newly installed cadmium retort furnaces, so that
it will be able to recover the cadmium directly on site.  In 1994,
INMETCO produced approximately  23,600 tons of remelt alloy ingots
containing approximately 2,400 tons  of nickel, 3,350 tons of
chromium, and 16,000 tons of iron, along with smaller quantities of
molybdenum. If they were not sent to INMETCO for processing, these
metals  would likely be buried in the ground, creating additional
risks to the environment, and would be lost to productive reuse in
commerce. Instead, INMETCO recovers them in the form of metal
ingots that are used as a feedstock in the production of stainless
and specialty steel.  In all, since 1978, INMETCO has prevented
                                           17

-------
more than 833,000 tons of metal-bearing materials from being
discarded as waste and buried in the ground.  INMETCO's HTMR process
provides important environmental, resource recovery, and other
benefits.  As EPA recognizes, metals cannot be destroyed, only
transformed, and they are a non-renewable resource.  At the same
time, metals are infinitely recyclable. By recycling metal-bearing
secondary materials in a recovery process, we can use the same
metal over and over again.  This allows us to achieve a number of
important environmental and other objectives, including the
following:

 •  Metals recovery, as EPA recognizes, "conserves nonrenewable resources". The conservation
of valuable material is one of the objectives explicitly articulated in Section 1003(a) of RCRA, and
metals probably are the best example
of anon-renewable resource whose conservation EPA should seek  to promote. For example, on
average, every ton of nickel recovered by INMETCO from secondary materials that otherwise
would be discarded as waste saves the mining of 110 tons of virgin nickel-bearing ore.

 •  Metals recovery "reduces energy demands
and pollution as compared to producing metal from virgin ore, "which
tend to have lower concentrations of metals than many reclaimable
secondary materials. INMETCO, for example,  uses less than half as
much energy to produce a pound of nickel and chromium from
secondary materials as its parent company does to produce a pound
of nickel and copper from virgin ore in Canada.  On average, the
generation of one kilowatt hour of electricity produces 1.5 pounds
of carbon dioxide, 5.8 grams of sulfur dioxide, and 2.5 grams of
nitrogen oxide.  At that rate, the fact that INMETCO's HTMR process
is 50 percent more energy efficient than metal production from
virgin ore means that in 1993, the energy savings alone allowed
INMETCO to produce its metal ingots while generating 32,675 fewer
tons of carbon dioxide, 278 fewer tons of sulfur dioxide, and 120
fewer tons of nitrogen oxide than would have been generated if the
same volume of metal had been produced from  virgin ore.  And, of
course, INMETCO's process avoids producing  the large volume of
tailings associated with the mining of ores.

•  By reducing the volume of metal-bearing hazardous and
non-hazardous wastes that are discarded, metals recovery "keeps
metal-bearing materials from being landfilled."  In addition to
preserving scarce landfill capacity, this "reduce[s] the mass loading
of hazardous constituents to the environment, "and is thus consistent
with an important national policy identified in the Hazardous and Solid
                                           18

-------
Waste Amendments of 1984.

• Metals recovery spurs pollution prevention efforts among waste
generators. Certain hazardous constituents present in a waste may
adversely affect the HTMR process or the metal produced in the process.
Consequently, waste generators have an incentive to eliminate or reduce
the presence of unwanted hazardous constituents in their waste in order
to make it amenable to recovery. This is true, for example, of
INMETCO's metal finisher customers, who may have to alter their
production processes in order to reduce the concentrations of
phosphorus, copper, and total cyanides in their F006 wastewater
treatment sludge, so that the material will meet INMETCO's
feedstock specifications. Metals recovery, as pointed out in EPA's
Metal Recovery Report to Congress, also "serves our country's
balance of payments and strategic interests." It does this by
expanding supplies of metals for use in the manufacture of
commercial and consumer products and by reducing U.S. purchases of
primary metals, such as chromium, from foreign sources.

In sum, as EPA has observed, high temperature metals recovery "is
important not only [to achieving the objectives set forth] in RCRA but
to the U.S. economy in general." For these reasons, among others,
EPA has judged HTMR to be the "besf'technology and "BOAT for
most metal constituents in nonwastewater forms of listed hazardous
wastes." INMETCO has been proud to serve our country's
environmental, economic, and strategic interests by providing
this technology since 1978, and we look forward to continuing to
do so in the future.

"RESPONSE

EPA thanks the commenter for providing background information on the high temperature
metals recovery process. The commenter's concerns regarding sampling are addressed separately
elsewhere in this document.
DCN PH4P096
COMMENTER INMETCO
RESPONDER AC
SUBJECT GRAB
SUBJNUM 096
19

-------
COMMENT

II.     The Underlying Data Do Not Support a Requirement
       that Nonwastewater UTS Limits for Metals Be Met on the Basis of Grab
       Sampling.

In setting the nonwastewater UTS limits for metals other than
arsenic, chromium, and mercury, EPA relied solely on the HTMR
performance data that were used to establish the BOAT treatment
standards for K061 — or, more specifically, on the data that the
Agency used to set the treatment standards for high zinc K061 in
August 1991, which were extended to low zinc K061  in August 1992.
No additional HTMR performance data we recollected in the Phase II
LDR rulemaking or, with the exception of certain grab sampling data
submitted by INMETCO late last year, in the Phase IV rulemaking
either.  As EPA acknowledges, the data used to set the BOAT treatment
standards for K061 included composite sample results. In  fact, all
of the INMETCO performance data that EPA used to set the K061  BOAT
limits (and the derivative UTS limits) for metals consisted of
monthly composite sampling results that INMETCO submitted to the
Agency in its May 13, 1991 Comments on the proposal to  revise the
high zinc K061 treatment standards.  Those data were used to set the
K061 BOAT and UTS limits for beryllium, nickel, and thallium.  Most
of the other BOAT metal treatment standards were based on composite
sample data from the HTMR zinc recovery process of Horsehead
Resource Development Company ("HRD"). Using its BOAT methodology,
EPA calculated a treatment standard limit for each metal by
applying a variability factor to the mean (average) treatment
performance value for the metal in the relevant data set.  The
variability factor accounts for"inherent mechanical limitations in
treatment control systems, treatability variations caused by
changing influent loads,  unavoidable variations in procedures for
collecting treated samples, or variations in sample analysis." To
develop the BDAT/UTS treatment standards for metals other than
arsenic, chromium, and mercury, EPA used four datasets — (1) a
data set from SKF Plasma Technologies, (2) a dataset from
International Mill Service, (3) the monthly composite sample data
set submitted by INMETCO, and (4) a combined data set from HRD.
EPA's objective was to set BOAT (and derivative UTS) standards for
metals that would be "achievable by all of the major HTMR
technologies." Accordingly, the Agency set the treatment standards
as follows:
                                          20

-------
"First, treatment standards we recalculated individually for each
HTMR process. Next the four sets of standards were compared to
one another. Based on this comparison, the Agency selected the
highest standard for each metal from each of the four processes to
allow for process variability and detection limit difficulties . . .
. [T]he Agency believed that this approach would derive limits
achievable by all of the major HTMR technologies. ..."

This is a logical approach to follow, as long as the nature of the
various data sets is taken into account. As noted above, the
FNMETCO data set consisted of composite samples. Thus, the
treatment standard values that EPA derived from this data set by
calculating variability factors and applying them to the mean
performance values for the various metals in the dataset may
adequately reflect variability within INMETCO's composite sample
data. They do not, however, adequately reflect the range of, or
variability within, grab sample results. Since all of the data
that EPA used from INMETCO to set BDAT/UTS standards and much of
the data that EPA used from HRD reflected composite samples, the
Agency was not in a position to say that the treatment standards
are "achievable by all of the major HTMR technologies" when
measured on the basis of grab sampling. EPA seemed to recognize
this point at the time the nonwastewater BDAT metal treatment
standards were set. In order to reconcile ~
(1) its intent that the BDAT standards would be achievable by all
the major HTMR technologies, and
(ii) the fact that much of the HTMR performance data on which the
standards were based reflected the testing of composite samples,
EPA made clear in the preambles to the final rules setting the BDAT
treatment standards for K061 that compliance with the standards was
to be based on the maximum for any single composite sample.
Through apparent inadvertence, this point got lost in the process
of transferring the K061 treatment standards into the table in §
268.41 of the Code of Federal Regulations, which covered hundreds
of other wastes for which the standards were based on grab samples.
That inadvertent mistake should now be corrected.

RESPONSE

This comment likewise indicates why some type of regulatory amendment is needed if
compliance with treatment standards for all nonwastewaters is to be based upon grab sampling.
In the second supplemental Phase IV proposed rule (62 FR 26047, May 12, 1997), EPA noted
that the current treatment standards for hazardous waste K061, K062, and F006 were based
21

-------
partially on the use of-composite rather than grab sampling. The Agency also noted that the
BOAT technology for K061 was HTMR and data used to develop parts of the standard came
from composite samples. Based on concerns raised by HTMR facilities regarding the treatability
of TC metals using grab samples, the Agency in the second supplemental proposal, allowed two
HTMR facilities. Horse head Resource Development Company Inc. and International Metals
Reclamation Company Inc.. to comply with the current treatment standards for K.061, K062, and
F006 through the use of composite samples — as an interim measure pending regulatory
amendments to develop standards for these wastecodes based upon performance ot the BOAT
technologies measured exclusively by grab sampling. However, EPA's ultimate intent is to
require compliance with UTS on a grab sampling basis for all facilities.

COMMENT

In the present rulemaking, EPA has implicitly conceded that,
because the database underlying the nonwastewater UTS limits for
metals included composite samples, the UTS limits may not be
consistently achievable on the basis of grab sampling. As noted
above, the Agency set the BDAT and UTS limits for beryllium on the
basis of monthly composite sampling data submitted by INMETCO.
Applying its BDAT standard-setting methodology, EPA multiplied the
mean value of the INMETCO monthly composite sample data for
beryllium (0.0073 mg/1) by a variability factor of 1.9,to produce
a BDAT (and derivative UTS) limit of 0.014 mg/1. That value may
reflect what is achievable, talcing account of variability in the
results of composite sampling, but it does not reflect the greater
variability that would be expected for grab sampling data. EPA has
acknowledged that fact in the present rulemaking and now proposes
to increase the UTS limit for beryllium to 0.04 mg/1. The new
limit is based upon a statistical analysis of grab sampling data
from INMETCO, showing that the UTS limit of 0.014 mg/1 (set on the
basis of composite sample results) cannot be met consistently on
22

-------
the basis of grab sampling. To further illustrate the problem of
measuring compliance with UTS limits on the basis of grab samples,
INMETCO collected random grab samples of its slag (every 20th slag
tap)during the months of September and October 1995 and analyzed
them for TCLP levels of various metals. The analytical results
for nickel and selenium are presented in Table 1 below.

Table 1  Random Grab Sample TCLP Results (in mg/L) for INMETCO's
Slag: September and October 1995

DATA ARE  NOT REPRODUCED HERE

Using EPA's  methodology, we calculated the mean and standard
deviation of the log transformed data for the nickel and selenium
values shown in Table 1 and applied EPA's BOAT formula to derive
the appropriate treatment standard level (i.e., the C99value below
which 99 percent of the TCLP performance results are estimated to
fall). For these random grab sample data, the BDAT/UTS treatment
standard value for nickel would be 7.1 mg/1, while for selenium it
would be 0.20 mg/1.  These values are higher than the comparable
UTS limits of 5.0 mg/1 for nickel and 0.16 mg/1 for
selenium. The values included in Table 1 above reflect random grab
sampling results.  INMETCO's experience suggests that slags having
the highest total chromium composition tend to exhibit somewhat
higher TCLP levels of various metals. Table 2 below shows the
nickel and selenium TCLP grab sample results for the four highest
chromium composition slags produced during each of the months of
September and October 1995. When the data in Table 1 are expanded
to include the data from Table 2, the calculated treatment standard
levels (i.e., the  C99 values below which 99 percent of the TCLP
performance results are estimated to fall) are found to be even
higher.

Table 2 Grab Sample TCLP Results (in mg/L) Four Highest
Total Chromium Composition Slags Produced at INMETCO in September
and October 1995

DATA ARE NOT REPRODUCED HERE.
These data confirm what EPA discovered in the case of beryllium.
They show that the BDAT/UTS nonwastewater limits that EPA
calculated for metals on the basis of data that included (or, in
some cases, consisted solely of) composite samples will not
be consistently achievable if compliance is measured on the basis
                                          23

-------
of grab samples. As noted above, a primary EPA objective
in establishing the BDAT/UTS limits was to ensure that they
are achievable by the major HTMR technologies. In light of the
very significant environmental and other benefits that
HTMR technologies provide, this is indeed an important objective. In
order to ensure that it can be met. compliance with
the nonwastewater UTS limits for metals should be determined on
the basis of composite samples, not grab samples. Sections
268.40 and 268.48 of the regulations should be revised to reflect
this point.

RESPONSE

The revised data base indicates that the standards, as amended, can be met consistently
using a grab sampling regime. In the second supplemental Phase IV proposed rule (62 FR 26047,
May 12, 1997), EPA noted that the current treatment standards for hazardous waste K061, K062,
and F006 were based partially on the use of composite rather than grab sampling. The Agency
also noted that the BOAT technology for K061 was HTMR and data used to develop parts of the
standard came from composite samples. Based on concerns raised by HTMR facilities regarding
the treatability of TC metals using grab samples, the Agency in the second supplemental proposal,
allowed two HTMR facilities. Horse head Resource Development Company Inc. and International
Metals Reclamation Company Inc., to comply with the current treatment standards for K061,
K062, and F006 through the use of composite samples. However, EPA's ultimate intent is to
require compliance with UTS on a grab sampling basis for all facilities.

COMMENT

III. Measuring Compliance with Nonwastewater UTS Limits
for Metals on the Basis of Composite Sampling Would Be consistent
With the Approach Followed Under the Delisting and Generic
24

-------
       Exclusion Programs, and Would Not Create Difficulties in
       Enforcement.

A. Consistency With Delisting and GenericExclusion Compliance
Determinations

In Part II above, we showed that the data underlying EPA's
establishment of nonwastewater BDAT/UTS
limits for metals do not support a requirement that compliance be
measured on the basis of grab samples.  To the contrary, as EPA now
recognizes in the case of beryllium (and as we have shown with the
examples of nickel and selenium), if the UTS limits for metals are
set to reflect grab sampling data, they would have to be higher
than when the limits are based on composite sampling data. That
alone is sufficient reason to measure compliance with the
nonwastewater UTS limits for metals on the basis of composite
sampling results.  Another reason is that composite sampling is the
basis that EPA has specified for evaluating compliance with TCLP
limits for metals in delisting determinations.  For example, in a
recent delisting decision for chemically stabilized electric arc
furnace dust, EPA specified that initial verification testing
for compliance with the designated TCLP limits must reflect
the results of composite sampling and that subsequent
verification testing must be based on monthly composite samples.
EPA also has specified composite sampling as the basis for
determining whether HTMR slags qualify for the generic exclusion
from the"derived-from rule." The composite sampling approach that
EPA has applied for determining compliance with TCLP metal
limits under the delisting and generic exclusion programs is sound.
It should be used to demonstrate compliance with UTS
nonwastewater limits for metals under the LDR program as well.  B.
Enforcement Considerations In the Background Document for this
proceeding, EPA suggests that evaluating compliance with UTS limits
on the Basis of grab sampling is necessary in order to avoid adding
"another layer of complexity to regulatory compliance
determinations." EPA appears to believe that if compliance were
based on composite sampling, "enforcement personnel would be faced
with having to perform composite sampling from a group of drums [or
other waste depositories] that may or may not have any correlation
with The composite sample technique used when treatment standards
we redeveloped.  On the other hand, . .  . if treatment standards
are based on grab samples . . . enforcement personnel could
simply take a grab sample from any drum [or batch] and, thereby,
                                          25

-------
make a compliance determination." EPA's analysis of this point
is flawed, and its concern is misplaced. We can sympathize with
EPA's reluctance to have compliance sampling conducted on a basis
that "may not have any correlation with the . . sample technique
used when treatment standards were developed." However, as shown
in Part II of these comments, this is an argument for using
composite sampling, rather than grab sampling, to determine
compliance with the UTS LIMITS for metals: EPA's calculation of
the BDAT/UTS limits for metals was based, to a large extent, on
composite sample data; accordingly, EPA enforcement personnel
should measure compliance with those standards on the basis of
composite samples as well. EPA's fear that composite sampling would
add a "layer of complexity" to regulatory compliance determinations
is overblown. After all, as noted above, compliance with TCLP
metal limits under both the delisting program and the generic
exclusion rule is determined on the basis of composite sampling.
The same is true for determining compliance with UTS limits
in wastewaters. We are not aware that this has made
compliance determinations particularly complex. Any legitimate
concerns that EPA might have about basing regulatory compliance
determinations on composite sampling can be addressed adequately by
requiring HTMR operators (and generators of other wastes subject to
the nonwastewater UTS LIMITS for metals) to preserve samples from
each slag tap (or each batch of waste generated) and to prepare a
composite of these samples covering an appropriate time period,
e.g., monthly composites. The generator could be required to
retain these samples for whatever period of time is reasonably
necessary to accommodate the needs of the enforcement program. EPA
personnel could then request one or more composite samples from
the generator for purposes of making a compliance determination. In
sum, EPA's concern about the potential complexity of enforcement is
misplaced. Compliance with the nonwastewater UTS LIMITS for metals
can be based on composite sampling without creating any special
difficulties for the enforcement program.

RESPONSE

As noted in response to the Chem Waste comment, the Agency did not intend to, and did
not reopen the issue of whether treatment standards can permissibly be based upon grab sampling,
or whether they should be. The question is, what standards are achievable if grab sampling is
used as the means of compliance. As the commenter indicated, certain standards would have to
be increased to reflect the greater variability inherent in this type of a sampling regime, and EPA
has done so in the final rule. The Agency reiterates that, the treatment standards for metal
26

-------
constituents are to be based on grab samples for compliance purposes, with the temporary
exception of K061, K062, and F006 wastes managed at two HTMR facilities, Horse head
Resource Development Company Inc. and International Metals Reclamation Company Inc. The
Agency prefers that compliance with LDR standards for nonwastewaters be based on grab
samples (a one-time sample taken from any part of the waste), rather than composite samples fa
combination of samples collected at various locations for a given waste, or samples collected over
time from that waste). The Agency believes that grab samples generally reflect maximum process
variability, and thus would reasonably characterize the range of treatment system performance.
(See discussion on this issue at 54 FR 26605, June 23, 1989 and 55 FR 22539, June 1, 1990).
The grab sampling also meets the ultimate objective of the LDR program, that all of the
hazardous waste to be land disposed be treated in a way that minimizes the threats that land
disposal could pose, not just the average portion of the waste to be so treated (a possible use of
composite sampling). In addition, since grab sampling is based on individual sampling events, it
facilitates enforcement proceedings for EPA and authorized states.

DCN PH4P097
COMMENTER Hazardous Waste Management
RESPONDER AC
SUBJECT GRAB
SUBJNUM 097

COMMENT

With regard to the grab verse composite sampling issue, the
Agency needs to reevaluate its position that wastewaters require
composite and nonwastewaters require grab samples. The Agency's
preamble discussion (60 FR 43683) which states:

"While UTS nonwastewater limits for metals specify a grab sample,
the data used to develop standards included both grab and composite
samples".

That UTS levels were not all established using grab samples,
indicates that the Agency recognizes that composite samples for
nonwastewaters may be appropriate. As such, the Agency
should allow the use of both sampling methods for nonwastewaters,
and for enforcement purposes, the Agency should utilize the same
method employed by the generator or TSDF.
27

-------
RESPONSE

The Land Disposal Restrictions LDR Program does not recognize that composite samples
for nonwastewaters may be approppriate. During the development of Federal Register 60 43683,
it simply reflected the best data EPA could obtain at that time. Then, consistent with the principle
that complaince is to be based upon the same mode of sampling used to develop the treatment
standards, compliance with vertain of the existing standards whish reflected composite sampling in
turn also had to be based on composite sampling. This does not affect EPA's policy preference
for grab sampling.

The Agency reiterates that the treatment standards for metal constituents are to be based
on grab samples for compliance purposes, with the temporary exception of K061, K062, and
F006 wastes managed at two HTMR facilities, Horsehead Resource Development Company, Inc.
and International Metals Reclamation Company, Inc.

The Agency prefers that compliance with LDR standards for nonwastewaters be based on
grab samples (a one-time sample taken from any part of the waste), rather than composite samples
(a combination of samples collected at various locations for a given waste, or samples collected
over time from that waste). The Agency believes that grab samples generally reflect maximum
process variability, and thus would reasonably characterize the range of treatment system
performance. (See discussion on this issue at 54 FR 26605, June 23, 1989 and 55 FR 22539,
June 1, 1990). The grab sampling also meets the ultimate objective of the LDR program, that all
of the hazardous waste to be land disposed be treated in a way that minimizes the threats that land
disposal could pose, not just the average portion of the waste to be so treated (a possible use of
composite sampling). In addition, because grab sampling is based on individual sampling events,
it facilitates enforcement proceedings for EPA and authorized states.

The Agency also notes that the revised UTS numbers promulgated in the final Phase IV
LDR rule for all the TC metals is based on treatment performance data from grab samples. (Please
refer to the May 12, 1997 Federal Register Notice for new data on grab samples.) Therefore, the
Agency believes that the UTS for the TC metals could be achieved through grab samples from
commercially available HTMR and/or stabilization technologies.

DCN PH4P113
COMMENTER Chemical Manufacturers Association
RESPONDER AC
SUBJECT GRAB
SUBJNUM 113

COMMENT

6. CMA recommends composite sampling and questions why four
samples are needed for each sampling event.
28

-------
The Agency has proposed to use annual sampling of the wastewaters
in the surface impoundment to determine if regulated constituents
are present at concentrations that exceed the trigger level. The
Agency has proposed that determinations of whether or not a
trigger level has been exceeded would be calculated from a minimum
of a four sample set on a four time per year basis (the Agency
notes quarterly). CMA concurs that seasonal sampling may be
appropriate but questions the requirement to use four samples per
event. The only basis CMA can determine for requiring four samples
per event is that that was what was finalized for the Subpart
CC regulations in December 1994. The rationale under that rule that
the wastes are potentially variable does not hold for wastewaters
treated in impoundments. The variability of
constituent concentrations in wastewaters in impoundments is slight
at best, especially on a short-term sampling event, and requiring
four samples per event is unnecessarily burdensome. If the
Agency finalizes a requirement to use four samples per event, CMA
requests that the Agency allow composite samples.

RESPONSE

The Agency appreciates the commenter's thoughts on sampling. However, the Agency
believes that demonstration samples must consist of enough representative samples, but in no case
less than four samples, taken over a period of time sufficient to represent the variability or the
uniformity of the waste.
(See BOAT QA/QC Methodology Background Document)
29

-------
Phase IV Second Supplemental Proposed Rule:
Comments and Responses on Demonstrating Compliance by Grab or Composite Sampling
1. 2P4P-00010 International Metals Reclamation Company, Inc. (ENMETCO) and
INCO United States, Inc.

As EPA acknowledges, the existing UTS/BDAT nonwastewater standards for most metals
were developed in whole or in part on the basis of composite sampling data from HTMR facilities,
including INMETCO. See 62 Fed. Reg. at 26047/2, And the standards were set on the
assumption that they could be achieved by all the major HTMR facilities. See 56 Fed. Reg.
41164, 41169 (August 19, 1991). But, because of the greater variability in grab sampling results,
a standard that can be met on the basis of composite sampling may not be achievable on the basis
of grab sampling. See 62 Fed. Reg. at 26047. That is a principal reason why EPA is revising the
metals standards to reflect solely grab sampling data. See id. In effect, the existing composite
sample-based standards are being translated into grab sample-based standards of equivalent
stringency. The new standards reflect the performance of the same technologies used to set the
old standards, but that performance is now measured on the basis of grab sampling, rather than
composite sampling. Thus, even in those cases where the numerical values are being revised
upward to reflect grab sampling data, the standard is not being made less stringent; it is just being
expressed in a different way — i.e. the equivalent degree of stringency is being "translated" from
composite to grab sampling.

EPA has recognized this point in the Phase IV Second Supplemental Proposal by
providing that until the new grab sample-based standards are adopted, INMETCO and the other
major HTMR operator, Horsehead Resource Development Company, may determine compliance
with the BDAT standards for metals on the basis of composite sampling. See id. at 26047-48,
26069-70. INMETCO is able to meet the existing standards on a composite sample basis, and it
expects to be able to meet the revised standards (assuming the adjustments suggested above are
made) on a grab sample basis — operating its HTMR process in precisely the same way. Thus, as
far as INMETCO is concerned, the HTMR-based standards will not have become less stringent
when the numerical values go up while, at the same time, the basis for determining the company's
compliance shifts from composite to grab sampling. The standards will have been restructured,
but their stringency will remain the same.

I. Proposed Revisions of the UTS and BDAT Standards

As explained in the Federal Register notice, EPA proposes to revise the present
UTS/BDAT nonwastewater TCLP leachate standards for twelve metals as follows:

• The standard for antimony would be decreased from 2.1 mg/L to 0.07 mg/L;
• The standard for barium would be increased from 7.6 mg/L to 21 mg/L;
• The standard for beryllium would be increased from 0.014 mg/L to 0.02 mg/L;
30

-------
The standard for cadmium would be increased from 0.019 mg/L to 0.02 mg/L;
The standard for chromium would be decreased from 0.86 mg/L to 0.85 mg/L;
The standard for lead would be increased from 0.37 mg/L to 0.75 mg/L;
The standard for nickel would be increased from 5.0 mg/1 to 13.6 mg/1;
The standard for selenium would be increased from 0.16 mg/L to 5.7 mg/L,
The standard for silver would be decreased from 0.30 mg/L to 0.11 mg/L;
The standard for thallium would be increased from 0.078 mg/L to 0.20 mg/L;
The standard for vanadium would be increased from 0.23 mg/1 to 1.6 mg/1; and
The standard for zinc would be decreased from 5.3 mg/L to 4.3 mg/L.

