t r:
UnkwIStatM
Environmental Protection
Agency
v>EPA
Report of
REVIEW OW REGION IV ' s
SIGNIFICANT SUE*EKFUND REMOVAL
ACTIONS
VOLUME I
KSEH VO401818 XQ3T7
.
'
'V? '" 1
;-V"1ifc.
-*&#:
«£
'^a
-..-*£
.Si!-'-.
ri«:v '-'H^
-------
-------
fJ.J
REVIEW OF REGION IV's MANAGEMENT OF
SIGNIFICANT SUPERFUND REMOVAL ACTIONS
E5EH7-04-0181-81927
September 26, 1988
HCADQUMTERSUMMY
ENVIRONMENTAL PROTECTION AGENCY
WASHINGTON, DC. ?0460
-------
TNMIU WHMUQQM3H
'<(;
-------
TABLE OF CONTENTS
VOLUME I
Page
SCOPE AND OBJECTIVES 1
SUMMARY OF FINDINGS 2
ACTION REQUIRED 7
BACKGROUND 7
FINDINGS AND RECOMMENDATIONS
1. IMMEDIATE REMOVAL AUTHORITY EXCEEDED 11
2. COMMERCIAL TESTING AND DEVELOPMENT FUNDED 25
3. INCINERATION DID NOT MEET STANDARDS 35
4. REGION CLEANUP PREEMPTED STATE ACTION ....... 47
5. ENFORCEMENT ACTION INCOMPLETE 51
6. QUESTIONABLE CONTRACTING METHODS 55
7. SITES REMAIN CONTAMINATED 71
APPENDIX
I DISTRIBUTION 93
APPENDICES IN VOLUME II
II ENGINEERING REPORT ON PEAK OIL HAZARDOUS WASTE SITE
III PEAK OIL SITE INVESTIGATION REPORT
IV ENGINEERING REPORT ON GENERAL REFINING HAZARDOUS WASTE SITE
V GENERAL REFINING SITE INVESTIGATION REPORT
VI REGION IV RESPONSE
-------
-------
UNITED STATES ENVIRONMENTAL PROTECTION AGENCY
OFFICE OF INSPECTOR GENERAL
SOUTHERN DIVISION. SUITE 276
1 375 PEACHTREE STREET. N.E
ATLANTA. GEORGIA 3O3Oa
REGIONAL OFFICE:
EARLE CABELL Bi-DG.. RM. 2F5O
DALLAS. TX 75143
(214) 767-2854 AUDIT
(214) 767-9OO1 INVESTIGATION;-
September 26, 1988
SUBJECT: Audit Report No. E5EH7-04-0181-81927
Review of Region IV 's Management of Significant
Superfund Removal Actions
FROM: Rathryn M. Kuhl-Inclan
Divisional Inspector General for Audits
Southern Division
TO:
Greer C. Tidwell
Regional Administrator, Region IV
SCOPE AND OBJECTIVES
We have completed an audit of Region IV's management of significant
Superfund removal actions. Our objective was to evaluate the Region's
effectiveness and efficiency in managing significant removal actions at
Superfund sites. Specifically, we determined whether:
a. Sites were properly evaluated and classified as a removal in accor-
dance with the guidelines set forth in the National Contingency Plan.
b. Immediate removal was the most effective and efficient option at the
time.
c. Cleanup method selected was appropriate for the site.
d. Actions taken mitigated the site hazards.
e. Environmental laws were adhered to during the removal action, or
deviations therefrom were properly justified.
f. Responsible parties were identified and appropriately pursued to
share in the cost of the cleanup.
To accomplish our objective, we selected the two significant (cost of
over $1 million) removal actions from nine on-going removals in Region IV
as of April 1987. The two actions selected were the Peak Oil site near
Tampa, Florida and the General Refining site near Savannah, Georgia. We
observed activities at the sites and reviewed documentation and conducted
interviews at Region IV, the States of Florida and Georgia, the offices
-------
Report of Audit E5EH7-04-0181-81927
of the contractors involved in site cleanup, and Headquarters EPA. In
addition, we conducted engineering evaluations of the two sites and
requested site investigations from the EPA National Enforcement
Investigations Center (NEIC). At General Refining, the NEIC objective
was to determine the extent of contamination in wastes remaining on site
and to identify possible migration of contaminants off site. At Peak
Oil, the NEIC objective was to obtain contaminant information on lagoon
soil, incinerator ash, scrubber effluent (blowdown water) and
groundwater.
We performed our review in accordance with the Standards for Audit of.
Governments! Organizations* PrnffrAims. Activities, and Functiop^ issued by
the Comptroller General of the United States. We reviewed activities at
the sites to determine if the actions were in accordance with Agency
guidance. Activities reviewed at the two selected sites were conducted
during the period August 13, 1985 to October 30, 1987. Our review was
accomplished during the period May 15, 1987 to February 29, 1988.
Our review included tests of management controls and procedures to the
extent we considered necessary under the circumstances. We did not
specifically review all of Region IV's internal controls in effect for
conducting removal actions. The findings section of this report
identifies the control problems identified durir.g our review and the
recommendations to strengthen management's internal control over
Superfund removal actions. No other issues came to our attention which
we believed were significant enough to warrant expanding the scope of
this audit.
SUMMARY OF FINDINGS
Region IV's management of the removals at the Peak Oil and General
Refining sites cannot be considered efficient or effective. Region IV
properly evaluated and classified both sites as removals, and initially
took appropriate actions to mitigate site hazards. However, subsequent
Region IV actions used technologies that were inappropriate for immediate
removal sites. Further, a Federally funded removal may not have been the
most cost-efficient option at the time, and one responsible party was not
appropriately pursued to share in cleanup costs. Finally, environmental
laws were not always adhered to during the activities at the Peak Oil
site and the deviations made were unjustified.
Regional personnel are frequently required to make short notice decisions
in the face of urgent site conditions; however, urgency was not the
primary concern at either site once the Region stabilized the site
threats. At both sites, Regional personnel continued to use emergency
policy and procedures to justify testing unproven technology. This
report is not meant to criticize using unproven technologies at Superfund
sites. The Agency has a program that tests and develops new
technologies. Further, although this report states the two technologies
used by Region IV failed to perform as claimed, either technology may
have worked at other sites with other types of waste or may have
eventually worked at the audited sites given unlimited funding and time.
-------
Report of Audit E5EH7-04-0181-81927
1. Immediate Removal Authority Exceeded
Region IV continued unwarranted immediate removal activities after
stabilizing the hazards at the General Refining and Peak Oil sites. At
both sites, the on-scene coordinators took immediate actions to stabilize
the site hazards. The hazards at General Refining were stabilized within
one month with less than $50,000 expended. At Peak Oil, stabilization
occurred within four months with less than $255,000 expended. In the
case of Peak Oil, the site should have been referred to the Region's
remedial section personnel for an extended evaluation and assessment.
The General Refining site should have been considered for a planned
removal or cooperative agreement with the State. However, instead of
taking the proper action, the Region arranged for the mobilization,
funding, and testing of unproven cleanup technologies selected for the
respective sites. The continuation of cleanup activities under the
auspices of immediate removal was not justified and should not have been
undertaken. The on-scene coordinators presented misleading reports and
gained approval for work beyond the statutory dollar ($1 million) and
time (6 months) limits allowed by the immediate removal authority. As a
result, the Region expended $5.3 million over two years which exceeded
the immediate removal authority.
Summary of Region 's Response. The Regional Administrator disagreed with
the finding stating the initial actions only slowed the potential impact
the sites were having on the surrounding public and environment while
other complex technical and administrative issues could be pursued.
Therefore, a conscious decision was made to minimize the most immediate
threat and then plan and implement a complete action which would result
in elimination of a continual source of pollution.
Stannary of Evaluation of Region 's Response. Key to the finding is
whether the site conditions constituted emergencies as envisioned by the
National Contingency Plan, i.e., an emergency, posing an immediate risk,
and no other assistance for the site would be provided on a timely basis.
The Region does not present any information to refute our position that
initial actions taken at both sites ended any emergency that may have
existed. Thus, the criteria needed to exceed CERCLA dollar and time
limits were not met, and what the Region describes as its conscious
decision approximates a planned removal or remedial action, rather than
| an immediate removal.
Summary of Recommendation. He recommended the Regional Administrator
ensure removals are not extended beyond the limits established by the
National Contingency Plan and establish a formal review process for
exemptions from statutory time and dollar limits.
2. (^pifli^f^cial Testing and Development",
Region IV funded commercial testing and development of two hazardous
waste treatment prototypes: SHIRCO's infrared incinerator and Resources
Conservation Company's Basic Extraction Sludge Treatment (BEST) unit. To
fund the tests, the Region side-stepped several internal controls; such
as permitting, delisting, and contracting regulations. Region IV 's
selection of the two technologies was speculative and unsupported by
-------
Report of Audit E5EH7-04-0181-81927
scientific or engineering fact. Nonetheless, both prototypes were used
to conduct full-scale operations at removal sites prior to evidence the
manufacturer's performance claims were true. In selecting the
technologies for testing, the Region did not apply appropriate funding
and testing controls. As a result of the Region's actions, the two
commercial enterprises benefited from tests costing the Superfund $5.3
million. The costs of the tests conducted should have been borne by the
technology manufacturers, thus the funds expended were misused.
Summary of Region 's Response. The Regional Administrator agreed testing
and development was carried out but contends such activities were
authorized by the National Contingency Plan. The Region further stated
that their innovative technology initiative in the removal program would
be of significant long-term benefit to the Superfund program and the
Agency.
Suamary of Evaluation of Region 's Response. The Region did not dispute
the two technologies selected were not known to be feasible, cost-
effective, and available.
Suamary of Recommendation. We recommended the Regional Administrator
ensure adequate testing and independent verification is conducted
whenever consideration is given to selecting technologies.
3. Incineration Did Hot Meet Standards
The prototype infrared incinerator did not meet requirements set forth by
the Toxic Substances Control Act (TSCA), the Resource Conservation and
Recovery Act (RCRA), and EPA's own policies. The Region, not following
its own stated testing plan for the incinerator, allowed production-level
operations to commence without first conducting required tests. The
first formal test burn was conducted after 5 months of burning 2,257 tons
of hazardous waste. Burning was then allowed to continue after formal
tests indicated the prototype incinerator did not meet incineration
standards. Consequently, PCBs and particulates (mainly lead) were
released into the air and thousands of gallons of wastewater containing
lead were sent to the local wastewater treatment plant.
Summary of Region 's Response. The Regional Administrator disagreed with
the finding. The Region states TSCA destruction standards did not apply
to the incineration process at Peak Oil. The Region states the less
stringent RCRA standards were applicable which, in addition to a lower
PCS destruction standard, allowed a 720 hour test period in which to get
the prototype working prior to conducting a test burn.
Suamary of Evaluation of Region 's Response. TSCA standards applied
because PCBs in excess of 50 ppm were found at the site (the high level
of PCB is the justification to use an on-site incinerator). TSCA
requires a more stringent PCB destruction standard and a test burn when
PCBs are first incinerated.
Suamary of Recommendation. We recommended the Regional Administrator
establish internal controls to ensure removal actions are conducted in
accordance with applicable environmental standards.
-------
Report of Audit E5EH7-04-0181-81927
4. Region Cleanup Preempted State Action
Region IV preempted the State of Florida's planned cleanup of the Peak
Oil site. Florida officials had notified the Region of their intentions
to conduct a removal at the Peak Oil site. The Region initially
indicated that time and budgetary constraints prevented EPA involvement.
However, the Region later became interested in the site. Due to
confusing discussions and letters, the State delayed its plan to
incinerate at Peak Oil under a false impression it needed an incineration
permit. The Region never offered a contractual agreement for a state-
lead cleanup, and eventually took over cleanup of the site. The Region's
actions discouraged the State's participation. If the State had funded
all or part of the cleanup, the EPA could have used the Superfund money
spent at Peak Oil on other Superfund sites.
Summary of Region 's Response. The Regional Administrator disagreed with
the findixig stating that significant negotiations remained between the
State of Florida and the State's cleanup contractor before the two
parties could sign a contract. The Region also contends that State
personnel assumed incineration permits would be waived and interpreted
the EPA's June 4, 1985 letter to mean the incinerator had to be fully
permitted .
Summary of Evaluation of Region 's Response. The Region did not refute
any of the facts or comment as to why it did not offer to waive the
permits or pursue a state- lead removal under a contractual agreement.
Summary of Recommendation. We recommended the Regional Administrator
ensure state cleanups are actively pursued and encouraged.
5. Enforcement Action
The EPA did not take adequate action to pursue a responsible party
cleanup for the General Refining site. Regional personnel did not follow
proper enforcement procedures after coming to a conclusion the respon-
sible party was financially incapable of cleaning up the site. The
decision to finance the cleanup with Superfund monies was based on verbal
representations of indigence made by the responsible party. However,
written correspondence from the responsible party is not as clear, since
he states he would help in any way with site cleanup. Nonetheless,
Regional personnel started cleanup without first giving the responsible
party an opportunity to respond to a legal notice. Further, the Region
continued to preclude /responsible party involvement after evidence was
available to indicate the responsible party may have had financial
assets. As a result, the Agency may have unnecessarily spent Superfund
monies to clean up the site, as well as possibly lost any opportunity to
recover treble damages from the involved responsible party.
Summary of Region 's Response. The Regional Administrator did not agree
with the finding and stated site conditions precluded using an
administrative order and that the verbal representations of the
responsible party were sufficient to rule out issuing an administrative
order.
-------
Report of Audit E5EH7-04-0181-81927
Summary of Evaluation of Region's Response. The Region's comments did
not refute any of the facts presented in the finding. Further, neither
the conditions of the site nor the responsible party's verbal
representations of insolvency were sufficient to preclude issuing an
administrative order in accordance with Agency policy.
Summary of Recommendation. We recommended the Regional Administrator
establish internal controls to ensure compliance with Headquarters
guidance .
6. Quest innflfole Contracting Methods
Contracting methods used to fund the testing and development of unproven
technologies did not follow established Agency procurement policies and
procedures. For instance, delivery orders did not include total known
costs or scope of work and cost comparisons were not properly supported.
Also, EPA personnel allowed contractors to start work prior (up to 8
months) to signing delivery orders and did not subject the technology
manufacturers to the competitive bidding process. As a result, several
procurement procedures, which should have prevented the activities
carried on at General Refining and Peak Oil after the stabilization of
the site hazards, did not affect the projects at all.
Summary of Region 's Response. The Regional Administrator agreed the
contracting methods did not follow established policies and procedures
but stated they did not apply. The Region's policy was to let Superfund
encourage field testing of alternative technologies. The Region pointed
out that full and open competition routinely took in excess of 9 months
and, for alternative technology, the method had no established history of
success. Therefore the Region felt justified in pursuing "creative
procurement procedures" to acquire alternative technology in a timely
manner.
Summary of Evaluation of Region's Response. The Region's "creative
procurement procedures" were, and are, not in accordance with the Federal
Acquisition Regulation, the EPA Acquisition Regulation, and the Emergency
Response Cleanup Services contract. The Region's expenditure of
Superfund monies to field test the infrared incinerator and the BEST unit
was an unauthorized use of government funds. Regardless, the Region
exceeded the 9 months it considered necessary for full and open
competition by taking 1 year at one site and 2 years at the other to
contract for the alternative technologies.
Summary of Recommendation. We recommended the Regional Administrator
establish and implement adequate internal controls to ensure that on-
scene coordinators do not go beyond the limits of their authority by
permitting contractors to begin work and incur expenses prior to
obtaining ~he contracting officer's approval.
7 , Sjtea Re.Bflin
Lead and PCB still contaminate the General Refining and Peak Oil sites.
Post-treatment testing has disclosed that the sites are not clear of the
hazards which originally attracted the EPA's attention; cleanup efforts
-------
Report of Audit E5EH7-04-0181-81927
are still needed after the expending of $6 million to dispose of on-site
contaminated materials. At both sites, the primary problem is waste
piles which contain excessive levels of lead. The problems at both sites
stem from EPA personnel misjudging the capabilities of the selected
treatment technologies and the fact that neither technology worked as
promised by the manufacturer. Three of four manufacturer-advertised
merits of the BEST system proved inaccurate during actual use of the
prototype. At Peak Oil, the on-scene coordinator did not adequately
address the problems associated with the site's lead contamination. The
infrared incinerator was advertised as offering a permanent solution to
the known lead hazard; a claijn which EPA personnel discounted from the
beginning. Nonetheless, the incinerator was used and lead remains a
hazard at the site.
Summary of Region 's Response. The Regional Administrator agreed waste
piles at both sites required further cleanup. However, the Regional
Administrator disagreed that testing the BEST unit and the SHIRCO
incinerator was either inappropriate or unsuccessful. The Regional
Administrator also stated the Region would continue to use innovative
technologies at Superfund sites.
Summary of Evaluation of Region 's Response. At issue in the finding is
the fact the Region elected, in the face of regulations to the contrary,
to use two unproven technologies at immediate removal sites. Had the two
technologies worked, the developers would have profited by the proven
technologies they could nave marketed. This latter fact would open the
government to complaints of unfair competition practices and possible bid
protests from parties that may have desired to compete for the
opportunity to test their technologies.
Summary of Recommendation. We recommended the Regional Administrator
monitor contractors to achieve treatment standards required by
contractual terms and take appropriate action with the hazardous
materials remaining at the two sites.
ACTION REQUIRED
In accordance with EPA Directive 2750, the action official is required to
provide this office a written response to the audit report within 90 days
of the report date.
BACKGROUND
On December 11, 1980, Congress passed the Comprehensive Environmental
Response, Compensation, and Liability Act (CERCLA), which established the
original "Superfund". The Superfund program was created to protect
public health and the environment from the release, or threat of release,
of hazardous substances from abandoned hazardous waste sites. A Trust
Fund was established to fund cleanups ranging from control of emergency
situations to permanent remedies. The Act required that payment for
cleanups be sought from those parties responsible for the problem,
including property owners, waste generators, and waste transporters.
