ILLINOIS POLLUTION CONTROL BOARD
August 4, 1988
REILLY TAR AND CHEMICAL CORP.
)
Petitioner,
)
V.
)
PCB 88—47
ILLINOIS ENVIRONMENTAL
)
PROTECTION AGENCY,
)
Respondent.
OPINION AND ORDER OF THE BOARD (by J. Anderson):
This matter comes before the Board upon a March 8, 1988
petition and an April 13, 1988 amended petition filed by Reilly
Tar and Chemical Corporation (Reilly) and the City of Granite
City (City) requesting a one year extension of its prior three
year variance granted in PCB 84—82, until August 31, 1989, from
35 Ill. Adm. Code 307.103 as it relates to Reilly’s mercury
discharge to the City’s sewer system. On July 11, 1988, the
Illinois Environmental Protection Agency (Agency) filed its
recommendation that variance be granted, subject to conditions.
Hearing was waived and none was held.
Reilly distills coal tar, which is a by—product of coke
making, at its refinery in Granite City. Its products include:
coal tar pitch, used primarily as a binder for anodes and
cathodes used in aluminum production; creosote oil, a pesticide
used to treat railroad ties, utility poles, etc; and pipeline
coatings, used on underground gas and oil transmission
pipelines. Reilly employs about 50 persons.
Reilly annually batch processes about 12—15 million gallons
of coal tar obtained from various sources, one source being
Granite City Steel, from which Reilly purchases their total
production of coal tar. The distillation process removes various
amounts of creosote oil to produce various grades of pitch. At
the start of distillation water contained in the coal tar is
first removed in a separate distillation cut. Wet scrubbers
prevent particles of creosote oil from polluting the air. Wet
scrubber water is recirculated until the creosote oil
concentration is about 50, at which point the mixture goes to an
oil—water separator. This separated water becomes part of the
wastewater produced.
In a March 29, 1985 quarterly progress report, Reilly’s
mercury concentration by wastestream is as follows (Pet. Appendix
B—5)
91—119
—2—
SOUI~ESAND QUANTITIES OF MERCURY
PRESENT IN WASTEWATER AT RI~C’S GRANITE CITY PLANT
Volurre of
Mercury
Percent
Wastewater
Load
of
Source
Mercury (rr~g/l)
(gal/week)
(g/week)
Total
I. Water Decantation
A. Tar Storage
0.006
750
0.017
0.08
B. Front End Oil Storage
0.008
9,000
0.273
1.25
C. #1 Creosote Oil Storage
0.081
5,250
1.612
7.40
II. Storirwater
(tank 100)
0.003
53,700
0.611
2.80
III. Wet Scrubbers
0.012
1,500
0.068
0.31
IV. Wet Distillate Cut
0.129
39,300
19.214
88.16
V. Miscellaneous Water
NA*
50,000
Total
100
*Not analyzed.
The discharge limit for mercury in 35 Ill. Adm. Code 307.103
is 0.0005 mg/l (0.5 ppb).
Reilly’s discharge limit in its PCB
84—82 variance is 0.035 mg/l, subject to the averaging rule of 35
Ill. Adm. Code 304.104(a).
The City’s NPDES effluent limit is
0.0005 mg/l. Reilly asserts that the mercury is
present in the
coal tar, regardless of source, and is known to occur naturally
in coal.
Reilly’s pretreatment system,
constructed in
1983, consists
of the following:
An industrial wastewater pretreatment facility
consisting of a 50,000 gallon primary settling
tank, a 100,000 gallon primary settling tank,
two settling pans with oil skimmers, a 50,000
gallon flow equalization tank with mechanical
mixing, three 250,000 gallon bio—oxidation
tanks, two 2,800 gallon rectangular
clarifiers, and all necessary pumping, piping
and appurtenances designed to treat wastewater
from a coal tar pitch and creosote oil
manufacturing operation with the discharge of
(30,000 gpd DAF; 45,000 gpd DMF; 300 PE) and
tributary to the Granite City Sewer Treatment
Plant. (Agency Rec. p. 3)
9 1— 120
—3—
On February 22, 1985, pursuant to its present variance,
Reilly commenced discharging to the City’s treatment plant.