In each case, the proposed standard reflects the higher of EPA's stabilization-based or
HTMR-based "C99" calculations, using solely grab sample data. See 62 Fed. Reg. at 26045/2.
This differs from the existing nonwastewater metals standards, most of which were based on
HTMR data reflecting, in whole or in part, the results of composite sampling. See id. at 26044/2,
26047. The proposed standards for antimony, barium, lead, selenium, thallium, vanadium, and
zinc are based on data from stabilization processes, while the proposed standards for beryllium,
cadmium, chromium, nickel, and silver are based on grab sample data from INMETCO.[fn2:See
March 10, 1997 Memorandum to Anita Cummings from Howard Finkel, Attachment 5 (Item
SOO11 in Docket F-97-2P4P-FFFFF); Memorandum to Anita Cummings from Stan Moore of
Versar Inc., enclosing Draft Report: Metals Treatment Standards Derived From Data Submitted
by Industry (Item S0012 in Docket F-97-2P4P-FFFFF).] Because the higher of stabilization-
based or HTMR-based calculations was used, EPA believes the standards will be achievable by
both stabilization and HTMR technologies. See 62 Fed. Reg. at 26045/2. Having reviewed the
data, we believe EPA has properly applied its BOAT methodology to calculate the proposed
revised standards for antimony, barium, cadmium, chromium, lead, nickel, selenium, thallium,
vanadium, and zinc. We caution, however, that the proposed revisions do not reflect grab sample
data from another HTMR operation which may submit such data within the next few months. See
62 Fed. Reg. at 26047-26048. It is possible that data from that HTMR process will support a
higher standard for one or more of these metals. Whatever may be the case for those metals, the
proposed standard for silver, as explained below, should be increased to reflect — at a minimum —
more complete grab sampling data from INMETCO. Moreover, EPA should consider whether
the stabilization data for beryllium are too limited to conclude that the proposed standard of 0.02
mg/L is achievable by stabilization when higher beryllium content wastes are involved.

Response

The Agency appreciates the support provided by the commenter on this issue. EPA also
largely agrees that the standards are equivalent, although the Agency has also gathered data on
stabilization operations treating particularly hard-to-treat wastes in order to assure that treatment
standards for all characteristic wastes are achievable. This methodology also accounts for some
of the increase in the standards (such as the standard for lead, for example). It is also noted that,
EPA has revised the treatment standards for antimony, beryllium, cadmium, chromium, nickel,
and silver. Treatment standards for antimony, beryllium, and silver were revised based on new
31

-------
data provided by commenters, that represented the '"most difficult to treat" wastes. Thus, the
Agency is promulgating revised UTS for antimony, beryllium, and silver at 1.15 mg/1, 1.22 mg/1,
and 0.14 mg/1 respectively. For cadmium, chromium, and nickel, the Agency has identified a
technical error in the BOAT determination of the proposed standards. In applying the BOAT
methodology for calculating the treatment standard. EPA failed to perform a "Z-score" outlier
test. The Agency corrected this error and has re-calculated the treatment standards for cadmium.
chromium, and nickel at 0.11 mg/1, 0.60 mg/1, and 11 mg/1 respectively. Although these standards
are more stringent than the proposed standards, based on the treatment performance data
reviewed, the Agency believes that these new standards are also achievable by commercial
treatment technologies such as stabilization and HTMR. (See the BDAT background materials in
the docket for today's rule, for additional information on the revised treatment standards.)

2. 2P4P-00012 The Ferroalloys Association (TFA)

The sampling technique proposed by this rule, grab sampling, contradicts procedures
given in the US EPA prepared Test Methods for Evaluating Solid Wastes - SW-846, chapter nine.
Chapter nine explains that it is imperative that composite sampling be used to eliminate selective
sampling and that randomness achieves accuracy. However, the language in the prop .sed rule
indicates that by requiring grab sampling, "there is an implementation advantage ... since
enforcement for EPA, authorized states, or citizen groups is facilitated if enforcement can be
based on individual sampling events." Therefore, the use of grab sampling will be required
because it is easier.

Response

The Agency prefers that compliance with LDR standards for nonwastewaters be based on
grab samples (a one-time sample taken from any part of the waste), rather than composite samples
(a combination of samples collected at various locations for a given waste, or samples collected
over time from that waste). The Agency believes that grab samples generally reflect maximum
process variability, and thus would reasonably characterize the range of treatment system
performance. (See discussion on this issue at 54 FR 26605, June 23, 1989 and 55 FR 22539,
June 1, 1990). The grab sampling also meets the ultimate objective of the LDR program, that all
of the hazardous waste to be land disposed be treated in a way that minimizes the threats that land
disposal could pose, not just the average portion of the waste to be so treated (a possible use of
composite sampling). In addition, since grab sampling is based on individual sampling events, it
facilitates enforcement proceedings for EPA and authorized states.
3. 2P4P-00017 Battery Council International (BCI) and Association of Battery
Recyclers (ABR)

Before publishing the proposed standards, EPA had before it at least two data sets: a
200-point set of analyses of composite samples taken from six facilities over a several month
32

-------
period of time[fn8: See comments of Battery Council International (Nov. 27, 1995) and
Association of Battery Recyclers (Nov. 20,1995) submitted in response to EPA Land Disposal
Restrictions Phase IV proposed rule (60 Fed. Reg. 43,654 (1995)). This data contained analysis
of composite sampling results on the treatment of secondary smelter slag and contaminated soils.]
and a much smaller set of analyses of grab samples taken over a much shorter period of time.[fn9:
See Memorandum of Howard Finkel to Anita Cummings, supra.] EPA disregarded the first and
relied solely upon the second. No effort was made, however, to select a representative set. For
example, EPA's treatment standard for lead was derived from twenty-seven grab sample data
points.[fnlO: Id.] 276 other data points submitted by BCI and ABR, taken over several months,
were ignored. Similarly, the Agency's treatment standard for beryllium was derived from only
four grab sample data points.[fnl 1: Id.] The treatment standards for antimony, silver, and thallium
were derived from nine, fourteen, and fifteen grab sample data points, respectively.[fnl2: Id.]

This is far too limited a set of samples from which to calculate treatment standards. Such
a limited sampling size will not adequately represent the variability of the D008 waste stream.
This is particularly so for secondary smelter slag. Because of its physical and chemical
composition, this slag is extremely variable, differing from one minute to the next and one week
to the next. BCI and ABR note that, in other contexts, the Agency requires many more samples
to be taken than were included in this set. For example, for municipal incinerator ash, a waste not
as variable as secondary smelter slag, the Agency requires a total of fourteen eight-hour
composite samples. These samples must be collected hourly for a total of eight hours for seven
days.[fn!3: Environmental Protection Agency, Sampling and Analysis of municipal Refuse
Incinerator Ash (June 1995).]

The Agency has compounded its error by analyzing only grab samples, a one-time sample
occurrence. Unless performed frequently over a several week period of time, and then analyzed
using appropriate statistical techniques, these samples will fail to accurately characterize highly
variable heterogeneous wastestreams. As discussed above, secondary smelter slag is extremely
heterogeneous. Thus, unless many grab samples are taken over an extended period of time
(which was not done by EPA in this instance), it will be only sheer coincidence if they are an
accurate measure of this wastestream. respond to why used here but not w municipal ash.

Finally, all of the grab .ample data points obtained from Rollins Environmental, which
comprised the vast majority 01 the data points used by EPA in calculating the treatment standards
for D008 wastes, were based on the treatment of remediation and other types of wastes, not of
secondary smelter slag. Smelter slag however, has chemical and physical characteristics distinctly
different from the remediation and other wastes obtained from Rollins. For example, secondary
smelter slag is partly or completely solid, and of variable particle size. Because of this physical
variably, its treatment through stabilization is much less effective than other types of D008 wastes.
[fnl 4: All of the remaining data points used by EPA in the calculation of treatment standards for
silver, antimony, beryllium, and thallium were collected by Rollins Environmental. See
Memorandum of Howard Finkel to Anita Cummings, supra] were based on the treatment of
remediation and other types of wastes, not of secondary smelter slag. Smelter slag, however, has
33

-------
chemical and physical characteristics distinctly different from the remediation and other wastes
obtained from Rollins. For example, secondary smelter slag is partly or completely solid, and of
variable particle size. Because of this physical variably, its treatment through stabilization is much
less effective than other types of D008 wastes.

EPA's reliance upon this data set ignored the much more comprehensive and
representative data submitted by BCI, ABR and others in response to EPA's original proposed
LDR Phase IV rule.[fn!5: See comments of Battery Council International (Nov. 27,1995) and
Association of Battery Recyclers submitted in response to EPA Land Disposal Restrictions Phase
IV proposed rule (60 Fed. Reg. 43,654 (1995)).] BCI and ABR's data contained 276 composite
data points for lead and 156 composite data points for selenium and barium in secondary smelter
slag.[fn 16: Id.] The data showed the 99th percentile confidence interval for stabilized slag to be
2.97 mg/1 for lead, 2.48 mg/1 for selenium, and 8.92 mg/1 for barium.[fnl7:Id] This level for lead
is well above the treatment level now proposed by EPA.

This data was generated over a several month period, from six secondary lead smelters,
and represents over one thousand individual grab sample points consolidated into composite
samples. EPA ignored this data on the grounds that it was not obtained from grab samples and
did not provide "in versus out data points, "[fnl 8: See Overview of Five Data Sets Submitted in
Response to Phase IV Proposed Rule: Treatment of Metals 4 (Dec. 1996).] This rationale fails for
several reasons. First, it is inconsistent with past practice. In past LDR rulemakings, EPA has
considered all of the relevant data in determining what treatment standards could typically be
achieved. For example, in the Third -Third LDR rule EPA stated with regard to D008 treatment
standards that although "[the majority of the data received by the Agency did not have the proper
QA/QC, corresponding influent and effluent data, and design and operating parameters
The Agency, nevertheless, evaluated all of the data to assess the range of waste variability and
what standard could typically be achieved."[fnl9: 55 Fed. Reg. 22,565 (1990] Had EPA
evaluated all of the data it received in response to the original proposed Phase IV rule, it clearly
would have recognized that the proposed treatment standard of .75 mg/1 for lead could not be
achieved.

Second, the use of grab samples is inappropriate for analyzing highly variable D008
wastes. [fn20:The issue of whether grab sampling is an inherently unreliable method for measuring
heterogeneous wastes, such as secondary smelter slag, for the development of treatment standards
has not previously been subject to judicial review. See e.g., Chemical Waste Management v.
United States EPA, 976 F.2d 2 (D.C. Cir. 1992), cert, denied 507 U.S. 1057 (1993).] Secondary
smelter slag is extremely variable and heterogeneous. Three factors principally affect the
variability of slags: (1) changes in concentration with respect to time; (2) changes in concentration
where the material is partly or completely solid; and (3) changes in concentration where the
material is of variable particle size. The constituents of waste materials comprised of particles
larger in size than dust, including crushed slag from a discontinuous process such as secondary
smelting of lead, are very inconsistent from one day to the next or one week to the next. Because
of the highly variable nature of secondary smelter slag grab sampling, unless done frequently over
34

-------
a extended period of time (which was not done by EPA), does not provide an accurate measure
for the development of treatment standards (or, for that matter, the enforcement of existing
treatment standards).
It is no doubt for this reason that SW-846 mandates that when analyzing heterogeneous
wastes resulting from discontinuous or variable processes (such as lead-acid battery and other
lead scrap smelting), statistically random samples of waste must be taken over extended periods
of time. [fn21:Environmental Protection Agency, Test Methods for Evaluating Solid Waste, SW
846 3rd Ed (Nov. 1986).] Indeed, for highly variable waste operations, such as secondary lead
smelting, EPA's SW-846 method further requires hundreds of samples to be taken and analyzed to
obtain an adequate data baseline. [fn22: Id]

In contrast, as discussed above, BCI and ABR's composite sampling method combined
over a thousand individual grab samples that were averaged into 276 composite samples for lead
and 156 composite samples for barium and selenium. These composite samples represent an
average concentration obtained over a several month period on the effectiveness of stabilization
technologies.
Finally, comparison of "in versus out data points" is not meaningful when analyzing
secondary smelter slag. As discussed above, secondary smelter slag is heterogeneous; its
composition varies over time and space. It thus is wholly coincidental if incoming and treated slag
samples are true pairs. A sample taken one inch from another can have dramatically different
parameter concentrations.

Response

With respect to the comment on the data ignored by the Agency, itis noted that all of the
data ignored came from composite sampling. The Agency has made a consistent policy decision
for LDR not to look at average properties of a waste, but the entire waste. This best assures that
treatment minimizes threats posed by that waste's land disposal. The Agency notes that
composite sampling data are not appropriate for setting a treatment standard that is to be met
based on grab samples. The Agency believes that grab samples generally reflect maximum
process variability, and thus would reasonably characterize the range of treatment system
performance. (See discussion on this issue at 54 FR 26605, June 23, 1989 and 55 FR 22539,
June 1, 1990). The grab sampling also meets the ultimate objective of the LDR program, that all
of the hazardous waste to be land disposed be treated in a way that minimizes the threats that land
disposal could pose, not just the average portion of the waste to be so treated (a possible use of
composite sampling). The Agency views the authority to establish treatment standards based on
grab sampling to a settled issue. 976 F. 2d at 34. Moreover, smelter slags were already
prohibited in the very rule at issue in the ChemWaste decision, which upheld aithority to require
compliance based on grab sampling where that is the basis of the sampling used in developing the
treatment standard. Secondary smelter slag is extremely variable and heterogeneous. In
addition, since grab sampling is based on individual sampling events, it facilitates enforcement
proceedings for EPA and authorized states.
35

-------
       The Agency also notes that the treatment performance data obtained and reviewed by the
Agency from HTMR and stabilization facilities indicate that the TC metal-bearing wastes.
including smelter slags can be treated to the UTS. EPA also believes this data based to be amply
sufficient to make this determination, based in particular on the fact that the model treatment
facilities were treating particularly difficult to treat wastes (including secondary  lead slags)
containing very high concentrations of various metals (including lead).  EPA notes that the Fourth
Circuit, in the Kennecott decision from 1986, has rejected an argument that EPA is compelled to
use a particular data set to establish treatment standards because of the amount  of data in the data
set.  Here, the commenter was amply on notice of the type of data the Agency would be using to
establish the treatment standards, but ofr its own reasons provided a  different type of data.  The
Agency thus had ample grounds in basing the treatment standard on properly-documented
performance of properly-operated stabilization technology and HTMR technology. Likewise,
sampling of the wastes used to establish the treatment standards confirms that the facilitiy was
treating wastes as hard, or harder to treat, then the secondary lead slags referred to by the
commenter.

        In addition, the Agency  notes that the mere existence of the secondary lead smelting
industry itself proves the applicability of HTMR to D008 wastes. EPA repeats that, to compile
further evidence regarding the treatability of TC metal wastes, including D008 wastes, to the
UTS, the Agency conducted site visits to commercial hazardous waste treatment facilities and
collected additional stabilization and HTMR treatment performance data that better represent the
diversity of metal wastes than those previously used.  The treatment performance data (based on
grab samples) represented a wide range of metal-bearing wastes (both listed and characteristic)
that the Agency believes represents the most difficult to treat metal-bearing wastes, based on the
waste characteristics that affect the performance (WCAPS) of the treatment technology. The
types of waste treated included battery slag, mineral processing wastes, baghouse dust, soils, pot
solids, recycling by-products, and sludge.  These waste streams contained multiple metals which
would be representative of a characteristic waste with UHCs, and significant concentrations of
combination metals including: lead and cadmium, barium and lead, and chromium and antimony.
The Agency compared the treatment standards developed based on stabilization and HTMR and
selected the highest (less stringent) standard for each metal to establish the proposed UTS, thus,
allowing for process variability and detection limit difficulties.

       In response to the commenters issue on data selection, the Agency reviewed the data
submitted by the commenter, and found the data to be seriously lacking in form and quality
assurance/quality control prerequisites.  Specifically, the data submitted to the Agency were: (1)
based on composite samples rather than grab samples, the latter being the only type used to
develop BDAT treatment standards; (2) lacking in any quality assurance/quality control
documentation; and (3) not accompanied with adequate indication that treatment process was in
fact well designed and operated. Therefore, the Agency was unable to use the data for developing
the BDAT treatment standards.

       Also, the reference to SW-846  when analyzing heterogeneous waste.  The Agency
                                            36

-------
believes that grab samples generally reflect maximum process variability, and thus would
reasonably characterize the range of treatment system performance. (See discussion of this issue
at 54 FR 26605, June 23. 1989 and 55 FR 22539, June 1, 1990.

The Agency also would like to note that if a particular waste possesses unique properties
that make it more difficult to treat than the waste on which the standards are based, the affected
party may petition the Agency for a treatability variance as per 40 CFR 268.44 on a case-by-case
basis.

4. 2P4P-00037 CF Industries, Inc.

CF Industries, Inc. submits the following comments on the Environmental Protection
Agency's (hereinafter EPA or the agency) May 12, 1997 proposal entitled "Land Disposal
Restrictions Phase IV. - Second Supplemental Proposal on Treatment Standards for Metal Wastes
and Mineral Processing Wastes, Mineral Processing and Bevill Exclusion Issues, and the Use of
Hazardous Waste as Fill" (hereinafter second supplemental proposal). In addition, CF Industries,
Inc. endorses by reference comments submitted by The Fertilizer Institute and the Florida
Phosphate Council.

CF Industries, Inc. (hereinafter CF or the company) is an interregional farm supply
cooperative which is owned by and serves eleven regional cooperatives. CF manufactures
nitrogen and phosphate fertilizers and purchases potash fertilizer for resale. We distribute these
fertilizer products to over one million farmers and ranchers in 48 states and two Canadian
provinces.

CF currently manufactures phosphate fertilizer in Florida and has two phosphogypsum
stacks; one at its Plant City Phosphate Complex and one at its Bartow Phosphate Complex, which
is operating at reduced capacity. As the agency is aware, through the process of manufacturing
phosphate fertilizer, large quantities of phosphogypsum and process water are created. As a
result, CF would be adversely affected by an agency decision to take regulatory action that would
alter the scope of the
Bevill exclusion, which covers solid wastes from the extraction, beneficiation, and processing of
ores and minerals, including phosphate rock, from EPA's hazardous waste management
requirements.

CF strongly supports the agency's earlier determination, following intense study and
review, that Subtitle C regulation of the phosphate fertilizer industry was essentially unworkable
and unwarranted. In addition, the company believes that, without additional Congressional
action, the agency lacks the statutory authority to revisit its earlier determination in this matter.
Further, in response to the agency's question regarding the need for additional study or regulatory
controls (62 FR 26054), it is CF's belief, as supported by this document, that the state and local
regulatory framework makes any additional Federal controls unnecessary.
37

-------
       The State of Florida has developed a comprehensive regulatory program for the
management of phosphogypsum and process water.  CF's comments will illustrate to EPA that
Florida has developed a regulatory program that addresses environmental concerns relating to the
phosphate industry without the need for an overriding, inflexible, command and control Federal
regulatory structure, such as RCRA Subtitle C.  The company firmly believes that the EPA's
proposed reexamination of the Bevill exclusion  is not warranted and that the agency should once
and for all recognize that the Florida regulations provide adequate controls on the industry.

       The following is a discussion of various  issues related to the EPA's request for comments
concerning potential changes to the Bevill exclusion for phosphate rock mining and mineral
processing secondary materials.

Grab Sampling Versus Composite Sampling Would Be Overly Restrictive

       On page 26047 of the second supplemental proposal, the agency discusses its decision to
continue the use of grab sampling rather than the composite sampling procedure to determine
compliance with RCRA.  The use of a single grab sample versus a composite sample over the
duration and expanse of a manufacturing process could potentially produce information that
misrepresents the true nature of the material and its potential effect on the environment.  Certainly
where leakage of the material into the soils or groundwater is the mechanism of concern,  a grab
sample could grossly misrepresent the potential environmental impact while a composite sample
would be appropriate.  This could then result in the loss of the Bevill exempt status for the
secondary material management system.  CF urges the EPA to reconsider its  decision.
Response

       The Agency prefers that compliance with LDR standards for nonwastewaters be based on
grab samples (a one-time sample taken from any part of the waste), rather than composite samples
(a combination of samples collected at various locations for a given waste, or samples collected
over time from that waste). The Agency believes that grab samples generally reflect maximum
process variability, and thus would reasonably characterize the range of treatment system
performance.  (See discussion on this issue at 54 FR 26605, June 23, 1989 and 55 FR 22539,
June 1, 1990).  The grab sampling also meets the ultimate objective of the LDR program, that all
of the hazardous waste to be land disposed be treated in a way that minimizes the threats  that land
disposal could pose, not just the average portion of the waste to be so treated (a possible use of
composite sampling). In addition, since grab sampling is based on individual sampling events, it
facilitates enforcement proceedings for EPA and authorized states.

5.  2P4P-00038  The Fertilizer Institute

       These comments are submitted by The Fertilizer Institute ("TFI") on behalf of its member
companies in response to the United States Environmental Protection Agency's ("EPA" or
                                           38

-------
"Agency") May 12, 1997 proposal entitled "Land Disposal Restrictions Phase IV: Second
Supplemental Proposal on Treatment Standards for Metal Wastes and Mineral Processing Wastes.
Mineral Processing and Bevill Exclusion Issues, and the Use of Hazardous Waste as Fill," 62 Fed.
Reg. 26,041 (May 12, 1997) (hereinafter "second supplemental proposal"). EPA solicits
comment on potential changes to the exemption from EPA's hazardous waste management
requirements for certain secondary materials generated by the phosphate mining and mineral
processing industry. The exemptions were based on EPA regulatory determinations that
hazardous waste management requirements were not warranted. Consistent with EPA's
conclusion underlying its prior regulatory determinations, application of those requirements could
potentially threaten the industry's viability. As TFI explains herein, although EPA suggests
revisiting its prior regulatory determinations, the Agency lacks the statutory authority to do so;
EPA provides no explanation of whether the bases for its earlier phosphate industry regulatory
determinations have changed; EPA fails to recognize the significant environmental controls
currently imposed on TFI members; and EPA suggests it may conduct a reexamination of its
earlier determinations based on purportedly new risk information that is either not new or is an
inappropriate basis for reexamination. The continued exemption is critical to TFFs members as
they have expended substantial resources in reliance on EPA's prior regulatory determinations and
a change in that status is simply not warranted.

TFI is a non-profit trade association of the fertilizer industry. Its more than 200 member
companies manufacture in excess of 90 percent of all domestically produced fertilizer. The
second supplemental proposal potentially impacts, inter alia, the regulation of the mining and
mineral processing of phosphate rock. Most of the phosphate rock mined and processed in the
United States is used in the production of phosphoric acid for use in phosphate fertilizer.

One final issue of concern to TFI's member companies is EPA's continued requirement of
demonstrating compliance by grab rather than composite sampling. Grab sampling, in the context
of the large volumes of wastes generated in mining and mineral processing operations, could
result in a single analytical outlier result implicating the Bevill status of extremely large scale and
long-standing management systems for secondary materials. TFI believes such a consequence
should not stem from a single grab sample.

Response

The Agency prefers that compliance with LDR standards for nonwastewaters be based on
grab samples (a one-time sample taken from any part of the waste), rather than composite samples
(a combination of samples collected at various locations for a given waste, or samples collected
over time from that waste). The Agency believes that grab samples generally reflect maximum
process variability, and thus would reasonably characterize the range of treatment system
performance. (See discussion on this issue at 54 FR 26605, June 23, 1989 and 55 FR 22539,
June 1, 1990). The grab sampling also meets the ultimate objective of the LDR program, that all
of the hazardous waste to be land disposed be treated in a way that minimizes the threats that land
disposal could pose, not just the average portion of the waste to be so treated (a possible use of
39

-------
composite sampling). In addition, since grab sampling is based on individual sampling events, it
facilitates enforcement proceedings for EPA and authorized states.

6. 2P4P-00039 American Iron and Steel Institute (AISI)

The American Iron and Steel Institute ("AJSI"), whose member companies account for
approximately two-thirds of the steel production capability of the United States, welcomes this
opportunity to submit comments on the second supplement to the proposed Phase IV land
disposal restrictions ("LDR") rule for metal wastes and mineral processing wastes, which was
issued by the U.S. Environmental Protection Agency ("EPA") pursuant to the Resource
Conservation and Recovery Act ("RCRA"). 62 Fed. Reg. 26,041 (May 12, 1997). During the
course of their iron and steel making operations, AISI member companies generate a variety of
metal-bearing secondary materials, some of which may be subject to regulation under the
proposed rule. For this reason, AISI is keenly interested in the current supplemental proposal.
AISI's main comments are summarized briefly below.

B. Compliance with Treatment Standards Should be Based on Composite Samples,
Rather than Grab Samples

EPA has proposed to require that compliance with the LDR treatment standards be
demonstrated on the basis of grab samples, rather than composite samples. The only exception to
this rule is for the HTMR residues generated by two facilities: Horsehead Resource Development
Company, Inc. ("Horsehead") and International Metals Reclamation Company, Inc.
("INMETCO"). Indeed, the Agency states that its "ultimate intent" is to require even these two
facilities to comply with the treatment standards based on grab samples, rather than composite
samples.

AISI strongly opposes the use of grab samples for determining LDR compliance. EPA
has developed the UTS to reflect the performance of the BOAT technology. The "best"
technology, in turn, is the one that minimizes threats to human health and the environment. The
extent of such threats generally depends not on the instantaneous concentration and mobility of
hazardous constituents in a specific waste sample, but on the concentration and mobility of
hazardous constituents in an entire waste stream over time. For this reason, the proper indicator
of environmental performance is composite sampling, rather than grab sampling, and compliance
should be judged on this basis.