Congress revised CERCLA with the 1986 enactment of the Superfund
Amendments and Reauthorization Act (SARA). The amendments included
funding for an $8.5 billion program for the period 1987 through 1991.
-------
Report of Audit E5EH7-04-0181-81927
The Trust Fund is named the Hazardous Substance Superfund. The
regulatory blueprint for the Superfund Program is the National Oil and
Hazardous Substances Contingency Plan (NCP) (40 CFR Part 300). ,
The NCP defines two broad categories for responses to identified
hazardous waste sites: removal and remedial. Removals are relatively
short-tern responses to emergency-type situations and remedials are long-
term actions to provide permanent remedies for serious abandoned or
uncontrolled hazardous waste sites. If after completing a removal
action, a site still poses long-term environmental problems, such as
area-wide groundwater contamination, removal action personnel refer the
site to remedial response program personnel for further investigation,
assessment, and cleanup.
j
Congress recognized the Federal Government could only assume
responsibility for remedial response at a limited number of |
sites representing the greatest public threat. Therefore, the law f
requires maintaining a National Priorities List (NPL). The NPL ranks
sites on the basis of a standard scoring system which evaluates the r
potential threat to public health. However, each State is allowed to add j
its highest priority site to the NPL without scoring.
CERCLA Section 104(c)(3) requires the state in which the release occurs
to enter into a contract or cooperative agreement with EPA, and to
provide certain assurances including cost sharing. At most sites, the
state must pay 10 percent of the cost of remedial action. Pre-remedial
activities (preliminary assessments, site inspections), remedial planning
(remedial investigations, feasibility studies, remedial designs), and
removal actions may be funded entirely by the EPA.
GENERAL REFINING SITE HISTORY
The General Refining Superfund site is located in Garden City, Georgia
near Savannah. The site is bordered on the west by a 35-home residential
community, on the north by a railroad yard, on the east by a wooded area,
and on the south by a railroad right-of-way. The area is characterized
by sandy soil and a shallow water table.
The site is an abandoned waste oil reclaiming facility. The General
Refining Company operated the facility from 1961 to 1978. Waste oil was
brought to the site and stored in bulk tanks until processed.
Processing involved heating the waste oil (to drive off water and higher
aromatics) and collecting solids with sulfuric acids, thus forming sludge
and oil layers. The sludge was stored in unlined lagoons. The oil was
mixed with kaolin, pressed through filter cake, and canned. The used
filter cake was stockpiled.
The site consisted of a back-filled pit, four unlined lagoons containing
sludge, a sludge mound, 27 tanks which held unprocessed waste oil, and
about 75 barrels (14 containing waste material). Oil was spilled around
the site and was also in a drainage ditch between the site and a railroad
track.
-------
Report of Audit E5EH7-04-0181-81927
Contaminants identified at the site were petroleum compounds, lead,
copper, arsenic, and polychlorinated biphenyls (PCBs).
The site was monitored by the State of Georgia but was not considered
hazardous enough to be placed on EPA's National Priorities List.
The Region's initial response started August 13, 1985 and ended September
6, 1985. The Region's secondary response began with the arrival of the
BEST unit in December 1985 and ended with the eventual removal of the
unit in March 1987. The Region is presently planning its next response.
PEAK OIL SITE HISTORY
The Peak Oil Superfund site is located in Brandon, Florida near Tampa.
The site is bordered on the north by a railroad, on the east by a
railroad and a galvanizing plant, on the south by an open field, and on
the west by a railroad spur and a drum recycling facility. The area is
characterized by a sandy aquifer about 10 feet below the surface.
The site is an idle oil re-refining facility. The Peak Oil Company,
under two ownerships, operated the re-refining facility from 1950 to
1979. Since 1979, site operations have been limited due to environmental
restrictions imposed by the State of Florida.
Used crankcase oil, hydraulic fluid, and transformer fluid were re-
refined at the site. The re-refining process included distillations, a
caustic wash, a sulfuric acid wash, and filtration. A base oil was
produced which was then mixed with additives to produce motor oil. Waste
sludge from the process was stored in a holding pond.
The site contained a wiaste sludge pond (80 by 110 feet). Water from the
pond drained through an oil/water separator into several smaller pools
which drained into drainage ditches south of the site. Another area of
pooled water was inside a diked area enclosing storage tanks. Site soil
was stained with oil and held standing oil.
Contaminants identified at the site were PCBs, pesticides, solvents,
extractable organic compounds, and heavy metals.
The site was monitored by the State of Florida and Hillsborough County,
Florida.
In June 1986, the EPA grouped the site with the adjoining drum recycling
facility (Bay Drum) and placed them as a single site on the National
Priorities List. As of July 1987, the Peak Oil/Bay Drum site was ranked
as the 53rd worst site in the country.
The Region's initial response started in October 1985 and ended in
January 1986. The Region's secondary response began with the arrival of
the SHIRCO incinerator in December 1986 and ended with the eventual
removal of the incinerator in October 1987. The Region is presently
planning its next response.
9
-------
Report of Audit E5EH7-04-Q181-81927
This page left intentionally blank.
10
-------
Report of Audit E5EH7-04-0181-81927
FINDING HQ, 3- - IMMEDIATE REMOVAL AUTHORITY EXCEEDfip
Region IV exceeded its Superfund authority at the General Refining and
Peak Oil Sites. The Region extended the immediate removal activity after
stabilizing the hazards at the two sites. The on-scene coordinators
presented misleading reports to justify approval for work beyond that
allowed by immediate removal authority. These reports misrepresented the
two sites as emergency sites posing immediate risks to the public health,
welfare and the environment. However, as discussed in engineering
reports on the two sites (see appendices II and IV), the sites did not
constitute emergencies after the Region had taken actions to stabilize
the sites. The $5.3 million in expenditures for the work beyond
stabilization activity exceeded the immediate removal authority.
Background
Removal actions are relatively short-term responses to control releases
from hazardous waste sites. Removals are distinguished from remedial
actions which are long-term responses to provide permanent remedies for
hazardous waste sites. Removal actions are intended to eliminate an
immediate threat posed by a site. Both CERCLA and SARA establish time
and monetary limits for removal actions. Under CERCLA, a removal was
limited to 6 months and $1 million; SARA changed the limits to 12 months
and $2 million. Without limitations, an inordinate amount of the
Superfund monies could be spent on sites which pose a less significant
threat than others placed on the National Priorities List. However, if a
particular site cleanup could not be completed within the limitations,
the 1982 version (in effect when the Region commenced its work at both
sites) of the National Contingency Plan Section 300.65(d) provides for
exceptions if the following conditions are met:
(1) Continued response actions are immediately required to prevent,
limit, or mitigate an emergency;
(2) There is an immediate risk to public health, welfare or the
environment; and,
(3) Such assistance will not otherwise be provided on a timely
basis.
If after completing a removal action, a site still poses long-term
environmental problems, such as area-wide ground water contamination,
removal action personnel refer the site to remedial response program
personnel for further investigation and assessment.
general Refining
The emergency conditions at General Refining ceased to exist after
September 6, 1985 (see Appendix IV). As depicted in the following graph,
the Region exceeded the $1 million limit by spending over $2 million and
exceeded the 6-month time limit for by spending a total of 25 months on
the site. The removal action started August 14, 1985 with a scheduled
completion date of November 30, 1985 (three-and-one-half months);
however, site treatment continued through to October 1987 (25 months).
11
-------
Report of Audit E5EH7-04-Q181-81927
After the initial start-up, removal activity was sporadic; the site was
inactive from August 27 to December 1, 1985, and again from February 8 to
April 6, 1986. Extended breaks in an immediate removal raise questions
regarding the immediacy of the site's hazards and the justifications for
extending the removal beyond statutory limits.
General Refining, CERCLA Limits
Coal limit,
11.000.000
Time - 25 Months
tl.036.7G8
Cost - $2,038,752
(Month*)
a. Removal Action. On August 13, 1985, the Region signed a delivery
order with its emergency response contractor (HAZTBCH, Inc) for immediate
removal actions at the General Refining site. The delivery order
authorized spending $200,000 and specified an estimated time period of
three-and-one-half months. Specifically,
- Mobilize equipment to site.
- Stabilize waste lagoons.
- Transport and dispose of solids pile in an approved waste disposal
site.
- Transport waste oil for disposal/recycling at an approved
facility.
- Pursue alternate technology for treatment/disposal of waste
lagoons.
By September 6, 1985, the hazards at the site were stabilized and an
emergency no longer existed. In the words of Region personnel (quoted
from the on-scene coordinator's action memorandum of February 26, 1986):
"These initial actions served to stabilize the sludge lagoons and bulk
oil storage areas to prevent further release into the surrounding
drainage system and to allow additional time to pursue available
12
-------
Report of Audit E5EH7-04-Q181-81927
technologies for treating/disposing of the wastes."
After stabilizing the site, the Region did not consider a planned removal
action in lieu of continuing the immediate removal. A planned removal
was a viable option under the 1982 NCP. Further, the Region apparently
did not consider using a contract or a cooperative agreement with the
State of Georgia.
b. First Modification. On September 30, 1985, the Region modified the
delivery order to commence "phase II" (a phase I was not mentioned in the
original delivery order). In a memo to the Regional Administrator, the
on-scene coordinator provided the following facts:
"The original scope-of-work included stabilizing acidic oil
sludge lagoons and bulk oil storage areas. An additional task
was to pursue treatment/disposal technology of the lagoons.
"The lagoons were secured and stabilized as were the bulk oil
storage areas and some of the waste oil was sent to a recycling
facility. A fence was put up around the site to restrict
access. Also, temporary monitoring wells were installed around
the lagoons to identify off-site migration of the wastes.
Phase I of the removal was completed on September 6, 1985.
"Since completion of Phase I operations, a treatnent tech-
nology, Basic Extractive Sludge Treatment (BEST), has been
identified that will be used as a hazardous sludge reduction
process for the lagoon wastes."
This first modification also added $400,000, raising the total cost of
the project to $600,000, an increase of 200 percent.
c. Second Modification. On February 18, 1986, the Region extended the
time limit from November 30, 1985 to February 13, 1986. In a memo dated
February 13, 1986, the on-scene coordinator provided the following facts
to the Regional Administrator:
"Continued response actions of a duration greater than six
months cannot be undertaken unless an exemption to Section 104
(c) of the Comprehensive Environmental Response, Compensation
and Liability Act of 1980 (CERCLA) is granted... The six month
time limit will expire on February 13, 1986."
The coordinator outlined the three requirements for the CERCLA tine
exemption in the memo. The coordinator provided statements to establish
the site required emergency mitigation, posed an immediate risk to the
public, and that no tunely alternatives were available. Specifically:
1) Rmfirg«*n
-------
Report of Audit E5EH7-04-0181-81927
2) Risk. The coordinator stated the site constituted "a health
hazard because of the presence of hazardous substances, ...
lack of containment ... and limited access restrictions." '
3) Alternatives. The coordinator noted: the site was not on the
National Priorities List, the owner had stated he lacked funds
to cleanup the site, and the State and local governments had no
funds.
The coordinator, despite the language in the first action memorandum that
announced the $200,000 thus far spent had stabilized the site, closed the
memo by recommending continued removal activities and exemption from the
CEHCLA time limit. i
The Region's action did not meet the criteria of the NCP because the site »
had been stabilized and there was no emergency. The initial measures j
implemented during the immediate removal accomplished the objectives for
a short-term stabilizing action to prevent an immediate risk to the
public. Further, the General Refining site had not been established as a I
source of water contamination. The tests taken at the site did not *
provide any conclusive indication of ground water or surface water
contamination, much less an emergency or significant health hazard (see I
Appendix IV). '
d. Third Modification. On May 27, 1986, the Region raised the dollar
ceiling for its contractor to $2.3 million and extended the time limit to
October 31, 1986. In the approval memorandum sent to EPA Headquarters,
the Region spelled-out the activities at the site and stated some
$665,000 had been spent on the first "three phases" of the on-site waste
treatment. This included "laboratory" and "component" testing. The
Region pointed out the remaining authorized $60,000 was insufficient to
conduct any actual on-site cleanup operations. The Region also said it
needed an increase to $2,650,000 because they discovered an additional
lagoon and had made more accurate calculations of lagoon waste volume
using aerial photography.
As part of its justification package, the Region presented the following
project schedule:
1. September 20, 1985 - January 31, 1986: Conduct laboratory
analyses and component pilot testing.
2. December 16, 1985 - February 10, 1986: Mobilization and
installation of BEST treatment unit.
3. February 10 - March 10, 1988: Pre-operation unit testing,
determination of sludge pretreatment requirements, and
finalization of BEST by-product disposal/treatment requirements.
4. March 10 - June 30, 1986: Treatment of wastes on site.
5. July 1-10, 1986: Demobilization.
14
-------
Report of Audit E5EH7-04-0181-81927
Region IV informed Headquarters that several "removal alternatives" were
identified and evaluated. The alternatives, and their respective costs
were: landfilling, $8.5 million; on-site incineration, $7 million; off-
site incineration, $4.8 million; and BEST, $2.6 million. However, the
only evaluation done was to compare costs; the Region's cost data could
not later be produced. The Region also built a case for the BEST saying
that although only recently applied to oil sludges, the technology was
over 10 years old. However, the Region's memo, entitled "Removal Action
Alternative Selection", did not mention that the BEST unit did not exist
and would be built and used for the first time at the General Refining
site.
The memo also stated phases 1 and 2 (pond composition analyses and
component testing) were complete and the unit was capable of handling all
waste on site. However, the memo did not mention that neither the Region
nor HAZTECH had verified the accuracy of the phase 2 results. Instead,
the Region submitted the project for approval based on the
representations made by BEST'S manufacturer that the system would work at
the General Refining site.
e. Fourth Modification. On August 8, 1986, the Region consummated the
modification to install the BEST unit at General Refining. The unit had
been on-site since December 1985 (Resources Conservation Company had
moved the unit voluntarily to the site while the Region pursued funding).
f. Fifth Modification. On October 30, 1986, the Region temporarily
closed the site for 90 days. The stated purpose was to "make every
attempt to resolve problems associated with the BEST unit. At the end of
that time period the OSC ... will determine if the BEST unit is
operational, and whether to reactivate the site." Resources Conservation
Company was not allowed to bill the costs for this time period; however,
the Region had HAZTECH remain on site to aid the subcontractor, and
continued to pay for utilities and supplies.
Pursuant to this modification, the coordinator eventually determined the
BEST unit would not continue processing. He cited the unit's inability
to treat the filter cake and lagoon sediment solids, reach operating
capacity of 100 tons per day, and maintain consistent operations for a
sustained period.
g. Sixth Modifioation. On January 24, 1987, Resources Conservation
Company (RCC), wishing to remain on site to continue testing their unit,
agreed to process sludge at no charge to the EPA. However, the EPA paid
HAZTECH $381,209 to support RCC's activities. RCC eventually withdrew
its BEST unit from the site.
Peak Oil
By January 14, 1986, the hazards at the site were stabilized and an
emergency no longer existed. As depicted in the following graph, the
Region exceeded the $1 million limit by spending over $3.6 million and
exceeded the 6-month time limit by spending a total of 25 months on the
site. Removal activity started in October 1985 and extended through
October 1987. After completing stabilizing activity and preparing for an
15
-------
Report of Audit E5EH7-04-0181-81927
incinerator, the site was closed down from February 28, 1986 until
September 14, 1986, when the site was reactivated to prepare for a
different incinerator. As stated earlier for the General Refining
removal, an extended break during an immediate removal raises questions
as to the immediacy of the site's hazards as well as the justifications
used to extend the removal beyond statutory time and dollar limits.
Peak Oil, CERCLA Limits
Bsoeeded
Co* Limit
11,000.000
I
Time - 25 Months Cost - $3,810,968 |
a. Delivery Order. On September 27, 1985, the Region signed a delivery
order with its emergency response contractor (HA2TECH, Inc.) for actions
needing $250,000 and 6 months. The delivery order authorized HAZTECH to:
- Provide men and equipment for preparation of site for incineration
of hazardous waste.
- Provide sampling and analytical support.
- Provide mobile (transportable) incinerator for on-site destruction
of hazardous waste.
- Provide transportation and disposal of hazardous waste, if
necessary.
b. First Modification. On September 30, 1985, 3 days after signing the
delivery order, the Region modified the delivery order raising the cost
ceiling to $1.5 million. No signed memorandum was located that
authorized exceeding the CERCLA dollar limit of $1 million. However, an
undated, unsigned memo was located in the Region's files. This memo
requested the waiver, but quoted an estimated price for the project of
$1.9 million. The memo was from the Regional Administrator and was
addressed to the Assistant Administrator for Solid Waste and Emergency
Response. The memo stated the CERCLA waiver was needed because "A
16
-------
Report of Audit E5EH7-04-0181-81927
significant threat to public health and the environment exists . . . and
may worsen if response is delayed..." The memo spelled out the following
conditions:
- Destruction of 3,000 cubic yards of oil and acid sludge
containing organic chemicals and PCB was required to prevent
further contamination of ground water resources which were used
for drinking water.
- A dike wall surrounding the sludge was deteriorating and
rupture would release the contents to the soil and ground
water.
2) Risk. The memo stated 56,000 people were served by ground water
wells located within a three-mile radius of the site. The memo
cited sampling tests conducted by the State of Florida which
"found that the wells are contaminated with volatile organic
compounds . "
3) Alternatives . The memo stated remedial action was not scheduled
until 1988, responsible parties did not have sufficient funds to
conduct a removal, and "the State of Florida has indicated that
they desire EPA to coordinate and accomplish this cleanup."