Prior to that time, Reilly had discharged to a lagoon, which is
undergoing RCRA closure. The RCRA closure plan requires that a
maximum of about 43,000 gpd of contaminated groundwater be
removed and treated through the pretreatment facility. The
Agency noted that the 43,000 gpd, when added to the 26,000 gpd
from process and stormwater (when it rains) exceeds the hydraulic
capacity of the facility, thus reducing its effectiveness.
Reilly, however, asserts that the system over the years has
exceeded the predicted efficiency and is able to effectively
treat the 70,000 gpd being fed to it. (Agency Rec. p.4, Pet. p.
4)
During the time of Reilly’s discharge, the City’s treatment
plant effluent did not exceed 0.0005 mg/i in 1986 (Appendix A, p.
A—i). However, in 1987 and through April, 1988 the Agency
compared three apparent excursions of the City’s treatment plant
effluent with Reilly’s discharge (Pet. Table 9, Agency Rec. p.
4,5) as follows:
Discharge Monitoring Granite City
Reilly
(Table 9)
Report
Feb. 1987
0.00084
0.007
Oct. 1987
0.0017
0.0022
10/2
0.0078
10/9
0.0039
10/16
0.0089
10/23
0.0069
10/30
Jan. 1988
0.0130
0.0014
1/8
0.0016
1/15
0.0038
1/22
0.0211
1/29
There is no firm correlation between discharges from Reilly
and excess discharges by Granite City. The Board also references
the PCB 84—82 Opinion, which notes that the City exceeded its
mercury effluent limit a number of times between October 1981 and
June 1984, all prior to the time Reilly started discharging to
the City’s treatment plant.
One underlying problem has been the tests for total mercury,
which can be uncertain at such low levels in a complex matrix
such as Reilly’s, where organic—mercury compounds are present.
Reilly searched for, and feels it has found, a laboratory that
can give reliable results; however, Reilly asserts that the
results are still questionable from a statistical standpoint.
The Agency agrees that the tests are complicated by the organic—
mercury compounds; however, the Agency believes that the Standard
9 1—12 1
—4—
Methods For Examination of Waste and Wastewater, 14th Edition
does contain methods to eliminate interferences and suggests that
Reilly’s statistical variability is probably caused by the
variability of the raw material and batch operations. The Agency
noted that both Reilly and the City have programs to improve
testing accuracy.
COMPLIANCE EFFORTS
Reilly’s pretreatment efforts have produced a discharge
meeting all pretreatment requirements except for mercury. The
pretreatment goal was to reduce incoming levels of 0.1 mg/i (100
ppb) down to 0.003 mg/i (3 ppb), presumably as a step to qualify
for the pretreatment exception in 35 Ill. Adin. Code 307.103(c),
which allows a 0.003 mg/i, instead of 0.0005 mg/i, discharge if
all other requirements of that subsection are met. Reilly
believes that it can now comply with a 0.025 mg/i monthly average
(as well as a .035 daily average) if variance were granted.