-------
       By focusing on grab samples, rather than composite samples, EPA may very well be
allowing treatment that does not minimize risks.  Consider, for example, a treatment technology
that generally reduces the concentration of a hazardous constituent to 1  ppm, but 10 percent of
the time can only reduce the concentration to 10 ppm. This technology  would clearly be better
from an environmental perspective than a different technology that consistently achieves a
concentration of 9 ppm.  Yet, a focus on grab samples would lead EPA to conclude that the latter
technology is "best."  The result would be increased environmental loadings of the hazardous
constituent by a factor of approximately 5.[fnlO: The first technology would result in an average
concentration of 1.9 ppm ((0.9xl.0)+(0.1x10.0)), while the second technology would result in an
average concentration of 9.0 ppm, an increase by a factor of 4.74 (9.0/1.9).]
       If EPA nevertheless persists in judging compliance with most LDR treatment standards on
the basis of grab samples, AISI urges the Agency to at least allow composite sampling for all
K061, K062, and F006 wastes. As EPA itself admits, the treatment standards for these waste
streams were based in part  on composite data. Thus, it would be inappropriate to require
compliance on a grab sample basis.  Moreover, EPA should not restrict the use of composite
sampling to two facilities, as it has proposed to do. The Agency has always maintained that once
the BOAT treatment standards are established for a waste, any technology that is capable of
meeting the standards (including both the BOAT and other technologies) can be used on the
waste.  However, by requiring different sampling protocols for different facilities, EPA is
effectively establishing different treatment standards for a single waste.  This approach is a
significant departure from prior EPA practice and could have the undesirable effect of favoring
one treatment technology over another, without  any environmental basis.  To avoid this problem,
EPA should allow all facilities (including those using both HTMR and stabilization technologies)
to demonstrate compliance with the treatment standards based either on grab samples or
composite samples.
Response

       The Agency prefers that compliance with LDR standards for nonwastewaters be based on
grab samples (a one-time sample taken from any part of the waste), rather than composite samples
(a combination of samples collected at various locations for a given waste, or samples collected
over time from that waste).  The Agency believes that grab samples generally reflect maximum
process variability, and thus would reasonably characterize the range of treatment system
performance. (See discussion on this issue at 54 FR 26605, June 23, 1989 and 55 FR 22539,
June 1,  1990). The grab sampling also meets the ultimate objective of the LDR program, that all
of the hazardous waste to be land disposed be treated in a way that minimizes the threats that land
disposal could pose, not just the average portion of the waste to be so treated (a possible use of
composite sampling).  In addition, since  grab sampling is based on individual sampling events, it
facilitates enforcement proceedings for EPA and authorized states.

       The BOAT scenario addressing two technologies where the first technology consistently,
90% of the time,  reduces the concentration of hazardous constituents to 1.0 ppm, and the second
                                           41

-------
technology consistently reduces the concentration of hazardous constituents to 9.0 ppm, both
based on grab sampling, the Agency would evaluate BOAT in the following manner. In
determining BOAT optimization of a treatment technology (process) is evident in the reduction of
hazardous constituents in the treated wastes. In determining process optimization we compare
levels concentrations of hazardous constituents in both the untreated and treated wastes. In the
evaluation of these two data results, after QA/QC, it would be determined that performance was
not optimized by the second technology, due to some type of interference. If the treated waste
results in a high level of hazardous constituents, it would establish that this technology has not
been optimized, and the waste should be retreated. (See BOAT Background Document QA/QC
Methodology dated June 23, 1991).

7. 2P4P-00041 Cyprus Amax Minerals Company

Cyprus Amax Minerals Company and its affiliates, including Cyprus Climax Metals
Company, Cyprus Miami Mining Corporation, Cyprus Sierrita Corporation, Cyprus Bagdad
Copper Corporation, Climax Molybdenum Company, Cyprus Foote Mineral Company and
Amax Gold, Inc. ("Cyprus Amax"), submit these comments on the Notice of Proposed
Rulemaking published by the U.S. Environmental Protection Agency ("EPA" or "the Agency")
entitled "Land Disposal Restrictions Phase IV: Second Supplemental Proposal on Treatment
Standards for Metal Wastes and Mineral Processing Wastes, Mineral Processing and Bevill
Exclusion Issues, and the Use of Hazardous Waste as Fill," 62 Fed. Reg. 26,041 (May 12, 1997),
commonly referred to as the "Second Supplemental Phase IV proposal." Cyprus Amax is a
diversified mining, mineral processing, and manufacturing company involved in the extraction,
beneficiation, and processing of a variety of minerals and metals, including copper,
molybdenum, coal, lithium, and gold.

VII. COMPLIANCE WITH LDR TREATMENT STANDARDS SHOULD BE
BASED ON COMPOSITE SAMPLES, RATHER THAN GRAB SAMPLES.

EPA has proposed to require that compliance with the LDR treatment standards be
generally demonstrated on the basis of grab samples, rather than composite samples. Cyprus
Ama x strongly opposes the use of grab samples for determining LDR compliance.

EPA has developed the UTS to reflect the performance of the "best, demonstrated,
available technology. The "best" technology, in turn, is the one that minimizes threats to human
health and the environment. The extent of such threats generally depend not on the instantaneous
concentration and mobility of hazardous constituents in a specific waste sample, but on the
concentration and mobility of hazardous constituents in an entire waste stream over time. For
this reason, the proper indicator of environmental performance is composite sampling, rather
than grab sampling, and compliance should be judged on this

EPA is simply wrong when it claims that "[grab] s'ampling is in keeping with the ultimate
objective of the land disposal restrictions program: that all of the hazardous waste to be land
42

-------
disposed be treated in a way that minimizes the threats that land disposal could pose." 62 Fed.
Reg. at 26,047. Instead, grab sampling focuses on the threats from a minute fraction of a total
waste of grab samples can be substantially skewed by outlier conditions, sample contamination
etc. Moreover, that small fraction is less likely from a statistical viewpoint to be representative
of the waste as a whole. EPA should allow composite sampling to demonstrate LDR
compliance.

Response

The land disposal restrictions (LDR) Program requires that all of the hazardous wastes to
be land disposed be treated in a way that minimizes the threats that land disposal could pose.
Compliance with LDR treatment criteria, being grab sampling, is based on the upper tolerance
limit for the pollutant concentration of the waste population; Whereas composite sampling or
average, is based on the upper confidence limit of the pollutant concentration of the waste. The
LDR Program prefers grab sampling because it is designed to try to assure that at least 99% of
the wastes hazardous pollutant concentration is below the limit. Because the LDR Program is
technology based, and reduces the concentration of hazardous constituents being land disposed
composite sampling, and averaging is unacceptable. (See Background Document BOAT QA/QC
Methodology, dated October 23, 1991). In establishing treatment standards the BOAT
Methodology is a statistical analysis that not only measures the reduction of hazardous constituent
concentrations in the untreated and treated residuals, but also reasonably characterizes the range
of treatment syste performance.
The Agency prefers that compliance with LDR standards for nonwastewaters be based on
grab samples (a one-time sample taken from any part of the waste), rather than composite samples
(a combination of samples collected at various locations for a given waste, or samples collected
over time from that waste). The Agency believes that grab samples generally reflect maximum
process variability, and thus would reasonably characterize the range of treatment system
performance. (See discussion on this issue at 54 FR 26605, June 23, 1989 and 55 FR 22539,
June 1, 1990). The grab sampling also meets the ultimate objective of the LDR program, that all
of the hazardous waste to be land disposed be treated in a way that minimizes the threats that land
disposal could pose, not just the average portion of the waste to be so treated (a possible use of
composite sampling). In addition, since grab sampling is based on individual sampling events, it
facilitates enforcement proceedings for EPA and authorized states.

8. 2P4P-00048 National Mining Association

The National Mining Association ("NMA") is the industry association representing the
producers of most of the nation's coal, metals, industrial and agricultural minerals; the
manufacturers of mining and mineral processing machinery, equipment, and supplies; and
the engineering and consulting firms, financial institutions and other firms serving the
coal and hard rock mining industry. These comments are submitted in response to the
Supplemental Notice of Proposed Rulemaking issued by the United States
Environmental Protection Agency ("EPA" or "the Agency") entitled "Land Disposal
43

-------
Restrictions Phase IV: Second Supplemental Proposal on Treatment Standards for Metal
Wastes and Mineral Processing Wastes, Mineral Processing and Bevill Exclusion Issues,
and the Use of Hazardous Waste as Fill," 62 Fed.  Reg . 26,041 (May 12, 1997),
commonly referred to as the "Second Supplemental Phase IV" proposal.

       Enclosed are the original and two copies of the comments of the National Mining
Association on the Environmental Protection Agency's May 12, 1997, proposed rule:
"Land Disposal Restrictions Phase IV: Second Supplemental Proposal on Treatment
Standards for Metal Wastes and Mineral Processing Wastes, Mineral  Processing and
Bevill Exclusion Issues, and the Use of Hazardous Waste as Fill" (62  Federal Register
26041, May 12, 1997) ("proposed rule").

       The National Mining Association (NMA) comprises the producers of most of the
nation's coal, metals, industrial and agricultural minerals; the manufacturers of mining
and mineral processing machinery, equipment and supplies; and the engineering and
consulting firms,  financial institutions and other firms serving the mining industry.

       The May  12 proposed rule is a complex and far-reaching piece of rulemaking.
Some of its major provisions include:

- prohibiting the use of non-virgin or "alternative" feedstocks in beneficiation and mineral
processing operations;

- prohibiting land storage of most mineral processing  secondary materials that are to be
recycled;

-"significantly narrowing the scope of the Bevill Amendment by altering the criteria for
determining what is or is not a Bevill waste;

- seeking comment on "reexamining" earlier RCRA Regulatory Determinations not to
impose RCRA hazardous waste regulation on extraction, beneficiation and 20 mineral
processing wastes; and

- establishing land disposal restriction (LDR) treatment standards for  mineral processing
wastes not subject to the Bevill Amendment (i.e., "newly identified" mineral processing
wastes)

       Overall, the proposal is environmentally counterproductive, exceeds the agency's
statutory authority, lacks virtually any support in the administrative record, and ignores
the demonstrated expertise of the  states in managing appropriate regulatory programs for
this industry's waste management  issues.  The proposed rule's impacts on NMA's
member companies, particularly those engaged in the extraction, beneficiation and
processing of non-coal minerals, would be direct, significantly adverse, and totally
                                           44

-------
unwarranted.
XI. COMPLIANCE WITH LDR TREATMENT STANDARDS SHOULD BE
BASED ON COMPOSITE SAMPLES. RATHER THAN GRAB SAMPLES.

EPA has proposed to require that compliance with the LDR treatment standards
be generally demonstrated on the basis of grab samples, rather than composite samples.
A grab sample is defined by EPA as a "one-time sample taken from any part of the
waste." NMA strongly opposes the use of grab samples for determining LDR
compliance.

EPA has developed the UTS to reflect the performance of the "best,
demonstrated, available technology." The "best" technology, in turn, is the one that
minimizes threats to human health and the environment. The extent of such threats
generally depend not on the instantaneous concentration and mobility of hazardous
constituents in a grab sample, but on the concentration and mobility of hazardous
constituents in an entire waste strewn over time. For this reason, the proper indicator of
environmental performance is composite sampling, rather than grab sampling, and
compliance should be judged on this basis.

EPA is simply wrong when it claims that "[grab] sampling is in keeping with the
ultimate objective of the land disposal restrictions program: that all of the hazardous
waste to be land disposed be treated in a way that minimizes the threats that land
disposal could pose." 62 Fed. Reg . at 26,047. Instead, grab sampling focuses on the
threats from a minute fraction of a total waste. Grab samples can be substantially
skewed by outlier conditions, sample contamination, etc. Moreover, that small fraction
is less likely from a statistical viewpoint to be representative of the waste as a whole.
Grab sampling is a non-probabilistic process which always introduces unaccountable
errors. EPA should provide that composite sampling be used to demonstrate LDR
compliance.

Response

Compliance with the land disposal restrictions (LDR) Program treatment criteria,
by grab sampling, is based on the upper tolerance limit for the pollutant concentration of
the waste population; whereas composite sampling or average, is based on the upper
confidence limit of the pollutant concentration of the waste. The LDR Program prefers
grab sampling because it is designed to try to assure that at least 99% of the waste
pollutant concentration is below the limit. (See Background Document BOAT QA/QC
Methodology dated June 23, 1991).
The Agency prefers that compliance with LDR standards for nonwastewaters be
based on grab samples (a one-time sample taken from any part of the waste), rather than
45

-------
composite samples (a combination of samples collected at various locations for a given
waste, or samples collected over time from that waste). The Agency believes that grab
samples generally reflect maximum process variability, and thus would reasonably
characterize the range of treatment system performance. (See discussion on this issue at
54 FR 26605, June 23, 1989 and 55 FR 22539, June 1, 1990). The grab sampling also
meets the ultimate objective of the LDR program, that all of the hazardous waste to be
land disposed be treated in a way that minimizes the threats that land disposal could
pose, not just the average portion of the waste to be so treated (a possible use of
composite sampling). In addition, since grab sampling is based on individual sampling
events, it facilitates enforcement proceedings for EPA and authorized states.

9.  2P4P-00054  Kennecott

VII.   Compliance with UTS Standards Should be Based on Representative Sampling.

       EPA proposes that grab samples (as opposed to composite or representative
samples) be used to judge compliance with the LDR'S.  A grab sample is defined by EPA
as a "one-time sample taken from any part of the waste." EPA attempts to justify this
requirement in two ways:
First, because "grab samples normally reflect maximum process variability, and thus
would reasonably characterize the range of treatment system performance," grab
sampling is somehow construed to be in keeping with the ultimate LDR objective of
minimizing the threat of land disposal, and
Second, grab  sampling evokes an "implementation advantage" since enforcement can be
based on individual sampling events by EPA, authorized states, or citizen groups.

       Unless amended, EPA's grab sampling proposal will unjustifiably tighten the UTS criteria
to which wastes must be treated and will deny the adoption of scientifically  sound sampling
protocol in preference to a sampling technique characterized by sampling experts as "a dangerous
form of gambling." [FN 25 25Pitard, Francis. Pierre Gy's Sampling, Theory and Sampling
Practice. Vol, II: Sampling, Correctness and Sampling Practice, CRC Press Inc., Boca Raton FL,
1989, pages 5-6.]

       In Chemical Waste Management vs- EPA, the Court set aside arguments that EPA's
decision to use grab samples to  enforce LDR standards was vague and arbitrary by relegating
sampling issues to individual facility disposal permits.  Unfortunately, the Court placed a
somewhat naive level of trust in EPA's ability to account for the inherent variability in waste
materials and different sampling methods "by adjustment of sampling results to account for the
inevitable variability of content." The Phase IV LDR proposal now offers EPA an opportunity to
clarify this issue by specifically addressing the technical defensibility (as opposed to vagueness or
arbitrariness)  of grab  sampling.

       The proposed UTS levels are based on grab samples of HTMR and  Stabilization wastes.
                                           46

-------
According to EPA, the demonstration of compliance with these proposed UTS criteria is,
therefore, also to be based on grab samples of the treated waste. Missing from this argument is
the fact that the initial determination of hazardous waste characteristics must be based on
representative samples of the original waste. A continuous string of logic can only be preserved if
the representativeness test is consistently applied, from initial hazardous determination, to UTS
development, to compliance determination. This is particularly important for mineral processing
wastes which are characterized by large volume, relatively low hazard, and variant composition
(both in particle size and chemical content).

A technically permissible grab sample from a mineral sample containing particles of at least
1.25 inches in diameter exceeds 2000 pounds. [FN 26 Newton, J., An introduction to Metallurgy,
John Wiley and Sons, Inc., London, page 464. ] To impose this requirement on regulatory
agencies or citizen groups seeking to determine compliance seems an unfair obligation. However,
without defining an adequate (or permissible) sample size, or even imposing the Agency's own
sampling guidance found in Chapter 9 of Test Methods for Evaluating Hazardous Waste, SW846,
grab samples could arguably be construed to consist of only the 100 grams required for a TCLP
test. If a grab sample weighing one pound is taken from a process treating 1000 tons per day of
waste, the sample represents only 0.04 seconds of production. Based on the inherent variability
of mineral processing wastes, combined with the stochastic variability of any treatment process,
an unfair obligation to always and uniformly achieve the UTS criteria is placed on the treatment
operator. The only way compliance could be achieved, so that "all of the hazardous waste to be
land disposed be treated in a way that minimizes the threats that land disposal could pose," would
be to design and operate treatment facilities to achieve average UTS levels far (perhaps orders of
magnitude) below those proposed by EPA. [FN 27 Federal Register, Vol. 62, No. 91, page
26047, emphasis added. ] Treatment processes are designed and operated to achieve average
results within a specified degree of variance. If a waste treatment operation is to be evaluated on
the basis of grab samples, 99% assurance that the UTS levels are being achieved in all portions of
the treated waste will force the facility to operate at average levels significantly below those for
which EPA has demonstrated feasibility. The very fact that UTS levels for these wastes are based
on grab samples from HTMR and Stabilization processes invalidates any assurance that the
proposed UTS levels can be achieved on either an average or maximum basis, or that UTS levels
are not arbitrary and capricious.
In its simplest form, grab sampling is described in sampling literature "as taking small,
equal portions by scoop or shovel at random or at regular intervals from the mass of material to
be sampled. When the amount of material is large and heterogeneous in nature (e.g. mineral
processing wastes), correct results can hardly be expected and it is more probable that samples
would run regularly high or regularly low." [FN 28 Behre, H.E. Handbook of Mineral Dressing -
Section 19, Sampling and Testing, John Wiley and Sons, NY, 1945, pages 24-25. ] In no sense
would the professional sampling community agree with EPA's definition that grab sampling is a
"one-time sample taken from any part of the waste."

Grab sampling is a non-probabilistic process which always introduces unaccountable errors.
"In other words, grab sampling delivers specimens not samples. Grab sampling was developed
47

-------
when sampling was a primitive art practiced only by a limited number of initiates who failed to
update their scientific knowledge. We definitely recommend to avoid them at any cost." [FN 29
Gy, Pierre M, Sampling, of Particulate Materials Theory and Practice, Elsevier Scientific
Publishing Co. Amsterdam - Oxford - New York, page 33. ] The errors incurred by grab sampling
are usually large enough to deprive the samples of any practical value. [FN 30 Ibid.]

Proper and scientifically defensible decisions, whether they are to build and operate a waste
treatment facility or to regulate the treated waste against a reasonable set of standards, can only
be supported by relying upon representative samples. EPA should not depart from this concept
(which is properly adopted in initially characterizing the wastes), for establishing LDR'S. The
formal adoption of "grab sampling," especially as defined by EPA, is scientifically indefensible and
imposes unfair and impractical obligations on both the regulated community and responsible
citizen groups.
Response

The commenter is somewhat confused with grab sampling, and the number of samples
required. Extent of sampling can be determined site-specifically, for example, in waste analysis
plans. See 976 F. 2dat31.
The proposed UTS levels are based on grab samples of HTMR and Stabilization wastes. Use
of grab sampling does not meas that only onw sample is taken. A grab sample, as defined by
EPA, as a one-time sample taken from any part of the waste. Grab sampling verifies consistency
and/or optimization in the process. Compliance with the LDR treatment criteria, being grad
sampling, is based on the upper tolerance limit for the pollutant concentration of the waste
population; whereas composite sampling or average, is based on the upper confidence limit of the
pollutant concentration of the waste. The LDR Program prefers grab sampling beause it is
designed to try to assure that at leas 99% of the hazardous waste pollutant concentration is below
the limit.
Treatment of the hazardous constituents prior to land disposal is preferred rather than
allowing dilution of hot spots. (See BDAT Background Documtne QA/QC Methodology dated
June 23, 1991)
The Agency prefers that compliance with LDR standards for nonwastewaters be based on
grab samples (a one-time sample taken from any part of the waste), rather than composite samples
(a combination of samples collected at various locations for a given waste, or samples collected
over time from that waste). The Agency believes that grab samples generally reflect maximum
process variability, and thus would reasonably characterize the range of treatment system
performance. (See discussion on this issue at 54 FR 26605, June 23, 1989 and 55 FR 22539,
June 1, 1990). The grab sampling also meets the ultimate objective of the LDR program, that all
of the hazardous waste to be land disposed be treated in a way that minimizes the threats that land
disposal could pose, not just the average portion of the waste to be so treated (a possible use of
composite sampling). In addition, since grab sampling is based on individual sampling events, it
facilitates enforcement proceedings for EPA and authorized states.!
48

-------
10.  2P4P-00056  Nexsen, Pruet, Jacobs & Pollard, LLP (Counsel to Macalloy
Corporation)

    As counsel for Macalloy Corporation ("Macalloy"), we are pleased to have this opportunity
to comment on the United States Environmental Protection Agency's ("USEPA's") proposed
Land Disposal Phase IV regulations (hereinafter referred to as the "Phase IV Proposal"). As the
only remaining manufacturer of ferrochromium in the United States, Macalloy is in a unique
position to address the impacts of these proposed regulations on mineral  processing operations
involving chromium-containing minerals. Since Macalloy does occupy a  unique position within
the mineral processing industry as a whole,  we will address these impacts and offer where
appropriate, suggested modifications, additions or deletions the Phase IV Proposal which will not
only be consistent with EPA's mandate under the Resource Conservation and Recovery Act and
the Hazardous and Solid Waste Management Amendments, [FN  1: For convenience, these  two
pieces of legislation will hereinafter be collectively referred to as RCRA.] but will also provide a
significant level of protection to human health and the environment. Macalloy has already
addressed its general objections to this, proposal under separate cover.
C.   USING INDIVIDUAL GRAB SAMPLING AS THE BASIS FOR LDR
COMPLIANCE  DETERMINATIONS IS IMPROPER, UNSCIENTIFIC, AND
CONFLICTS WITH EPA     GUIDANCE ON SAMPLING AND ANALYSIS OF
SOLED WASTE STREAMS AND    DATA QUALITY OBJECTIVES

     In the preamble, EPA states that individual grab samples, rather than composite samples or
statistically reduced results from multiple samples, are to be used for determining compliance with
LDR standards. Macalloy believes that relying on single, individual  samples for the purpose of
making such a compliance determination- is improper, unscientific, and in conflict with EPA
guidance on  sampling and analysis of solid waste streams and EPA guidance on data quality
objectives. EPA justifies its position by stating that "grab samples normally reflect the maximum
process variability, and thus would reasonably characterize the range of treatment system
performance. "  However, in allowing compliance with LDR standards to be determined on the
basis of a single grab sarr pie EPA fails to account for variations inherent in sampling and
analytical techniques. It  s quite possible that analysis of a single grab sample which indicates that
LDR standards have not been achieved could be the result of sampling error or analytical error
rather than any variability in the treatment process itself. Further more, allowing compliance
determinations to be made on the basis of a single grab sample contradicts EPA's diligent efforts
in other areas to ensure that data quality objectives are identified and achieved.

Response

     The commenter is somewhat confused with grab sampling, and the number of samples
required. Use of grab sampling does not mean that only one or a single grab is taken. A grab
sampl as defined by EPA, is a one-time sample taken from any part of the waste, thereby verifying
                                           49

-------
consistency or optimization of the technology and the process.
Compliance with LDR treatment criteria, being grab sampling, is based on the upper
tolerance limit for the pollutant concentration of the waste. The LDR Program prefers grab
sampling because it is designed to try to assure that at least 99% of the waste pollutant
concentration is below the limit. Therefore, since the LDR Program is interested in the upper
bound of the pollutant concentration, composite sampling, and averaging technique, in
unacceptable. Treatment of the hazardous pollutant prior to land disposal is preferred rather than
allowing dilution of hot spots. (See BOAT Background Document QA/QC Methodology dated
June 23, 1991).
The Agency prefers that compliance with LDR standards for nonwastewaters be based on
grab samples (a one-time sample taken from any part of the waste), rather than composite samples
(a combination of samples collected at various locations for a given waste, or samples collected
over time from that waste). The Agency believes that grab samples generally reflect maximum
process variability, and thus would reasonably characterize the range of treatment system
performance. (See discussion on this issue at 54 FR 26605, June 23, 1989 and 55 FR 22539,
June 1, 1990). The grab sampling also meets the ultimate objective of the LDR program, that all
of the hazardous waste to be land disposed be treated in a way that minimizes the threats that land
disposal could pose, not just the average portion of the waste to be so treated (a possible use of
composite sampling). In addition, since grab sampling is based on individual sampling events, it
facilitates enforcement proceedings for EPA and authorized states.
Furthermore, the Agency's policy for compliance with LDR requires that a waste be treated
in its entirety to the required treatment standards, not that the waste be treated on an average
basis. Consistent with this policy, the Agency has calculated treatment standards based on
individual grab samples such that the standards developed reflect the maximum variability of the
treated wastes and not the average. By setting standards at the upper bound of performance, any
random sample from adequately treated wastes should be found to be in compliance.

The facility bears the burden to show that all wastes have been treated in compliance with
the promulgated standards. Treatment facilities must test their wastes according to the frequency
specified in their waste analysis plans (40 CFR 268.7(b)). In developing the plan, verification
frequency should avoid the generation of significant amounts of non-compliant wastes, which
would require retreatment or additional treatment.

11. 2P4P-00068 Horsehead Resource Development Company, Inc.

Horsehead Resource Development Company, Inc. ("FIRD") submits these comments on
the U.S. Environmental Protection Agency's ("EPA" or "Agency") Land Disposal Restrictions
("LDR") Phase IV supplemental proposed rule. 62 Fed. Reg. 26041 (May 12, 1997). HRD
operates high temperature metals recovery ("HTMR") recycling operations in several states, and
is nation's largest recycler of electric arc furnace dust (KO61). F£RD represents a substantial
portion of the domestic recycling capacity for K061. Indeed, FIRD provides an essential recycling
service, having recycled over 3.8 million tons of K061 since its inception, and is a substantial
contributor to U.S. zinc production.
50

-------
EPA's Phase IV proposal includes two major elements upon which HRD will comment.
First, HRD will comment on EPA's proposal to modify the universal treatment standards ("UTS")
for inorganic constituents in two respects, including changing the UTS levels and requiring use of
grab sampling data as the basis for implementing and enforcing the UTS for K061, K062, and
F006. Second, HRD will comment on EPA's proposed ban on "fill material.

EPA's Phase IV LDR supplemental proposed rule includes two proposals that EPA should
reconsider. First, EPA should retain the composite sampling requirement for determining
compliance with the inorganic UTS for K061, K062, and F006, since composite sampling
provides more accurate and representative information on material composition, and the regulated
community and EPA have successfully implemented and enforced the UTS on the basis of
composite sampling for a number of years. Second, EPA should withdraw the proposed
prohibition on a 11 fill material, since the proposal in unnecessarily duplicative of other policies and
regulations, ambiguous and unduly expansive, inconsistent with other Agency rulemakings, and
without scientific basis. More specifically, HRD provides the following recommendations on the
LDR Phase IV supplemental proposed rule:

•Should EPA proceed with the grab sampling protocol, the Agency must ensure that all major
technologies, including HTMR, can achieve the revised UTS on the basis of grab sampling.

Proposed UTS Revisions

• EPA should retain composite sampling as the basis for compliance with the UTS for K061,
K062 and F006. Grab sampling is technically unsound, since only composite sampling
results in an accurate characterization of the material.
• Should EPA proceed with the grab sampling protocol, the Agency must ensure that all major
technologies, including HTMR, can achieve the revised UTS on the basis of grab sampling.
• EPA's current plan for implementing the UTS will be impracticable to implement and
unlawful. As explained below, the regulated community will be subject to UTS that are
based on a mix of composite and grab sampling data, yet are implemented and enforced
solely on the basis of grab sampling, which is contrary to EPA policy and unlawful.
• EPA should rectify this problem by continuing to allow compliance with the current UTS for
K061, K062 and F006 on the basis of composite sampling until all relevant states have
adopted all of the UTS.
• Alternatively, should EPA decline to extend the availability of composite sampling during the
transitional period, EPA should determine that all of the revised UTS are more stringent than
the current UTS, since grab sampling methodology is more stringent than composite
sampling methodology. This would ensure that all of the new UTS are implemented
concurrently.