The foregoing memo did not address Florida's plan, which had been
submitted to the EPA for approval, to incinerate the sludge at a cost of
$1.4 million using State funds.
c. Second Modification. On February 18, 1986, the Region extended the
period of performance to February 13, 1986.
d. Third Modification. On May 31, 1986, the Region extended the period
of performance to November 30, 1986. The on-scene coordinator, in a memo
dated May 19, 1986 to the Regional Administrator, pointed out the 6-month
time period had elapsed on March 30, 1986. One reason cited for the
delays was that the EPA's contracting mechanism was not conducive to
alternative or innovative technologies being approved in a quick and
timely fashion for the immediate removal program. In justifying a time
extension, the coordinator repeated all the points made in the previously
discussed memorandum except one. The memo of May 19th did not mention
the deteriorating dike wall because the EPA had stopped lateral
movement of both the contaminated sludges and the contaminated ground
water by January 14, 1986. The Region had fenced the Peak Oil site and
had driven a sheet piling cofferdam around the sludge lagoon down to and
into the impervious clay under layer 10-20 feet below the surface. The
coordinator did not mention that after confinement of the surface
contaminants, the site no longer posed a serious imminent threat to the
public (see Appendix II). Further, tests conducted by the State and the
EPA did not find PCBs in the deep potable water wells, but did find trace
chemical elements in a non-potable well at Peak Oil (see Appendix II).
e. Fourth Modification . On April 14, 1987, the Region modified the
17
-------
Report of Audit E5EH7-04-Q181-81927
delivery order to increase the cost to $1.8 million and extend the
allowed time to July 28, 1987.
f. Fifth Modifixation. On August 25, 1987, the Region modified the
delivery order for the installation and operation of a SHIRCO incinerator
at the site and increased the dollar limit to $3,484,700. The effective
date of the modification was November 1, 1986; the SHIRCO incinerator had
started operations on December 31, 1986. This modification set forth the
operating standards for the incinerator which had been operating for 8
months.
g. Sixth Modification. On October 28, 1987, the Region increased the
dollar limit to $3,723,388. The last incinerator activity occurred on
October 13, 1987.
Delayed Actions.
Both the General Refining and Peak Oil sites were subjected to
significant delays. In both removals, the on-scene coordinator took I
immediate action to mitigate the site hazards. Once mitigated, activity '
at both sites was delayed for a significant period of time while the
coordinators arranged for the funding, testing, and mobilization of the I
unproven technologies they had selected for the respective sites. '
The continuation of cleanup activities under the auspices of emergency
response was not justified by a "time-critical" risk and should not have
been undertaken. Once the contamination was mitigated, the removal
actions should have been terminated and long-term remedial action
instituted. The following graph illustrates the number of days which
elapsed in each phase of the removal action.
Removal Activities
fnp. 107
vn
Peak Oil General Refining
Dayn
18
-------
Report of Audit E5EH7-04-0181-81927
Conclusion
Region IV continued unwarranted activities after stabilizing the hazards
at the General Refining and Peak Oil sites. Stabilization was immediate
and well within the time and dollar limitations set forth in CERCLA. The
on-scene coordinators justified continued spending by presenting
misleading data to approval authorities. In the case of Peak Oil, the
site should have been referred to the Region's remedial section personnel
for an extended evaluation and assessment. The General Refining site
could have been considered for planned removal action or cleanup with a
contract or cooperative agreement with the State.
At the General Refining site, stabilization occurred by securing the
sludge lagoons, bulking and disposing of waste oil to a recycling
facility, fencing the site, and installing monitoring wells around the
lagoons. General Refining was stabilized within the first month and with
less than $50,000 spent. At the Peak Oil site, stabilization occurred
when the site was fenced and the sheet piling installed; the piling was
installed within four months, and less than $255,000 had been expended.
Region IV far exceeded immediate removal action requirements as set forth
in CERCLA and the NCP. Total funds committed for both sites exceeded
$6.8 million. Actual combined time from start to apparent cessation of
activities (both sites have yet to be officially closed) was 50 months.
The following graphs illustrate Region IV's cost and time used after the
site hazards were stabilized.
Activity Beyond Stabilization
CMtAboT*
6,361.600
Combined Cost
Combined Time
(Month*)
The efforts expended at the two sites were not for the sole purpose of
stabilizing the immediate hazard, but rather to test unproven treatment
technologies (see finding number 2). As a result of Region IV's effort
to test unproven technologies, the EPA spent at least $5.3 million above
19
-------
Report of Audit E5EH7-04-0181-81927
This page intentionally left blank.
20
-------
Report of Audit E5eH7-04-0181
Peak Oil
Sheet pilings at Peak Oil
driven around sludge lagoon
-------
-------
Report of Audit E5EH7-04-0181-81927
that necessary to stabilize the hazards. Unfortunately, both sites
remain contaminated (see finding number 7). Region IV's actions used
funds for testing technologies that could have been used to respond to
other critical sites.
DRAFT RECOMMENDATION
The Regional Administrator should:
a. Establish and implement written policies that ensure future removal
actions are not extended beyond mitigation.
b. Establish and implement written policies that provide for formal
reviews of sites proposed by on-scene coordinators for exemption
from statutory time and dollar limits.
REGION IV RESPONSE
The Regional Administrator disagreed with the finding and recommendation.
In addition to specific statements provided for each site (see Appendix
VI) the Administrator provided the following general comment:
"Initial actions were taken at both sites to slow the potential
impact the sites were having on the surrounding public and environment
while other complex technical and administrative issues could be pursued.
In no way was it intended that the initial measures taken at the sites
constituted total stabilization of the threat which existed. On the
contrary, criticism had been levied in the past when such limited
responses were taken without eliminating the original cause of the
problem, thus resulting in repeated reactivations to address
reoccurrences of the problem (i.e., pumping down a lagoon that was
threatening to breach without implementing more permanent actions to
prevent refilling and overflowing at a later date). This issue was
partially clarified in Draft Removal Guidance from [name deleted] ERD,
dated March 7, 1986, under a section entitled "Termination of Removal
Action;" Therefore, a conscious decision was made to minimize the most
immediate threat and then plan and implement a complete action which
would result in elimination of a 'continual source' of pollution. It was
felt if the removal was left incomplete, it could result in a much
broader and significantly more expensive remedy in the future. This
approach was later reflected in OSWER Directive #9360-0-13 entitled
"Guidance on Implementation of the "Contribute to Remedial Performance"
Provision" - Section 3.1-Purpose. The lack of reauthorization and
funding provided the planning time needed to explore the feasibility of
using these technologies."
EVALUATION OF REGION'S RESPONSE
In all documents justifying exceeding the dollar and time limits set by
CERCLA, including the specific comments for each site in Appendix VI, the
Region states the sites met the criteria for continued immediate removal.
Specifically, the sites were an immediate emergency, posed an immediate
risk, and no other assistance for the site would be provided on a timely
basis.
23
-------
Report of Audit E5EH7-04-0181-81927
Key to the finding is whether the two sites constituted an emergency site
as envisioned by Section 300.65 of the 1982 National Contingency Plan.
The Region does not present any information to refute our position that
initial actions taken at both sites ended any emergency that may have
existed. The Regional Administrator's response says the initial actions
slowed the hazards but did not constitute total stabilization of the site
threats. However, the specific responses say that the sites continued to
pose an emergency after the initial actions (see, Appendix VI) and that
the initial action at General Refining did stabilize the site but the
actions taken at Peak Oil never abated the emergency which allegedly
existed there.
In his response, the Regional Administrator quotes from a draft policy
document and a final policy document published by Headquarters after the
decisions affecting the two audited sites were made. The Regional
Administrator infers that Regional actions at the two sites anticipated
changing EPA policy towards exceeding CERCLA limits or using alternative
technologies at immediate removal sites. However, neither document
supports the actions taken by the Region. For example, the draft
guidance included a requirement to determine if non-NPL sites were
expected to be placed on the NPL. The quoted OSWER Directive states:
"... The major objective of this requirement is to provide
maximum protection of public health and the environment at
minimal cost by avoidance of removal restarts. ... a removal
action may be a phased response. The first removal action
might involve site stabilization and waste characterization.
The site may then be demobilized and closed out to allow
removal personnel to prepare an analysis of waste
treatment/disposal options. Once an option is selected, a
removal restart would actually contribute to achieving a more
efficient cleanup. Removal restarts may also occur in an
attempt to meet other program goals, such as pursuing
responsible party (RP) cleanups or State assumption of removal
action operation and maintenance requirements..."
The foregoing does not change the criteria to continue a removal action
beyond the CERCLA dollar and time limits. Furthermore, the directive
does not address the use of alternative technologies at immediate removal
sites.
RECOMMENDATION
The Regional Administrator should:
a. Establish and implement adequate controls to ensure that removals
did not continue beyond limits established by CERCLA, as amended,
and the National Contingency Plan.
b. Establish and implement a formal review process for proposed
exemptions from statutory time and dollar limits to ensure that the
conditions for the exemptions have been met. This process should
include verification of site conditions where appropriate.
24
-------
Report of Audit E5KH7-04-0181-81927
in the $200,000 cleanup delivery order (August 1985) with HAZTECH. The
delivery order, besides directing immediate cleanup actions, required
HAZTECH to "Pursue alternate technology for treatment/disposal of waste
lagoons." Competitive bidding was not conducted and HAZTECH hired
Resources Conservation Company as a HAZTECH subcontractor, thus
fulfilling its commitment to "pursue alternate technology". Contractual
dollar and time limitations were waived as discussed in finding number 1.
Since the BEST unit was a prototype, cost and performance data was not
available. Because of the lack of information, Region IV was not able to
adequately make a cost comparison of BEST to other available
technologies. The Region's analysis of BEST was misleading because the
costs of up-front material handling and eventual disposal of BEST'S
residues were not included.
b. Selection of the SHIRCQ Incinerator
HAZTECH, with the encouragement of regional personnel, agreed to purchase
and use the SHIRCO incinerator at the Peak Oil site. SHIRCO had not yet
built a production model. However, a pilot-scale model of the
incinerator was taken to the Peak Oil site, and SHIRCO conducted a test
burn. The test did not establish the incinerator's ability to destroy
PCBs in accordance with EPA standards. However, the on-scene coordinator
and a member of the EPA's Environmental Response Team wrote the following
to the EPA contracting officer as part of the justification to fund
HAZTECH's purchase:
"We feel that the incinerator has a great commercial potential
in the hazardous waste industry. Based on the initial cost
estimates, it appears that [the] infrared incinerator could be
[the] salvation for EPA in terms of cost control."
SHIRCO was only a manufacturer and did not perform site cleanups.
Therefore, it was not one of the two companies that originally bid to
incinerate PCBs at Peak Oil (see finding number 4). However, the
successful bidder was not accepted, and some time later, purchase
negotiations were started involving HAZTECH, SHIRCO, and the EPA.
Competitive bidding was not required because a purchase was not outside
the scope of the original $1.5 million delivery order which required
HAZTECH to provide an incinerator (see finding number 1). HAZTECH
protected its position by making the purchase contingent on the EPA
providing a cleanup site and by signing a performance agreement which
spelled out SHIRCO's promised capabilities of the incinerator.
Approval to exceed the dollar and time limitations of a removal was
obtained as discussed in finding number 1. Further, the incinerator did
not have to undergo permitting or delisting because operations were to be
carried out on site as an EPA removal action. However, as discussed in
finding 3, the incinerator did not pass permitting test burns and the
lead in its ash meant its residue could not be delisted.
HAZTECH submitted a proposal to the EPA for a cleanup of Peak Oil using a
SHIRCO incinerator which HAZTECH would purchase for the job. However,
the approval process was extremely slow and prompted the on-scene
coordinator to complain that the contracting process was untimely and not
27
-------
Report of Audit E5EH7-04-0181-81927
conducive to approving alternative or innovative technologies. Both the
on-scene coordinator and the emergency response team member supplied the
EPA's contracting officer with written praise of the incinerator's
assumed capabilities. The contracting officer also sought advice from
the EPA's chief of cost advisory operations. The chief said that
accepting the proposal without an audit would be "taking a tremendous
risk". The chief further stated there was insufficient information to
express an opinion on the reasonableness of HAZTECH's proposal citing
such specifics as: no invoice or written proposal from SHIRCO to support
the purchase price; no computations to support depreciation, property
insurance, capital cost of money, and liability insurance; and no
documentation to support expected maintenance, personnel, and overhead
costs. The chief also asked if it were possible for the EPA to purchase
the incinerator directly (such a purchase would have required competitive
bidding - see finding number 6).
Although required for proposals over $500,000, the contracting officer
did not request an audit of HAZTECH's $2 million-plus cost proposal. In
justifying not requesting the audit, the contracting officer cited, among
several points, contracting's review of the purchase agreement between
HAZTECH and SHIRCO, the eagerness of region personnel to "get the unit on
site to see if it would work as advertised", and HAZTECH's refusal "to
accept a firm-fixed rate with audit or renegotiation provisions". The
contracting officer approved the purchase arrangement stating:
"... it is in the EPA's best interests for HAZTECH to own a
SHIRCO incineration unit. The primary reason for this
conclusion is that it gives the EPA a 'real' alternative to
other firms which own the SHIRCO unit, and/or other units ...
This alternative creates competition and will hopefully cause
other firms to reduce their rates to reasonable levels."
Conducted Testing and Development
Prior to Region IV's activities at the General Refining and Peak Oil
sites, neither the BEST unit nor SHIRCO incinerator had been used to
treat hazardous wastes. The units used at the two sites were full-scale
prototypes, the first ever manufactured. Neither had undergone full-
scale field testing. Region IV's use of the technologies in the removals
allowed both Resources Conservation Company and SHIRCO to test their
technologies using Superfund monies.
As discussed in findings number 3 and 7, both technologies experienced
significant problems and required on-site modifications. Further, both
technologies failed to operate as advertised and as contracted for by
HAZTECH. The on-scene coordinator at General Refining eventually ordered
Resources Conservation Company to remove the difficulty-plagued BEST unit
from the site. Following repeated failures of SHIRCO's incinerator to
meet permitting requirements, HAZTECH filed suit for non-performance and
SHIRCO filed for bankruptcy. The following are examples of some of the
testing and development:
a. Testing BEST- Resources Conservation Company, along with the on-
scene coordinator and the EPA's Technical Assistance Team contractor,
-------
Report of Audit E5EH7-04-0181-81927
authored a paper extolling the BEST unit at General Refining. Some
extracts from the paper (published in the proceedings for the 7th
National Conference on Management of Uncontrolled Hazardous Waste Sites,
December 1-3, 1986, Washington D.C.) which point out problems with the
prototype are:
"Unfortunately, the BEST unit itself has not been free of
mechanical difficulties. Three problem areas have been the
centrifuge seals, the dryer solids conveying system and the
control of solvent stripper.
"Unfortunately, the seals never worked and when they failed
they caused the bearings to fail. After many centrifuge
outages, a suitable material for the seal was discovered ...
A search for an appropriate seal design continues ...
"The first [dryer solids conveying system] design failed ...
Two changes are being made to the dryer to alleviate the
situation ... It is anticipated that these modifications will
result in a controlled exit of solids.
"[The solvent still controls used live steam]. Control of the
live steam proved to be difficult. As a result... RCC also
is considering the use of a direct water injection design
instead of the live steam design."
However, the article went on to say such statements as:
"... [The] full-scale 100 ton/day solvent extraction unit is
being used to process approximately 10,000 tons of hazardous
waste which were abandoned in 1975." (The BEST actually
processed only 3,448 tons at 16 tons/day - see finding no, 7)
" The oil is trucked off-site and sold as bunker fuel; the
water is treated on-site and trucked to a large local
wastewater treatment facility; and the solids, having been
rendered non-hazardous, are disposed of on-site." (Instead of
selling the oil, EPA paid for its removal; the water was
treated on-site and trucked out; and the solids, instead of
being rendered nonhazardous, became RCRA hazardous waste and
must still be treated and disposed of - see finding no. 7)
b. Testing SHIRCQ's Incinerator. During operations at the site, HAZTBCH
made numerous modifications and design changes to the incinerator (see
finding number 3). HAZTBCH communicated its difficulties to SHIRCO in
March 1987 and demanded that SHIRCO correct the defects:
"... if SHIRCO is unable to meet the performance levels and
deadlines ... significant additional damages will be incurred.
Not only will the EPA's loss of confidence in the SHIRCO unit
reduce SHIRCO's chances for future sales ... but it will
certainly preclude the use of the unit now on the Peak Oil
site on EPA projects, significantly damage HAZTECH's reputation
with EPA, and jeopardize HAZTECH's future revenues from public
and private work."
29
-------
Report of Audit E5EH7-04-0181-81927
i
I
Defining the Technologies
If defined using the SITE program definitions, neither technology was
ready for the full-scale operations Region IV subjected them to. The
BEST unit fit the term "emerging"; whereas the SHIROO incinerator may
have been on the threshold of being an "innovative" technology.
a. BEST An Emprgjr\g Technology- The BEST unit was an emerging
technology. It was not developed enough for a full-scale field
demonstration. As noted in the EPA's November 1987 Super fund Tnnovative
Technoloy Evalua
The BEST unit is "... not scheduled for demonstration [because]
the technology is not at full scale... The BEST system was used
as part of an EPA removal action at a Savannah, Georgia site.
The developer collected sampling and performance data that ORD
[EPA's Office of Research and Development] is evaluating. If
the results show the unit is able to operate at full scale, a
demonstration will be conducted, once a suitable site is
selected."
The BEST unit was evaluated by EPA's Office of Research and Development
(ORD) during its use at General Refining. ORD evaluated data from a 24-
hour period in February 1987. The testing was requested by Region 10.
ORD's report evaluating the BEST unit is currently in its second draft.
The report does not indicate the BEST unit was ready for demonstration.