However, Reilly has concluded that its only relief now is site
specific relief, for which it has applied. (R88—9) Reilly
asserts that its treatment system was one of only two systems out
of eleven studied for regulating a large number of priority
pollutants that was considered “by USEPA to be classified under
the category of Best Available Technology Economically Achievable
(BAT) for the Organic Chemicals, Plastics and Synthetic Fibers
(OCPSF) industrial category” (Pet. p. 3). The Agency stated that
the fact that USEPA has considered Reilly’s treatment facility to
be BAT does not apply to mercury, since there is no federal law
regulating mercury discharges by petitioners as a priority
pollutant. In the USEPA’s notice of final rule 52 F.R. 42522
(November 5, 1987), certain priority pollutants, including
mercury, were listed as not regulated, and the reasons why not;
in the case of mercury, the reason given was that mercury was
“Detected in Treated Effluents From a Small Number of Discharge
Sources and Uniquely Related to Those Sources” (see Appendix D—
Toxic Pollutants of the Notice, Id. 42568). The Agency correctly
asserts that mercury was not excluded because of other potential
reasons, i.e., trace amounts and insignificant toxic effects, or
sufficient controls by existing technologies. The tJSEPA’s
authority to make such a determination is contained in paragraph
8(a)iii, 1842 of a modified settlement agreement in NRDC v.
Train, 12 ERC 1833, March 9, 1979.
The Agency asserts that the necessary degree of treatment is
contained in the above mentioned exception provisions in the
Board’s rules for mercury where, in Section 307.l03(c)(3) the
requirement is the “best degree of treatment consistent with
technological feasibility, economic reasonableness and sound
engineering judgment”. The Agency does not at this time agree
that Reilly is providing the best degree of treatment.
91—122
—5—
The record addresses a number of alternatives. The Board
will address the compliance efforts in summary, recognizing that
many treatment strategies under these general headings were
discussed in the Petition, and others were summarized in the PCB
84—82 Opinion:
1)
Use of filters to
replace the wet scrubbers. Reilly
rejected this option because of minimal reductions, and short
life due to continual
plugging. The Agency feels filter design
has recently improved but agrees that greater potential
reductions exist in other
waste streams, particularly diversion
of stormwater and separate
treatment of the wet distillate cut.
2)
Diversion of
Stormwater. The stormwater averages 0.003
mg/i and might be directly discharged pursuant to the
Section
307.103(c) and subject to City approval of treatability
for
phenols, thus reducing hydraulic overload.
However, the Agency
feels this option is not
essential since the limit is exceeded
without the stormwater flow.
The Board notes that the City now
has an Agency approved Pretreatment Program, so the City will
approve or disapprove Reilly’s discharges and thus the Agency
need not renew Reilly’s permit, due to
expire on August 31, 1988.
3) Increase mercury precipitation. Reilly has undertaken a
number of efforts in this area.
The Agency had no comment except
to note that Reilly now is able to convert 75 of the mercury to
precipitate
as compared to 52
in 1985—86 (Pet. B—9, Agency Rec.
p. 7), allowing removal by the
clarifier.
4)
Increased clarifier efficiency.
Changes in design,
operation, settling aids and flow have been attempted. The
Agency believes a larger and newer clarifier could reduce the
present solids concentration in the discharge by 90, although
0.003 mg/i mercury would not be achieved.
5) Source separation. The Agency notes that the Petition does
not discuss the possibility of treating the wet cut separately
and before treatment for organics in the bio—reactors.
ENVIRONMENTAL ASSESSMENT
Reilly asserts that since grant of variance in 1984,
Mississippi water quality has not degraded, referencing the U.S.
Geological Survey’s (USGS) monitoring data collected 20 miles
upstream from St. Louis (Alton) and 137 miles downstream from St.
Louis (Thebes). (Pet. B—3,4) The Agency notes that no
statistical variation would be expected, given the large flow in
the Mississippi River. The Agency also presents other
environmental and health effects data concerning mercury: The
Food and Drug Administration guideline limit for fish flesh is
1.0 ppm (1 mg/i), which was not exceeded during the present
91—123
—6—
variance*; the USEPA published a document in 1980 r”Ambient Water
Quality Criteria for Mercury” (EPA 440/5—80—058, October 1980),
which established criteria of a) 0.00057 g/1 (0.57 mg/i) for
total recoverable mercury as a
24 hour average, with no
exceedences of 0.0017 g/l (1.7 mg/i) at any time to
protect
freshwater aquatic life and b) an
ambient water criteria of 144
ng/l (0.000144 mg/i) to protect human health.