HTMR's significant role in the development of the UTS is important because EPA has
proposed to change the UTS in two respects that affect HTMR facilities. First, EPA proposes to

-------
require that compliance with and enforcement of the revised UTS for inorganic constituents be
based exclusively on grab sampling, even though composite sampling was used to develop most
of the inorganic UTS, and composite sampling ensures accurate characterization of the material.
Second, EPA proposes to change many of the levels for inorganic UTS, in large part to allow
stabilization technologies to meet the UTS. As explained below, HRD recommends that EPA
retain composite sampling as the basis for compliance with the current inorganic UTS, particularly
for K061, K062 and F006. Should EPA nevertheless decide to revise the UTS, EPA should
ensure that any changes (including changes to either the levels or the sampling basis) be
achievable by all major technologies, including HTMR. HRD also explains that EPA's proposed
changes to the UTS will be impracticable to implement, and unlawful, unless the Agency takes
steps (explained below) to ensure consistent implementation of the UTS by requiring a single
sampling methodology during the transition period for the proposed new UTS.

A. EPA Should Retain Composite Sampling as the Basis for Compliance with the UTS.

EPA should retain the composite sample requirement for K061, K062 and F006. As EPA
has acknowledged, grab sampling does not provide accurate and representative information on
material composition. In contrast, a composite sampling protocol ensures that the samples
accurately reflect the material's composition. Indeed, EPA has established a general requirement
for representative sampling, which is possible only through composite samples. As EPA explains
in its sampling protocol manual (SW-846), " [the EPA, in its hazardous waste management
system, has required that certain solid wastes be analyzed for physical and chemical
properties [and] requires that representative samples of waste be collected and defines
representative samples as exhibiting average properties of the whole waste. [FN2: Test Methods
for Evaluating Solid Waste, Physical/Chemical Methods § 9. 1. 1.1 (EPA Pub. SW-846 Third
ed.).]

Despite EPA's assertion that it prefers grab sampling (62 Fed. Reg. at 26047), composite
sampling is the basis for most of the existing inorganic UTS, which first became effective in 1988.
EPA and the regulated community have implemented and enforced composite sampling since that
time without any environmental issues or implementation problems. Composite sampling also is
consistent with the sampling requirements contained in most waste analysis plans of companies
subject to regulation under RCRA. Moreover, EPA requires and/or encourages composite
sampling in many situations where representative sampling is performed for verification against
particular EPA specifications and standards. For example, EPA delistings require composite
sampling to verify compliance with the exclusion. [FN3: 1 E.g., 58 Fed. Reg. 6925, 6933 (Feb.
3, 1993); 57 Fed. Reg. 37927, 37935 (Aug. 21, 1992).]

Finally, no environmental issues have arisen from the use of composite sampling to determine
compliance with the UTS. Accordingly, EPA should allow use of composite sampling data to
determine compliance with the UTS.

B. The UTS Must Be Achievable by all Major Technologies.
52

-------
Should EPA nevertheless choose to revise the UTS, and to require grab sampling as the basis
for compliance with them, it must ensure that all major technologies can achieve the revised UTS
on the basis of grab sampling. This prerequisite is consistent with EPA's explicitly stated policy
that the UTS be "achievable by all of the major HTMR technologies." 56 Fed. Reg. 41164, 41169
(Aug. 18, 1991) (K061 standards). Inconsistently, however, EPA states in the Phase IV LDR
proposal that "it appears that with the new UTS levels proposed in this notice, . . . HTMR
facilities should be able to meet UTS on a grab sampling basis. " 62 Fed. Reg. at 26047. EPA is
not currently in a position to state that all major technologies and facilities are able to meet the
UTS on the basis of grab sampling data. Given EPA's stated objective that all major technologies
meet the treatment standards, promulgating the proposed UTS without a demonstration that all
major technologies can meet the standards would be arbitrary and capricious.
Response

The commenter is somewhat confused with grab sampling, and the number of samples
required. Use of grab sampling does not mean that only one or a single grab sample is taken. A
grab sample as defined by EPA, is a one-time sample taken from any part of the waste, thereby
verifying consistency or optimization of the pppprocess.
Compliance with LDR treatment criteria, being grab sampling, is based on the upper tolerance
limit for the pollutant concentration of the waste pollutant; whereas composite sampling or
average, is based on the upper confidence limit of the pollutant concentration of the waste. The
LDR Program prefers grab sampling because it is designed to try to assure that at least 99% of
the waste pollutant concentration is below the limit. Therefore, since the LDR Program is
interested in the upper bound of the pollutant concentration, composite sampling, an averaging
technique, is unacceptable. Treatment of the hazardous pollutants ppprior to land disposal
preferred rather that allowing dilution of the hot spots.
(See BOAT QA/QC Methodology Background Document, dated October 23,1991).
The Agency prefers that compliance with LDR standards for nonwastewaters be based on
grab samples (a one-time sample taken from any part of the waste), rather than composite samples
(a combination of samples collected at various locations for a given waste, or samples collected
over time from that waste). The Agency believes that grab samples generally reflect maximum
process variability, and thus would reasonably characterize the range of treatment system
performance. (See discussion on this issue at 54 FR 26605, June 23, 1989 and 55 FR 22539,
June 1, 1990). The grab sampling also meets the ultimate objective of the LDR program, that all
of the hazardous waste to be land disposed be treated in a way that minimizes the threats that land
disposal could pose, not just the average portion of the waste to be so treated (a possible use of
composite sampling). In addition, since grab sampling is based on individual sampling events, it
facilitates enforcement proceedings for EPA and authorized states.
This commenter submitted some 152 data point based only on grab sampling. Of these, 100%
showed complaince with the proposed metal standard after removal of the high statistical outliers.
Indeed the great majority of the data submitted showed as much as an order of magnitude below
53

-------
the promulgated standard. (See Phase IV Final rule Background Document, Memorandum dated
December 17, 1997 and data evaluation of the Horsehead Research Development grab data
submittal of 152 data points). Horseheads own data corrobarates the Agency proposed
standards.
In addition, EPA has revised the treatment standards for antimony, beryllium, cadmium,
chromium, nickel, and silver. Treatment standards for antimony, beryllium, and silver were
revised based on new data provided by commenters, that represented the "most difficult to treat"
wastes. Thus, the Agency is promulagting revised UTS for antimony, beryllium, and silver at 1.15
mg/1, 1.22 mg/1, and 0.14 mg/1 respectively. For cadmium, chromium, and nickel, the Agency has
identified a technical error in the BOAT determination of the proposed standards. In applying the
BOAT methodology for calculating the treatment standard, EPA failed to perform a "Z-score"
outlier test. The Agency corrected this error and has re-calculated the treatment standards for
cadmium, chromium, and nickel at 0.11 mg/1, 0.60 mg/1, and 11 mg/1 respectively. Although these
standards are more stringent than the proposed standards, based on the treatment performance
data reviewed, the Agency believes that these new standards are also achievable by commercial
treatment technologies such as stabilization and HTMR. (See the BDAT background materials in
the docket for today's rule, for additional information on the revised treatment standards.)

12. 2P4P-00069 Macalloy Corporation

The proposed rule requires that grab sampling shall be used instead of composite sampling
when analyzing for compliance with the LDR standards. Yet it excludes certain wastes (K061,
K062, and F006) since composite data was used instead of grab sampled data for the creation of
the current treatment standards. What kind of data was used for D007? This aspect of the
proposed rule not only contradicts but invalidates the U.S. EPA authored Test Methods for
Evaluating Solid Waste - SW 846, Chapter 9. By using the analysis from a grab sample instead of
a composite sample, the technique cannot account for unexplained data anomalies. SW 846
Chapter 9 explains that "It is important to emphasize that a haphazardly selected sample is not a
suitable substitute for a randomly selected sample. That is because there is no assurance that a
person performing undisciplined sampling will not consciously or subconsciously favor the
selection of certain units of the population, thus causing the sample to be unrepresentative of the
population". One of the reasons stated for requiring grab sampling is because "there is an
implementation advantage to use of grab sampling, since enforcement for EPA, authorized states.
or citizen groups is facilitated if enforcement can be based on individual sampling events (as
occurs with grab sampling)". SW 846 Chapter 9 goes on to say that "accuracy is achieved by
incorporating some form of randomness."
Response

The Agency notes that all the UTS for TC metals, including chromium (D007), promulgated
in today's rule are based on treatment performance data obtained from grab samples. In addition,
the Agency has identified a technical error in the BDAT determination of the proposed chromium
54

-------
standard. In applying the BOAT methodology for calculating the treatment standard, EPA failed
to perform a "Z-score" outlier test. The Agency corrected this error and re-calculated the
chromium treatment standard and is promulgating the UTS for chromium at 0.60 mg/1 TCLP.
Although this standard is more stringent than the proposed standard, based on the treatment
performance data reviewed, the Agency believes that this new standard is also achievable by
commercial treatment technologies such as stabilization and HTMR.

With respect to the comment of sampling, the Agency prefers that compliance with LDR
standards for nonwastewaters be based on grab samples (a one-time sample taken from any part
of the waste), rather than composite samples (a combination of samples collected at various
locations for a given waste, or samples collected over time from that waste). The Agency
believes that grab samples generally reflect maximum process variability, and thus would
reasonably characterize the range of treatment system performance. (See discussion on this issue
at 54 FR 26605, June 23, 1989 and 55 FR 22539, June 1, 1990). The grab sampling also meets
the ultimate objective of the LDR program, that all of the hazardous waste to be land disposed be
treated in a way that minimizes the threats that land disposal could pose, not just the average
portion of the waste to be so treated (a possible use of composite sampling). In addition, since
grab sampling is based on individual sampling events, it facilitates enforcement proceedings for
EPA and authorized states.

13. 2P4P-00075 United States Department of Defense (DoD)

DoD Comment - DoD opposes the use of grab sampling to determine compliance with LDR
treatment standards. As noted in the preamble, EPA's current approach considers any single grab
sample excedence of a standard to be a violation, rather than basing violation determinations on
an average or composite sampling. DoD is concerned that this enforcement approach is overly
stringent and is inconsistent with the concept of "representativeness" that is the basis for other
sampling and analysis requirements in the RCRA program. In the preamble to the HWIR
.proposed rule (60 FR 66344), EPA addresses a similar issue concerning sampling to determine
compliance (pages 66386-66387). In that discussion, EPA acknowledged, that a composite
sample could be used to determine an excedence of a standard (in that case, an "exit level"):

EPA believes s important to retain the practical approach whereby a single composite
sample of a waste a^ some arbitrary point in time or space during a short visit is considered
sufficient for enforcement purposes. (60 FR 66387; emphasis added)

EPA's stringent approach may also serve as a disincentive to voluntary contamination clean up
efforts. Contamination in soil and other media generated through remediation may be highly
variable. Concern that large amounts of treated soil could be deemed as exceeding the treatment
standard because of one grab sample might discourage some clean-up efforts that would
otherwise provide a net benefit to the environment.

Even if use of a composite sample failed to show an excedence of a treatment standard
55

-------
somewhere in the waste stream, the excedence would represent only a small portion of the overall
waste stream. The presence in a landfill of such a minor amount of waste above the treatment
standard would not pose a threat to human health or the environment, since any liquid passing
through the landfill cell would not concentrate on that particular portion of the waste.

Response

The Agency prefers that compliance with LDR standards for nonwastewaters be based on
grab samples (a one-time sample taken from any part of the waste), rather than composite samples
(a combination of samples collected at various locations for a given waste, or samples collected
over time from that waste). The Agency believes that grab samples generally reflect maximum
process variability, and thus would reasonably characterize the range of treatment system
performance. (See discussion on this issue at 54 FR 26605, June 23, 1989 and 55 FR 22539,
June 1, 1990). The grab sampling also meets the ultimate objective of the LDR program, that all
of the hazardous waste to be land disposed be treated in a way that minimizes the threats that land
disposal could pose, not just the average portion of the waste to be so treated (a possible use of
composite sampling). In addition, since grab sampling is based on individual sampling events, it
facilitates enforcement proceedings for EPA and authorized states.

14. 2P4P-OL003 Environmental Technology Council

The Environmental Technology Council (ETC) is a national trade association of firms
engaged on a commercial basis in the recycling, treatment, and disposal of hazardous and
industrial wastes. The Council's member companies provide high-technology methods for the safe
and effective treatment, reclamation and disposal of waste materials; remedial response at
uncontrolled waste sites; and waste treatment equipment. The Council has been actively involved
in every major LDR rulemaking since the inception of this regulatory program. Council members
provide incineration, treatment, and stabilization services that will be directly affected by the
issues raised in this LDR Phase IV supplemental proposal.

6. COMPOSITE VERSUS GRAB SAMPLE TO DEMONSTRATE LDR COMPLIANCE

The ETC feels that for practicality's sake, EPA should revisit its position on grab samples
versus composite samples as the basis for LDR treatment standards. EPA maintains that grab
sampling is required to demonstrate compliance with LDR standards. The Council feels strongly
that composite samples are better indicators of compliance and a more accurate basis for setting
treatment standards.

Historically, EPA has maintained that a number of grab samples taken from a continuous
process stream over a period of time will give better information for calculating process variability
than a composite sample of the same process over the same period of time. However the EPA
should consider that LDR compliance for the treatment of waste streams that are not uniform in
their physical and chemical properties is best determined by composite sampling. Reliance on a
56

-------
grab sample taken from a common point in the process is less representative of the treatment
residue than a composite sample taken from a number of points that represents a cross-section of
the process stream.

This is particularly true for "dewatered.1 sludges, incinerator ash, debris, and other mixtures of
solids that are typically handled and processed in batches rather than continuously flowing
streams. Considering that most waste material destined for landfill disposal will be transported in
a roll-off box or similar transportation equipment, representative sampling requires taking
individual samples of the material from different sections of the container. These samples are then
composite into a single sample to fairly represent the waste stream in the container. Analysis of
each separate sample that comprises the composite does not adequately tell the final treatment
operator if the waste stream meets LDR standards or how the waste stream will respond to
treatment if it doesn't already meet standards. Only the analysis of a representative composite
sample provides this information to the treater or land disposal facility.

In the past, EPA has maintained that grab sampling is required for the purposes of
enforcement. While ETC understands the Agency's position in this area, we feel it should not
preclude the use of composite samples to demonstrate LDR compliance. Scientific validity, not
legal simplicity, should be the basis for establishing waste treatment standards and compliance
with them.

Response

The Agency prefers that compliance with LDR standards for nonwastewaters be based on
grab samples (a one-time sample taken from any part of the waste), rather than composite samples
(a combination of samples collected at various locations for a given waste, or samples collected
over time from that waste). The Agency believes that grab samples generally reflect maximum
process variability, and thus would reasonably characterize the range of treatment system
performance. (See discussion on this issue at 54 FR 26605, June 23, 1989 and 55 FR 22539,
June 1, 1990). The grab sampling also meets the ultimate objective of the LDR program, that all
of the hazardous waste to be land disposed be treated in a way that minimizes the threats that land
disposal could pose, not just the average portion of the waste to be so treated (a possible use of
composite sampling). In addition, since grab sampling is based on individual sampling events, it
facilitates enforcement proceedings for EPA and authorized states.
57

-------
Phase TV Second Supplemental Proposed Rule:
Comments and Responses on Applying UTS, LDR to Mineral Processing Wastes
1. 2P4P-00048 National Mining Association

New Information on the Inappropriateness of Applying the Toxicity Characteristic to Mineral
Processing. Wastes:

EPA has not expressly requested comments on applying the RCRA Toxicity Characteristic
(TC) to mineral processing wastes. Nonetheless, new developments have occurred since EPA's
January 25, 1996 proposal to reapply the TC to mineral processing wastes. In deciding whether
to make that proposed rule final, EPA must consider these new developments, including EPA's
final military munitions rule and the D.C. Circuit's decision in Dithiocarbamate Task Force v. EPA
(98 F. 3d 1394, 1996).

As NMA's comments make clear, in light of these new developments there can no longer be
any doubt that EPA should abandon its misguided efforts to apply the TC to mineral processing
wastes. Instead, the Synthetic Precipitation Leachate Procedure (SPLP), modified as suggested in
NMA's comments, should be used to evaluate the hazardousness of wastes from the mining and
mineral processing industry.

XIV. THE TOXICITY CHARACTERISTIC CANNOT BE APPLIED TO MINERAL
PROCESSING WASTES.

NMA recognizes that application of the Toxicity Characteristic ("TC") to mineral processing
wastes is not one of the issues that EPA has explicitly requested comments on in its May 12, 1997
proposal. However, a number of recent developments have occurred that the Agency must take
into account as it decides whether to finalize its January 25, 1996 proposal to reapply the TC to
mineral processing wastes. These developments include the decision of the U.S. Court of Appeals
for the District of Columbia Circuit in Dithiocarbamate Task Force v. EPA. 98 F.3d 1394 (D.C.
Cir. 1996), and the final military munitions rule promulgated by EPA on February 12, 1997, 62
Fed. Reg. 6622. The relevance of these developments to the current rulemaking is discussed
below.

A. EPA Must Consider the Full Range of Factors Relating to the Potential
Hazardousness of Mineral Processing Wastes, Not Just Waste Mismanagement Scenarios.

In proposing to apply the TC to mineral processing wastes, EPA has focused exclusively on
one issue ~ whether the mismanagement scenario that underlies the TC is plausible for such
wastes. 61 Fed. Reg. 2338, 2354 (Jan. 25, 1996). The statute and EPA's own regulations,
however, require more. Because the Agency has not considered all of the required factors, it
cannot finalize its proposal.
58

-------
The D.C. Circuit in Dithiocarbamate Task Force held that, in deciding whether to designate a
waste as hazardous, EPA must consider a broad range of factors. Although this conclusion was
reached in the context of listing hazardous wastes, it applies equally to the identification of
hazardous waste characteristics. The statute itself provides that EPA's identification and listing
decisions must "take] into account toxicity, persistence, and degradibility in nature, potential for
accumulation in tissue, and other related factors." 42 U.S.C. § 6921 (a). EPA has specified at
least some of these other factors in its regulations. See 40 C.F.R. § 26 1.1 I (a)(3).

Although the full list of factors is set forth in a provision relating to hazardous waste listings,
the D.C. Circuit noted that "the very question that the ten factors ... are supposed to help answer"
is whether the particular wastes meet the statutory definition of hazardous waste. Dithiocarbamate
at 1400. Because that definition applies to wastes that are designated as hazardous by
characteristic, as well as by listing, the full list of factors clearly applies to both identification of
characteristics and listing of hazardous wastes.

EPA itself has recognized this fact since the very beginning of the hazardous waste regulatory
program. In 1980, the Agency explicitly rejected certain characteristic tests because "it
questioned whether these tests sufficiently took into account the multiple factors which bore on
the question of the hazardousness of [the] wastes." 45 Fed. Reg. at 33,105 (May 19, 1980).
EPA, in fact, acknowledged that it was more important to consider a broad range of factors when
identifying a hazardous waste characteristic than when listing a waste. According to the Agency:

The identification mechanism has its limitations and must be used with considerable care. First
and foremost, this mechanism must only employ characteristics which, in and of themselves,
sufficiently define the properties of a waste that cause the waste to meet the definition of
hazardous waste prescribed in Section 1004(5) of RCRA. In other words, a solid waste that
possesses the property or properties described in an established characteristic must meet the
statutory definition of hazardous waste without consideration of any other properties of or factors
-about the waste or its constituents. When properties or factors outside those defined in the
characteristic must be considered to reach a determination that a waste meets the statutory
definition of hazardous waste, a characteristic is not sufficient for the purpose of designating a
hazardous waste.

Criteria Background Document at 16. [FN 23: EPA explained that one of the main reasons for
requiring all factors to be fully considered in identifying a characteristic is that the characteristic
mechanism, unlike the listing mechanism (which includes a delisting procedure), provides "no
opportunity" for generators to "demonstrate ... that their waste is in fact not hazardous." 45 Fed.
Reg. at 33,107. Instead, "the test prescribed in the characteristic constitutes a final determination
of hazard." Id. Moreover, wastes that otherwise would have to be listed based on consideration
of all the specified factors can be listed solely on the grounds that they exhibit a characteristic.
See 40 C.F.R. § 261.1 l(a)(l).The only way to ensure that the characteristics do not become a
means for circumventing the listing factors is to require those factors to be considered in the
59

-------
identification of the characteristics.] Applying this principle in practice, EPA placed more of an
emphasis on at least two factors (i.e... waste constituent migration and subsequent environmental
fate) when developing the hazardous waste characteristics than when developing the hazardous
waste listings. See 45 Fed. Reg. at 33,107.

EPA cannot now read the criteria for identifying hazardous waste characteristics in a different
way, because doing so would make those criteria unlawful. In the absence of the factors, EPA's
criteria for the hazardous waste characteristics would do little more that parrot the statutory
definition of hazardous waste. Compare 40 C.F.R. § 261.10(a)(l) with 42 U.S.C. § 6903(5).
However, the mere articulation of the statutory definition of hazardous waste as the basis for
identifying a characteristic would be circular and would constitute an abrogation of EPA's
statutory duty to establish criteria for characteristics which expand on the statutory definition.
See 45 Fed. Reg. 33,106 (citing with approval comments that made this point in the context of
the listing criteria). Moreover, this approach would fail to take into account the factors that are
specifically mentioned in the statute. Id.

Clearly, EPA is required to consider the full range of factors specified in 40 C.F.R. § 26 1.1 I
(a)(3) when identifying a hazardous waste characteristic. Because the Agency has only
considered a single factor (i.e.. mismanagement scenarios) in its proposed rule to apply the TC to
mineral processing wastes, it cannot go forward with the proposal. [FN 24: EPA also cannot rely
on past consideration of the factors when the TC was first developed. As an initial matter, it is
not clear that the factors were properly considered in the earlier rulemakings. Moreover, many of
the factors may have changed since the time of those rulemakings (i.e... other regulations affecting
mineral processing wastes may have been amended, and scientific knowledge about the migration
of constituents in the environment may have improved).] If EPA plans to remedy its effort and
consider the full range of factors before issuing a final rule, it must provide the public notice of
and an opportunity to conunent on the Agency's analysis of the factors, as required by the
Administrative Procedure Act.

B. EPA Has Failed to Consider Properly the Plausible Management Scenarios for Mineral
Processing Wastes.

Not only has EPA failed to consider virtually all of the requisite factors in determining
whether to apply the TC to mineral processing wastes, but it also has failed to consider properly
the one factor that it did address — mismanagement scenarios. In Dithiocarbamate Task Force the
D.C. Circuit declared that "EPA seems to have turned the mismanagement factor upside down,
from an inquiry into whether dangerous management practices are 'plausible'... into an inquiry
into whether they have been ruled out absolutely." 98 Fed. Reg, at 1405. The Agency is making
precisely the same error in attempting to apply the TC to mineral processing wastes.

The TC is premised on the assumption that the wastes being evaluated under the characteristic
will be disposed in a municipal solid waste ("MSW") landfill. In an effort to manufacture a
rationale for applying the TC to mineral processing wastes, EPA has searched extensively and
60

-------
come up with only a few alleged examples of what it refers to as "likely," "possible," or
"potential" cases of co-disposal with municipal wastes. As NMA noted in its comments on the
January 25, 1996 Supplemental Phase IV proposal, however, none of EPA's alleged examples
demonstrate that co-disposal has ever taken place. For that reason, EPA cannot apply the TC to
mineral processing wastes.

Moreover, EPA's efforts to find a few cases where mineral processing wastes have been
disposed in MSW landfills are completely misguided, because even if such cases exist they would
not justify application of the TC to mineral processing wastes. The D.C. Circuit made clear in
Dithiocarbamate Task Force that a mismanagement scenario is not "plausible" just because it has
not been "ruled out absolutely." 98 FR at 1405. The Court explicitly recognized that "complete
certainty is not possible." Id. at 1404.

According to the D.C. Circuit, "one should bear in mind that the ultimate question ... is
whether [a] waste poses a 'substantial1 hazard in light of the various possibilities of improper
management." 98 F.3d at 1400; See also Criteria Background Document at 54 ("The Agency ...
recognizes that it must be reasonable in selecting the [mismanagement scenario]; the statute
requires that the potential hazard posed by the waste be substantial before a waste is hazardous").
Since the beginning of the RCRA regulatory program, EPA has acknowledged that "[hazards]
arising from wholly unrealistic or improbable waste management... are not substantial." Criteria
Background Document at 54. For this reason, the Agency has concluded that "a scenario of
improper management has to be developed for each waste, based upon the types of management
the waste could normally undergo. ... In all cases, the scenario should be one that is reasonably
possible or plausible, not one that very rarely would occur or is otherwise] unrealistic." Id. at 54-
55 (emphasis added). [FN 25: EPA similarly stated that "the Agency would not examine possible
hazards arising from improper waste incineration if the waste in question is not likely to be
incinerated." 45 Fed. Reg, at 33,113 (emphasis added).] Clearly, a few isolated examples of
disposal of mineral processing wastes in an MSW landfill (if they indeed do exist) would not
justify application of the TC to mineral processing wastes under the Dithiocarbamate case and
-EPA's past practices.

Indeed, applying the TC in such a manner would run afoul of the Dithiocarbamate Task Force
decision in another WAY. The Court in that case stated that:

Where it is reasonable to consider the factors in relation to a class of chemicals, EPA may do so.
... [That means essentially that if the known similarities of members of a class are such that it is
reasonable to infer the presence of a disputed characteristic throughout the class (not just among
the members for which it has been shown), the EPA is free to draw that inference.

98 F.3d at 1399. In the present case, EPA is distorting this standard beyond recognition.
According to EPA, as long as it can demonstrate (or at least allege) that one particular type of
waste has been managed in a certain way on one occasion, the Agency is free to assume, without
more, that all mineral processing wastes may always be managed in that same way. Clearly, this is
61

-------
not what the Court had in mind.