For example:
"The General Refining operation provided the opportunity to
compare a prototype full-scale commercial facility's data with
the laboratory-scale data for treatment of hazardous waste
sludges. Further testing is needed to confirm the system
efficiencies and develop complete operating and cost data."
b. SHIRCQ Neither Emerging Hor Innovative. Region IV personnel
considered the SHIRCO incinerator an innovative technology. The on-scene
coordinator says the following about the incinerator prior to its
purchase by HAZTECH:
"Based on the pilot plant data results, the infrared technology
destroyed the pollutants found at the Peak Oil site as required.
Since conditions and parameters of the pilot plant unit are
similar to those of the full sized unit, the results should be
the same".
Considering the foregoing, the next logical step would be a demonstration
to accumulate cost data. However, the incinerator, as evidenced by its
performance at the site, was not fully developed, and did not produce the
desired results.
Both Resources Conservation Company and SHIRCO entered their respective
30
-------
Report of Audit E5EH7-04-0181-81927
technologies into the SITE program; but the activities at General
Refining and Peak Oil were not part of the program. However, the
activities at Peak Oil have frequently been represented as part of SITE.
For example, the following was printed in the November 1987 Superfund
Innovative Tecimology Evaluation Program conference paper:
"The first innovative technology site demonstration was
completed at the Peak Oil Superfund site in Brandon, Florida.
This demonstration took place during a removal operation by EPA
Region IV. The Region had contracted with HAZTECH, Inc. of
Atlanta, Georgia to incinerate approximately 6,000 cubic yards
of waste oil sludge contaminated with PCBs and Lead."
The foregoing listed HAZTECH and Region IV as the technology developers,
and the on-scene coordinator as the developer point-of-contact. The
"demonstration" the article refers to was conducted under the SITE
program. However, the program personnel stated they would have been
foolish to pass up the opportunity to test the incinerator while it was
being used.
In contrast to Region IV's "demonstration", which involved burning 7,110
tons of contaminated soil over a period of 287 days using a full-scale
100-250 ton-per-day unit, an "official" planned demonstration of the
SHIRCO technology is described in the SITE program newsletter as follows:
"A portable pilot (one ton-per-day) SHIROO unit is being
evaluated at the Rose Township-Demode Road Superfund site in
Michigan." At the site, "... contaminated soil contains high
concentrations of organics, PCBs, and metals, principally lead.
The unit will operate at the site for approximately 2 weeks
and will treat about 10 cubic yards of contaminated soil."
Link to SITE Program
Both projects were started prior to the formal establishment of the SITE
program. However, even if the projects were later included in the
program, the Region failed to use the proper authority and funding
controls for the demonstrations.
a. Authorization. SITE authority was not available to any of the regions
for the removal activities, including Region IV. The "demonstrations"
commenced prior to SARA enactment on October 17, 1986. At General
Refining, site preparation for the BEST unit began in December 1985 and
operations started in July 1986. At Peak Oil, the pilot-scale model was
operated at the site in May 1986. Site preparation for the full-scale
incinerator began in September 1986, and the full-scale incinerator was
mobilized to the site in October 1986, just prior to SARA.
b. Developer Funding. Region IV did not require Resources Conservation
Company, SHIRCO, or HAZTECH to fund the costs of the "demonstrations".
Under the SITE program, demonstration projects are carried out at the
expense of the developer, except that EPA can provide up to 50 percent of
project costs if the developer can show the inability to obtain private
funding. A total of $5.3 million of Superfund monies were paid to test
31
-------
Report of Audit E5EH7-04-0181-81927
the technologies. The on-scene coordinators approved direct costs of
$400,000 for the BEST unit and $1.6 million for the incinerator. In
addition, HAZTECH was paid an additional $3.3 million for work at the
sites to support the two technologies.
Conelas ion
The activities at the General Refining and Peak Oil sites were not SITE
program demonstrations. Instead, the activities represent Region IV's
use of removals to test and develop two unproven technologies. Region IV
failed to apply appropriate funding and testing controls. Region IV's
selection of the two technologies was speculative and unsupported by
scientific or engineering fact.
DRAFT RECOMMENDATION
The Regional Administrator should:
a. Employ SITE program authorization and controls to test and develop
alternative treatment technologies.
I
b. Create and implement written procedures governing the involvement of
EPA in any form of endorsement regarding commercial testing of I
technologies. Written guidelines should preclude EPA officials from
authoring or co-authoring articles in trade journals, as well as
being quoted as endorsing one technology or company over another.
REGION IV RESgQNSE
The Regional Administrator disagreed with the finding and had no comment
on the recommendation. In addition to specific statements provided for
each site (see Appendix VI), the Regional Administrator provided the
following general comments (see Appendix VI):
"Prior to 1985, another significant piece of legislation was passed
that would have a profound impact on the entire Superfund program. An
information paper, dated December 22, 1986 from the Director of the
Emergency Response Division (ERD) summarized the impact as follows:
'The initiative to develop alternative technology policy and
guidance in the removal program had it's origin in legislation,
regulation, and EPA management. The first mandate came from
the Hazardous and Solid Waste Amendments of 1984 (HSWA),
Section 3004(d), in which Congress created a strong presumption
against land disposal of hazardous waste as being protective of
human health and the environment (unless the waste has been
previously treated). National Contingency Plan (NCP) revisions
proposed in February 1985 and finalized in November 1985 stated
in Section 300.61(c)(5) that in 'undertaking Fund-Financed
action, the lead agency shall, to the extent practicable:...
consider alternative and innovative technology.' The Hay 1985
'Procedures for Planning and Implementing Off-Site Response
Actions' directive (commonly called the 'Off-Site Policy') from
the Assistant Administrator/OSWER also strengthens the "policy
32
-------
Report of Audit E5EH7-04-0181-81927
to pursue response actions that use treatment, reuse, or
recycling over land disposal to the greatest extent
practicable... EPA requires that such alternatives be
considered for all Fund-financed and private party removal and
remedial actions.
"This memo although written in December 1986, presents the direction to
the removal program that had been forthcoming since HSWA was signed in
1984. The emphasis was on utilization of alternative technology to
destroy/reuse CERCLA waste, but the RCRA program had not been tasked by
Congress to develop these new technological alternatives; this was left
to industry to develop as the most frequently used disposal options of
the past were phased out. In addition, the SITE program within Superfund
had not been authorized when these two removals were initiated, thus
leaving no program within EPA actively working on the development of
alternative technologies."
"At this time, Region IV felt an innovative technology initiative in the
removal program would be of significant long term benefit to the
Superfund program and the Agency... The Region elected to identify two
sites for utilization of alternative technology."
EVALUATION OF REGION'S RESPONSE
The Regional Administrator does not dispute that the Region elected to
fund testing and development of two technologies which were not known to
be feasible and cost-effective. Both the BEST unit and the SHIRCO
incinerator were unbuilt prototypes for which no field cost data or field
operating data existed. Further, we are not advocating land disposal.
RECOMMENDATION
The Regional Administrator should ensure that adequate testing and
independent verification is conducted whenever consideration is given to
selecting a technology that has not been proven for the type of waste
being treated.
33
-------
Report of Audit E5EH7-04-0181-81927
This page left intentionally blank.
34
-------
Report of Audit E5EH7-Q4-0181-81927
FINDING NO. 3 - INCINERATION DID NOT MEET STANDARDS
Region IV's on-site incineration at Peak Oil did not meet EPA
incineration standards. Specifically, the on-scene coordinator allowed
HAZTECH (the Region's emergency response contractor) to operate an
untested and unpermitted incinerator which did not meet the EPA's
incineration standards. Further, the planned sequence of test burns was
not carried out because of difficulties with the prototype. The on-scene
coordinator did not require adequate testing for compliance with the
standards and allowed continued incineration when test burns showed the
standards were not met. Consequently, Region IV's use of on-site
incineration at Peak Oil resulted in PCBs and particulates being released
into the air, thousands of gallons of wastewater containing lead being
sent to the local wastewater treatment plant, and thousands of tons of
incinerator ash containing lead and possibly PCBs being left on site.
Background
Superfund removals often involve actions dealing with hazardous wastes
that are regulated under other environmental laws. EPA policy is to
pursue on-site removal actions that will meet the applicable or relevant
and appropriate requirements of other Federal environmental and public
health laws to the maximum extent practicable, considering the urgency of
the situation. (EPA Office of Solid Waste and Emergency Response policy:
CERCLA Compliance With Other Environmental Statutes; October 2, 1985).
At Peak Oil, the on-scene coordinator determined that incineration would
be the best means for removing and disposing of the source of
contamination. HAZTECH conducted the incineration, using a prototype
mobile infrared incinerator purchased from SHIRCO, under the supervision
of the on-scene coordinator. Incineration began on December 31, 1986 and
continued until October 13, 1987 with 7,110 tons of hazardous waste
burned.
Hazardous wastes (lead and PCB) regulated under the Resource Conservation
and Recovery Act (RCRA) and the Toxic Substances Control Act (TSCA) were
present at the site. Accordingly, EPA contracted with HAZTECH to operate
its infrared incinerator in accordance with the RCRA and TSCA standards.
TSCA requires monitoring of stack emissions when an incinerator is first
used to burn PCBs.
Incineration Stantjprjds
EPA's incineration standards are found in RCRA and TSCA regulations and
EPA policy. Although an incineration permit can be waived for on-site
incineration, the standards must still be met in the absence of an
overriding emergency. At Peak Oil, the Region had stabilized the
immediate hazard (see finding no. 1) by January 15, 1986. Therefore, no
emergency existed at Peak Oil that would have justified the failure to
attain incineration standards. The standards of interest at Peak Oil
were:
35
-------
Report of Audit E5EH7-04-0181-81927
1) PCS Emissions. TSCA severely limits the amount of PCB allowed to
escape out an incinerator's stack. Almost complete destruction of
PCBs is required (99.9999%). The standard is measured by comparing
the PCBs fed into the incinerator to the PCBs emitted out the stack.
(40 CFR Part 761)
2) Participate Emissions. RCRA limits the release of particulates,
mostly heavy metals, into the air. Emissions are measured by
determining the amount of particulate matter in an air sample of the
stack gas. (40 CFR Part 264)
3) PCB Residues. EPA policy limits the release of PCBs into the
environment from the ash and wastewater produced by the incineration
process. The standard is measured by the amount of PCBs in samples
of ash and wastewater. (Office of Toxic Substances Policy)
Test^Burn Results
HAZTECH began incinerating on December 31, 1986. A total of 7,110 tons
of material were burned in the incinerator when operations ceased on
October 13, 1987. Five test burns were conducted on the incinerator.
The incinerator did not meet the PCB and particulate emission standards
when burning feed material from the on-site sludge ponds. (The
incinerator did meet certain standards in selective testing, but only
when using test conditions not reflective of normal operating conditions
at the site.) Results of the five test burns follow.
Test Month
No. (1987)
Standard Met
Run Number
PCB Particulate
Emissions Emissions
1
2
June
1,2,3
4
5
August 1,2
3,4
August 1,2,3
1,2,3,4
October 1,2,3
October Preliminary
No
No
No
Yes
No
No
Yes
No
No
No
Test Ho1. Initial test burn required by TSCA. This test was
conducted 5 months after incineration began and after burning 2,257
tons of hazardous material. The test was scheduled several times,
but cancelled for various operating reasons related to the
performance of the incinerator. PCB and particulate emissions
exceeded the standards.
Test No 2. Test conducted for the EPA's SITE Program (see finding
no. 2). PCB and particulate emission standards were not met in the
first two runs of the test. In the last two runs, PCB emissions
36
-------
Report of Audit E5EH7-04-0181-81927
continued to be excessive, but particulate emissions were
acceptable. However, the acceptable level was attained only when
the flow of gas out the stack was restricted, a condition which was
not equivalent to normal operating conditions for the site.
Test No 3. PCB emissions were acceptable but particulate emission
standards were not met. However, the entire test was conducted with
material not similar to actual site composition. The material used
was spiked with PCB-laced oil, imported to the site in a tanker, to
create a more burnable feed.
Test No 4. Conducted on the day before incineration was completed
at the site. Particulate emission standard was not met. PCB
emissions were not tested.
Test Ho 5. Conducted after work at the site was completed and for
the purpose of obtaining a national TSCA permit to incinerate PCBs.
The permit would have enabled HAZTECH to use the incinerator
elsewhere in the country. During preparation for the test, a TSCA
permitting official noted the stack's smoke suggested excessive
emissions. A preliminary stack test found particulate emissions
three times the allowable standard. HAZTECH personnel cancelled the
official permit test burn.
Qperat i on A1\pwed
The on-scene coordinator allowed production-level operations to commence
without first conducting appropriate tests. Specifically, the TSCA
required test burn was not performed. The coordinator's project
documentation shows a test burn was to occur within the first 10 days of
incineration. The documentation also reflects the coordinator's
intention to shut down the incinerator following the test bum and await
the test results before restarting operations. The following are
pertinent extracts from the site's daily pollution reports:
December 31, 1986
"Full scale test burn expected in approximately 10 working
days."
January 2, 1987
"-12/31/87 [sic] - HAZTECH completed a limited burn of the
incinerator today. A total of 3 tons of material was
incinerated. Samples of the feed, ash, and blow-down water
were taken and will be analyzed.
"During the burn a number of problems were identified.
"Begin full scale burn as soon as all technical problems are
corrected".
37
-------
Report of Audit E5EH7-04-0181-81927
January 6, 1987
I
,1
"Three tons of lagoon material were processed and successfully \
incinerated on 31 December 1986. '
"Full scale test burn expected to be underway on or before 15
January 1987."
January 25, 1987
i
"Two tons of material were processed in the 7 hours of running."
February 9, 1987
"On February 5, 1987 HAZTECH started up the incinerator ...
[and] in order to get a better handle on the problems HAZTECH
operated the incinerator until the next day.
"The schedule now anticipates completion of all the major
modifications and start up later this week."
February 15, 1987
"On February 13 ... approximately 12.5 tons of material were
burned. However, a problem developed with the screw auger ...
and the unit was shut down. It was decided that SHIRCO
personnel would be called in to run the unit until after
completion of the demonstration burn."
March 31, 1987
"Daily ash samples have been collected for analysis.
"On 30 March a meeting was held with EPA, HAZTECH, and SHIRCO
to review problems with the incinerator.
"To date 242.32 tons of material have been incinerated.
"Future plans ... continue incineration of material"
April 24, 1987
"... Incineration continues.
"Approximately 270 tons of slightly contaminated ash has been
returned to the lagoon area, this material will be deducted
from [HAZTECH's bill to the EPA].
"A meeting of representatives from HAZTECH, SHIRCO, and EPA was
held on 22 April, 1987. The demonstration burn is tentatively
scheduled for 4 May 1987."
38
-------
Report of Audit E5EH7-04-0181-81927
May 22, 1987
"TAT [Technical Assistance Team] member noted a release of
brown smoke from incinerator on May 13. HAZTECH's explanation
was that 'subatomic' carbon was being emitted during bakeout
as a result of low oxygen in the primary.
"The demo burn is now tentatively scheduled for ... 30 or 31
May 1987."
May 29, 1987
"The demo has been delayed until 8 June to allow . . . time to
make necessary equipment adjustments and calibrations."
June 15, 1987
"On June 14 , ... a five hour test run was done ..."
The first test burn was conducted after 5 months of site operations
and the burning of 2,257 tons of hazardous waste. The incinerator
did not meet the test burn standards.
In March 1987, HAZTECH notified SHIRCO that the incinerator was not
working in accordance with the purchase agreement and said:
"... if SHIRCO is unable to meet the performance levels and
deadlines ... significant additional damages will be incurred.
Not only will the EPA's loss of confidence in the SHIRCO unit
reduce SHIRCO 's chances for future sales ... but it will
certainly preclude the use of the unit now on the Peak Oil site
on EPA projects, significantly damage HAZTECH's reputation with
EPA, and jeopardize HAZTECH's future revenues from public and
private work."
Testing
The on-scene coordinator did not test the incinerator's residues for
compliance with the PCB residues standard. Neither the ash nor the
incinerator's wastewater were properly sampled. The on-scene coordinator
frequently had the ash sampled for PCBs, but rarely tested the
wastewater .
Project documentation indicates the on-scene coordinator intended to test
the ash for PCBs on a daily basis. Yet, tests were not made on 46 (37
percent) of the 126 days on which ash was produced. On tested days, 15
percent of the ash did not meet the PCB residue standards and required
reburning.
Wastewater was transferred to a local treatment plant. Tests for PCBs in
the wastewater were not made on 3,026,320 gallons (98 percent) of the
3,084,740 gallons sent to the plant. The wastewater that was tested met
the standard.
39
-------
Report of Audit E5EH7-04-0181-81927
Prior to June 29, 1987, the on-scene coordinator allowed HAZTECH to
transfer 1,029,520 gallons of untreated incinerator wastewater to the
local treatment plant. Samples of the wastewater showed its lead level
was as high as 100 milligrams per liter; the local treatment plant could
only accept 1 milligram per liter. The treatment plant objected to the
high lead level in the wastewater and refused to accept any more
wastewater from the site. The on-scene coordinator was forced to shut
down the incinerator. After June 29, HAZTECH added on-site treatment to
reduce heavy metals. The treatment reduced the lead levels to acceptable
levels, according to tests performed on three of the 57 transfers made to
the plant after June 29th.
Test Results not
The on-scene coordinator could not explain why tests were not always
taken, but did state that test burn reports were not available.
Therefore, we asked the EPA's National Enforcement Investigation Center
(NEIC) to do on-site sampling including tests of the operating
incinerator. However, when the NEIC team arrived from Denver, the
incinerator was down for maintenance to cure problems which had left PCBs
in the ash. The on-site HAZTECH engineer stated he was on 24-hour orders
to get the unit operating. The on-site technical assistance team member
confirmed the status of the incinerator and stated it would be operating
the next day. The on-scene coordinator called from Atlanta and asked the
NEIC team leader why the team was on the site. Following that
conversation, the HAZTECH site manager informed the NEIC team that the
incinerator would be down for at least a week, and that the HAZTECH
engineer had returned to Atlanta. Therefore, NEIC's tests could not be
conducted .