Regarding the latter study, the authors used incidences of
mercury poisoning to determine a lowest observable effect level
(LOEL) of 200
g/day/70 kg for ingested mercury. The mercury
concentration in the average diet for food from various species
was calculated using
a “Practical Bioconcentration Factor” (PBCF)
to quantify
a species—specific ability to concentrate mercury.
The median PBCF was then used to back calculate a water mercury
concentration that would produce the fish mercury concentration,
and that figure was given a safety factor of 10.
The Agency, referenced the
following assumptions in the
criteria document and made the following observations (citations
to record omitted, Agency Rec. 9,10.):
Amount of
Mercury Concentration
Fish Eaten
in Edible Fish Tissue
Species
g/day
ppm (mg/i)
PBCF
Trout
0.561
0.240
6000
Bass
0.467
0.200
5000
Catfish
0.467
0.070
1750
Pike
0.224
0.390
9750
Total
1.719
5500
Median
a. The total amount of freshwater fish is 1.719 g/day which is
less than 10 of the 18.7 g/day/70 kg of the average total
fish and shellfish intake.
b The mercury concentration of fish tested in the Mississippi
River upstream and downstream of the facility is generally
less than that assumed in the guidance document.
C.
The PBCF calculated in the guidance document uses a
freshwater total mercury concentration of 40 ng/1 (0.000040
mg/i or 0.000000040 g/l). From the very limited data, it
*
The Agency is presumably referring to the FDA action level for
mercury in fish and shellfish of 1.0 mg/kg referenced in 49 FR
45663, November 1984.
91—124
—7—
appears that the actual Mississippi water concentration is
higher but the fish concentration is lower i.e. that the
assumed PBCF is too high to describe the Illinois data.*
The Agency also repeats, as part of its recommendation,
specific findings contained in the Board’s R70—5 Opinion by Mr.
Currie adopted March 31, 1971:
1. Although various mercury compounds are of varying
toxicity, all are subject to bacterial conversion into
the highly toxic methyl compounds. This conversion is
likely to occur under conditions common to Illinois
stream beds and to soils. Accordingly it makes sense to
deal with all discharges of mercury and its compounds on
the assumption that they may end up in methyl form.
2. Once mercury gets into the environment it neither
degrades to harmless substances nor ceases to exist.
Mercury deposits presently existing on stream beds will
be a continuing source of methyl mercury to the water
for many years. Dredging of deposits has so far proved
of doubtful value, since it stirs up the mercury and
increases water concentrations for the short term.
Morever, mercury in bottom sediments is often converted
to the volatile dimethyl mercury, which escapes to the
atmosphere and comes down in the rain many miles away.
In short, mercury once put into the environment remains
where it can do harm for a very long time.
3. Mercury is biologically concentrated by fish on the
order of 3000 times. This means that very low
concentrations in water will result in substantially
higher concentrations in fish, which people eat.
(citations to record omitted)
The Agency requests the Board’s official notice of the rest
of its R70—5 opinion and wishes to emphasize the following quote
from Mr. Currie at p. 4:
*
The Board notes that the 1980 document cited by the Agency and
discussed above has been updated (“Ambient Water Quality Criteria
for Mercury
—
1984”, EPA 44015—84—026, January, 1985), and
references procedures described in the “Guidelines for Deriving
Numerical Water Quality Criteria for the Protection of Aquatic
Organisms and their Uses” by Stephans, C.E. et al. 1985
(available from NTIS, Springfield, VA). The Board requests that
these documents (or updates, if any) be addressed in Reilly’s
site—specific proceeding or other subsequent proceedings, if any.