The D.C. Circuit also stated that "[where EPA is confronted with evidence challenging its
classification, it must respond, either by altering its classification or by reasonably defending its
choices." Id. at 1402 In the present case, perhaps the strongest evidence that it is inappropriate
to apply the co-disposal mismanagement scenario to mineral processing wastes is the fact that
EPA, despite considerable effort, has only been able to muster a few wholly deficient allegations
that co-disposal has ever taken place. Assuming, for the purposes of argument only, that EPA
can somehow identify a few true examples of co-disposal, the Agency must limit application of
the TC to those few situations. Alteratively, EPA must show that it is reasonable to infer that
other mineral processing wastes will also be co-disposed. Such a showing, however, will be
extremely difficult, if not impossible, to make.

C. EPA Should Adopt a Modified SPLP, Rather than the TCLP, to Evaluate Mineral
Processing Wastes,

One of the key problems with mechanically applying the TC that was designed for general
industrial wastes to mineral processing wastes is the leaching test that is used under the existing
characteristic, the TCLP. [FN 26: As NMA noted in its comments on the January 25, 1996
Supplemental Phase IV proposal, the regulatory levels that are used under the TC to evaluate the
TCLP extract from a waste and thereby determine whether the waste is hazardous (See 40 C.F.R.
§ 261.24, Table 1) are also problematic. NMA maintains that these levels are unlawful, whether
the TCLP or the (modified) SPLP is used as the leaching procedure.] As discussed in NMA's
comments on the January 25, 1996 Supplemental Phase IV proposal, the TCLP is overly
aggressive with regard to certain metals and waste types, thereby resulting in inflated teachability
results. Much more accurate results (ie.. results that reflect more closely the actual degree of
leaching that is likely to occur in the environment) can be obtained with the SPLP, with certain
specified adjustments to the particle size requirements in that test.

Because the modified SPLP is more accurate than the TCLP with respect to mineral
processing wastes, it must be adopted by EPA for such wastes. Congress in 1984 specifically
directed the Agency to "make changes in the [toxicity characteristic], including changes in the
leaching media, as are necessary to insure that it accurately predicts the leaching potential of
wastes which pose a threat to human health and the environment when mismanaged." 42 U.S.C. §
6921 (g) (emphasis added). In fact, EPA's own regulations require the Agency to adopt the
testing method that is "superior ... in terms of sensitivity, accuracy, and precision (i.e.,
reproducibility)." 40 C.F.R. § 261.21(a). [FN 27: Under the regulations, if the two testing
methods are "equal to" one another in terms of sensitivity, accuracy, and precision, both methods
must be approved. See 40 C.F.R. § 261.21 (a). In such an event, generators can use either test
method to demonstrate whether their wastes are hazardous] Because both the TCLP and SPLP
have been developed by EPA, utilize the same analytical methods, and have been used by the
Agency in various regulatory contexts, they should be comparable in terms of sensitivity and
precision. Because the SPLP is more accurate than the TCLP for mineral processing wastes.
62

-------
however, EPA must adopt the SPLP for such wastes.

At most, the TCLP should be used in those cases (if any) where mineral processing wastes
are actually co-disposed in MSW landfills. As discussed in detail in its comments on the January
25, 1996 Supplemental Phase IV proposal, NMA would support a contingent management
approach in which the leaching test to be applied to a mineral processing waste would depend
upon the type of landfill in which the waste is disposed.

EPA's recent rule on military munitions provides legal support and precedent for just such
an approach. In that rule, EPA conditionally excluded from the definition of hazardous waste
military munitions that are stored in accordance with the standards of the Department of Defense
Explosives Safety Board ("DDESB"). See 40 C.F.R. § 266.205(a). EPA based its approach on its
"belief that RCRA section 3001 (a) provides the Agency with the flexibility to consider good
management practice in determining the need to regulate waste as hazardous." 62 Fed. Reg, at
6636. EPA acknowledged that "in the early 1980's its interpretation of RCRA's definition of
hazardous waste focused on the inherent chemical composition of the waste, and assumed that
mismanagement of such waste would occur and would result in threats to human health or the
environment." Id. at 6637. Based on its many years of experience with the management of
hazardous wastes, however, "EPA believes that it is no longer required ~ nor is it accurate or fair
— to assume that all inherently hazardous wastes will be mismanaged, thus creating the necessity
to regulate them under subtitle C. "Id. Instead, wastes can be conditionally excluded from the
definition of hazardous waste in those instances where they are managed in a protective manner.

NMA wholeheartedly agrees with EPA that RCRA provides the Agency authority to make
the regulatory definition of hazardous waste contingent on the way that wastes are actually
managed. Indeed, NMA believes that the statute may require such a "contingent management"
approach, in order to ensure that only those wastes as generated that meet the statutory definition
of hazardous waste are subjected to the management requirements of Subtitle C. NMA also
agrees with EPA that it is neither accurate nor fair to assume that wastes will be mismanaged if
"they are not, in fact, being mismanaged. As the D.C. Circuit held in Dithiocarbamate Task Force.
such an assumption would be arbitrary and capricious.

These conclusions are no less true in the case of mineral processing wastes than they are in
the case of military munitions. Accordingly, to the extent that EPA decides to apply the TC to
mineral processing wastes at all, it must do so only in those circumstances where the wastes are
actually managed as assumed under that characteristic (i.e, disposed in an MSW landfill). In other
cases, EPA should adopt the SPLP as modified per NMA's suggestion.
Response
For EPA's response to this issue, see RCRA Docket # F-98-2P4F-FFFFF
Response to Comments Document, Volume 14: Comments Related to Second Supplemental
Proposed Rule (May 12. 1997): Treatment Standards for Metal Wastes and Mineral Processing
63

-------
Wastes. Mineral Processing and Bevill Exclusion Issues, and the Use of Hazardous Waste as Fill.
Land Disposal Restrictions—Phase IV: Final Rule Promulgating Treatments Standards for Metal
Wastes and Mineral Processing Wastes; Mineral Processing Secondary Materials and Bevill
Exclusion Issues; Treatment Standards for Hazardous Soils; and Exclusion of Recycled Wood
Preserving Wastewaters. (April 30, 1998).
2. 2P4P-00074 New York State Department of Environmental Conservation

Enclosed are the comments of the New York State Department of Environmental
Conservation (DEC) on the proposed supplemental Land Disposal Restriction (LDR), Phase IV
Rule, published in the May 12, 1997 Federal Register.

DEC supports the delegation of authority to states for certain aspects of the LDR. New
York State, for instance, has been authorized for the LDR program since May 1992. In its
continuing effort to improve and update the LDR and associated hazardous waste programs, EPA
may wish to consider delegating authority to states with approved LDR programs for at least
case—by-case variances, extensions, and exemptions. DEC believes that improvement to the LDR
program has its base in the ability of regulatory agencies to quickly and efficiently administer the
program. This will improve compliance and expedite environmentally safe solutions to hazardous
waste management.

We appreciate the opportunity to participate in this rulemaking process. If there are any
questions, please contact Normart H. Nosenchuck, Director of our Division of Solid &
Hazardous Materials, at (518) 457-6934. The following comments apply to the proposed rule,
published in the May 12, 1997 Federal Register. The proposed rule establishes treatment
standards for metal wastes and mineral processing wastes. It also addresses mineral processing,
Bevill exclusion issues and the use of hazardous waste as fill material.

EPA has in the past received comments on proposed rulemakings which emphasize that
"data used in setting treatment standards is not all inclusive. DEC would caution against an
approach which sets standards, based on limited waste matrices that may not fully represent
significant waste streams, generated in a particular category. Technology-based standards can be
inequitable and overregulate certain wastes and generators. DEC agrees with EPA that revising
standards should be ongoing and reflective of "new" data, but increasing stringency should, at this
point in time, be weighed against criteria which can determine if the intent of RCRA is met or
exceeded. As an example, DEC would disagree with the lowering of the Universal Treatment
Standard (UTS) for silver, most available data would indicate that silver has a very stringent
proposed UTS (0.11 mg/1) in relationship to health-based risks and environmental mobility. EPA
also believes that most silver wastes are recycled and, therefore, are not land disposed, and yet
proposes to lower the UTS level, based on technology alone. Because available technology may
be able to achieve a particular level of treatment, this should not dictate whether or not a standard
is needed to protect human health and the environment. DEC does not believe that statutory
64

-------
mandates are best served by such overregulation. The current UTS for nonwastewater silver
wastes (0.30 mg/1 by TCLP) is sufficiently protective in light of economic incentives for recycling
silver and statistically established low toxicity

2.      Proposed Treatment Standards for Mineral Processing Wastes

       DEC agrees with EPA that mineral processing wastes should be subject to LDR treatment
standards as proposed, previous comment excepted.

Response

       In today's final rule, EPA is promulgating a nonwastewater treatment standard of 0.14
mg/1 TCLP for both the characteristic and listed silver (D011) constituents.  For wastewaters,
EPA is promulgating a treatment standard of 0.43mg/l as proposed in the original Phase IV
proposal on August 22, 1995 (60 FR 43684). The Agency is aware of the issues surrounding the
recycling of silver therefore, EPA is currently is the process of determininig whether silver should
be altered or revisited to determine whether silver should be on the TC list.
                                           65

-------
Phase IV Second Supplemental Proposed Rule:
Comments and Responses on Ignitable, Reactive, and/or Corrosive (ICR) Wastes
1. 2P4P-00081 Eastman Chemical Company

As stated previously, Eastman's Tennessee facility operates two rotary kilns and a liquid
chemical destructor under a RCRA permit issued in 1989. These units manage both hazardous
and nonhazardous incinerable wastes. Because many of the manufacturing processes generate
corrosive D002 wastes which are managed in these incineration units, it is necessary for
Eastman's to demonstrate compliance with underlying hazardous constituent UTS before the ash
from the combustion of these decharacterized corrosive wastes can be land disposed. There are
approximately 80 approved waste streams at this facility, which are acetic liquid waste steams.
Many of these waste streams have as many as 40 analyses in the database for antimony. Using
SW-846 Method 601 0, more than 99% of the data for these streams show values lower than the
detection limit in the data system. For these highly organic waste streams, the detection limits
range from 1.28 mg/1 to 2.0 mg/1 which is below the current UTS of 2.1 mg/1. Again, these
detection limits are based on an extensive analytical testing program for multiple wastes and waste
matrices. In addition, incinerator ash samples yield antimony TCLP values ranging from 0.016
mg/1 to 1.3 mg/1 which is again below the current UTS. However, neither database is adequate to
demonstrate compliance with the proposed UTS of 0.07 mg/1.
Response

In response to comments regarding the treatability difficulties associated with antimony,
the Agency reviewed additional data submitted by the commenters in response to the second
supplemental Phase IV proposed rule. The Agency has determined that the antimony treatment
standard (0.07 mg/1) proposed in the second supplemental Phase IV proposal (62 FR 26041, May
12, 1997) does not represent BOAT with a "most difficult to treat" waste. Therefore, the Agency
has re-calculated the BOAT treatment standard for antimony at 1.15 mg/1 based on newly
obtained treatment performance data from Waste Management, Inc., that represented the "most
difficult to treat" waste. The Agency believes that the new antimony standard can be achieved
through well operated commercially available stabilization and HTMR technologies, and can be
measured reliably be existing analytical procedures. The Agency notes that SW-846 Method 704
achieves detection limits of 0.003 mg/L via GFAA, 0.2 mg/L via FAA, and 0.032 mg/L via ICAP.
The Agency also would like to note that if a particular waste is unique or possesses properties
making it difficult to treat, the affected party may petition the Agency for a treatability variance as
per 40 CFR 268.44 on a case-by-case basis.

2. 2P4P-00088 FMC Corporation

The starting point for this capacity determination is the understanding that the composition
66

-------
of these three additional waste streams — the Non-Hazardous Slurry Assurance Process (NOSAP)
Slurry, Precipitator Slurry, and Phossy Water — is identical to the three waste streams for which
EPA has proposed an NCV in all respects that are relevant to national capacity determinations.
As with the three waste streams for which EPA has proposed a two-year NCV, the newly
identified streams contain varying amounts of both naturally occurring radioactive materials
(NORM) and elemental phosphorous. Like the three waste streams addressed in the original
petition, it is possible that NOSAP Slurry, Precipitator Slurry, and Phossy Water could exhibit the
Toxicity Characteristic (TC) [FN 6: 40 C.F.R. ° 261.24.] in the event of process upsets due to the
presence of heavy metals. In addition, they all contain a variety of other metals, albeit below TC
concentrations. Unlike the three waste streams addressed in the original petition, these waste
streams are defined as non-wastewaters[FN 7: In 40 C.F.R. ° 268.2, EPA defines a non-
wastewater to be wastes that do not meet the criteria for wastewaters. Wastewaters are wastes
that contain less than 1 percent by weight total organic carbon (TOC) and less than 1 percent by
weight total suspended solids (TSS).] under the LDR program.

FMC is applying the DOO1 and D003 waste codes for the ignitable and reactive
characteristics to certain waste streams based on preliminary EPA test results that EPA claims
demonstrate the presence of these characteristics. These waste characteristic determinations are
contrary to those FMC previously made. These determinations and other information FMC has
obtained lead it to believe that EPA's results are neither representative nor valid. Nonetheless,
FMC is prepared to manage these wastes as ignitable and reactive as part of an overall RCRA
compliance program. That FMC will manage these streams as if these characteristics apply does
not constitute any agreement on the part of FMC that the D001 or D003 waste code designations
are accurate or appropriate.

Response

The Agency believes that phosphorous and phosphine generating wastes are hazardous
waste. These wastes are D001 because the phosphorus is capable of causing fire through
spontaneous chemical changes and D003 because they generate phosphine - a toxic gas vapor or
fume in quantities sufficient to present danger to human health or the environment. Therefore, in
managing these wastes, FMC should take adequate precautions to avoid worker injury.

(See Phase IV Final Rule Background Document Land Disposal Restrictions TC Metals and
Mineral Processing Waste)
67

-------
DCN PH4P010
COMMENTER Coalition on West Valiey
RESPONDER AC
SUBJECT RADI
SUBJNUM 010

COMMENT

The Coalition is, and always has been, opposed to burial of radioactive material of any
kind. Experience at West Valley mirrors experience at other sites; radioactive waste has not been
isolated by burial. Although burial technology could be improved and might slow down the
migration, it will not eliminate it.

We oppose the proposed exemption of West Valley Demonstration Project wastes from
the proposed Land Disposal Restrictions for burial of mixed waste. At West Valley 21,000 drums
have already been filled with cemented mixed waste but not yet disposed of. These drums do not
meet the proposed new restrictions. Since retreatment of those wastes to meet the new standard
(opening the drums and grinding the cement) would pose a significant risk, EPA proposes that
they be exempted from the new restrictions. We agree that retreatment is not a reasonable option.
But to us the burial of waste which does not meet the new standards is not reasonable either.
Therefore it should not be exempted from the restrictions. If these restrictions are necessary to
protect the environment from contamination then they should be met unconditionally. I am not
clear as to whether your LDRs consider above-ground tumulus disposal as burial, but we do
consider a tumulus to be burial.

RESPONSE

The Agency notes that radioactive mixed wastes are subject to dual regulation. Thus,
radioactive wastes mixed with metal characteristic waste have to comply with the LDR treatment
standard for the metal characteristic wastes established by EPA, as well as any requirements set
forth by the NRC for the radioactive component of the mixed.

In today's rulemaking, the Agency is promulgating treatment standards for radioactive
wastes mixed with metal characteristic waste. The Agency believes that requiring facilities to re-
treat radioactive mixed wastes could pose significant threates to human health and the
environment (worker exposure, environmental releases), so that further treatment would increase,
not minimize, potential threats. Therefore, the Agency will allow characteristic metal mixed
wastes that have undergone stabilization prior to the effective date of Phase IV final rule, to
comply with the LDR metal standards that were in effect at the time the waste was stabilized.
Mixed radioactive/characteristic metal wastes that are stabilized after the effective date of the
Phase IV final rule will be subject to the metal treatment standards promulgated in today's
rulemaking.
68

-------
       In addition, the Agency recognizes the lack of available treatment capacity for previously
untreated wastes. Therefore, the Agency is granting a two-year national capacity variance for
mineral processing wastes mixed with radioactive wastes.  While EPA appreciates the
commenter's further concerns, this rule does not address the general rules for disposal of
radioactive wastes. EPA believes that the hazardous waste portion of the material has been
adequately treated to allow disposal of hazardous waste, however

DCN      PH4P013
COMMENTER  New York DEC
RESPONDER  AC
SUBJECT   RADI
SUBJNUM   013

COMMENT

       DEC agrees with the proposed handling of previously treated mixed
       wastes, contaminated with characteristic metals. Mixed wastes that
       have met the applicable standard at the time they were treated
       should be considered in compliance even if the standards were to
       change before actual disposal takes place. Disposal capacity
       inadequacy for mixed wastes has been and continues to be a major
       problem for LDR compliance. The retreatment of previously
       stabilized mixed wastes to address new standards for CMW will
       benefit no one while increasing the dangers of handling these
       wastes.

RESPONSE

       The Agency thanks the commenter for their support.
DCN     PH4P031
COMMENTER  Department of Energy
RESPONDER  AC
SUBJECT   RADI
SUBJNUM   031

COMMENT

On August 22, 1995, the Environmental Protection Agency (EPA) published a Notice of
Proposed Rulemaking (NPRM) to amend the regulations, for implementing the Land Disposal
Restrictions (LDR) program (known as the LDR Phase IV proposed rule). As part of this
NPRM, EPA presents and requests comment on options for regulating potential releases of
                                          69

-------
hazardous constituents from surface impoundments treating wastes that were hazardous when
generated, but have been diluted to render them nonhazardous (i.e., decharacterized wastes). In
addition, the NPRM includes proposed treatment standards for toxicity characteristic (TC) metal
wastes. Furthermore, the NPRM suggests a number of regulatory modifications intended to
clarify and "clean up" existing LDR requirements, and proposes procedures for streamlining state
authorization to implement certain LDR regulations.

The Department of Energy (DOE) appreciates the opportunity to raise concerns and provide input
in response to the LDR Phase IV proposed rule. The enclosed comments refer to potential
regulatory approaches and topics covered by the NPRM, and are presented for your consideration
in finalizing changes to the LDR requirements. These comments combine the viewpoints and
concerns identified by DOE Field Organizations and Program Offices.

Many of DOE's comments on the LDR Phase IV proposed rule relate to the options being
considered to control releases of hazardous constituents from surface impoundments that manage
decharacterized wastes.

Regarding the proposed LDR treatment standards for TC metal wastes (D004 - D011), DOE fully
supports EPA's proposal not to require re-treatment prior to final land disposal of mixed
radioactive and characteristic metal wastes that are treated to meet existing LDR standards before
the LDR Phase IV rule becomes effective, but that are not disposed until after the effective date.

The enclosed comments have been divided into two sections: general and specific. The general
comments address broad concerns. The specific comments relate directly to potential regulatory
approaches and issues raised in particular sections of the NPRM. For clarity, each specific
comment is preceded by a reference to the section of the NPRM to which it applies and a brief
description in bold-face type of the issue within that section to which DOE's comment is directed.

RESPONSE

The Agency thanks the commenter for providing comments. The Agency notes that the
commenter's specific comments are addressed separately in this document. (See the following
responses).
DCN PH4P031
COMMENTER Department of Energy
RESPONDER AC
SUBJECT RADI
SUBJNUM 031

COMMENT
70

-------
DOE's primary comments in this regard relate to concerns that dual
regulation may be proliferated, concerns that implementation of the
second option that EPA is considering would be overly complex, and
concerns about possible delays in operating the high-level
radioactive waste vitrification system at the Savannah River Site
if the LDR Phase III and Phase IV rules are applied to some of the
component facilities.

The SRS has designed and constructed an elaborate inter-connected
treatment system for its mixed high-level wastes. The system has
several integral facilities which perform various  separation and
treatment processes, some with an element of land  disposal.  The
majority of this integrated system is operated under Industrial
Wastewater Treatment Facility Permits issued by the South Carolina
Department of Health and Environmental Control (SCDHEC).  One of
the system's treatment facilities, the Effluent Treatment Facility
(ETF), has a National Pollutant Discharge Elimination System
(NPDES) Permit, while another treatment facility, the
Saltstone Processing Facility,  is permitted under the CWA.  The
Saltstone Disposal Facility is permitted as an Industrial Waste
Disposal Facility (Subtitle D).  A third very important part  of
the treatment system, the Defense Waste Processing Facility (DWPF),
will vitrify high-level waste under a CWA permit.  Attachment A
describes in detail the  relationships of and permits held by
the various integral facilities of the SRS mixed high-level waste
treatment system.
 DOE believes that, as proposed, the LDR Phase III and Phase IV
rules could have a detrimental impact on the ability of some of the
integral facilities of the SRS mixed high-level waste treatment
system to support operation of the system as currently planned.
This comment centers primarily on the Saltstone Processing and
Disposal Facilities, although the DWPF is also of concern.
This discussion does not address the DWPF at length because it is
felt that EPA does not intend  the LDR Phase III and Phase IV rules
to impose treatment restrictions on high-level waste beyond
the existing requirement to apply the specified technology of
vitrification.

 DOE Recommendations
DOE's primary concern in the  LDR Phase III and Phase IV
rulemakings with regard to the SRS is the status of the Saltstone
Processing and Disposal Facilities which are currently
permitted, respectively, under the CWA and as a Subtitle D
                                     71

-------
Industrial Waste Disposal Facility. The Saltstone Processing
Facility treats a mostly inorganic, characteristically hazardous,
radioactive wastewater (i.e., mixed waste) to form a non-hazardous,
pumpable, low-level radioactive waste known as saltstone. The
saltstone is pumped from the Processing Facility into the
Disposal Facility, where it is placed into covered, above-ground
concrete vaults.  The pozzolanic saltstone solidifies within these
vaults into a monolithic, non-hazardous waste.  However, there is
a potential for this stabilized waste to contain total
concentrations of at least two organics at levels slightly above
the UTS. Neither the Saltstone Processing Facility nor the
Saltstone Disposal Facility contains surface impoundments.
According to the proposed LDR Phase III rule, end-of-pipe
equivalency requires that discharges from CWA and CWA-equivalent
wastewater treatment facilities (including zero  dischargers)
meet the limitations and standards imposed on  UHCs by an applicable
CWA permit or authorization.  If no such limitations or standards
have been set, the UTS apply.  Therefore,  if the SRS
Saltstone Processing  Facility were to be governed by the LDR Phase
III Rule (because this facility was construed to be an affected
zero-discharge wastewater treatment facility
administering CWA-equivalent treatment), then DOE believes the
pumpable saltstone and/or the solidified saltstone could be
required to meet the UTS since placement in the SRS Saltstone
Disposal Facility represents ultimate land disposal. However,
application of these new requirements to The Saltstone Processing
and Disposal Facilities seems to go beyond the type of waste
treatment systems EPA intended to cover (i.e., wastewater treatment
facilities with surface impoundments).	
 The proposed regulatory language [60 FR 11742] that appears to
 encompass the Saltstone Processing and Disposal Facilities within
 the scope of the LDR Phase III rule, and subsequently within the
 scope of the LDR Phase IV rule, even though neither facility
 contains surface impoundments, reads as follows:
 §268.39 Waste specific prohibitions — spent aluminum potliners,
 carbamates  and organobromine wastes.
 *   *    *
 (b) On [Insert date two years from date of publication of the
 final rule],characteristic wastes that are managed in systems
 whose discharge is regulated under the Clean Water Act (CWA), or
 that are zero dischargers that engage in CWA-equivalent treatment
 before ultimate land disposal, are prohibited from land disposal.
                                      72

-------
       DOE is of the opinion that such language could require the SRS
       (under the LDR Phase III rule) to demonstrate that all underlying
       constituents in the pumpable saltstone and/or solidified saltstone
       meet the UTS through some treatment mechanism other than dilution,
       prior to disposal in the permitted Subtitle D Saltstone Disposal
       Facility. Further, DOE believes that such language could also
       require the SRS, under the LDR Phase IV rule, to comply with
       additional controls on air emissions, sludges and leaks from the
       SRS Saltstone Processing and Disposal Facilities.- Response? The wastes
managed in the SRS Saltstone Facilities are byproducts of
       pretreating high-level radioactive waste before vitrification,
       which is the LDR-specified treatment technology for high-level
       waste.  If process changes were required in the SRS Saltstone
       Facilities to comply with the LDR Phase III and Phase IV proposed
       rules, delays in the high-level waste treatment program would
       undoubtedly result.  Hence, DOE requests that EPA consider
       clarifying the proposed regulatory language quoted above in order
       to make clear that it did not intend to include facilities
       such as the SRS Saltstone Processing and Disposal Facilities
       within  the scope of the LDR Phase III and Phase IV final rules.

       b.  As the discussion above indicates, DOE is  extremely concerned
       that the LDR treatment requirements established under the LDR Phase
       III and Phase IV rules might be applied to outputs from certain
       integral facilities of the SRS treatment system for mixed
       high-level wastes (particularly, the Saltstone Processing and
       Disposal Facilities). Certain of the proposed requirements have
       the potential to detrimentally impact the ability of these
       facilities to support operation of the mixed high-level waste
       treatment system as currently planned. As also indicated in the
       preceding discussion, the Department believes that EPA does not
       intend  the LDR Phase III and Phase IV rules regarding equivalency
       to apply to Clean Water Act permitted facilities that have no
       surface impoundments (such as the SRS Saltstone Processing
       Facility). This aside, DOE remains somewhat  unclear as to the
       exact applicability of the LDR program to
       characteristically hazardous wastes being managed by the SRS
       Saltstone Processing and Disposal Facility. Therefore, the
       Department requests that EPA provide clarification and describe
       its regulatory interpretation of the specific applicability of the
       LDR requirements to such facilities (i.e.,non-surface impoundment
                                            73

-------
wastewater treatment facilities operating under State-issued CWA
and Industrial Waste Disposal Facility (Subtitle D) permits).
Specifically, the SRS  Saltstone Processing Facility conducts
CWA-equivalent treatment of characteristically hazardous,
radioactive wastewaters (i.e., the waste stream contains
corrosive, EP  and TC metals, and TC organic wastes) in a
RCRA-exempt CWA-permitted tank-based centralized wastewater
treatment system. This treatment system forms a
non-hazardous pumpable low-level radioactive waste stream which is
pumped into the Saltstone Disposal Facility where it is placed into
covered, above-ground concrete vaults.  The waste solidifies
within these vaults into a monolithic, non-hazardous waste.  In
other words, the CWA-equivalent  treatment administered by this
treatment system produces no liquid effluents.  This
particular component of the high-level waste treatment system,  the
Department requests that EPA clarify how the LDR program will apply
to the treated wastes produced by  these facilities (i.e.,
pumpable and/or solidified Saltstone) after the LDR Phase III and
Phase IV rules are finalized?
As previously stated,  if process changes were required in the SRS
Saltstone Facilities to comply with certain requirements of the LDR
Phase III and Phase IV proposed rules, delays in the high-level
waste treatment program would undoubtedly result. When the  SRS
vitrification system for managing high-level radioactive wastes was
designed, those portions of the system for managing the low-level
waste fraction (i.e., the Saltstone Processing and Disposal
Facilities) were planned and permitted in accordance with existing
standards. Now, as years of design and construction are nearly
complete, it appears that new LDR treatment  standards could be
imposed on the low-level waste component of the high-level waste
vitrification system, jeopardizing the schedule and viability of
the entire project. Furthermore, in light of the design of the
Saltstone Disposal Facility (i.e., above-ground concrete vaults
with ground water monitoring), compliance with the LDR treatment
requirements applicable to UHCs is unlikely to significantly
reduce risks to human health and the environment associated with
the solidified saltstone.
DOE urges EPA to consider implementing the following regulatory
approach to ensure that the SRS high-level waste treatment system
is not unintentionally  or unduly impacted by these new LDR rules.
The suggested approach would be to extend the LDR treatment
standard for high-level mixed wastes to recognize" the
inter-dependence of the specified technology for
                                    74

-------
treating high-level waste (i.e., vitrification) and the technology
for managing the low-level waste fraction of such high-level waste
(i.e., immobilization/stabilization). In other words, DOE
suggests that EPA adopt a specified technology treatment standard
of immobilization/stabilizatio for the low-level waste fraction of
high-level mixed wastes being treated by vitrification at SRS.
DOE also requests that EPA clarify that treatment of UHCs is not
required for either treated high-level mixed wastes, or the treated
low-level waste fraction, from the SRS vitrification system.
To explain the Department's concerns related to this issue in
further detail and to support the proposed approach discussed
above, DOE would welcome the opportunity to provide
additional technical information about the SRS vitrification
system, and to meet with EPA staff regarding such information (at
the Agency's request and convenience).