Test burn results were provided for our review after incineration at the
site was completed.
Incineration Presented s
The on-scene coordinator and the EPA's Superfund Innovative Technology
Evaluation (SITE) program office took part in implying the incineration
process at Peak Oil was a success. For example, the following was
printed in the October 17, 1987 National Journal and quoted the on-scene
coordinator as saying "It took quite a while to debug it ... It was quite
an ordeal getting it ready, but we turned a corner in June, and it's been
doing real well since then" .
The SITE program mentioned the SHIRCO incinerator in its November Program
newsletter:
"The first innovative technology site demonstration was
completed at the Peak Oil Superfund site . . . The Region had
contracted with HAZTECH, Inc, of Atlanta, Georgia to incinerate
approximately 6,000 cubic yards of waste oil sludge contamin-
ated with PCBs and lead . . . The main components of the demon-
stration were to evaluate the reliability of the unit in
destroying PCBs, and to validate the manufacturer's claim that
40
-------
Report of Audit E5EH7-04-0181-81927
May 22, 1987
"TAT [Technical Assistance Team] member noted a release of
brown smoke from incinerator on May 13. HAZTECH's explanation
was that "subatomic" carbon was being emitted during bakeout
as a result of low oxygen in the primary.
"The demo burn is now tentatively scheduled for ... 30 or 31
May 1987."
May 29, 1987
"The demo has been delayed until 8 June to allow ... time to
make necessary equipment adjustments and calibrations."
"On June 14,
June 15, 1987
a five hour test run was done
The first test burn was conducted after 5 months of site operations
and the burning of 2,257 tons of hazardous waste, The incinerator
did not meet the test burn standards.
In March 1987, HAZTECH notified SHIROO that the incinerator was not
working in accordance with the purchase agreement and said:
"... if SHIFCO is unable to meet the performance levels and
deadlines ... significant additional damages will be incurred.
Not only will the EPA's loss of confidence in the SHIRCO unit
reduce SHIRCO 's chances for future sales ... but it will
certainly preclude the use of the unit now on the Peak Oil site
on EPA projects, significantly damage HAZTECH's reputation with
EPA, and jeopardize HAZTECH's future revenues from public and
private work."
The on-scene coordinator did not test the incinerator's residues for
compliance with the PCB residues standard. Neither the ash nor the
incinerator's wastewater were properly sampled. The on-scene coordinator
frequently had the ash sampled for PCBs, but rarely tested the
wastewater .
Project documentation indicates the on-scene coordinator intended to test
the ash for PCBs on a daily basis. Yet, tests were not made on 46 (37
percent) of the 126 days on which ash was produced. On tested days, 15
percent of the ash did not meet the PCB residue standards and required
reburning.
Wastewater was transferred to a local treatment plant. Tests for PCBs in
the wastewater were not made on 3,026,320 gallons (98 percent) of the
3,084,740 gallons sent to the plant. The wastewater that was tested met
the standard.
39
-------
-------
Report of Audit E5EH7-04-0181-81927
lead compounds could be converted from a soluble to an insoluble
form by the SHIRCO process."
Incineration Results
Although some PCBs were destroyed, the incineration process did not meet
TSCA and RCRA requirements. Specifically, stack emissions were
unacceptable and included PCBs and lead. Wastewater sent to the local
treatment plant contained unacceptable lead levels, and ash produced by
the process was a hazardous waste.
During the time period of the attempted destruction of the hazardous
wastes, particulate matter in the incinerator stack exhaust gases was
well in excess of the maximum amount permitted by RCiRA standards.
Further, although continuous sampling for chemical composition of the
exhaust gases was not performed, tests were taken at different times over
a 3-month period. The average of the test results indicates that
approximately 60 percent of the particulate matter being exhausted was
lead.
Analysis of the incineration records reveals that the SHIRCO unit was
processing hazardous wastes for approximately 1,875 hours. Available
test data indicates that during this time the particulate emissions rate
was approximately 3.5 pounds per hour over the RCRA allowed maximum.
This is equivalent to over three tons of excessive particulate matter
emitted to the atmosphere during the total operation of the incineration
unit, the majority of which was lead.
Prior to adding a water clarifier for on-site treatment, the 1,029,520
gallons of wastewater sent to the local wastewater treatment plant
contained lead as high as 100 times what the plant could accept for
processing. The on-scene coordinator also allowed 3,026,320 gallons of
wastewater to be sent to the plant without testing for PCBs.
Over 7,000 tons of incinerator ash were left on the site. NEIC, in
limited sampling, found the ash was hazardous because of the lead content
(see finding no. 7). Complete sampling results are not available on the
ash; therefore, unknown quantities of PCBs and hazardous wastes other
than lead may exist in the ash.
The release of PCBs through the stack gases, and possibly through the
incinerator wastewater, is of concern. In TSCA, Congress singled out
PCBs for both immediate regulation and phased withdrawal from the market.
PCBs are of concern because tests on laboratory animals show that chronic
exposure to PCBs may cause reproductive failures, gastric disorders, skin
lesions, and tumors. Moreover, PCBs persist and when released into the
environment tend to accumulate in tissues of living organisms. This
means that as PCBs move up in the food chain toward humans their
concentration and thus their potential for harm increases.
41
-------
Report of Audit E5EH7-04-0181-81927
I
It
This page intentionally left blank. \
42 [
-------
Report of Audit E5eH7-04-0181
Peak Oil
Incinerator before water clarifier was added
to reduce lead sent to treatment plant
Incinerator after water clarifier
43
-------
-------
Report of Audit E5EH7-04-0181-81927
Conelus ion
Although EPA incineration permits could be waived, the EPA incineration
standards had to be met. Thus, the following actions taken at the site
were not in accordance with sound judgment:
A test burn was not performed prior to full scale operations.
Incineration continued without a test burn.
Incineration continued after test bums indicated EPA
incineration standards were not met.
Testing for contaminants in the wastewater sent to the local
wastewater treatment plant was inadequate.
. Transfers of contaminated wastewater were made to the local
wastewater treatment plant.
DRAFT RECOMMENDATION
The Regional Administrator should establish and implement written
procedures that ensure on-scene coordinators conduct removal actions in
accordance with the standards of applicable environmental laws.
REGION IV RESPONSE
The Regional Administrator disagreed with the finding and therefore had
no comment on the recommendation. The Region's verbatim comments are in
Appendix VI.
In the response, the Region states a lower PCB destruction standard
(99.99%) applied to the incineration at Peak Oil and that the Region was
allowed 720 hours of incineration prior to performing the first test
burn. The Region also says the incinerator ash and wastewater were
adequately tested and that the discharge of blowdown water was made with
the local wastewater treatment plant's knowledge. In the response to
finding number 7, the Region agrees the particulate standard was
exceeded, but not by a significant amount. The Region also says Peak Oil
was in an area where no people were living within one mile of the
downwind plume of the incinerator stack.
EVALUATION OF REGION'S RESPONSE
The finding stated EPA standards for incineration were violated,
quantified the extent of some of the violations, and cited data which
shows releases of contaminants from the incinerator exceeded EPA's
limits. The Region's response quotes the incineration standards of RCRA
(99.99 percent destruction and 720 test hours). However, as the Region
states in its response to finding number 2, TSCA standards applied
because PCBs in excess of 50 ppm were found at the site (the high level
of PCB is the justification to use an on-site incinerator). TSCA
requires a 99.9999 percent destruction and a test burn when PCBs are
first incinerated. In addition, the standards set for the incinerator in
45
-------
Report of Audit E5EH7-Q4-0181-81927 j
the contract between HAZTECH and the EPA were those of TSCA a 99.9999 ^^
percent destruction of PCBs. Further, the on-scerie coordinator planned
for a successful test burn at that level of destruction prior to allowing I
production-level incineration. *
While saying the incinerator ash and wastewater were adequately tested, I
the Region did not provide any test results in addition to those |
mentioned in the finding. Therefore, as stated in the finding, 37
percent of the ash and 98 percent of the wastewater was not tested. *
The Region states the discharge of blowdown water was made with the local
wastewater treatment plant's knowledge. However, site documentation
indicates the plant was not aware of the lead content of the wastewater. !
Only after the treatment plant objected to the high lead levels in the I
wastewater and stopped accepting the discharge did the Region take
action. f
The Regional Administrator should establish and implement adequate I
internal controls to ensure that removal actions are conducted in *
accordance with applicable environmental standards. These controls
should include frequent monitoring of affected media and treatment I
products, and provision for suspension of operations until needed I
corrective measures are implemented to the extent the immediacy of the
hazard permits.
46
-------
Report of Audit E5EH7-04-0181-81927
FINDING NO. 4 - REGTON CLEANUP PREEMPTED STATE ACTION
Region IV preempted a state cleanup of the Peak Oil site. Specifically,
the State of Florida was prepared to cleanup the site, but Region IV
personnel delayed the State's plans by saying the State needed an
incineration permit. Region IV did not offer a contractual agreement to
the State for a state-lead cleanup, but eventually took over the site
from the State. As a result, Region IV performed a removal which could
have been performed by the State.
Background
EPA regulations (40 CFR Part 300) encourage the use of state cleanups.
The regulations allow states to perform cleanups under state authorities
without Federal funding, or state-lead cleanups under EPA agreements with
Federal authority and funding. Superfund authorizes the EPA to enter
into a contract or cooperative agreement with a state when the Agency
determines a state has the capability to undertake a cleanup. EPA policy
specifically waives the application for and receipt of permits for on-
site emergency response actions conducted pursuant to CERCLA.
State Cleanup
On August 15, 1984, the State of Florida notified Region IV of the
State's intentions to conduct initial actions at the Peak Oil site.
Region IV indicated that time and budgetary constraints prevented EPA
involvement. The State intended to neutralize, solidify, and remove the
contents of the acid sludge pond. At the time, the State planned to use
a landfill for disposal of the hazardous waste.
The State proceeded to plan a cleanup with their cleanup contractor, O.H.
Materials, and with funds from the State's Water Quality Assurance Trust
Fund. However, landfill disposal fees increased at the planned disposal
site and the cost of the project increased significantly. As a result,
the State searched for a cost-effective alternative to the use of the
landfill.
The State decided the cost-effective alternative would be on-site
incineration and selected a mobile incinerator manufactured by
Environmental Systems Company (ENSCO). The State went through its
contracting procedures and by June 1985 was ready to begin a $1.4
million, 18-week cleanup involving 12 weeks of on-site incineration.
However, the State was unsure whether the incinerator needed a permit and
wrote the following to the EPA on May 15, 1985:
"... request an expeditious clarification of some recent
revisions of EPA guidance which state that removal actions at
Superfund sites are required to meet the criteria and standards
of other Federal laws but do not require permits. Although we
believe that this permit exemption applies to state funded
actions at NPL sites, we would like it confirmed before we
begin this project.
47
-------
Report of Audit E5EH7-04-0181-81927
"The only mobile incinerator ... which is capable of this
project, is one owned and operated by ENSCO ... This system is
designed to meet all destruction and emissions standards. The
company's liquids incinerator is TSCA certified for PCB
destruction, but the solids incinerator, which is needed at
Peak Oil, is still awaiting this certification.
"... we are excited about the prospect of taking
aggressive, innovative action to destroy contaminants from a
Superfund site rather than landfilling them in another state
or 'stabilizing' on site. We plan to begin sludge incineration
in June, so we need confirmation of the permit exemption as
soon as possible."
State personnel say the Region verbally implied the permit exemption did
not apply, and that the permitting process would take 8 months to 2
years. In an internal memo dated Hay 15, 1985, Region personnel discuss
"... if they [Florida] were going to carry out the clean-up work, the
Superfund permitting exemption did not apply." On June 4, 1985, the EPA
Director of the Office of Emergency and Remedial Response wrote the
following to the State:
"Members of my staff have spoken with representatives of
the Region IV removal program in an effort to assist you in
your cleanup efforts. As you know, the Region will be sending
an On-Scene Coordinator (OX) to the site in early June to
conduct an on-site assessment. If conditions at the site are
determined by the OSC to present an immediate and significant
risk to public health and the environment in accordance with
Section 300.65 of the National Contingency Plan, Region IV will
undertake a Federally-funded removal at the site. A permit
will not be required to implement this Superfund response
action, including on-site incineration of PCB waste, as you have
proposed."
In an internal memo dated October 22, 1985, the State summarizes its
$56,022 spent to date and says: "The EPA has taken the lead on this site
and has budgeted $1.9 million for sludge treatment and $700,000 for a
remedial investigation and feasibility study. Final cleanup costs have
not been estimated. Future [Florida] costs will be limited ... to track
activities."
Region IV brought in its emergency cleanup contractor, HAZTECH, Inc., and
solicited bids from incineration contractors. In a letter dated December
6, 1985, ENSCO quoted the Peak Oil incineration, with an on-site test
burn, would cost $2.6 million. On February 6, 1986, a Region press
release said a contract had been awarded to ENSCO to incinerate PCBs and
other hazardous wastes at the Peak Oil site. However, ENSCO's bid was
later deemed unacceptable and the cleanup was not started until December
1986 using an incinerator manufactured by SHIRCO.
48
-------
Report of Audit E5EH7-04-0181-81927
Permit Waiver
After ignoring the Florida permit exemption inquiry, the EPA granted a
waiver for a responsible party cleanup at a site referred to as Florida
Steel. The following table illustrates the inequity of permits waived.
Site.
Peak Oil
Peak Oil
Florida Steel
Lead Party
State of Florida
Region IV
Responsible Party
Permit
Waived
No
Yes
Yes
Inc inerator
ENSCO
SHIRCO
SHIRCO
The ENSCO mobile incinerator is a rotary kiln, which is a traditional
incineration technology. The incinerator was permitted for liquid PCB
disposal but not for solid PCB disposal as found at the Peak Oil site.
The SHIRCO mobile incinerator is an infrared model, a unique and patented
incineration technology. SHIROO is the sole manufacturer of infrared
incinerators. The two SHIROO incinerators used at the Peak Oil and
Florida Steel sites were the first two manufactured. Neither were
permitted for the disposal of any kind of hazardous waste.
Conclusion
Region IV preempted the State's cleanup of Peak Oil. If the State had
funded all or part of the cleanup, the EPA could have used the Superfund
money spent at Peak Oil on other national priority sites. The Region's
actions actively discouraged State participation in cleaning up hazardous
waste sites.
DRAFT RECOMMENDATION
The Regional Administrator should implement written procedures to ensure
that state cleanups are not preempted but actively pursued and
encouraged.
REGION IV RESPONSE
The Regional Administrator disagreed with the finding and therefore did
not comment on the recommendation. The Region's verbatim comments on the
finding can be located in Appendix VI. In the response, the Region said
State of Florida personnel thought ENSCO's cost estijnate was unreasonably
low and that significant negotiations remained before the two parties
could sign a contract. The Region also contends State personnel assumed
incineration permits would be waived under CERCLA and interpreted EPA's
June 4, 1985 letter to mean ENSCO must be permitted.
EVALUATION OF REGION'S RESPONSE
The Region did not explain why they did not confirm a permit waiver or
why they did not pursue a state-lead removal under a contractual
agreement. Instead, the Region states the State dropped the project
49
-------
Report of Audit E5EH7-04-0181-81927
because of contracting difficulties with ENSCO. However, the State
actually dropped the project because a TSCA permit for incineration was
not waived by Region IV.
There is no evidence, nor did the Region provide any, that the State was
having or anticipated contracting difficulties with ENSCO. On the
contrary, the State accepted ENSCO's formal proposal (not an informal
cost estimate) and was ready to sign a contract.
The State did interpret EPA's June 4, 1985 letter as meaning ENSCO needed
permits. However, as stated in the finding, the State's project manager
confirmed that Region IV also verbally said ENSOO would have to get a
permit and that would take 8 months to 2 years. The State assumed the
TSCA permit would be waived for ENSCO because the permits are waived when
EPA conducts a removal.
RECOMMENDATION
The Regional Administrator should establish and implement adequate
internal controls to ensure that state-funded responses at removal sites
are facilitated appropriately, and that Federal funds are reserved for
sites where response by another party is not feasible.
50
-------
Report of Audit E5EH7-04-0181-81927
FINDING NO. 5 - ENFORCEMENT ACTION INCOMPLETE
Region IV began removal actions at the General Refining site without
first exhausting appropriate enforcement options. Specifically, regional
personnel decided not to issue an administrative order, contracted for
immediate removal without first giving the responsible party an oppor-
tunity to respond to a notice of unilateral clean up of the site, did not
act on a written representation of the owner, and planned and started
long-term removal actions without any responsible party involvement.
Regional personnel did not follow proper enforcement procedures because
they assumed the responsible party was financially incapable of cleaning
up the site. Further, the Region failed to pursue appropriate
responsible party involvement even after evidence was available to
indicate the responsible party may have had financial assets. As a
result, the Agency may have unnecessarily spent Superfund monies to clean
up the site. In addition, by not issuing an administrative order, the
Agency may have lost any opportunity to recover treble damages from the
responsible parties.
Backgr ound
Responsible parties, accountable for environmental damage, should fund
clean-up efforts. When regional personnel determine a responsible party
is unable or unwilling to voluntarily clean up a site, the Region should
issue an administrative order to compel the clean up. If the responsible
parties do not agree to the terms of the order, Superfund monies are used
for site clean up and the responsible parties can later be sued for cost
recovery.
An administrative order sets forth the imminent and substantial dangers
of a site and the risk of harm of a site as determined pursuant to
Section 106 of CERCLA and Section 300.65 of the National Contingency Plan
(NCP). An administrative order also establishes the government's claim
to punitive (treble) damages and the responsible party's liability under
Section 107 of CERCLA.