91—125
—8--
Furthermore, even apart from the nondegradable
aspect of mercury pollution, it would be folly
to set effluent standards at such a level as
to permit existing pollution sources in every
case to degrade the water to the level set by
the standard. To do so would transform
standard designed to protect the environment
into licenses to degrade. It would ignore the
fact that a water quality standard prescribes
not the ideal condition of the environment,
but an outer limit of dirtiness that should be
avoided if it reasonably can be. It would
commit us to the philosophy of allowing the
environment to be as dirty as we can bear it,
when our correct philosophy should be to make
the environment as clean as we reasonably
can. Finally, to allocate to existing users
the entire
waste—diluting
capacity
of the
environment would leave no room for new
industry, encourage inefficient practices, and
either discriminate against new entrants or
require a re—examination and tightening of
effluent limit whenever a new facility was
contemplated.
ECONOMIC INFORMATION
Reilly asserts that its pretreatment system’s original
capital cost exceeded $900,000. Subsequent improvements have
increased the capital investments to over $1,300,000 with
operating costs running $11.20/bOO gallons of treated water.
Since Reilly asserts it is able to treat up to 70,000 gal/day,
and the system was designed to treat about 40,000 gal/day, the
daily operating costs are presumably about $784 and $448
respectively. Reilly also stated that its estimate when it
sought its present variance was $15,000 to reach the 0.003 mg/i
level; however, it has spent about $100,000 to date (Pet. p. 21,
B—l5) and, although some progress has been made, Reilly has found
no technology to date to reach that level. Reilly has
concluded:
Reilly, during the course of investigating the
various technologies found they would either
1. Not reduce the mercury concentration below
3 ppb. 2. Create additional waste which would
have to be disposed of (a problem not
currently existing since Reilly can recycle
all the sludge presently being generated in
the treatment plant)
3. Cause Reilly’s
product to not meet customer specifications.
4. Increase operating cost at Reilly’s Granite
City plant to a point that Reilly could not
91—126
—9—
recover the cost and remain competitive in its
area of marketing (i.e. Reilly’s Granite City
plant would no longer remain a profitable
operation).
5. A combination of some or all
of the above four. (Pet. p. 21)
Reilly has also committed to continue search for and evaluate any
new technology that may become available.
Reilly also has
informed the
Agency that
it will continue the compliance plan of
the present variance, specifically changes in clarifier
operation
to increase solids capture as required by
Granite City’s
Pretreatment Program (Agency Rec. 7).
BOARD CONCLUSIONS:
At the outset, the Board emphasizes, as it did in PCB 84—82,
that variance is not requested from the City’s effluent standard,
nor is such variance being considered here.
Petitioners, as the Agency correctly asserts, have provided
no new compliance plan but the adoption of a site specific rule
in R88—9; however, a compliance plan that consists solely of a
site specific rulemaking is not acceptable (Alton Packaging
Corporation v. IEPA, PCB 83—49, May 5, 1988). Nevertheless, the
Board has occasionally granted short term “more information” or
“more research” variances for the purpose of developing a
workable compliance plan, and it will so grant a one year
variance here, with conditions. The Board is not persuaded that
Reilly has fully exhausted its design and operating options that
particularly could lead to compliance with 35 Ill. Adm. Code
307.103(c), which allows a concentration of 0.003 mg/i if other
provisions of that subsection are met.
This grant of variance is a close call. While the
environmental and human health effects are not substantial during
this grant of variance, mercury discharges are inherently a cause
of concern. On the other hand, Reilly appears to have in good
faith under its present variance actively pursued treatment
strategies, so its hardship is not self imposed. Under no
circumstances, however, should Reilly construe this one year
extension, explicitly or implicitly, as allowing a relaxation of
Reilly’s efforts to come into full compliance during the pendancy
of the site—specific proceeding. The Board particularly notes
the Agency comments that the use of filters to replace the wet
scrubber has not been persuasively shown to be infeasible, and
that the feasibility of separately treating the wet cut for
mercury before treating for organics in the bio reactors has not
been addressed. Efforts to improve clarifier design and
operation should be continued. While not essential, the option
of direct diversion of stormwater flow would reduce hydraulic
load on the treatment facility; the record indicates, however,
that this strategy might not enhance compliance. The direct
91—127
—10—
diversion question aside, the removal and treatment of the
contaminated groundwater from the lagoon site must continue as
part of Reilly’s RCRA closure plan.