RESPONSE

The Agency is not imposing additional treatment requirements on wastes for which BOAT
is established as a specified method of treatment (i.e., HLVIT). That is, these wastes do not have
to be further treated for UHCs.

The commenter also raised a question regarding a residual from the treatment of a
characteristically hazardous wastewater (hazardous due to high pH, metals and benzene). The
residual is then stabilized and does not exhibit a characteristic. The stabilized waste is considered
to be a new point of generation. Assuming the residual does not exhibit a characteristic, it is not
considered a hazardous waste, and any UHCs present would not be required to be treated.
DCN PH4P031
COMMENTER Department of Energy
RESPONDER AC
SUBJECT RADI
SUBJNUM 031

COMMENT

V.D.3. Treatment Standard for Previously Stabilized Mixed
Radioactive and Characteristic Metal Wastes
1. p. 43683, col. 3 — EPA proposes to allow characteristic
metal mixed wastes (i.e.,radioactive wastes which" exhibit the
toxicity characteristic for metals) that have undergone
stabilization prior to the effective date of the Phase IV final
75

-------
rule, to comply with the LDR treatment standards for metals that
were in effect at the time the waste was stabilized.  Mixed
radioactive/characteristic metal wastes that are stabilized after
the effective date of the LDR Phase IV final rule would be subject
to the metal treatment standards in the Phase IV rule.
a.   DOE fully supports EPA's proposal not to require re-treatment
of mixed radioactive and characteristic metal wastes that have been
treated (i.e., undergone stabilization) to meet currently
applicable treatment standards, but that may not be land disposed
until after the new treatment requirements of the LDR Phase IV rule
become effective.  Requiring re-treatment of previously stabilized
wastes that have been in storage simply because of a change in
standards that occurs before disposal can be accomplished makes no
sense unless the benefits associated with re-treatment outweigh the
risks to workers and the costs.  DOE agrees with EPA's conclusion
that in the case of previously stabilized mixed radioactive and
characteristic metal wastes, the opposite may be true because the
hazards from added worker radiation exposure associated with
re-treatment would probably offset any gain in protection of human
health and the environment resulting from compliance with the new
metal treatment standards in the Phase IV rule.
b.   As recognized in the preamble [60 FR 43683, col. 3], DOE has
provided information to the Agency concerning treatment and storage
of the low-level waste fraction of high-level reprocessing wastes
at the West Valley Demonstration Project (WVDP). The
Department brought this specific mixed waste example to the
Agency's attention because it involves mixed radioactive and
characteristic metal wastes that have been treated and stored
to await the development of disposal capacity.  DOE was concerned
that if, as is proposed, the LDR Phase IV rule were to adopt more
stringent treatment standards for Characteristic Metal wastes than
the standards applicable when the WVDP wastes were treated,
re-treatment might be necessary.  This could occur unless disposal
were accomplished before the LDR Phase IV rule became effective.
DOE's concerns stemmed from the hazards (from added worker
radiation exposure) and expense associated with re-treatment of
these already treated wastes. With these concerns in mind, DOE
requested that language be incorporated into the LDR Phase IV rule
that would prevent any mixed radioactive and characteristic metal
wastes that have been treated to previously applicable
treatment standards, and that are being stored until disposal
capacity is available, from having to be re-treated:
DOE believes EPA's proposed approach will address the Department's
                                    76

-------
concern regarding the WVDP situation.  As such, DOE supports the
proposed approach, as indicated in the preceding comment.  The
Department also believes, however, that the scope of
this regulatory proposal is too limiting. Therefore, the
Department recommends that the proposed provision be broadened to
cover mixed radioactive and characteristic metal wastes that have
been treated by treatment methods other than stabilization
(e.g.,macroencapsulation of non-debris wastes). Stabilization is
only one of many types of treatment that may be performed on
characteristic metal wastes to comply with LDR treatment standards.
c.  Because DOE supports EPA's proposal in this section of the
preamble, the Department is concerned that EPA has not proposed any
corresponding regulatory language to implement the section. DOE
suggests that EPA add appropriate regulatory language to the
proposed rule as follows:
(1) A section should be added to proposed 40 CFR 268.30  which
would read:
§268.30 Waste specific prohibitions — wood preserving wastes,
and characteristic wastes that fail the toxicity characteristic.
*****
(d) The requirements of paragraphs (a), and (b) of this section
do not apply if:

(5) The wastes are radioactive wastes mixed with or containing
D004-D011 wastes, which have been treated to meet Subpart D
treatment standards in effect prior to [insert effective date of
Phase IV regulations (including any applicable national capacity
variance) for radioactive wasted mixed withD004-D011]. Such
wastes must have been treated prior to [insert effective date of
Phase IV regulations (including any applicable national
capacity variance) for radioactive wastes mixed with D004-D011] to
be excluded from application of paragraph (b).
d.  Given the lack of sufficient disposal capacity for mixed
wastes, DOE urges EPA to consider including exceptions similar to
the one proposed in the LDR Phase IV proposed rule (i.e., in
section V.D.3 of the preamble) in future LDR treatment standard
proposals.  An important lesson DOE learned from the WVDP situation
is that more treated DOE mixed wastes (both characteristic  and
listed) could fall into similar situations if EPA continues to
promulgate increasingly more stringent LDR treatment standards in
the future without grandfathering treated wastes that are being
stored while they await disposal capacity. DOE recently signed
compliance orders that set up schedules to treat mixed wastes at
                                     77

-------
       over twenty-five sites.  As a result, DOE will soon be treating
       thousands of cubic meters of mixed waste to the existing LDR
       treatment standards and storing the treated wastes (potentially
       over the next several decades) until mixed waste disposal caACity
       becomes available.  Therefore, if EPA changes applicable LDR
       treatment standards to more stringent levels while such treated
       wastes are in storage, then without an exception, DOE and the
       commercial sector (which also has insufficient disposal capacity),
       could potentially be required to re-treat the stored wastes. This
       re-treatment could not only be expensive, but also cause
       unnecessary worker radiation exposure.

RESPONSE

       The Agency agrees with the commenter that requiring  facilities to re-treat radioactive
mixed wastes could pose significant threats to human health and the environment (worker
exposure, environmental releases).  Therefore, the Agency will allow characteristic metal mixed
wastes that have undergone stabilization prior to the effective  date of Phase IV final rule, to
comply with the LDR metal standards that were in effect at the time the waste was stabilized.
Mixed radioactive / characteristic metal wastes that are stabilized after the effective date of the
Phase IV final rule will be subject to the metal treatment standards promulgated in today's
rulemaking. The Agency thanks the commenter for their support on this issue.

       As for those wastes that have  been treated by treatment methods other than
stabilization (e.g., macroencapsulation of non-debris waste); Because the exposure concerns of
re-treating the previously stabilized waste primarily center around the idea of first grinding up the
stabilized material to retreat it and the potential added radiological exposures, the broadening  of
this exemption without more specific  information is not warranted at this point. Of course, if any
wastes already meet the applicable treatment standards, for example, macroencapsulation, then
there is no need to initiate further treatment.
DCN     PH4P031
COMMENTER  Department of Energy
RESPONDER  AC
SUBJECT   RADI
SUBJNUM   031

COMMENT

COMMENTS ON PROPOSED REGULATORY LANGUAGE NOT DISCUSSED IN
PREAMBLE
                                           78

-------
1.  p. 43694, col.3, 40 CFR 268.40(e) -- EPA proposes the
following language for 40CFR 268.40(e):
(e)  For characteristic wastes subject to treatment standards
in the following table "Treatment Standards for Hazardous Wastes,"
all underlying hazardous constituents (as defined in§268.2(I))
must meet Universal Treatment Standards, found in§268.48, Table
UTS, prior to land disposal.
The regulatory language in existing §268.42(d) states that "where
treatment standards are specified for radioactive mixed wastes in
the Table of Treatment  Standards, those treatment standards will
govern."  In the existing table "Treatment Standards for Hazardous
Wastes" [40 CFR 268.40], specified technologies are identified that
apply to certain radioactive mixed wastes containing D004-DO11
(e.g., radioactive high level wastes generated during the
reprocessing of fuel rods). Under the LDR Phase IV proposed rule
(i.e., the changes proposed to §268.40),however, the specified
technologies for mixed radioactive wastes containing D004-DO11 are
not listed.
The wording in proposed §268.40(e) states (as noted above) that
"for characteristic wastes subject to treatment  standards in the
following table "Treatment Standards for Hazardous Waste,"all
underlying  hazardous constituents (as defined in  §268.2(1)) must
meet Universal Treatment standards, found in  §268.48, Table UTS,
prior to land disposal."  This specific language does not exclude
characteristic mixed radioactive waste for which EPA has
established a technology-based standard. As such,  radioactive
mixed wastes containing characteristic metal wastes
[e.g., high-level radioactive wastes generated during the
reprocessing of fuel rods and carrying waste codes D002 or D004
through D011, for which the "treatment standard" is vitrification
(HLVIT)] would also be required to meet UTS for underlying
hazardous constituents.  DOE does not  believe EPA intended such a
requirement. With respect to high level mixed wastes, for
instance, EPA stated in the Third Third  final rule preamble that
the potential hazards associated with exposure to radioactivity
during the analysis of high-level waste precluded the
establishment of concentration-based treatment standards to be
applied to the final treated glass form resulting from
vitrification [55 FR 22520, 22627 (06/01/90)].
Therefore,  to avoid confusion, DOE suggests that the wording of 40
CFR 268.40(e) in the final LDR Phase IV rule be. modified to read as
follows:
For characteristic wastes subject to treatment standards in
                                     79

-------
       the following table "Treatment Standards for Hazardous Wastes,"
       all underlying hazardous constituents (as defined in §268.2(1))
       (except for those waste codes for which the treatment standard is
       a specified technology) must meet Universal Treatment
       Standards, found in §268.48, Table UTS, prior to land disposal.

RESPONSE

       The Agency notes that today's rulemaking revises the treatment standards for metal
characteristic wastes, including revision of certain metal numeric treatment standards and
application of UTS levels to UHCs in the characteristic waste.  (For additional information on
applying UTS levels to UHCs in the characteristic waste, see the background material on UHCs in
the RCRA docket for today's rulemaking.) Therefore, the treatment standards for radioactive
wastes mixed with metal characteristic waste that were not specifically subcategorized in the
Third Third rule are also affected. In other words, unless specifically noted in Sections 268.41,
268 42, or 268.43, the standards promulgated today apply to all mixed wastes.   The underlying
hazardous constituents requirements only applies to a specified method when in 40 CFR 268.40
the requirements includes "and must meet 268.48 standards."
       Thus, the commenter is correct that if the treatment standard is a specified method of
treatment, this is the only treatment  requirement applicable to the wastes.

DCN     PH4P031
COMMENTER  Department of Energy
RESPONDER   AC
SUBJECT   RADI
SUBJNUM   031

COMMENT

       2. p. 43695, 40 CFR 268.40, table of Treatment Standards for
       Hazardous Wastes —EPA proposes to modify the LDR treatment
       standard for waste code D003,Explosives Subcategory based on
       §261.23(a)(6),(7)and(8).
       EPA has proposed that the columns of the Treatment Standards table
       labeled  "Wastewaters" and"Nonwastewaters" be modified for waste
       code D003, Explosives Subcategory to read: "DEACT and meet §268.48
       standards."  DOE requests clarification as to why the D003,
       Reactive Sulfides Subcategory and the D003, Other Reactive
       Subcategory were not also modified to require meeting §268.48
       standards. Will "DEACT" be the only required LDR treatment for
       such wastes?

RESPONSE
                                          80

-------
The Agency notes that, as stated in 40 CFR 268.40 (e) Applicability of Treatment
Standards, characteristics wastes (D001 -D043) that are subject to treatment standards in the
table "Treatment Standards for Hazardous Wastes," all underlying hazardous constituents (as
defined in 268.2(i)) must meet Universal Treatment Standards found in 268.48, "Table UTS,'
prior to land disposal as defined in 268.2(c). Thus, all reactive wastes must meet this
requirement. For additional information on applying UTS levels to UHCs in the characteristic
waste, see the background material on UHCs in the RCRA docket for today's rulemaking.
DCN PH4P049
COMMENTER Molten Metal Technology
RESPONDER AC
SUBJECT RADI
SUBJNUM 049

COMMENT

Molten Metal Technology, Inc. (MMT), a company that provides an innovative source
reduction and recycling technology known as Catalytic Extraction Processing (CEP), is pleased to
submit an original and two copies of these comments on EPA's Proposed Rule on Land Disposal
Restrictions - Phase IV. See 60 Fed. Reg. 43654 (Aug. 22, 1995). These comments address two
issues: First, EPA's approach to cross-media transfer of hazardous constituents from surface
impoundments, and second, EPA's proposal to grant a capacity variance for certain mixed wastes
subject to the Phase IV Rule.

Mixed waste, like most hazardous waste, is subject to the LDR
regulations. The LDR regulations require that hazardous waste be
treated to specified standards prior to land disposal. If there
is insufficient treatment capacity to treat a particular waste, EPA
may grant a "capacity variance'Tor up to three years. With few
exceptions, EPA has not established specific treatment
standards for mixed wastes, but rather has specified that the LDR
treatment standard for a given hazardous waste applies to the
hazardous component of a mixed waste.
Over the years, EPA has regularly issued capacity variances for
mixed wastes. In 1990, EPA granted a two-year national capacity
variance for radioactive waste mixed with most hazardous waste.
This variance was subsequently extended for one year but has now
expired. Another capacity variance was granted in August 1994, as
part of the LDR "Phase II" rulemaking. At that time, EPA noted
that "DOE is in the process of increasing its capacity to manage
mixed RCRA / radioactive wastes" but that a "significant capacity
shortfall currently exists." Continuing the pattern, EPA proposed
81

-------
another capacity variance in the Phase III ruiemaking (see 60
Fed.Reg. 11702, 11734 (March 2, 1995)), and is now proposing to
grant a two-year national capacity variance to radioactive waste
mixed with hazardous wastes affected by the Phase IV
ruiemaking.  See 60 Fed.  Reg. at 43686.
The traditional technologies that can comply with the LDR
standards (e.g., incineration and combustion) are considered to be
mature, demonstrated technologies that have been "available'Tor
treatment of hazardous wastes for many years.  Nevertheless,  as a
practical matter, these treatment technologies have not been
successfully applied on a large scale to mixed waste.  As a result
of the lack of suitable treatment capacity, most mixed  waste is
currently in storage facilities awaiting treatment, which is not
permissible under RCRA. This mixed waste storage dilemma was one
of the significant drivers leading to enactment of the Federal
Facility Compliance Act. which  obligates the Department of Energy
to develop mixed waste treatment plans and develop mixed waste
treatment capacity.
MMT does not understand why EPA apparently continues to  believe
that traditional treatment technologies such as incineration and
combustion are appropriate or realistically will ever be available
for  mixed waste. There are significant technical, political, and
regulatory difficulties associated with these technologies. From a
technical standpoint, they do not address or minimize radiological
hazards. In fact, they typically disperse radioactive
constituents into the air, and create a radioactive ash that
requires further treatment prior to disposal as a mixed waste.
Also, combustion technologies are meeting with increasing  public
opposition and EPA, state and local regulatory agencies are
implementing programs and policies  designed to encourage the use
of alternatives to combustion for treatment of hazardous waste
(see, e.g., EPA's "Hazardous Waste Combustion Strategy"). When
attempts are made to apply combustion technologies to mixed waste,
these political and regulatory difficulties are likely to
intensify, and MMT does not believe
that these technologies can be relied  upon to provide any
significant treatment capacity for mixed waste.
There is no reasonable basis for assuming that traditional
combustion technologies will in  the foreseeable future provide
adequate or appropriate treatment capacity for mixed waste. Thus,
the  continuing practice of granting capacity variances amounts to
little more than a holding action which will not allow time for
mixed waste combustion capacity to develop, but rather just
                                            82

-------
delays the time when it will become obvious that these wastes must
be treated using new, innovative technologies that specifically
address and minimize all the hazards presented by mixed
waste,  including radioactive hazards. MMT therefore urges EPA to
reconsider its practice of assuming that LDR requirements for mixed
waste can be achieved by using traditional hazardous
waste treatment technologies to treat the hazardous portion of
mixed waste,  and instead identify and designate as BOAT those
technologies which address and effectively minimize all the
hazards presented by mixed waste, including radioactive hazards.
MMT recognizes that the needs of DOE and others who currently
store mixed waste because of the current shortage of available
capacity must be responsibly addressed in accordance with current
laws and regulations. Thus, at  this time MMT does not question
the need for, or oppose the granting of, mixed waste capacity
variances. However, MMT strongly urges EPA to condition such
variances upon a commitment to deploy and use technology which can
address both the RCRA and radiologic hazards posed by mixed waste
and realistically can be sited and permitted.
In this  regard, M4 Environmental L.P., a partnership of MMT and
Lockheed Martin Corporation,  has presented information and data to
the Department of Energy (DOE) which demonstrate that CEP is
available and  can provide a unique innovative solution to many of
the most difficult mixed waste management issues that the DOE
faces.  M4 is currently constructing a CEP facility in Oak Ridge,
Tennessee that will have the capability to process significant
quantities of mixed wastes for which there is currently no available treatment capacity, while
meeting and surpassing any applicable LDR standards.  This facility will become operational
within the next six months.  Thus MMT
believes that substantial effective mixed waste treatment capacity
will in  fact became available in  the very near future, obviating
the need for future mixed waste capacity variances or extensions.

RESPONSE

              There presently is no available treatment capacity for most radioactive mixed
waste, and EPA is consequently extending the enforcement policy on storage of these wastes six
(6) more months.  This policy indicates that if treatment capacity becomes available it must be
used immediately.  Thus, if the commenter's system comes on line and can treat effectively, it
would have to be  utilized.

DCN     PH4P056
COMMENTER  Westinghouse
                                            83

-------
RESPONDER  AC
SUBJECT    RADI
SUBJNUM   056

COMMENT

6.   Westinghouse also recommends EPA clarify the requirements
for obtaining treatability variances for stored mixed wastes.  Our
primary concern is that treatability variance applicants should not
be required to resample and analyze the waste only because
the treatment standard has been revised.  Requesting additional,
unnecessary or marginally useful information could result in
workers being unnecessarily exposed to radiation.

RESPONSE

       Treatability variances  as stated in 40 CFR 268.44 are not specific to radioactive mixed
wastes.  The treatability variance may be granted for wastes that have LDR standards that are
expressed as concentrations of hazardous constituents in the waste or waste extract (i.e., a
numerical standard). Under the language of the treatabilility variance, petitions must demonstrate
that the waste of concern cannot be treated to the specified levels because its physical or chemical
properties differ significantly from the waste used to establish the LDR treatment standard.  Given
the hazards of handling mixed waste,  however, it may be that no further  sampling and analysis
might be needed.  ( Hazardous waste destined for land disposal but stored prior to meeting
existing treatment requirements are still subject to meeting LDRs' UTS requirements before
disposal of these wastes can occur.)  In addition the Agency will allow  characteristically
radioactive mixed metal waste that have been stabilized prior to the effective date of Phase IV
final  rule, to comply with the  LDR standards that were in effect at the time the waste was treated.
DCN     PH4P056
COMMENTER  Westinghouse
RESPONDER  AC
SUBJECT   RADI
SUBJNUM   056

COMMENT

Issue 4: Treatment Standard for Previously Stabilized Mixed
Radioactive and Characteristic Metal Wastes Reference. Preamble at
Section V.D.3, page 43683
Westinghouse supports the EPA proposal to exempt characteristic
metal mixed waste from the Phase IV LDR treatment standards if the
                                          84

-------
mixed waste was stabilized prior to the effective date of the Phase
IV LDR final rulemaking and if the stabilized waste complies with
the LDR TREATMENT standards that were effect when the waste was
stabilized.  However, this proposal should be expanded to include all
previously stabilized characteristic and listed mixed wastes that
met the standards in effect at the time of treatment but, due to
regulatory changes, may no longer meet the standards.  The same
rationale used to justify the West Valley example provided in the
preamble would apply to other characteristic and listed mixed
wastes that have been previously stabilized; opening drums,
grinding already treated masses of mixed waste to treat them again.
could expose workers to unnecessary and unacceptable levels of dust
containing metals, organics and radioactivity without a
commensurate environmental benefit.
An example would be ash that has been stabilized from an
incinerator that burns listed/characteristic mixed waste. At the
Savannah River Site (SRS), which is managed by Westinghouse on
behalf of the Department of Energy (DOE), the Consolidated
Incineration Facility (GIF) generates ash and blowdown residues.
The stabilization unit (called the ashcrete unit) treats the ash
and blowdown to meet the LDR treatment standards, producing an
average of about 100 drums of stabilized waste monthly.  This mixed
waste is being stored in RCRA storage facilities until RCRA
Subtitle C disposal vaults can be built. It is possible that
waste generated prior to Phase IV promulgation may not meet the
Phase IV requirements; (or future unknown requirements) thus
possibly prohibiting  disposal of the stabilized waste without
further treatment.
Additionally, with the evolution of the land disposal restrictions
program, it is possible to have standards change numerous times
prior to final disposal. Treatment and disposal capacity for mixed
waste is already lacking, and constantly changing standards will
only makes matters worse.  We support expansion of the concept that
wastes treated to  meet current standards will not have to be
further treated if standards change prior to disposal.
RESPONSE

        The Agency agrees with the commenter that requiring facilities to re-treat radioactive
mixed wastes could pose significant threats to human health and the environment (worker
exposure, environmental releases).  Therefore, the Agency will allow characteristic and listed
radioactive mixed metal mixed wastes that have undergone stabilization prior to the effective date
                                            85

-------
of Phase IV final rule, to comply with the LDR metal standards that were in effect at the time the
waste was stabilized.  Mixed radioactive/characteristic and listed metal wastes that are stabilized
after the effective date of the Phase IV final rule will be subject to the metal treatment standards
promulgated in today's rulemaking.
DCN     PH4P056
COMMENTER  Westinghouse
RESPONDER  AC
SUBJECT    RADI
SUBJNUM   056

COMMENT

Issue 9: Regulatory Language Found in Section 268.40(e) Reference:
Regulatory text at page 43694
In the Phase II and Phase III rules, high level radioactive wastes
generated during the reprocessing of fuel rods and having the waste
codes D002, D004-D011, have specified technologies listed in
§268.40. As stated in §268.42(d), where the treatment standards
are specified for mixed radioactive wastes  in the Table of
Treatment Standards, those treatment standards will govern.  In
Phase IV, the specified technologies for mixed radioactive
wastes containing D004-D011 are no longer listed. The wording in
§268.40(e) states that "for characteristic wastes subject to
treatment standards in the following table,  Treatment
Standards for Hazardous Wastes,' all UHC (as defined in §268.2(1))
must meet Universal Treatment standards found in §268.48, Table
UTS, prior to land disposal." This specific language does
not exclude characteristic mixed radioactive waste for which EPA
has assigned a technology-based standard,  such as the vitrification
of high level fuel rod reprocessing waste. In the Third Third rule
preamble, EPA stated that the potential hazards associated with
exposure to radioactivity during the analysis of high level waste
precluded the establishment of a concentration-based treatment
standard for the final treated glass form (55 FR 22627).  It
appears that this  1990decision supports the conclusion that EPA
does not intend to require the management of UHC in such high-level
waste since compliance with associated concentration-based
standards would  require the analysis of the high-level glass form.
To clarify the final rule, it is suggested that the wording to
§268.40(e) be modified to read: For characteristic wastes subject
to treatment standards in the following table "Treatment Standards
                                           86

-------
for Hazardous Wastes," all UHC (as defined in §268.2(1), except for
those waste codes that have a specified technology) must
meet Universal Treatment Standards (UTS) found in §268.48, Table
UTS, prior to land disposal.