The EPA's policy (contained in the August 1984 removal program guidance)
is to issue administrative orders before using Superfund monies.
However, an administrative order can be bypassed if the time needed to
issue it would pose an unreasonable risk of harm to public health,
welfare, or the environment.
Enforcement Actions
The on-scene coordinator followed the appropriate procedures to identify
parties responsible for the conditions at the General Refining site. The
site's owner (who had also operated the refinery on the site) verbally
told the coordinator that he lacked the financial resources to clean up
the site.
On August 9, 1985, the on-scene coordinator met with representatives of
the Region's investigation and compliance section and regional counsel.
These individuals agreed to commence with cleanup without issuing an
51
-------
Report of Audit E5EH7-04-0181-81927
administrative order for two reasons. First, the owner did not "appear"
to be financially capable and had verbally refused to do the action based
upon lack of adequate funds. Second, "immediate" site action was
necessary and negotiations with the owner would delay actions at the
site. However, no attempt was made to determine the actual financial
resources of the owner. Further, while the site was in need of immediate
action, not all planned actions were to be conducted immediately.
On August 12, 1985, the Region issued a notice letter to the owner
calling for a voluntary cleanup. The notice cited the verbal discussions
between the owner and the on-scene coordinator, asked if the owner was
willing to conduct or participate in the site cleanup, and gave the owner
2 days in which to respond. The letter stated the EPA was "planning to
conduct the following activities at the General Refining Site":
- Stabilize the sludge ponds and treat and/or solidify sludges to ,
lessen likelihood of releases to ground water. j
- Remove and dispose of sludge piles.
- Dispose of all hazardous substances which are not stabilized on
the site.
- Remove and dispose of waste oil from tanks and containment areas.
- Install a ground water monitoring system.
On August 13, 1985, the Region signed a delivery order with its emergency
response cleanup contractor (HAZTECH, Inc.) for immediate removal actions
needing $200,000 and three-and-one-half months. Specifically:
- Mobilize equipment to site.
- Stabilize waste lagoons.
- Transport and dispose of solids pile in an approved waste disposal
site.
- Pursue alternate technology for treatment/disposal of waste
lagoons.
- Transport waste oil for disposal/recycling at an approved
facility.
On August 15, 1985, the owner responded in writing that the General
Refining Company's "assets are without much value", but said "I will
continue to assist in any way that I can". The owner further stated he
had taken actions at the site which "have been completely successful in
stopping any oil from entering the Savannah River. The sludge ponds have
been monitered (sic) for over flow and none has been noted".
On September 6, 1985, HAZTECH had stabilized the immediate hazards at the
site (see finding number 1). In December, the Resources Conservation
Company (ROC) brought its unproven BEST System to the site (see finding i
52
-------
Report of Audit E5EH7-04-0181-81927
number 1). RCC waited for 5 months while the Region came up with the
funds necessary to begin processing the site's sludge using the BEST
system,
On January 29, 1986, the regional counsel reported to the on-scene
coordinator that:
- The site owner also owned Comet Oil Company, a small chain of gas
stations. (The original meeting between the site owner and the
on-scene coordinator was at the offices of Comet Oil.)
- The owner had asserted that Comet and General Refining were
separate and distinct companies; however, regional counsel had
information that indicated General Refining was a subsidiary of
Comet.
Prospect of Cost Recovery
The EPA did not take adequate action to pursue a responsible party
cleanup. The August 9th memorandum ruled out issuing an administrative
order. The memo was prepared prior to any written communication to or
from the site owner. The contract with HAZTECH was signed prior to the
end of the 2-day deadline given to the site owner to respond to the
Region's notice letter. Further, unilateral short-term cleanup action
was taken at the site, despite written representation from the owner that
he would help in any way. Finally, long-term action was started
(introduction of the BEST system to the site) without any negotiation
with the responsible party, despite indications he had assets that may
have provided him a basis for funding a responsible party cleanup.
Immediate removals often require discrete units of work. Thus owners who
lack sufficient financial resources for an entire cleanup, can accomplish
certain tasks at a moderate cost. At General Refining, the construction
of a fence around the site and the removal of drums are examples of such
aforementioned tasks. However, the Region's actions shut-out any owner
participation in the cleanup. The failure to follow the proper
procedural steps may have hampered the EPA's ability to recover Superfund
monies in a cost recovery action.
Conclusion
Region personnel decided to finance the cleanup using Superfund monies
based solely on verbal'representations made by the site owner regarding
General Refining Company's lack of financial resources. The Region's
failure to issue an administrative order contradicted EPA policy and may
have limited the possibility of collecting treble damages provided for by
law. Further, the actions of the Region, when taken as a whole, have
provided the owner with an argument that may insulate him from bearing
the costs in a future civil action to recover the $2 million spent
attempting to clean the General Refining site.
DRAFT RECOMMENDATION
The Regional Administrator should establish and implement written
53
-------
Report of Audit E5EH7-04-0181-81927
policies and procedures that provide for proper enforcement actions
during removals and ensure the Agency's legal position for cost recovery
is not compromised. Such procedures should include the proper sequencing
of events as well as the events themselves.
REGION'S IV RESPONSE
The Regional Administrator did not agree with the finding and therefore
did not comment on the recommendation. The Regional Administrator stated
that the site conditions at the General Refining precluded the use of an
administrative order. He further stated that the verbal representations
of the responsible party was sufficient to rule out issuing an adminis-
trative order. The Regional Administrator's verbatim comments for this
finding can be located in Appendix VI.
EVALUATION OF REGION'S RESPONSE
The Region's comments did not refute any of the facts presented in the
finding. Further, as discussed in Appendix VI, neither the conditions of
the site nor the responsible party's verbal representations of insolvency
were sufficient to preclude issuing an administrative order in accordance
with Agency policy.
RECDMMENDATIQH
The Regional Administrator should establish and implement adequate
internal controls to ensure compliance with Headquarters enforcement
directives and guidance. This includes taking all appropriate measures
to obtain response from potentially responsible parties and to provide
the strongest possible basis for cost recovery.
54
-------
Report of Audit E5EH7-04-Q181-81927
FINDING NO. B - Questionable Contracting Methods
Contracting methods used to obtain funding for the activities at the
General Refining and Peak Oil sites did not always follow established
policies and procedures. Specifically, the initial delivery orders did
not include the total estimated costs or the scope of work for the
cleanups to be accomplished. Also, documentation establishing site
dollar ceilings set by the Assistant Administrator was not properly
maintained. The cost comparisons made to justify using the two unproven
technologies, were not supported by any documentation. The contractors
were allowed to start work prior to the completion of formal delivery
orders and neither manufacturer was required to submit to the competitive
bidding process. Lastly, the contracting officer was provided
information that was misleading.
Baokground
a. Spending Authority. On-scene coordinators are authorized to spend up
to $250,000 at a removal site without prior approval. Only a contracting
officer can approve spending in excess of the $250,000.
b. Site Ceilings. Spending for removals is limited by CERCLA to $1
million per site with a time limit of 6 months (SABA later changed the
limits to $2 million and 12 months). The Regional Administrator can
approve increases in the site ceiling between $250,000 and $1 million.
He can also approve time limit extensions. The Assistant Administrator
for Solid Haste and Emergency Response is authorized to exceed the law's
limits in special circumstances. To obtain such approval, the on-scene
coordinator submits an action memorandum, through channels, to the
Assistant Administrator. In this memo, the proposed scope of work is
described and the needed costs estimated. The Assistant Administrator's
approval of the action memorandum establishes the estimated costs as the
new site ceiling for costs. Additional approval is required to exceed
either the cost or scope of work set in the memo.
c. Competitive Pricing. The EPA's general procurement policy states
that in the fulfillment of national policy, acquisitions by EPA shall be
conducted utilizing full and open competition with all responsible
sources (48 CFR 1506.301). The Emergency Response Cleanup Services
(ERGS) contract allows the prime contractor to subcontract services and
equipment. Exhibit E of the ERGS contract with HAZTECH states: "For
non-emergency needs, HAZTECH will use a competitive solicitation
procedure to assure cost-effective use of government awards". Article
XII, paragraph C of the contract also states that items of equipment not
specifically included in the contract "are to be treated as a
subcontracted item". The contract directs HAZTECH to prepare a bid
package for subcontractors that is to include an introduction giving "A
brief description of services required and sufficient background
information so that interested suppliers can accurately respond".
d. Audit Requirements. The Federal Acquisition Regulation (FAR 15.805-
5) requires an audit review when any contract or modification involves a
proposal in excess of $500,000.
55
-------
Report of Audit E5EH7-04-G181-81927
Delivery Orders
The initial delivery orders did not include all costs or scope of work.
Further, vague wording in the scope of work was later used to justify not
using required competitive procedures.
At General Refining, the initial delivery order was priced at $200,000
but included the following in its scope of work: "pursue alternate
technology for treatment/disposal of waste lagoons" . This phrase was
later used to justify HAZTECH's subcontract with Resources Conservation
Company, the introduction of the BEST unit to the site, and an additional
$2 million expenditure. However, the ERCS contract required competitive
bidding for services of subcontractors that were not in the contract or
on an approved subcontractor list. Neither the BEST unit nor Resources
Conservation Company were included in the ERCS contract or on an approved
subcontractor list. Despite their absence, competitive bidding was never
considered during the procurement of the BEST unit.
At Peak Oil, the initial delivery order was priced at $250,000, which is
the maximum an on-scene coordinator can obligate the government to spend.
However, the scope of work directed HAZTBCH to "Provide mobile
(transportable) incinerator for on-site destruction of hazardous wastes" .
The first delivery order modification, submitted 3 days later, increased
the delivery order amount to $1.5 million and was forwarded to the
Assistant Administrator for ceiling approval as well as to the
contracting officer for obligation of funds. When the initial $250,000
delivery order was prepared, regional personnel knew the cost of a mobile
incinerator would exceed $1 million. Further, any cost estimates were
subject to competitive bids because the ERCS contract did not include an
incinerator as a listed piece of equipment and the contract directed that
unlisted equipment be treated as ~ subcontracted item. After receiving
the delivery order, HAZTECH advertised the incineration service for bids.
However, as discussed later in this finding, HAZTECH decided to buy its
own incinerator and was able to circumvent the competitive bidding
process .
Also at Peak Oil, the known hazard of lead at the site was never
mentioned in the scope of work or the first action memorandum.
Therefore, the Assistant Administrator approved a new site ceiling
without information discussing a hazard that would not be treated during
the removal and that would later require additional funds and an increase
in the site ceiling.
Ceilin
The Region did not properly maintain documentation that established its
authority to exceed the statutory dollar limitations. The Region's
project files contained various copies of action memoranda requesting
authorization to increase the cost and extend the time limits; however,
there was no evidence the Assistant Administrator for Solid Waste and
Emergency Response had approved the requests as required. The following
schedules set forth the delivery orders, as approved by the contracting
officer, and the corresponding action memoranda.
56
-------
Report of Audit E5EH7-04-0181-81927
a. Peak Oil Delivery Order and Modifications.
Date
Prepared
Date
Approved
ERGS Cost Time
Ceiling Limit
09/27/85
09/30/85
not dated
not dated
11/01/86
10/27/87
09/27/85
09/30/85
05/31/86
04/07/87
08/25/87
10/28/87
$250,000
$1,500,000
$1,500,000
$1,804,700
$3,484,700
$3,723,688
03/27/86
03/27/86
11/30/86
07/28/87
07/28/87
07/28/87
Delivery Order
Modification 1
Modification 3
Modification 4
Modification 5
Modification 6
The foregoing should reflect actions approved by the contracting officer
and reflecting costs approved for the ERGS contractor. However, in the
case of the initial delivery order, the contractor is required to provide
an incinerator, the cost of which is known to exceed $250,000. The
ceiling of $1.8 million was exceeded in June of 1987. By the time
modification 5 was signed, costs incurred exceeded $2.5 million.
b. Peak Oil Action Memoranda.
Date Prepared:
Site Ceiling:
Submitted by:
Submitted to:
Approved by:
Date Approved:
Date Prepared:
Site Ceiling:
Submitted by:
Submitted to:
Approved by:
Date Approved:
Date Prepared:
Site Ceiling:
Submitted by:
Submitted to:
Approved by:
Date Approved:
Date Prepared:
Site Ceiling:
Submitted by:
Submitted to:
Approved by:
Date Approved:
Not dated
$1,925,975
Not signed
Assistant Administrator
Not signed
Not dated
05/19/86
$1,925,975 (asked for time extension)
On-scene Coordinator
Regional Administrator
Deputy Regional Administrator
05/20/86
06/02/87
$4,188,871
Director, Region IV Waste Management Division
Assistant Administrator
Not signed
Not dated
06/05/87
$4,188,871
Director, Emergency Response Division
Assistant Administrator
Acting Assistant Administrator
06/05/87
The foregoing should reflect approvals to exceed the $1 million CERCLA
limit approved by the Assistant Administrator, and later approvals for an
increase in the site ceiling. The costs are total site costs
contractor and EPA. However, the Region's documents were not originals
57
-------
Report of Audit E5EH7-04-0181-81927
or photo copies of originals, thus actual approval signatures and dates
did not always appear on the documents.
c. General Refining Delivery Order and Modifications.
Delivery Order
Modification 1
Modification 2
Modification 3
Modification 6
d. General Refining Action Memoranda.
Date
Prepared
08/13/85
09/30/85
Not dated
Hot dated
01/24/87
Date
Approved
08/13/85
09/30/85
02/18/86
05/27/86
03/10/87
ERGS Cost
Ceiling
$200,000
$600,000
$600,000
$2,300,000
$2,300,000
Time
Itimj-t
11/30/85
11/30/85
02/13/86
10/31/86
02/28/87
Date Prepared:
Site Ceiling:
Submitted by:
Submitted to:
Approved by:
08/13/85
$325,000
On-Scene Coordinator
Regional Administrator
Deputy Regional Administrator (stamped signature)
Date Approved: Not dated
Date Prepared:
Site Ceiling:
Submitted by:
Submitted to:
Approved by:
Date Approved:
Date Prepared:
Site Ceiling:
Submitted by:
Submitted to:
Approved by:
Date Approved:
Date Prepared:
Site Ceiling:
Submitted by:
Submitted to:
Approved by:
Date Approved:
Date Prepared:
Site Ceiling:
Submitted by:
Submitted to:
Approved by:
Date Approved:
09/27/85
$750,000
On-Scene Coordinator
Regional Administrator
Regional Administrator
09/30/85
02/13/86
$750,000 (asked for time extension)
On-Scene Coordinator
Regional Administrator
Assistant Regional Administrator
02/21/86
02/26/86
$2,650,000
Assistant Regional Administrator
Assistant Administrator
Hot signed
Not dated
04/10/86
$2,650,000
Director, Emergency Response Division
Assistant Administrator
Acting Assistant Administrator
04/15/86
58
-------
Report of Audit E5EH7-04-0181-81927
Work Started Without Signed Delivery Orders
There were instances where the on-scene coordinator allowed the
contractor to perform work on site before a delivery order was signed.
a. Peak Oil.
- Prior to a signed contract between HAZTECH (as prime contractor)
and ENSCO (as subcontractor) for use of ENSCO's rotary kiln incinerator
(see finding number 4), and while negotiations were being conducted among
EPA personnel, HAZTECH, and ENSCO; HAZTECH installed a concrete pad and
prepared the site for the rotary kiln incinerator.
- The SHIRCO incinerator started operations on December 31, 1986.
The delivery order modification for the installation and operation of the
incinerator was not signed by the contractor and contracting officer
until August 25, 1987.
The foregoing modification (number 5) set forth the
operating standards for the incinerator. Therefore, the incinerator had
operated for 8 months without any contractual requirements.
The foregoing modification also raised the site ceiling
approved by the contracting officer. However, HAZTECH had already billed
the EPA for much of the costs for tonnage processed.
The required incineration standards, which were signed after
8 months of incineration, did not include the manufacturer's claim his
incinerator would bond lead to the ash.
b. General Refining.
- Resources Conservation Company was allowed to mobilize its BEST
unit to the site in December 1985. A modification to the delivery order
for the installation and operation of the BEST unit was not signed until
August 1986.
- HAZTECH prepared the site for the BEST unit prior to having a
modified delivery order with the EPA, which specified the use of the Best
unit at the site. At the time, HAZTECH, as prime contractor, did not
have a contract with Resources Conservation Company, the subcontractor.
- The following manufacturer claims for the BEST unit were not
included in the contract: (1) Recovered oil would be sold as fuel; (2)
Dried solids would pass an EP Toxicity test (refers to Extractive
Procedure toxicity test, a method of testing material for toxicity); and
(3) The solvent used in the process would be recovered and reused.
- On August 4, 1986, modification 4 set forth the installation and
operating conditions for the BEST unit at General Refining. The system
had been on site since December 1985 awaiting funding and an ERGS
subcontract. The funding was provided in modification 3, dated May 27,
1986. The unit began operations in July of 1986.
59
-------
Report of Audit E5EH7-04-0181-81927
C_QSt Comparisons
a. Peak Oil. The only cost comparison made of the infrared incinerator
was to a conventional rotary kiln model owned by ENSCO, Inc. HAZTECH's
proposed a rate of $225 per ton for the infrared incinerator and ENSCO
proposed $650 per ton for the rotary kiln. EPA personnel compared the
two rates and stated the infrared incinerator was "one- third the cost" of
the rotary kiln. However, the comparison was significantly misleading.