Therefore, the Board finds that Reilly would suffer an
arbitrary or unreasonable hardship; Reilly’s hardship resulting
from denial of variance outweighs the injury of the public from
grant of the petition. The Board will also set the discharge
concentration at the 0.025 mg/i (25 ppb) limit proposed by Reilly
rather than the 0.035 mg/i limit in the present variance. The
Board will not order Reilly to comply with its proposed daily
limit of 0.025 mg/b (25 ppb); rather the Board will apply the
averaging rule “as is”, since there is insufficient information
in the record to evaluate this limit. The Agency’s proposed
conditions will also be included with little change. As in the
prior PCB 88—47 proceeding, Granite City also will be granted
variance from Section 307.103 so the City can allow Reilly to
discharge into its sewer system.
This Opinion constitutes the Board’s findings of fact and
conclusions of law in this matter.
ORDER
Reilly Tar and Chemical Corporation and the City of Granite
City are hereby granted variance from 35 Ill. Adm. Code 307.103,
as it applies to Reilly’s discharge, until August 31, 1989,
subject to the following conditions:
1. Reilly Tar and Chemical Corp. shall not discharge
mercury at a concentration greater than 0.025 mg/i,
subject to the averaging rule of 35 Ill. Adm. Code
304.104(a), to the City of Granite City sewer system,
nor shall the City of Granite City allow Reilly Tar and
Chemical Corp. to discharge mercury at a concentration
greater than that level.
2. Reilly Tar and Chemical Corp. shall conduct a compliance
program to reduce the mercury concentration in its
effluent consistent with the above Opinion and improve
the accuracy of its mercury analysis.
3. Reilly Tar and Chemical Corp. shall continue to submit
quarterly progress reports during the variance period
to:
Mr. Mark Books
Illinois Environmental Protection Agency
Division of Water Pollution Control
Compliance Assurance Section
2200 Churchill Road
Springfield, IL 62706;
9 1—128
—11—
4. Within 45 days of the date of this Order, Petitioner
shall execute and forward to Thomas Davis, Enforcement
Programs, Illinois Environmental Protection Agency, 2200
Churchill Road, Springfield, Illinois 62794—9276, a
Certification of Acceptance and Agreement to be bound by
the terms and conditions contained herein. This
variance will be void if the Reilly Tar and Chemical
Corp. fails to execute and forward the certificate
within the 45 day period. The 45 day period shall be
held in abeyance for any period during which the matter
is appealed. Failure to execute and forward the
Certificate within 45 days renders this variance void
and of no force and effect as a shield against
enforcement of rules from which variance was granted.
The form of the certification shall be as follows:
CERTI FICATION
I, (We), _____________________________, having read the
Order of the Illinois Pollution Control Board, in PCB 88—47,
dated August 4, 1988, understand and accept the said Order,
realizing that such acceptance rendrs all terms and conditions
thereto binding and enforceable.
Petitioner
By: Authorized Agent
Title
Date
5. Section 41 of the Environmental Protection Act (Ill.
Rev. Stat. 1985, ch. ll~/~par. 1041) provides for
appeal of final Orders of the Board within 35 days. The
Rules of the Supreme Court of Illinois establish filing
requirements.
IT IS SO ORDERED.
9 1—129
—12—
I, Dorothy NI. Gunn, Clerk of the Illinois Pollution Control
Board, hereby certify that the above Opinion and Order was
adopted on the ~ day of ~
,
1988, by a vote
of
~
.
~
Dorothy M. G~inn, Clerk
~
Illinois Pollution Control Board
9 1—130