SPECIFIC COMMENTS ON THE IMPACT OF THE PROPOSED LDR - PHASE   IV
RULEMAKING ON THE SAVANNAH RIVER SITE HIGH LEVEL WASTE SYSTEMS

As mentioned in our first general comment, because of the
inconsistencies between the preamble language and the regulatory
language,  it is unclear whether the Phase III and IV LDR proposals
apply only to CWA and CWA-equivalent systems that include surface
impoundments or to all types of CWA and CWA equivalent systems.
This was a concern that was expressed in our comments on the Phase
III proposal and it remains with the Phase IV.
The SRS has designed and constructed an inter-connected treatment
system for mixed high-level wastes.  The system has several
integral facilities which perform various separation  and treatment
processes, some with an element of land disposal. The majority of
this integrated system is operated under Industrial Wastewater
Treatment Facility Permits issued by the South Carolina Department
of Health and Environmental Control (SCDHEC).  One of the
system's treatment facilities, the Effluent Treatment Facility
(ETF), has a National Pollutant Discharge Elimination System
(NPDES)  Permit. Another treatment facility, the Saltstone
Processing Facility, is also permitted under the CWA.  The
Saltstone Disposal Facility is permitted as an Industrial Waste
Disposal Facility (Subtitle D).  A third, very important part of
.the treatment system, the Defense Waste Processing Facility (DWPF),
will vitrify the high level waste under a CWA permit, (see the
attached Addendum for a detailed description of these facilities).
These new regulations could potentially impact the ability of these
facilities to continue to operate  indirect support of the SRS's
high level  waste system as currently planned.  The discussions in
this comment package center primarily on impacts to the Saltstone
Processing and Disposal facilities under the proposed regulation.
Although  the DWPF is also of concern, it is felt that EPA does not
intend to impose additional treatment restrictions on high level
waste other than the requirement to meet the specified technology
of vitrification. However, due to certain ambiguities in the
proposed  rule, additional clarification is needed on  this point.
The primary concern at the SRS with these two rulemakings is the
status of the Saltstone Disposal Facility currently permitted  as a
                                           87

-------
Subtitle D Industrial Waste Disposal Facility.  This facility
treats an inorganic mixed wastewater, which is characteristically
hazardous. After pozzolonic stabilization, the waste is rendered
non-hazardous and is disposed in above-ground concrete vaults.
There is a potential for this stabilized waste to contain at least
two organics slightly above the UTS  Regulatory language in Phase
III and the likelihood of carry-over to the Phase IV rule at
§268.39 (b) are of principal concern.  It states:
On [Insert date two years from date of publication of the final
rule],characteristic wastes that are managed in systems whose
discharge is regulated under the Clean Water Act (CWA), or that are
zero dischargers that engage in CWA-equivalent treatment before
ultimate land disposal, are prohibited from land disposal.
Westinghouse is of the opinion that such language could require
the site to demonstrate that all underlying constituents of this
waste stream meet the UTS through some treatment mechanism other
than dilution, prior to disposal in the permitted Subtitle D
disposal facility.  We request clarification of the applicability
of these two proposals to this particular treatment system. If
EPA truly intends these regulations to include treatment in CWA
systems that involve an element of land disposal other than a
surface impoundment, then EPA must carefully consider the
impacts the proposed rulemakings on the SRS's high level mixed
waste treatment system and facilities, taking into account the
unique characteristics and risks associated with liquid mixed
wastes

Recommendation D:  Extend the Specified Treatment Technology for
High level Wastes to Stabilization of Low-Level Wastes Derived From
High level Wastes
When vitrification was named the specified technology for high
level wastes, the portion of the SRS system for treating the
low-level fraction of high level waste was not an issue.
This portion of the system was in compliance with all regulations
that were applicable at that time. Now, as years  of design and
construction are nearly complete, a new requirement is
suddenly potentially applicable to this portion of the high-level
waste treatment system. One  option for relief would be to extend
the applicability of the specified technology for treating high
level waste to the low-level fraction of that waste stream, and
determine that an immobilization/stabilization technology is a
specified technology for that low-level fraction.  The waste
treated in The Saltstone facilities are produced during the
                                            88

-------
pretreatment step to treat high level waste by vitrification which
is the specified technology.  If process changes were required to
comply with this proposed rule, delays in the high level waste
treatment program would undoubtedly result. Additionally, the EPA
should clarify that high level and low level mixed wastes with a
specified method of treatment do not need to be treated for
underlying constituents.

TECHNICAL JUSTIFICATION

At the SRS, additional removal or treatment of underlying
hazardous constituents in the high level radioactive waste
facilities in operation,  as well as those about to startup, beyond
that currently being performed would provide little or no value since
the intent of the proposed regulations will be satisfied by the
current operational plans for these facilities.  Line
breaks associated with new equipment tie-ins to existing
radioactive components  caused during construction, and emissions of
both radioactive and hazardous pollutants into the atmosphere from
monitoring requirements contained in the proposed regulations
could actually cause an increase in releases of radioactive and
hazardous constituents.
The Saltstone Processing and Disposal Facilities, which mixes
low-level mixed waste with fly ash,  slag, and cement to produce a
concrete-like grout, accomplishes the intent of the
proposed regulations by immobilizing the underlying hazardous
constituents (UHC).  As discussed in more detail in the Addendum,
the organics in the wastewater treated at the Saltstone Processing
Facility are immobilized  with the saltstone such that they meet the
'concentration based standards of the UTS when subjected  to a TCLP.
These facilities already have as their principal design tenet,
the protection of the public and the environment by preventing
contamination to the groundwater.  This protection is the essence of
the Phase III and Phase  IV proposed regulations.

ADDENDUM

Detailed Analysis of the Impact of the Proposed  Phase III  and
IV Land Disposal Restrictions to the Savannah River Site High Level
Waste System Facilities
Introduction The Environmental Protection Agency has proposed new .
Resource Conservation  and Recovery Act (RCRA) regulations known as
LDR Phase III [60 FR 11702 (March 02, 1995)],and LDR Phase IV [60
                                            89

-------
FR 43654 (August 22, 1995)] which would require hazardous or
mixed wastes that are treated and de-characterized by an approved
treatment method be further treated if there are any "underlying"
hazardous constituents in the untreated waste stream
in concentrations greater that the Universal Treatment Standard
(UTS) limits. Although the Phase III rules were intended to be
applicable to discharges managed under the Clean Water Act or Clean
Water Act-equivalent systems, the clear language of the proposed
regulations make the restrictions applicable to zero dischargers
that engage in CWA-equivalent treatment before ultimate land
disposal. The Defense Waste Processing Facility (DWPF) and
Saltstone Disposal Facility (SDF) at the Department of Energy's
Savannah River Site (SRS) discussed in this report could be
construed as being zero discharger facilities engaged in
CWA-equivalent treatment.  These new regulations could potentially
impact the ability of these two facilities to continue to operate
in direct support of the SRS's High Level Waste treatment system
as currently planned.  The discussions in this report center
primarily on the Saltstone Processing and Disposal Facility impacts
under the proposed regulations.
I. Overview of Potential Impacts
Currently, approximately 34 million gallons of high level mixed
(radioactive and hazardous) wastes are contained in 50 large, but
aging, underground storage tanks. Some storage tanks were installed
in the 1950s.  This high level waste will be treated by the LDR
specified technology of vitrification by way of complex, integrated
treatment facilities at the Savannah River Site.  Vitrification has
been determined to be the safest, long-term method for
management of such mixed waste. The proposed LDR Phase III and IV
regulations could require that certain de-characterized hazardous
wastes be further treated to remove underlying
hazardous constituents. Such regulations do not provide any
additional benefit to the safe operation of the SRS facilities, or
to human health or the environment.  Modifications to existing
facilities involved in radioactive operations to accommodate
additional treatment of constituents before they are to be
immobilized  would be very costly and cannot be justified from a
technical standpoint.  Such action could delay for many years the
removal of liquid high level mixed wastes stored in tanks that have
been in operation for more than forty years.  More  importantly,
such attempts could introduce a very real danger from potential
personnel exposure and emissions to the environment. In preparation
for startup, the DWPF Vitrification Facility, Saltstone Processing
                                           90

-------
Facility, and The Saltstone Disposal Facility have undergone
numerous independent and in-house rigorous examinations to
demonstrate their ability to safely treat, store, and dispose of
mixed wastes under their current configurations.

II. High Level Waste System Description Overview

A. Major Facilities. As a result of several decades of nuclear
materials production, approximately 34 million gallons (129 million
liters) of high level radioactive waste have been accumulated at
the SRS. The high Level Waste (HLW) System seeks to maintain safe
storage until such time as the waste shave been treated and
properly disposed.  This system includes operations which receive
various waste streams, places them into large underground storage
tanks, transfers the waste to pretreatment facilities, and which
treats the waste at the Defense  Waste Processing Facility (DWPF).
The DWPF is a one of a kind vitrification system. The entire
system includes the following key processing facilities:
Waste Storage:  Underground Storage Tanks (fifty-one tanks, one out
of service)
Volume Reduction:  Evaporator Systems
Pre-Treatment:In-Tank Precipitation Facility Extended Sludge
Processing Facility
Waste Treatment:   Defense Waste Processing Facility Saltstone
Processing Facility F-Area and H-Area Effluent Treatment Facility
Waste Disposal:   Saltstone Disposal Facility NPDES Permitted
Outfall Federal Repository
A schematic diagram has been included with this report as Figure 1
to depict the relationships between the various facilities.
B.  Underground Storage Tanks Fifty large underground storage tanks
located in two tank farms store approximately 34million gallons of
high level mixed waste produced from a number of sources as a
result of the separation of useful products from spent
aluminum-clad nuclear fuels and targets. Each tank has a capacity
of approximately one million gallons. Some of these tanks have
been in service since 1954.  The newest tank was placed into
service in 1986. [Ref. 2],  The primary objective of the high
Level Waste System is shifting from waste storage to removal of
mixed waste from the older tanks to prepare the waste fo"r feed to
the Defense Waste Processing Facility (DWPF). The wastes in these
tanks are in the form of saltcake (approximately 14.1 million
                                            91

-------
gallons), sludge including soluble salts and aluminum
(approximately 4 million gallons),  and salt
supernate (approximately 15.9 million gallons) which contains the
highest radioactive constituency.
C. Evaporator Systems The High Level Waste System utilizes two
evaporators to reduce the inventory of wastes in the tanks for the
near term.  Their operation is crucial to the success of the
mission to remove wastes from the older tanks.  To reduce the cost
of storage and improve safety, liquid supernate in the tanks is
evaporated to reduce its volume and mobility.  The overheads are
condensed and monitored to ensure that excessive amounts of
radionuclides are not being entrained.  [Ref. 3 at23]. When
necessary, the overheads are passed through an ion exchange column
to remove radioactive cesium. Following the condensing and
monitoring,  the overheads are transferred to the Effluent Treatment
Facility for further decontamination and eventual release to
the environment.  The concentrated supernate is transferred to the
In-Tank Precipitation Facility (ITP) where it is treated to remove
the majority of the radionuclides so that the salt solution can be
treated in the Saltstone Processing Facility. This ability to
reduce the waste volumes is a key
planning assumption for the Fligh Level Waste System due to the
need for tank space  capacity at the time of the Defense Waste
Processing Facility startup.
D. In-Tank Precipitation Facility  The In-Tank Precipitation
Facility (ITP) treats the salt solution from the evaporators
to remove the soluble radioactive metal ions such as cesium,
strontium, uranium, and plutonium.  This  removal is accomplished
through the precipitation of the salt solution with
sodium tetraphenylborate or adsorbed on  monosodium titanate to form
insoluble solids. The resulting precipitate, which is filtered to
concentrate the solids, contains most of the radionuclides. It is
then sent on to the Defense Waste Processing Facility for
vitrification in glass.  The decontaminated salt solution or
filtrate is passed through a stripper column to remove the
benzene generated from the decomposition of the tetraphenylborate.
The filtrate containing primarily sodium salts of hydroxide,
nitrate, and nitrite is then transferred to the Saltstone facility
for stabilization and disposal.
E. Extended Sludge Processing Facility The insoluble sludge stored
in the aging underground tanks is transferred to the
Extended Sludge Processing Facility where it is washed fo remove
the aluminum and soluble salts resulting from the addition of
                                            92

-------
sodium hydroxide to the fresh waste being sent to the tanks. If
the aluminum and soluble salts were not removed, then the Defense
Waste Processing Facility would have to increase the ratio of frit
to waste which would increase the number of glass canisters
produced. The washed sludge will be transferred to the Defense
Waste Processing Facility for vitrification. The wash water will be
returned to an evaporator system or reused to dissolve saltcake
depending on the salt concentration.
F. Defense Waste Processing Facility The one-of-a-kind Defense
Waste Processing Facility was  designed in the late  1970s
with construction started in 1984 and completed in 1989. This
facility has gone  through numerous mechanical check-outs,
modifications to  existing systems and cold chemical runs to
prepare for the start radioactive operations in December of 1995.
High level mixed waste streams (the precipitate from the In-Tank
Precipitation Facility and the washed sludge from the
Extended Sludge Processing Facility) are immobilized by melting the
waste with borosilicate glass frit and pouring the radioactive
mixture into stainless steel canisters.  This process, known as
vitrification, is the prescribed treatment standard for high level
mixed waste. The canisters will then be placed temporarily into a
temperature controlled glass waste storage building adjacent to
the vitrification facility where they will remain until transported
for disposal to a permanent federal geological repository.
Within the Defense Waste Processing Facility, the removal of
organics from the cesium-containing aqueous stream produces benzene
in the salt process cell. The precipitate slurry feed would
undergo acid hydrolysis to separate the low radioactive organic
portion from the high  radioactive water based portion.  The low
radioactive organic portion containing principally benzene would be
sent to the Organic Waste Storage Tank outside of the
vitrification facility where it will later be destroyed by
incineration at the Consolidated Incineration Facility.
[Ref. 4]. The high radioactive portion will go into the melter to
be vitrified.
The chemical process cell (also known as melter feed preparation)
within the Defense Waste Processing Facility will receive the
washed sludge from the Extended Sludge Processing Facility.  In
this  cell, a condenser is used to remove moisture and mercury from
the waste stream. This mercury is acid washed, water washed, and
vacuum distilled in preparation for eventual disposition. The
Department of Energy has explored the sale of this product to
off-site vendors  and is reviewing other options including
                                            93

-------
amalgamation.
G.  Saltstone Processing and Disposal Facilities The Saltstone
Processing and Disposal Facilities, a major part of this
integrated radioactive waste management system, were permitted for
construction by the South Carolina Department of Health and
Environmental Control (SCDHEC) in October of 1986 and construction
was completed in 1988.  The Saltstone facilities began processing
mixed waste on June 14, 1990. The land upon which these facilities
were constructed was chosen because of the water table
depth, distance to surface water and the public, and its proximity
to the waste generation site.
Pursuant to the policies and guidelines of the Department of
Energy and other regulatory agencies, a radiological performance
assessment (RPA) was prepared for the Saltstone facilities to
provide reasonable assurance that the design and disposal methods
would comply with the performance objectives of the Department of
Energy in a manner that would protect human health and safety, the
environment, and the ground water resources.
Low level mixed liquid (process aqueous salt solution) wastes from
the ITP facility and the ETF are solidified in a safe and efficient
pozzolanic process at the Saltstone Processing Facility. These
wastes are combined with a blend of cement, flyash, and slag to
generate anon-hazardous, pumpable, low level waste known as
saltstone.
The saltstone grout is then pumped to the Saltstone Disposal
Facility consisting of covered, above-ground concrete vaults where
the grout solidifies into a monolithic, non-hazardous solid waste.
H.  Effluent Treatment Facility Wastewater consisting of evaporator
overheads and other low level waste streams are sent to the
Effluent Treatment Facility. At this facility, the wastewaters
undergo decontamination through  a series of pH adjustments, solids
filtration, heavy metal and organic  adsorption, reverse osmosis,
and ion exchange. After these steps have been completed, the
treated effluent is analyzed and discharged to the environment
through a National Pollutant Discharge Elimination System permitted
outfall.  Each of the treatment processes has varying attributes.
The spent resin used in the removal of heavy metals, such as
mercury and lead, is not a hazardous waste since it does not fail
the Toxicity Characteristic Leaching Procedure. The activated
carbon is used for organics removal and could potentially become a
mixed waste if the concentrations of benzene in the influent
wastewaters are high. In the Effluent Treatment Facility .
evaporator, most of the contaminants go to the concentrate bottoms,
                                           94

-------
but some species such as metallic mercury and ammonia are volatile
if the pH is not low enough. The ETF evaporator bottoms are
eventually transferred to the Saltstone facility for disposal in
the grout.

III.  Existing Treatment of Hazardous Constituents

The quantities of the metals and organics in the waste streams fed
to the Saltstone Facility are quite small. The treatment to
produce saltstone at the Saltstone Processing Facility
converts characteristically hazardous wastewater (hazardous due to
high pH, metals and benzene) into a non-hazardous solid waste that
does not fail the TCLP.  After the stabilization process
is complete, the saltstone may possibly exceed the UTS limits for
only two organics, n-butyl alcohol and phenol. Both of these
organics result from a slow breakdown of chemicals used to
treat HLW in the ITP facility to enable separation into high-level
and low-level fractions.  The n-butyl alcohol is a product of the
hydrolysis of tributylphosphate, which is added to prevent
foaming during processing and filtration to concentrate HLW solids
in the ITP process.  The phenol is generated as a by-product of
radiolysis of tetraphenylborate, which is used as the
precipitating agent to convert radioactive cesium into an insoluble
compound that is removed by filtration. Inthe strong caustic
environment of the salt solution (ph >  13), both phenol and
n-butyl alcohol are actually present as soluble sodium salts of the
parent organic compound, and are thus not amenable to removal using
demonstrated techniques such as distillation or adsorption on
carbon beds that are normally used on wastewater containing only
trace levels of contaminants. The phenol is a by-product of
radiolysis of sodium tetraphenylborate. In other words, the
action of the radionuclides in the solution actually breaks down
the sodium tetraphenylborate to produce phenol as a by-product.
These chemical reactions occur continuously such that even
if/when these two organics are removed, they would "grow" back into
the solution.  It is conceivable that even if a very costly (in
terms of both time and money) addition was made to the facility
to remove  these two organics,  they could not be removed to the
levels required  by the UTS because of the continuous "growing in".
Furthermore, these two organic constituents are chemically bound
into the saltstone as they are converted to salts during the
stabilization process.
They will not leach  out at levels above the UTS limits when
                                            95

-------
subject to the TCLP, however, they will probably appear (converted
back to solution) when subject to a Totals analysis as required
by section 268.48.  If the EPA intends that the Phase III and Phase IV
regulations apply to this type of facility, it is possible that the
situation just described could be an inconsistency between
the Atomic Energy Act and RCRA.
SRS removes these organics to the greatest extent possible based
upon current proven technologies. During the design phase, limits
were placed on the concentrations of these constituents acceptable
by the Saltstone facility to meet regulatory requirements defined
by the EPA and adopted by the SCD HEC at the time of construction.
Nitrate was selected for modeling studies due to its high
concentration in the waste and its solubility in water. The
saltstone was tested for toxicity over a range of metal
concentrations to establish limits  in the feed that would be
blended with flyash, cement, and  slag.  As a result of these
limits, the saltstone produced was shown to be a non-hazardous
waste as defined by regulatory tests which allowed the facility to
be permitted as an  industrial solid waste disposal landfill site.
These permits require periodic reports to the state describing the
wastewater composition, both chemical and  radioactive, test
results on the saltstone produced, and the results of groundwater
monitoring. To date, the Saltstone production has complied fully
with all of the terms and conditions contained in the permits.
The performance objectives of the Saltstone facilities for
protection of the groundwater resources require that concentrations
of chemicals and radioactive contaminants at any point
of compliance not exceed the standards for public drinking water
supplies as established by the EPA. Since the Saltstone facilities
safely commenced  low level radioactive waste operations in
1990, groundwater monitoring at the points of compliance (located
at the boundary of the disposal vault facility) have met or
exceeded the safe drinking water standards.  A detailed
radiological performance assessment on the SDF vaults clearly
demonstrates that 1) organics are not released through leaching
(TCLP results show leaching well below UTS for organics even
though "total"organics in the waste form are above UTS levels), 2)
the concrete vault provides  an effective engineered barrier  against
migration of releases, 3) hydrogeological computer modeling
has estimated the time required for nitrates to reach maximum
groundwater contamination concentration levels is 7,100 years, 4)
at no time will the groundwater contamination exceed
State groundwater standards,  and 5) the risk for the existing
                                            96

-------
disposal system is less than 1E-06. Groundwater monitoring is
installed to verify compliance with groundwater standards and
provide detection of any releases that may occur.
Benzene and other organics are generated as by-products from
operations such as the radiolysis of sodium tetraphenylborate.
Portions of these organics are released into the atmosphere. For
example, the benzene generated at the In-Tank Precipitation
Facility (ITP) is released through a tall stack at the Benzene
Stripper Building and through the purge exhaust vents of the tanks
associated with the ITP system. These emissions are regulated
under permits issued by the South Carolina Bureau of Air Quality
Control.
Wastewaters consisting of evaporator overheads and other low level
waste streams are sent to the Effluent Treatment Facility. At this
facility, the wastewaters undergo decontamination through a series
of pH adjustments, solids filtration, heavy metal and organic
adsorption, reverse osmosis, and ion exchange. After these
treatment steps have been completed, the treated effluent is
analyzed and discharged to the environment through an NPDES
permitted outfall. These treatment steps concentrate the
contaminants into smaller volumes of waste which are
further concentrated by evaporation. The ETF evaporator bottoms
are eventually transferred to the Saltstone facility for disposal
in the grout.
EPA has indicated that situations which are expected to pose
little risk should be excluded from regulatory controls, and that
regulations should provide flexibility in dealing
with site-specific factors and cost-effective control alternatives.
Treating and disposing of the LLW fraction with the current
•Saltstone treatment process results in no significant risks or
impacts to human health and the environment. Installing additional
treatment for removing organics to UTS levels would actually result
in increased risk due to 1) releases and exposure resulting from
the organic treatment process and from the treatment and disposal
of the subsequent mixed wastes that would be generated from this
process, 2) increased risks associated with the extended time the
high level radioactive waste would have to remain in the aging
million gallon storage tanks, and 3) environmental impacts
associated with construction of additional process facilities.

RESPONSE

The Agency is not imposing additional treatment requirements on wastes for which BOAT
97

-------
is established as a specified method of treatment (i.e., HLVIT). That is, these wastes do not have
to be further treated for UHCs.

The commenter also raised a question regarding a residual from the treatment of a
characteristically hazardous wastewater (hazardous due to high pH, metals and benzene). The
residual is then stabilized and does not exhibit a characteristic. The stabilized waste is considered
to be a new point of generation. Assuming the residual does not exhibit a characteristic, it is not
considered a hazardous waste, and any UHCs present would not be required to be treated.
DCN PH4P089
COMMENTER ASTSWMO
RESPONDER AC
SUBJECT RADI
SUBJNUM 089

COMMENT

11. The proposed handling of previously treated mixed wastes
contaminated with characteristic metal is suitable. Mixed wastes
that have met the applicable standard at the time they were treated
should be considered in compliance even if the standards were
to change before actual disposal takes place. This is consistent
with the Storage Prohibition [40 CFR 268.50(e)], where wastes that
have met the applicable treatment standards are excluded from the
storage prohibition. Disposal capacity inadequacy for mixed
wastes continues to be a major problem for LDR compliance. The
retreatment of previously stabilized mixed wastes to address new
standards for characteristic metal waste will benefit no one while
increasing the dangers of handling these wastes.

RESPONSE

The Agency thanks the commenter for their support.
DCN PH4P113
COMMENTER Chemical Manufacturers Association
RESPONDER AC
SUBJECT RADI
SUBJNUM 113

COMMENT
98

-------
C. The treatment standards for previously stabilized mixed
radioactive and characteristic metal wastes should not be changed.
As the Agency points out,  to require more stringent treatment
standards for radioactive wastes that have been previously
stabilized could increase threat to human health and
the environment. Knowing this and taking into consideration that
the only reason these wastes have not already been land disposed is
that capacity has not been available makes it quite reasonable and,
in fact necessary, for the Agency to accept those LDR standards
which existed at the time of stabilization for these mixed wastes
that have been stabilized at "pre Phase IV" standards.

RESPONSE

The Agency thanks the commenter for their support.
                                            99

-------
Phase IV Second Supplemental Proposed Rule:
Comments and Responses on TC Metals Mixed with Radioactive Wastes
1. 2P4P-00014 Westinghouse Electric Corporation

Section III.D.6., Proposed Revision for UTS for Silver, page 26046

EPA requested comment- on quantities of silver non-waste waters that would be land
disposed. The September 1996 Mixed Waste Inventory Report for the Savannah River Site, a
Department of Energy (DOE) facility managed by Westinghouse, shows 7,240 m3 of mixed low
level wastes that include D011. There is also 126,325 m3 of high level wastes that include D011.
Since these wastes are radioactive, it is unlikely that any silver would be reclaimed or recycled.
Some of these waste streams will be macro encapsulated for land disposal. Others will be
incinerated or vitrified into glass.

Response

The Agency thanks the commenter for providing the data on quantities of radioactive
mixed wastes. The Agency recognizes the lack of available treatment capacity for characteristic
and listed mixed radioactive metal bearing wastes and believes that any new commercial capacity
that is available for mixed radioactive wastes must be used for mixed wastes that were regulated
in previous LDR rulemakings and whose variances have already expired. Therefore, the Agency
is granting a two-year national capacity variance for mineral processing wastes mixed with
radioactive wastes.

Stabilization of characteristic and listed mixed radioactive metal bearing wastes that have
previously achieved the statutory treatment standard requirements do not have to be retreated.
Likewise, for those characteristic and listed mixed radioactive metal bearing wastes that will not
be treated until after the promulgation of Phase IV, will be required to meet the revised treatment
standards, including being promulgated in today's Phase IV rulemaking. For those radioactive
mixed metal bearing wastes that can not meet these revised UTS if any should petition the Agency
for a treatability variance under 40 CFR 268.44.
2. 2P4P-00023 U.S. Department of Energy (DOE)

On May 12, 1997, the Environmental Protection Agency (EPA) published a Supplemental
Proposed Rule regarding revision of the universal treatment standards (UTS) for twelve metal
constituents and corresponding revisions to the waste-specific land disposal restrictions (LDR)
treatment standards for nonwastewater forms of characteristic metal wastes, mineral processing
wastes, and other metal-bearing wastes. The Supplemental Proposed Rule also: (1) requested
comment on several specific options concerning the regulation of secondary materials from
100

-------
mineral processing and the regulation of materials being co-processed in Bevill-exempt mining
units; (2) presented new information on threats to human health and the environment from Bevill
mining and mineral processing wastes and requested comment on whether such wastes,  which are
currently excluded from federal hazardous waste regulations, warrant regulatory controls; (3)
proposed an exclusion from the definition of solid waste for certain materials reused by wood
preserving operations; (4) proposed regulations clarifying the standards used by EPA for deciuing
whether to grant variances from LDR treatment standards; and (5) proposed regulations to
implement a prohibition on the use of most hazardous wastes as fill material.