The $650 per- ton rate was a proposal to burn 2,?pO cubic yards of waste
material whereas the $225 per- ton rate was a proposal to burn 5,000 cubic
yards. Further, HAZTECH's bid did not include material handling cost;
ENSCO 's bid is unclear as to what it included or excluded. Finally, the
two companies were not asked to submit bids in competition. When
attempting to justify the costs in HAZTECH's proposal, EPA personnel used
an old proposal made by ENSCO which the Agency had rejected.
b. General Refining. The cost comparison of the BEST unit consisted of
a comparison to costs for off-site incineration, on-site incineration,
and off-site landfilling. According to the on-scene coordinator, he
obtained the comparison cost figures by talking to other on-scene
coordinators who had experience with the various methods.
ft of Incinerator Without Bids
As described in the following paragraphs, the Region arranged to fund the
purchase of the SHIRCO infrared incinerator by the Region's contractor,
HAZTECH. Despite Federal procurement regulations requiring Competition
under the these circumstances and a competition requirement ..i the
emergency response contract with HAZTECH, no bids were solicited from
incinerator manufacturers or operators to compete with SHIRCO.
a. Heed for Incinerator. The State of Florida initially decided the
Peak Oil site was suited for on-site incineration. Prior to EPA's
involvement at the site, the State was negotiating to use the ENSCO
incinerator (see finding number 4). When Region IV took the site from
the State, the Region continued negotiations with ENSCO. ENSCO agreed to
provide the necessary labor, equipment, materials, and supplies to
conduct incineration at the site. ENSCO 's technical proposal stipulated
that prior to mobilization, additional site work would be required for
the placement of the equipment. The additional work included a concrete
pad for the incinerator, a sump tank, secondary combustor, ash drag and
storage bunker, feed hopper, and fuel dike for spill containment.
HAZTECH accomplished this work prior to any contract being signed with
ENSCO. On February 6, 1986, the Region issued a press release announcing
its contract with ENSCO to incinerate the wastes at Peak Oil.
b. 5HIRCQ replaces ENSCQ. In early 1986, HAZTECH submitted a proposal
to purchase an infrared incinerator from SHIRCO for use at Peak Oil. The
proposal received serious consideration by the contracting officer in May
1986. He sent copies of the proposal to those in a position to advise
him on its merits. According to a site report dated September 19, 1986,
the ENSCO incinerator contract was canceled due to insufficient funds.
HAZTECH'S proposal was accepted in the fall of 1986 and HAZTECH mobilized
its SHIRCO incinerator to the site.
1
60
-------
Report of Audit E5EH7-04-0181-81927
c. HA2TECH'S Proposal. HAZTECH proposed to use its incinerator on EPA
contracts at a daily cost to the EPA of $22,461. The daily rate was
based on operating the incinerator 210 days for the first year. A per-
ton rate of $225 was computed assuming the incinerator would operate for
20 hours each day and process 100 tons. The foregoing rates were
eventually accepted by the EPA. The rates were based on the following
costs computed by HAZTECH.
Equipment Depreciation $1,666,667
Property Insurance 100,000
Cost of Money 259,375
Maintenance 150,000
Personnel 568.339
Subtotal $2,744,381
General and Administrative 274,438
Profit 754.705
Subtotal 3,773,524
Liability Insurance 943.381
Total $4,716,905
The depreciation was figured using the double declining balance method, £
life of 3 years, and a cost of $2,500,000. SHIRCO advertises its
incinerators as having a useful life of 12 to 15 years, depending on how
frequently the unit is moved.
d. HAZTECH 'S Proposal AgpnupHnnc? The following were the major
assumptions used by HAZTECH to prepare its proposal. The proposed rate
did not include material handling equipment, cooling water, personnel
needed to prepare waste for incineration, expendable materials,
chemicals, mobilization and demobilization cost, power, and permits.
1) Twenty-four hour operations, with 4 hours per day allowed for
maintenance and adjustments.
2) Property insurance at 4 percent of the cost of the incinerator.
3) Liability insurance not available from a commercial carrier.
HAZTECH to self insure. Estimated cost for the risk 25 percent
of gross revenues.
4) Cost of money for the investment at 10.375 percent.
5) Maintenance contract with SHIRCO to cost $150,000 per year.
6) Twelve people needed for operation: consisting of four shifts
each having an operator, mechanic, and technician. Respective
annual salaries to be $26,000, $20,000, and $15,000. Employee
benefits and overhead to be computed at 35 percent and 95
percent respectively of direct labor costs.
61
-------
Report of Audit E5EH7-04-0181-81927
7) General and administrative expenses added to total cost at 10
percent.
8) Profit rate of 25 percent. HAZTECH argued the rated reflected
"the risk of ownership of a piece of equipment which has not
been tested nor has been proven to operate under the specified
conditions".
e. Cost Assistance. The contracting officer submitted HAZTECH'S
proposal to the EPA's Cost Advisory Group for evaluation. They responded
with the following concerns on April 24, 1986.
1) "There is insufficient information for us to express an opinion
on the reasonableness of the proposed cost".
2) HAZTECH should provide a copy of the invoice for the incinerator
or a written quote from SHIRCO.
3) Need computation support for the depreciation.
4) HAZTECH should provide support for the proposed 4 percent
property insurance, e.g., a quote from an insurance company.
5) In HAZTECH'S ERCS contract with the Agency, HAZTECH waived the
right to claim capital cost of money.
6) HAZTECH should provide an invoice or quote to support the
$150,000 proposed maintenance contract.
7) Detailed information on the proposed personnel is required. For
example, what is HAZTECH'S man year? At 8 hours per day, the
proposed individuals would only be working 1,680 hours per year.
"The IG [Inspector General] says HAZTECH does not keep very good
records. It is our opinion that to accept their rates without
an audit would be taking a tremendous risk".
8) What justification exists for charging the 25 percent ($943,381)
for self-insured liability insurance? "Will a reserve fund be
set up to cover this amount and what guarantee do we have that
HAZTECH would not have access to the fund and use it for other
uses?"
9) Can the EPA purchase the incinerator instead of paying HAZTECH
to purchase it?
f. Technical Assistance. To better understand the incinerator, the
contracting officer wrote the on-scene coordinator in May 1986 and asked
several questions. The on-scene coordinator and the EPA emergency
response team member responded to the questions in October. The
following are those questions and answers, with audit comments in
italics.
62
-------
Report of Audit E5EH7-04-0181-81927
1) Will the incinerator work?
Yes, based on the available data the incinerator should destroy
the hazardous wastes that are at Peak Oil. The rate of operation
is between 100-250 tons per day. The HAZTECH incinerator will
not destroy the metals that are in the waste, nor handle liquids
or thin sludges. Liquid and sludges can be incinerated in the
HAZTECH unit after they have been solidified. (No mention is
made of the fact a full-scale test has never been conducted)
2) What are some of the problems that might be encountered?
Materials handling. The waste at Peak Oil does not appear to
offer any substantial problems with regards to material handling.
The waste at Peak Oil is the ideal consistency for the
incineration process proposed by HAZTECH. (The fact they ffill be
using a prototype, which may have design or operating problems,
is not addressed)
3) What is the preparation for the material to be processed?
The contaminated material will be mixed with lime and/or fly ash
to produce a consistent mixture. The incinerator process will
remove the organics (PCB's etc) from the waste material. The
resulting ash will have to be addressed with regard to metals
content. Handling the metals in the ash will not be part of the
incineration process proposed by HAZTECH.
4) What is the assessment of the unit for a successful burn?
Based on the pilot plant data results, the infrared technology
destroyed the pollutants found at Peak Oil as required. Since
the conditions and parameters of the pilot plant unit are similar
to those of the full sized unit, the results should be the same.
(A pilot plant trial burn was performed with an apparent total
destruction of PCBs; however, the TSCA required destruction was
not calculated - see page 8 appendix II)
5) Need a technical evaluation of the costs?
As discussed at the meeting on October 4, 1986, the cost outlined
in HAZTECH'S proposal appears to be reasonable. The number and
categories of the personnel listed are necessary for the
successful operation of the incinerator. We feel that the
utilization rate of 65 percent and a maintenance time of 20
percent are reasonable, but have no comments on the other rates
such as general and administrative and profit. (The foregoing is
the extent of the Region 's technical cost evaluation, despite the
Cost Advisory Group's advice to get detailed cost information.)
6) Utilization rate?
Based on the unit moving three times in a year, it is not
unreasonable to assume that the unit will be used at least 210
63
-------
Report of Audit E5EH7-04-0 18 1-81927 |
days in that year. The most economical utilization of the ^^
incinerator will be to remain at a single location as long as
possible . *
7) Depreciation rate schedule?
An acceptable write off rate for this type of mobile equipment is
3 years. However, the method of depreciation should be
determined by IRS schedules. The life, salvage value, and
component value should be addressed by HAZTECH and SHIRCO. (In
the first year, the J5R4 ffould pay for too- thirds of the
incinerator 's cost)
The contracting officer accepted the foregoing answers to his questions
as meeting the technical analysis required by the Federal Acquisition
Regulation 15.805-4. However, the on-scene coordinator's answers were ?
generalities base on the contractor's representations and did not (
adequately deal with the specific concerns raised by the Cost Advisory
Group .
g. Contract Negotiations between EPA and HA2TECH. According to the *
contracting officer's documentation, HAZTECH proposed to buy the SHIRCO
incinerator for use at Peak Oil provided EPA would waive the permit I
requirements and allow HAZTECH to recover the majority of the incinerator j
cost on this one job. Negotiations eventually resulted in an agreed per-
ton rate of $225. Material handling and other direct costs were provided
for under the existing ERGS Zone II contract.
At the time of the negotiation, the contracting officer intended to
negotiate a provisional rate (with, if possible, a ceiling per-ton rate)
based upon incurred cost with an estimate for categories for which
verification was not possible, but for which a technical evaluation had
been performed. The delivery order would include a provisional rate with
provisions for audit and renegotiation. However, complete documentation
for cost incurred was not furnished by HAZTECH on a timely basis; thus
sufficient information was not available for a pre-negotiation memorandum
to be completed. Further, HAZTECH did not accept a provisional rate with
a ceiling amount.
HAZTECH argued they were accepting all the risk with no opportunity to
maximize their profit. HAZTECH stated they would accept a provisional
rate subject to audit and re-negotiation, only if the EPA agreed to
absorb all costs associated with the SHIRCO unit if the estimates were
exceeded. Since EPA's previous experience with innovative technology has
shown that the down side risk (high cost with little or no beneficial
results) can be considerable, the contracting officer decided a fixed
rate would be preferable. HAZTECH then refused to accept a contract with
a fixed rate if it had provisions for an audit and renegotiation.
h. Purnha^fl Approved- The arrangement was approved by the contracting
officer. He prepared the following memorandum to the file. It explains
the contracting officer's logic for not requesting the required audit.
\
64 ?
-------
Report of Audit E5EH7-04-0181-81927
1) "Much of the front-end equipment had not been received and final
payment had not been made. Purchase orders which had been issued
were received and amounts verified. While some amounts may vary,
the amount would be insignificant since the difference would
exist primarily due to difference in the cost of freight,
increases in cost of equipment, etc. The purchase agreement
between SHIRCO and HAZTECH was reviewed. The purchase price
stated in the agreement was $2,600,000.
2) "The depreciation method was agreed to by PCMD [EPA's procurement
division] prior to HAZTECH'S purchasing the unit in order to
induce HAZTECH to purchase the unit. This would allow EPA the
opportunity to see if a full-sized production unit would
effectively destroy hazardous waste materials at the manufacturer
stated processing rates. Pilot tests of the infrared technology
had successfully destroyed various hazardous waste materials and
the price range for the per-ton rate, if accurate, could
significantly reduce hazardous waste removal costs. Since
HAZTECH was willing to accept a fixed rate per processed ton,
this significantly reduced EPA's risk should the unit fail to
work as anticipated. Additionally, there would be a unit
available for EPA use (unexclusive) for site projects.
3) "The Overhead and G&A rates submitted by HAZTECH were the audited
recommended rates for 1985. HAZTECH claims that all direct costs
associated with the operation of the incinerator will be
accumulated in a separate cost center (at least for 1987).
Therefore, the allowable indirect costs associated with the
incinerator would be at a different rate than the indirect rate
allowable to general direct labor. However, HAZTECH had neither
performed an analysis nor provided any estimates. Since regional
personnel were anxious to get the unit on site to see if it would
work as advertised, it was decided to allow HAZTECH the
unapproved audit recommended indirect rates (72.42% & 15.36%) for
1986. While the general overhead and G&A rates will undoubtedly
be higher than the specific indirect rates for the incinerator
unit, the following factors tend to minimize any duplication or
over estimate:
"A. The cost of money for equipment under construction was not
considered. This is an allowable cost under the FAR and is not
specifically prohibited by EPA. Therefore, a good argument could
be made for including it as an allowable cost.
"B. PCMD's estimate of personnel costs are believed to be on
the low side. This is the first time HAZTECH personnel will
assemble and operate the unit and there are bound to be
inefficiencies which will cause HAZTECH to incur more cost than
PCMD has allowed."
The memorandum clearly shows the EPA was willing to fund HAZTECH's
purchase of the SHIRCO incinerator to see if the infrared system would
work. Adequate purchasing safeguards, including technical evaluations
65
-------
Report of Audit E5EH7-04-0181-81927
and full-scale testing for innovative technologies, competitive bidding,
pre-award audits, and documentation for cost incurred, were not followed.
Conclusion
Several established procurement policies and procedures to ensure
appropriate services and materials are purchased at a reasonable price
were not followed. On-scene coordinators permitted the contractors to
mobilize and initiate action at the sites prior to having a modified
delivery order. Contracting officers accepted questionable cost
comparisons and technical evaluations and allowed circumvention of
competitive pricing and auditing requirements.
DRAFT RECOMMENDATION
The Regional Administrator should:
a. Emphasize to on-scene coordinators the existing rules regarding the
expenditure of funds prior to obtaining appropriate approvals from a
contracting officer.
b. Emphasize to on-scene coordinators the existing rules regarding
conducting work at a site prior to written approvals for exceeding
site ceilings, time limits, and established scope of work.
c. Establish and implement procedures for obtaining and retaining
written approvals.
d. Establish written procedures prohibiting the use of open-ended
wording to describe the scope of work in delivery orders.
e. Establish written procedures to preclude the participation of on-
scene coordinators in identifying and justifying the use of unproven
technologies at removal sites.
f. Establish a written policy to include manufacturer or contractor
verbal claims as part of the standards for the scope of work.
REGION IV RESPONSE
The Regional Administrator agreed the contracting methods did not follow
established policies and procedures. However, the Regional Administrator
states the established policies and procedures did not apply; therefore
the Regional Administrator did not comment on the recommendations. The
Regional Administrator stated Region IV's policy was to let Superfund
take the lead to encourage field testing of alternative technology. The
Regional Administrator's verbatim comments, along with specific comments
for both sites, are located in Appendix VI. The following is taken from
the Regional Administrator's comments regarding the procurement methods
used for the immediate removals at Peak Oil and General Refining.
"An additional challenge was to find an appropriate contractual
vehicle that would provide a reasonable procurement time frame
with compensation based on performance. In this way, the
66
-------
Report of Audit E5EH7-04-0181-81927
Agency would only pay for accomplishments (processed waste)
while the technology supplier benefited from actual filed
applications on real waste (this later proved to be somewhat
flawed due to the need to use the ERGS contractor to perform
materials handling in preparing the site's waste for feeding
the treatment units). Many discussions were held with
Procurement and Contracts Management Division (PCMD) to
identify the possible contractual vehicles which could be
used. Full and open competition was felt to be inappropriate
to meet the clean-up needs at both sites due to extensive time
requirements (estimated nine months to procure), lack of any
known or very limited competition in these areas of technology
and difficulty procuring new technology where the Scope of
Work was difficult to define.
"Sole source negotiation was ultimately selected for General
Refining because no other competitive technology was known to
exist. For Peak Oil, limited competition with negotiation was
implemented by subcontracting through the ERGS contract. EPA
contracts specialists were consulted throughout the process
because of the nature of the procurement. It became obvious
that Superfund required some creative procurement procedures
to acquire alternative technology in a timely manner.
Established procurement procedures did not provide the
flexibility required to deal with such unique technology in a
satisfactory manner. The best available effort was devoted to
negotiating a contract that would favor the government by
requiring performance as a basis for reimbursement.
"It has been suggested that the best procurement vehicle
within the federal government is full and open competition
(which routinely takes in excess of nine months to procure),
but consideration must be given to the fact that the
government can normally provide clear, concise requirements
for the service it is seeking to procure from a multitude of
contractors. For alternative technology, this methodology has
no established history of success, in part, due to the lack of
significant competition, as well as the variation in waste
characteristics and technologies encountered. The OIG report
suggests a "wait and see" approach while the technology is
proven by private industry. The Region felt there were too
many barriers at that time which prevented private industry
field testing of alternative technology and therefore
Superfund would have to take the lead to encourage the use of
these technologies (subjecting the Agency to some degree of
risk) in an effort to demonstrate their effectiveness to
remediate hazardous waste site threats. Without complete
removal of the source of the threat, it was felt the sites
would be addressed with less consistent partial remedies that
would result in more complex, extensive, and costly repetition
in the future, the cost of which would be borne by the
taxpayer."
67
-------
Report of Audit E5EH7-04-0181-81927
EVALUATION OF REGION'S RESPONSE
The Region's expenditure of Superfund monies to field test the infrared
incinerator and the BEST unit was an unauthorized use of government
funds.
The Region states the normal procurement time frames were inappropriate
because they would require 9 months. In fact, the Region actually did
much worse as one site took 1 year to contract and the other took 2
years. Specifically, the Region had Resources Conservation Company
sample the General Refining site on August 14, 1985; the Region signed a
contract to use the BEST system on August 8, 1986. The Region decided to
use an incinerator at Peak Oil and signed a delivery order with HAZTECH
on September 27, 1985; the Region signed a contract to use the SHIROO
incinerator August 25, 1987. Therefore, the Region's argument that they
needed a contractual vehicle with a reasonable time frame does not hold
up under scrutiny.