      The Department of Energy (DOE) appreciates the opportunity to provide comments on
the May 12,  1997 Supplemental Proposed Rule (hereinafter referred to as the Phase IV second
supplemental proposal).  The enclosed comments combine the viewpoints and concerns identified
by DOE field organizations and program offices.  This consolidated DOE response to the Phase
IV second supplemental proposal focuses on several issues raised in the LDR Phase IV
rulemaking series [consisting of the initial LDR Phase IV proposed rule (60 FR 43654, August
22, 1995), the Phase IV first supplemental proposal (61 FR 2338, January 25, 1996), a notice of
data availability on Phase IV issues (61 FR 21418, May 10, 1996),  and the Phase IV second
supplemental proposal] that are of particular concern to the Department.

      DOE reiterates its support for EPA's proposal in the initial Phase IV proposed rule to
allow characteristic metal mixed wastes that have undergone stabilization prior to the effective
date of the Phase IV final rule, to comply at disposal with the LDR metal treatment standards that
were in effect at the time the wastes were stabilized (notwithstanding that actual disposal may
occur after the effective date of the Phase IV final rule).  The enclosed comments also respond to
information provided in the second supplemental proposal (62 FR 26063) concerning the
protective capability of vitrification technology. Furthermore, DOE comments support a number
of EPA's contemplated regulatory amendments regarding revised treatment standards for
characteristic metal wastes, clarification of treatment variance rules, and banning use of prohibited
.hazardous waste as fill material. Finally, a number of comments and suggestions are provided
relative to the proposed regulatory language presented in the second supplemental  proposal.

       The enclosed comments have been divided into two sections; general  and specific. The
general comment addresses a broad concern.  The specific comments relate directly to potential
regulatory approaches and issues raised in particular sections of the Phase IV second
supplemental proposal.  For clarity, each specific comment is preceded by a reference to the
section of the second supplemental proposal to which it applies and a brief description in bold-
face type of the issue within that section to which DOE's comment is directed.

 1.     DOE reiterates its support for EPA's proposal in the initial  Phase IV proposed rule to
allow characteristic metal mixed wastes, that have undergone stabilization prior to  the effective
date of the Phase IV final rule, to comply at disposal with the land disposal restriction (LDR)
metal standards that were in effect at the time the wastes were stabilized, notwithstanding that
actual disposal may occur after the effective date of the Phase IV final rule.
                                           101

-------
       In the preamble to the initial LDR Phase IV proposed rule, the U.S. Environmental
Protection Agency (EPA) discussed its intention to allow characteristic metal mixed wastes (i.e.,
radioactive wastes that exhibit the characteristic of toxicity for one or more metals) that have
undergone stabilization prior to the effective date of the Phase IV final rule, to comply at disposal
with the LDR metal standards that were in effect at the time the waste was stabilized [60 FR
43654, 43683; Aug. 22, 1995]. EPA acknowledged that there is a good possibility that when
these previously treated characteristic metal mixed wastes  are disposed, the Phase IV final rule
will be in effect and the metal portion of such wastes will be subject to revised, potentially more
stringent, treatment standard levels.  Even so, EPA stated  its belief that the prior stabilization of
such wastes achieves the statutory minimized threat standard, and to require re-treatment would
not minimize threat, but could increase it. DOE recognizes that EPA has not revisited this aspect
of the Phase IV proposed rule in the Phase IV second supplemental notice because no new
information has become available prompting the Agency to modify or augment its earlier
proposal. Nevertheless, DOE considers this issue to be very important and wants to take this
opportunity to again support EPA's August 1995 proposal and to restate some of the
Department's concerns.

a.      As was stated in DOE's comments in response to the initial Phase IV proposed rule,
[FN1: DOE Comments, Land Disposal Restrictions ~ Phase IV: Issues Associated with Clean
Water Act Treatment Equivalency, and Treatment Standards for Wood Preserving Wastes and
Toxicity Characteristic Metal Wastes, Specific Comment V.D.3, Item 1, p. 33-35 (Nov. 20,
1995).] requiring re-treatment of previously stabilized wastes that are in storage awaiting
development of disposal capacity, simply because a revision to the LDR treatment standards
becomes effective before the stored wastes can be disposed, makes sense only if the benefits
associated with re-treatment outweigh both the risks to workers and the costs. DOE agrees with
EPA's conclusion in the case of previously stabilized characteristic metal mixed wastes, that the
hazards from  added worker radiation exposure associated  with re-treatment  would probably
offset any gain in protection of human health and the environment that could result from
compliance with the Phase IV final metal treatment standards.

b.      Because stabilization is only one possible type of treatment that could be performed on
characteristic  metal mixed wastes to achieve compliance with LDR treatment standards, DOE
recommends that EPA broaden the final Phase IV compliance exception to cover previously
treated, stored, characteristic metal mixed wastes that comply with the LDR treatment standards
applicable at the time of treatment, regardless of the treatment method used to achieve
compliance (e.g., stabilization, macro encapsulation of non-debris wastes).

c.   DOE is particularly concerned that the final Phase IV compliance exception for previously
treated characteristic metal mixed wastes be appropriately codified. Therefore, DOE suggested
the  following  possible  regulatory language in its comments in response to the Phase IV proposed
rule,[FN2: DOE Comments, Land Disposal Restrictions —Phase IV: Issues Associated with Clean
Water Act Treatment Equivalency, and Treatment Standards for Wood Preserving Wastes and
Toxicity Characteristic Metal Wastes, Specific Comment V.D.3, Item 1, p. 34 (Nov. 20, 1995).]
                                           102

-------
and continues to advocate that this or similar language be added to 40 CFR 268.30(d) [as
proposed at 60 FR 43654, 43694; Aug. 22, 1995].

§268.30 Waste specific prohibitions — wood preserving wastes, and characteristic
wastes that fail the toxicity characteristic.

(d)The requirements of paragraphs (a) and (b) of this section do not apply if:

5) The wastes are radioactive wastes mixed with or containing D004 -D011 wastes, which
have been treated to meet Subpart D treatment standards in effect prior to [insert effective date of
Phase IV regulations (including any applicable national capacity variance) for radioactive wastes
mixed with D004 - D011], Such wastes must have been treated prior to [insert effective date of
Phase IV regulations (including any applicable national capacity variance) for radioactive wastes
mixed with D004 - D011] to be excluded from application of paragraph (b).

d. Finally, in future proposals of LDR treatment standards (whether such proposals address
listed or characteristic wastes), DOE urges EPA to consider including, if appropriate, compliance
exceptions for affected mixed wastes similar to the one put forth in the initial Phase IV proposed
rule. Such compliance exceptions would preclude re-treatment to meet revised LDR treatment
standards of affected mixed wastes that have been treated before the effective date of a revised
standard to meet the applicable standards at the time of treatment, and are in storage awaiting
development of disposal capacity. In 1995 and early 1996, DOE signed 31 compliance orders
(pursuant to the Federal Facility Compliance Act) with EPA or States that require compliance
with approved Site Treatment Plans (STPs). The approved STPs provide overall schedules for
achieving compliance with LDR storage and treatment requirements for mixed waste at each site.
As a result, DOE is moving forward with actions (e.g., appropriate NEPA reviews and RCRA
permitting activities) to support treatment of thousands of cubic meters of mixed waste to meet
the LDR treatment standards applicable at the time of treatment. Once treated, the wastes will be
stored (potentially for an extended period) until suitable mixed waste disposal capacity becomes
available. If EPA revises otherwise applicable LDR treatment standards during the period that
such wastes are in storage, DOE could be required to re-treat the stored wastes before disposal,
unless EPA also adopts an exception. For the same reasons that justify granting an exception
from revised LDR treatment standards for previously treated characteristic metal mixed wastes
(see item a., above), EPA should consider incorporating similar exceptions into future
rulemakings (where appropriate and where revised standards are more stringent). Such
exceptions would allow previously treated mixed waste affected by a revised LDR treatment
standard to comply with the standard that was in effect at the time the waste was treated rather
than the revised standard because the hazards from added worker radiation exposure associated
with re-treatment would probably offset any gain in protection of human health and the
environment that could result from compliance with the revised standard.

1. p. 26046. col. 3 — EPA states its belief that silver wastes are generally recycled due to
their economic value. As a result, the Agency speculates that there may be little or no land
103

-------
disposal of silver wastes, hence little or no impact from applying a new treatment standard. EPA
is seeking information on quantities of silver nonwastewaters that would be affected by LDR
treatment standards.

Many DOE sites generate, or have stored inventories of mixed wastes assigned the D011
waste code, among other codes. Data reported in the 1995 DOE Mixed Waste Inventory Report,
as revised by updated information submitted to the states during 1996, indicate that approximately
41,350 cubic meters of nonwastewater D011 mixed wastes (in the form of mixed low level wastes
and mixed transuranic wastes) were stored at 25 DOE sites as of December 1996. Additional
nonwastewater D011 mixed wastes are projected to be generated at 20 of these sites during the
coming 5 years. If practicable, DOE may use metal removal/recovery technologies to manage
some nonwastewater D011 mixed waste streams. However, due to the radioactive nature of
nonwastewater D011 mixed waste streams, DOE is treating or plans to treat most such streams
using macro encapsulation, stabilization, vitrification, or incineration technologies. Treatment
residues from these technologies that meet LDR treatment standards will be land disposed as
appropriate when capacity becomes available.

page 6, section b.

In General Comment 1 (p. 1), DOE discusses its support for EPA's proposal in
the initial Phase IV proposed rule to allow characteristic metal mixed wastes, that have undergone
stabilization prior to the effective date of the Phase IV final rule, to comply at disposal with the
LDR metal standards that were in effect at the time the wastes were stabilized. However, if EPA
decides for some reason not to grant such an exception, DOE believes a treatment variance based
on the "inappropriate" test might be fitting for situations such as the previously
stabilized mixed wastes located at the Department's West Valley Demonstration Project
(WVDP). [FN3: As discussed in the preamble to the initial LDR Phase IV proposed rule [60 FR
43654, 43683, col. 3], and in DOE's comments submitted in response [Specific Comment V.D.3,
item l.b, p. 33 (Nov. 20, 1995)], approximately 21,000 drums being stored at the WVDP contain
formerly characteristic metal mixed wastes that have been stabilized to meet currently applicable
LDR treatment standards. Continued storage of such drums in accordance with the WVDP Site
Treatment Plan is expected until approved disposal capacity becomes available. However, at the
present time it appears that disposal capacity is unlikely to be approved before the final LDR
Phase IV rule becomes effective. Therefore, unless EPA adopts an exception or grants a
variance, the final LDR Phase IV rule, if promulgated as proposed, would mandate re-treatment
of the previously stabilized wastes at WVDP because it would impose revised treatment standards
for characteristic metal wastes that would be more stringent than existing standards. DOE is
concerned that re-treatment of the previously stabilized wastes would create hazards to workers
from added radiation exposure, and would be expensive compared with any gain in protection of
human health and the environment resulting from compliance with the revised LDR treatment
standards.] Therefore, should EPA decide against the exception, DOE requests that the Agency
consider mentioning the WVDP situation in the preamble to the final revised 40 CFR 268.44 as an
example of circumstances that should qualify for an LDR treatment variance based on the
104

-------
"inappropriate" test.
Response

        The Agency agrees with the commenter that prior stabilization of such wastes achieves
the statutory minimize threat standard, and to require re-treatment could increase risk
significantly.  Therefore, no further treatment is required for mixed waste where already
stabiilized to meet an existing treatment standard but have not yet been disposed. For those
characteristic and listed mixed radioactive metal bearing wastes that will  undergo treatment post
promulgation of the Phase IV, they will be required to meet the standards being finalized in
today's rule.  In addition, the Agency recognizes the lack of available treatment capacity for
previously untreated wastes and, therefore, is granting a two-year national capacity variance for
mineral processing wastes mixed with radioactive wastes.
                                             105

-------
DCN PH4P067
COMMENTER Horsehead/Zinc Corp.
RESPONDER AC
SUBJECT SULF
SUBJNUM 067

COMMENT

EPA should increase the UTS level for sulfide to ensure that
wastewater treatment systems can both meet the UTS for sulfide and
effectively treat characteristic metal wastewaters.

RESPONSE

The Agency reviewed the basis for the sulfide standard in 40 CFR 268.48 and found it
was only intended to apply to F039. Today, the Agency is clarifying and correcting 268.48 UTS
table to limit sulfide to F039 and not to the mineral processing wastes identiifed by commenters.
These mineral processing wastes will of course be subject to the "DEACT" standard for reactive
sulfide waste streams, as set out in 268.40(a)(l) and (a)(5). The Agency will continue to
determine if a separate UTS standard for sulfide from reactive wastes from the mineral
processing and other sectors is needed and if so, will propose a standard in the near future.
DCN PH4P067
COMMENTER Horsehead/Zinc Corp.
RESPONDER AC
SUBJECT SULF
SUBJNUM 067

COMMENT

III. EPA SHOULD INCREASE THE SULFIDE UTS LIMIT FOR
WASTEWATERS.

One of the current thrusts of the Clinton Administration is to
introduce some common-sense flexibility in RCRA's LDR program (for
example, earlier this year the President proposed several"rifle
shot" legislative reforms to RCRA). One example of the exercise of
such flexibility already available in the current program, without
any need for legislation, would be to retract the applicability of
the UTS to Class I underground injection control ("UIC") wells. As
ZCA pointed out in the comment it submitted on-EPA's proposed Phase
III LDR rule, the decision in Chemical Waste Management v. EPA, 976
106

-------
      F.2d 2 (D.C. Cir.  1992), does not compel the application of the UTS
      to Class I UIC wells. Neither the Chemical Waste Management
      decision nor RCRA section 3004(m) requires EPA to impose rigid
      treatment standards for decharacterized wastes managed in Class I
      wells. Section 3004(m) requires only that EPA establish land
      disposal treatment standards that "substantially diminish the
      toxicity of the waste or substantially reduce the likelihood
      of migration of hazardous constituents from the waste so that
      short-term and long-term threats to human health and the
      environment are minimized." 42 U.S.C. § 6924(m). Thus, the
      treatment of hazardous constituents in characteristic wastes is not
      mandated unless those constituents are present at levels that
      threaten human health or the environment. ZCA and HRD therefore
      support the Administration's efforts to exempt decharacterized
      wastes from LDR treatment requirements if they are managed in Clean
      Water Act or Clean Water Act-equivalent systems.
      In a more specific context raised in the Phase IV Proposal, EPA
      should exercise its flexibility by increasing the UTS level for
      sulfide in wastewaters, which is currently 14 mg/1.
      Non-reactive sulfides commonly are used to treat wastes by chemical
      precipitation, which frequently increases the sulfide level in the
      treated waste stream. EPA has even identified chemical
      precipitation wastewater treatment as BOAT for several
      metal-bearing hazardous wastes, including K048-K052. See 57 Fed.
      Reg. 958, 964 (January 9,  1992). Since the Phase IV rule
      represents the first widespread application of the UTS to
      characteristic metal wastes, it is likely that many characteristic
      metal wastes will  be treated with sulfides in  chemical precipitation.
      Retaining the current sulfide level  of 14 mg/1 for UTS wastewaters
      in the Phase IV Proposal may significantly and unnecessarily
      increase the treatment costs for metal wastewaters because of
      the possibility that current treatment systems, which are fully
      protective of human health and the environment, would not achieve
      the sulfide wastewater UTS limit,  principally because the treatment
      itself may increase the sulfide level. As EPA recognizes, chemical
      precipitation using sulfide is a longstanding, effective, and
      environmentally protective method of waste treatment. 57 Fed. Reg.
      at 964. Therefore, EPA should increase the  UTS wastewater sulfide
      limit to ensure that wastewater treatment systems are able to
      continue using an effective treatment agent and consistently comply
      with the UTS.

RESPONSE


                                          107

-------
The Agency reviewed the basis for the sulfide standard in 40 CFR 268.48 and found it
was only intended to apply to F039. Today, the Agency is clarifying and correcting 268.48 UTS
table to limit sulfide to F039. Therefore, the current UTS standard for sulfide is applicable only to
F039 and not to the mineral processing wastes identified by commenters. These mineral
processing wastes will, of course, be subject to the "DEACT" standard for reactive sulfide waste
streams as set out in 268.40 (a)(l) and (a)(5). The Agency will continue to determine if a
separate UTS standard for sulfide from reactive wassste for mineral processing and other sectors
is needed, and if so will propose one in the future.

With respect to the comment on exempting decharacterized wastes from LDR treatment
requirements when they are managed in Clean Water Act or Clean Water Act-equivalent systems,
the Agency notes that this issues is not addressed in this rulemaking. EPA recognizes, chemical
precipitation using sulfide is a longstanding effective, and environmentally protective method of
waste treatment 57 FR 954. The Agency is currently conducting a 5-year study to assess the
effects of decharacterized wastes placed in surface impoundments, and the issue of RCRA
Subtitle C controls to such surface impoundments will be addressed at the completion of that
study.
DCN PH4P115
COMMENTER Courtaulds Fibers
RESPONDER AC
SUBJECT SULF
SUBJNUM 115

COMMENT

CFI also wishes to comment upon the designation of sulfide-bearing
waste streams as not amenable to biological treatment. The
designation of sulfide as a constituent that is not amenable to
biological treatment is based on a list submitted by the Chemical
Manufacturers Association (CMA) to EPA in response to EPA's March
1993 Supplemental Information report on potential responses to
Chemical Waste Management, Inc. v. EPA. 5
While CMA may have created the list and submitted it on behalf of
its members, CFI's experience is that some sulfide-bearing
waste streams are amenable to biological treatment and thus it is
inappropriate to classify all sulfide-bearing wastes as
nonamenable. CFI's wastewater treatment system has achieved
consistently high treatability for sulfide-bearing waste streams.
While CFI can report only on its own experience with treating
sulfide-bearing wastestreams, it is likely that other manufacturing
entities are achieving similar or better treatability efficiencies.
108

-------
       CFI would be pleased to provide whatever data it has available on
       this subject to EPA, if EPA would find these data useful.

RESPONSE

       The Agency thanks the commenter for providing additional information on sulfide-
bearing waste streams and, if appropriate,  might consider this issue in a separate rulemaking.
                                          109

-------
Phase IV Second Supplemental Proposed Rule:
Comments and Responses on Issues Related to Sulfide
1. 2P4P-00046 Mineral Policy Center

The lack of regulations on the handling of acid generating materials at mines is a huge gap
in RCRA. These acid generating sulfide materials (primarily pyrites) are abundant in many metal
ores. When exposed to air and water, they generate acid, which is harmful to aquatic life. This
acidity, in turn, enhances the toxicity of heavy metals, by dissolving them and making them more
available for uptake by organisms. According to the Bureau of Mines, mining contamination,
much of it from acid mine drainage, has polluted more than 12,000 miles of river and streams and
180,000 acres of lakes. [FN 1: Robert L.P. Kleinmann, Acid Mine Drainage, ENGINEERING
AND MINING JOURNAL, July 1989, p. 161.] Another harmful characteristic of acid mine
drainage is its insidiousness and longevity. Acid mine drainage can begin many years after a mine
closes, but once triggered, can last for centuries and is very difficult to control. Several experts
have reported that at present pollution rates, the abandoned Iron Mountain Superfund site in
Northern California can be expected to leach acid for at least 3,000 years before the pollution
source is exhausted. [FN 2: D. Kirk Nordstrom, et al.. The Production and Seasonal Variability
of Acid Mine Drainage From Iron Mountain, California: .A Superfund Site Undergoing
Rehabilitation, in ACID MINE DRAINAGE: DESIGNING FOR CLOSURE, p. 18 (21 June 199
1).] Thus acid mine drainage can plague the public with environmental and human health risks —
as well as cleanup costs ~ in perpetuity.

RCRA's failure to deal with acid generating materials is especially serious because current
federal and states regulation do not prescribe adequate requirements for monitoring wastes to test
for their potential to generate acid. . Therefore, imposing such testing requirements and other
controls on acid generating materials through a RCRA Subtitle C mining program would be a
positive step in addressing the long-term environmental and health threats posed by mining
wastes.

Response

The Agency notes that Section 3001(b)(3)(A)(ii) of RCRA excludes "solid wastes from
the extraction, beneficiation, and processing of ores and minerals" from regulation as hazardous
waste under Subtitle C of RCRA||. absent special study by the Agency and a specific
determination that regulation as a hazardous waste is warranted. Therefore, regulation of acid
mine drainage and similar mining waste is not within the scope of today's rulemaking.
7. 2P4P-00088 FMC Corporation

Clarifier Underflow Processing

110

-------
Clarifier underflow, a non-wastewater, will be pumped to a surge tank and then to a
series of lime-treatment (caustic hydrolysis) reactors for phosphorus removal and metals
stabilization. The reactors may be followed by gas separation units designed to remove phosphine
from the reactor slurry.

Specific reactor and degreaser conditions (residence time, temperature, and degree of
agitation, rate, along with distribution design for nitrogen sparging, etc.) must be determined in
laboratory tests. Because UTS for non-wastewaters have been lowered for several critical metals
in the proposed Phase IV Land Disposal Restriction for Mineral Processing Wastes, required
conditions will probably be significantly different than those defined in our previous lab and pilot
tests. Additional reagents other than lime, such as sulfides or chlorides, may be required for
treatment and are planned for evaluation in the laboratory tests. Residuals from these reagents
may affect gas- and solids-handling processes, including sizing and materials of construction,
downstream of the reactors. Nitrogen sparging requirements in the reactors and in possible
downstream de-gassing units must be defined.

Deficient reactor design could result in a treated non-wastewater stream in which
phosphorus removal is inadequate or that fails the new UTS standards. Inadequate gas removal
may result in inordinate amounts of phosphine being discharged from reactor solids if no
downstream solids handling system is installed (see Limed Solids Handling System discussion
below), or could significantly increase the design challenge for the solids handling system. Again,
laboratory testing is currently on hold pending "Treatability Study" notification of EPA, and the
required 45-day waiting period following such notification.

Lime Treatment Process Phosphine Handling System

Gases from the Caustic Hydrolysis reactors, and from other Lime Treatment System
process equipment, will be vented through ducts to a Phosphine Combustion/Scrubbing system.
Conceptually, this system will consist of a phosphine burner followed by a combustion-gas
•scrubber. Specific engineering details that must be established include phosphine burner
characteristics, capacity and materials of construction, and combustion gas scrubber type,
characteristics, capacity, and materials of construction.

System design and capacity will depend not only on estimated gas generation rate and
composition from the caustic hydrolysis reactors but on gases generated in other process systems,
including the Limed Solids Handling System (see discussion below). The scrubber design will be
particularly critical in terms of adequate removal of gas impurities before discharge. Design of
both the burner and scrubber, including combustion conditions, scrubber removal efficiency, and
materials of construction for both units, will also depend on specific reagents employed in the
reactors and other process. Any use of sulfide or chloride reagents, for example, could have a
major impact on materials selected for the gas handling system, and the required removal of any
gases associated with these reagents could alter the planned gas handling process.
Ill

-------
A technology development effort to define specific burner and scrubber conditions is
underway, and includes outside engineering experts in the design of burner/scrubber systems as
well as FMC engineering resources. A regulatory review of relevant design requirements,
standards, and considerations is also underway with our environmental consultants and attorneys.
These efforts will result in a system that is adequately designed to handle system gas flows
properly and that meets required regulations for combustion systems of this type.
Response

The Agency thanks the commenter for providing supporting information for its request
for a capacity variance. The Agency notes that, in response to the first supplemental proposal,
reviewed the first three waste streams—Medusa scrubber blowdown, Anderson filter media
(AFM) rinsate, and furnace building washdown—for which FMC requested a national capacity
variance. EPA subsequently proposed to grant a two-year capacity variance for these waste
.streams. In response to the second supplemental proposal, FMC requested a national capacity
variance for three additional waste streams—NOSAP slurry, precipitator slurry, and phossy
water—and stated that it has eliminated the generation of the AFM rinsate waste stream. After
careful review of FMC's petition, the Agency has determined that the five waste streams being
generated at the Pocatello facility would require a national capacity variance and, therefore, is
granting today a two-year national capacity variance for these wastes.

Two commenters raised issues related to the D003 subcategory reactive sulfide standard in 40
CFR 268.48 and its applicability to their processes including mineral processing wastes. We
reviewed the basis for the sulfide standard and found that it was only intended to apply to F039
wastewaters.

Today, we are clarifying and correcting 268.48 UTS table to limit sulfides to F039.
Therefore, the UTS standard for sulfide, today, will only apply to F039 and not to the mineral
processing wastes indicated by commenters. These mineral processing wastes will, of course, be
subject to the "DEACT"standard for reactive sulfide waste streams, as set out in 268.40 (a)(l)
and (a)(5). The Agency will continue to determine if a separate UTS number for reactive sulfide
from mineral processing and other sectors is needed and if so will propose a treatment standard in
the future

8. 2P4P-00103 Mining Impact Coalition of Wisconsin Inc.

Since acid mine drainage and resulting Metals contamination of surface and groundwater
often take years to (develop, how many current mining operations will ultimately become
polluters due to today's inadequate regulatory program? We are extremely concerned with the
ever-rising costs associated with almost intractable clean-ups such as the Sumnitville disaster.
Due to the difficulty and costs associates with remediation of acid mine drainage should be the
first goal. And just like wastes likely to produce acid, any other hazardous materials produced by
112

-------
mining must be treated in accordance with federal hazardous waste rules.  The mining industry
loophole must be repealed to protect our surface lakes, streams, rivers and our air and drinking
water.

        In Wisconsin, a 55-million-ton, polymetallic massive sulfide deposit (zinc and copper)
mine proposed by Exxon and Rao Algom's Crandon Mining Company is of major concern.  This
proposal includes an estimated 22 million tons of waste tailings to be impounded in a waste dump
at the site, with an additional 22 million tons to be backfilled underground. These waste tailings
are by definition, high sulfide and capable of producing large amounts of acid mine drainage for at
a minimum, hundreds of years.  The waste heavy metals known to consist in this company's
proposal include: mercury, lead, copper, zinc, cadmium, selenium, and many more.
Response

        The Agency notes that Section 3001(b)(3)(A)(ii) of RCRA excludes "solid wastes from
the extraction, beneficiation, and processing of ores and minerals" from regulation as hazardous
waste under Subtitle C of RCRA.  Therefore, regulation of acid mine drainage and similar mining
waste is not within the scope of today's rulemaking.
                                           113

-------