The Region states that limited competition was used at Peak Oil, but none
actually took place. The "limited competition" the Region refers to was
the advertisement for bids to incinerate the hazardous wastes at Peak
Oil. HAZTECH advertised the bids and two companies responded. While the.
EPA was considering the bids from the two companies, HAZTECH submitted a
proposal to buy its own incinerator for the EPA's annual use. The low
bidder for Peak Oil was ENSCO, who bid $2.1 million to incinerate 2,000
tons and $650 per ton for any incineration over the 2,000. HAZTECH
proposed to purchase an incinerator for Region IV's annual use at $225
per ton, but at as many sites as the Region wanted. Further, when
HAZTECH made its proposal, the estimated waste at Peak Oil had risen from
2,000 to 5,000 tons and HAZTECH had full access to the bids submitted by
the two companies. Had ENSCO been allowed access to the same insider
information made available to HAZTECH, ENSCO may have demanded the
opportunity to change their bid.
The Region states sole source negotiation was "ultimately" selected for
General Refining since no other competitive technology was known to
exist. The Region and EPA files on the BEST selection never mention sole
source procedures for procurement. Regardless, even if sole source had
been used, no other technology was known because the Region never asked
the industry what was available. In fact, had the BEST technology
manufacturer not told the Region they were working on an innovative
technology, the Region would have never known it existed. In July 1985
the State of Georgia suggested the Region perform a removal action at the
site. The Region was on the site by August 7th and Resources
Conservation Company took its samples of the site on August 14th. No
other manufacturer was invited to take samples at the site. Therefore,
the lengthy selection process suggested by the Region actually involved
only one manufacturer. The Region's sole source competition results
solely from the written promises of one manufacturer who held a patent on
a process that had not yet been tested.
The Region also claims to have written contracts that were based on
performance, thus insulating the Agency from risk. As pointed out in the
finding, neither contract included the manufacturer's claimed merits.
68
-------
Report of Audit E5EH7-04-0181-81927
Regardless, neither technology worked as contracted, but the on-scene
coordinators allowed the manufacturers to continually test and make
design changes at the EPA's expense in hopes of getting the technology to
work. At General Refining, the on-scene coordinator finally gave up and
asked the manufacturer to leave the site after the technology had
processed only 3,000 of the 10,000 tons of waste. At Peak Oil, the
SHIRCO incinerator was allowed to incinerate all the wastes on site
despite the fact it did not pass any of the test burns. HAZTECH sued
SHIRCO for failure to perform. In both instances, because the on-scene
coordinators did not insist the technologies perform as claimed and
allowed the continued testing, the performance factor in the contract
became moot. Therefore, the Agency, since it supplied all materials
handling and other site costs, actually bore a significant risk.
The Region also says full and open competition is too slow but does not
offer any explanation as to how it became aware of the SHIRCO and BEST
technologies. The Region, the Office of Program Management and
Technology, and the Office of Environmental Engineering and Technology
Demonstration all say in their respective responses to this audit that
the two technologies were identified as "fast track" projects with "high
potential". However, both technologies had not yet been built.
Therefore, the only source for identifying the technologies were the
respective manufacturers, a process potentially open to fraud and abuse.
REGQMMENDATIQB
The Regional Administrator should:
a. Establish and implement adequate internal controls to ensure that on-
scene coordinators do not go beyond the limits of their authority by
permitting contractors to begin work and incur expenses prior to
obtaining the contracting officer's approval.
b. Establish and implement standard documentation and filing procedures
to ensure project files contain complete documentation of the history
of site planning, authorizations, and response actions taken.
c. Establish and implement adequate controls to ensure that contract
delivery order scopes of work prepared by regional personnel are
clearly defined, avoid open-ended wording and require contractors to
meet performance objectives for proprietary technologies.
69
-------
Report of Audit E5EH7-04-0181-81927
I
This page intentionally left blank. I
70
-------
Report of Audit E5EH7-04-0181-81927
NO. 7 - SITES REMAIN
Lead and PCB still contaminate the General Refining and Peak Oil sites.
EPA personnel misjudged the capabilities of the treatment technologies
selected for use at both sites. At Peak Oil, the technology selected
offered no solution to the known lead hazard. Neither technology
operated as advertised. However, the manufacturer was not at significant
risk as claimed merits were not included in the cleanup contracts.
Nonetheless, extensive work was undertaken at both sites while the
manufacturers attempted to resolve their equipment design problems.
Unfortunately, post- treatment testing has disclosed the sites are not
clear of the hazards that originally attracted the EPA's attention. As a
result, further cleanup efforts are needed to dispose of contaminated
materials remaining on site.
Background
a. BEST Unit. According to its manufacturer, the BEST system is a phase
separation process which uses a solvent to separate oily sludges into
oil, water, and solids. Through the separation process, contaminants can
be concentrated into components, such as PCBs into the oil, and heavy
metals into the solids. Disposal of the components can be cheaper than
disposal of the original oily sludge.
b. Infrared Incinerator. According to its manufacturer, the SHIRCO
incinerator is a thermal destruction unit which uses electric infrared
heat to burn contaminants in solid wastes and sludges. The incinerator
can destroy PCBs, dioxins, and pesticides; while certain heavy metals,
including lead, can be thermally fixed in the incinerator's ash. Thus
the ash is tendered nonhazardous and suitable for on-site disposal.
Region IV personnel misjudged the capabilities of the technologies
selected for the General Refining and Peak Oil sites. At General
Refining, three of four manufacturer-advertised merits of the BEST system
proved inaccurate during actual use of the prototype. At Peak Oil, the
on-scene coordinator did not adequately address the problems associated
with the site's lead contamination. Destruction (or removal) of the lead
was not even included in the contract between the EPA and HAZTECH.
a. The BEST System.
In his request to exceed the $1 million cost ceiling (see finding no.
1), the on-scene coordinator described the BEST unit's merits as:
1) "Water, which comprises about 70% of the sludge, can be
separated and treated by conventional methods.
2) "Recovered oil can be sold as fuel due to its high BTU value,
thereby reducing treatment costs.
3) "Dried solids may pass the EP Toxicity test [refers to
71
-------
Report of Audit E5EH7-04-0181-81927
Extractive Procedure toxicity test, a method of testing
material for toxicity] which reduces disposal costs.
4) "The solvent used in the process can be recovered and reused."
The cleanup contract did not include the last three foregoing merits.
The oil was not required to be of saleable quality and the dried solids
were not required to pass an EP Toxicity test. Therefore, the
manufacturer's financial risk was not as severe as the EPA's because the
EPA, besides paying to test the system, also wou'ld have to pay if it
failed to operate as advertised. In actuality, the last three listed
merits did not materialize. The EPA had to pay to remove the oil and now
has to deal with the lead remaining on site that did not pass the EP
Toxicity test. Specifically:
1) Saleable Oil. Recovered oil from the system was not sold. Two
off-site storage tanks (capacity of 319,200 gallons) leased to store the
treated oil, were never used because the BEST system did not produce
enough treated oil. Oil from the BEST system and the lagoons was stored
in five on-site storage tanks (65,600 gallons). The oil was not sold
because of its heavy metals content. Instead, the EPA had to pay for a
company to remove oil from the site. Further, 3,000 gallons of oil was
even returned because the purchaser could not get it out of the tank
cars.
2) Dried Solids. The waste pile containing the dried solids did
not pass an EP Toxicity test. The on-scene coordinator had planned to
dispose of the solids on-site. Disposal could still be on-site, but tn?
site would have to be turned into a Resource Conservation and Recovery
Act (RCRA) landfill. Off-site disposal at a RCRA landfill is also an
option.
3) Solvent Recovery. Most of the solvent used (Triethylamine) in
the process was not recovered. During processing, solvent vapors leaked,
resulting in an estimated ten times the planned amount actually being
needed. When the BEST system was closed down, 67 drums, each containing
42 gallons of solvent, were recovered at the site; 713 drums were used
during processing.
b. The SHIKCQ Incinerator.
In February 1984, a Peak Oil/Bay Drum1 site investigation identified lead
as one of the five primary site contaminants. In his request to begin
on-site incineration in September 1985, the on-scene coordinator wrote
that incineration would "... offer a permanent solution to the issue of
PCB disposal inasmuch as the [site] material will be rendered
nonhazardous ...". The implications of the lead on site were not
addressed in the request.
i The EPA grouped the Peak Oil site with an adjoining drum recycling
facility (Bay Drum) and placed them as a single site on the National
Priorities List.
72
-------
Report of Audit E5EH7-04-0181-81927
SHIROO claimed their incinerator would not only burn PCBs, but also
thermally fix lead in the incinerator ash, thus rendering lead in the ash
nonhazardous . Despite the claim, the on-scene coordinator reviewed the
incineration proposal in October 1986 and informed the contracting
officer that the incinerator would not destroy the metals in the waste
and the lead was not mentioned in the ensuing contracts. The on-scene
coordinator further stated the ash would have to be addressed with regard
to its metals content. In the June 1987 request to increase the project
cost ceiling, the on-scene coordinator noted "additional monies will also
be necessary for immobilizing the ash on site to insure that the lead in
the ash will not recontaminate the groundwater at the site" .
Oeratin
The operations at both sites were continually plagued with problems
associated with using unproven treatment technologies (see finding no.
2). Neither technology performed as projected by its manufacturer. At
the General Refining site, the on-scene coordinator eventually cutoff
funding for the unsuccessful BEST system, but at the Peak Oil site, the
on-scene coordinator continued funding the infrared incinerator despite
unsuccessful operations.
a. BEST System.
The following problems persistently hampered the cleanup:
1) Leaks in the system's seals allowed solvent vapors to escape
into the air.
2) Mechanical problems with the dryer solids conveying system.
3) Detergents in the sludge hindered the oil/water separation.
In October 1986, the on-scene coordinator temporarily closed the site and
gave the manufacturer 90 days to "make every attempt to resolve problems
associated with the BEST unit". After the 90 days, the on-scene
coordinator allowed the manufacturer to continue processing lagoon
sludge, but at no cost to the EPA. The system was eventually shut down
on February 28, 1987.
The BEST system did concentrate, to some degree, the contaminants in the
sludge components. According to the EPA's Office of Research and
Development, PCBs were concentrated in the oil product and lead was
concentrated in the oil and solids products. The system was to have
processed 10,162 tons over a period of 152 days (67 tons per day). Site
reports show 3,448 tons were processed from July 30, 1986 to February 25,
1987 (211 days). By January 23, 1987, the date the EPA stopped paying
for processing, 1,874 tons had been processed. The other 1,574 tons were
processed after that date.
The following graph shows the planned (which was to begin in March and
end in August) and actual volume processing of the BEST system that began
in July and ended in February the following year.
73
-------
Report of Audit E5EH7-04-Q181-81927
ffoate Proceaoed By Beat System
2000
1600
1200
000
400
Tona
liar. Apr. May June July Aug. Sept. Oct. NOT. Dec. Jan. Feb.
Planned ESJ3 Actual
The on-scene coordinator, an EPA incineration consultant, EPA SITE
program personnel, and HAZTECH personnel all documented problems with the
infrared incinerator. For example,
1) Waste feed system.
- Crusher and weight hopper clogged.
- Pins sheared.
- Material spilling and sticking.
- No feed rate adjustment.
- Motors overloading and burning out.
2) Prijnary chamber.
- Belt stopping and breaking.
- Glow bars breaking.
- Cake breakers clogging.
3) Secondary chamber.
- Burner block failure.
74
-------
Report of Audit E5EH7-04-0181-81927
4) Quenching, scrubber, and ash handling.
- High temperature cracks.
- Inadequate quenching.
- Poor water spray distribution and coverage.
- Excessive volume of wastewater.
- Acidic wastewater containing excessive lead and zinc.
- Particulate and lead carry over to the stack.
- Vibration and noise in draft fan.
- Platting of salt.
- Ash removal auger clogging.
- Significant odor and dusting in the ash outlet.
- Motors overloading and burning out.
In addition to the foregoing operating problems, the incinerator did not
meet EPA incineration standards (see finding no. 3). The operating
problems required maintenance, repairs, and redesign resulting in a
deflated processing rate and inflated downtime. In October 1986, the on-
scene coordinator had informed the contracting officer the infrared
incinerator's 90 tons per day was four-and-one-half times the volume of a
conventional rotary kiln rate and only one-third the cost. The
incinerator actually managed only 24 tons per day.2
The following graph shows the planned (which was to start in December
1986 and end in March 1987) and actual volume processing of the infrared
incinerator (which began on December 31, 1986 and ended in October 1987).
2 A total of 7,110 tons was burned over the 287 days from December 31,
1986 to October 13, 1987. However, 355.6 tons had to be reburned because
the PCBs were not destroyed; therefore, only 6,754.4 tons were actually
treated, resulting in the actual 24 ton-per-day rate (6,754.4 tons
divided by 287 days). The incinerator did not operate on 133 of the 287
days; however, it did meet the 90 ton-per-day rate on 24 of the 154
operating days.
75
-------
Report of Audit E5EH7-04-0181-81927
Wost« Processed By Shiroo Incinerator
1
2800
2400
2000
1600
1200
800
400
TOM
Dec. Jan. Feb. Mar. Apr. May June July Aug. Sept. Oct
^1 Planned ESS Actual
Site Status
During August of 1987, personnel from the EPA's National Enforcement
Investigation Center (NEIC) conducted a site investigation at our
request. At General Refining, NEIC sampled the soil, water, oil, and
waste. At the time of NEIC's sampling, processing with the BEST systen
and lagoon solidification was completed. At Peak Oil, NEIC sampled the
incinerator feed, incinerator ash, incinerator wastewater, and
groundwater. Testing at the Peak Oil site took place during on-site
operations (see finding number 2). The following is a description of the
two sites as described in the NEIC reports (see appendices II and IV).
a. General Refining.
Four waste piles at the site contained high concentrations of lead
(between 640 and 41,900 parts per million (ppm)) and some PCBs at
acceptable levels (0.6 -5.8 ppm). One of the waste piles, which
contained the solids produced by the BEST system, exhibited the hazardous
waste characteristic of EP Toxicity for lead (8.1 ppm 5 ppm would be
an acceptable level). Another waste pile was pushed up against the site
fence and was eroding into the off-site drainage ditch.
The drainage ditch outside the site fence contained lead in
concentrations similar to the nearest on-site waste piles (650 ppm).
Water in the drainage ditch contained lead (between 0.53 and 10.3 ppm).
The soil between the waste piles and the drainage ditch contained lead in
concentrations similar to the waste piles (2,910 - 4,910 ppm).
The on-site storage tanks contained oil produced by the BEST system as
well as oil pumped directly from the site lagoons. The oil contained
76
-------
Report of Audit E5EH7-04-0181-81927
PCBs (between 13.3 and 24.2 ppm) and lead (3,820 - 4,470 ppm). Within
the diked area beneath the tanks, an oil layer floated on top of a liquid
pool. The soil beneath the pool was saturated with oil. Adjacent to the
diked area, and outside the site fence, the soil was stained with oil and
the water in the drainage ditch also contained oil. Three additional
tanks, located in another corner of the site, contained oil and oily
sludge. Further, 20 to 30 whole tanks and tank sections were still on
the site; most were empty or contained water, but some were leaking oil
onto the ground.
The fence was cut, lying down, or missing at several points around the
site. Warning signs were not posted at the site or on the site fence to
deter entrance to the site. (While visiting the site, we observed a
local power company employee enter the site through a gap in the fence
and walk to the top of a waste pile to read a power meter).
b. Peak Oil.
The incinerator ash contained lead (between 5,710 and 10,100 ppm) and
PCBs (less than 2 ppm an acceptable level for PCB in the ash). The
ash also exhibited the hazardous waste characteristic of EP Toxicity for
lead (10.9 - 76.2 ppm 5 ppm would be an acceptable level). Attempts
to contain the ash piles had been made; however, ash was spilled and
spread over parts of the site. The ash storage areas were not covered.
Storage tanks, located at several points throughout the site, were used
to store waste oil containing solvents and other contaminants. Both
rainwater and groundwater, which had collected in the lagoon, were also
stored in some of the tanks.
Groundwater from monitoring wells was visibly contaminated and contained
heavy metals and volatile organic constituents. No PCBs were detected in
the groundwater.
Remaining Cleanup
a. General Refining.
Future actions at the site are needed to address the lead and PCB in the
7,000 yards of material in the site waste piles, the drainage ditch, and
the 3,000 gallons of returned waste oil. The one waste pile which
exhibited the RCRA hazardous waste characteristic of EP Toxicity for lead
is a hazardous waste. Therefore, the pile is subject to disposal in
accordance with RCRA guidance.
b. Peak Oil.
Future actions at the site are needed to address the incinerator ash and
on-site storage tanks. All the ash, approximately 7,000 tons, from the
incineration process is stored on site. The ash exhibited the RCRA
hazardous waste characteristic of EP Toxicity for lead. Thus the ash is
a RCRA hazardous waste and is subject to disposal in accordance with RCRA
guidelines.
77
-------
Report of Audit E5EH7-04-0181-81927 I
Technology Advertised
Despite a lack of successful operations, both manufacturer's benefited I
from the EPA's endorsement of the new products. The activities at both *
General Refining and Peak Oil have been recognized in EPA and industry
publications EPA officials, including the on-scene coordinators, are <
frequently quoted and even co-authored one industry article. For ,
example:
The January/February 1988 issue of
Magazine (a member of the magazine's editorial advisory board is
listed as also being a member of the EPA's Office of Waste Programs
Enforcement):
"A recent independent research report funded by the US EPA
rated the BEST process No.l in desirability for treating
PCB-contaminated sediments. . . [the] 100-ton-per-day BEST
unit has been successfully demonstrated in a Superfund
cleanup of a PCB-contaminated, used oil refining site near
Savannah, Georgia."
The December 1986 Treatment & Disposal (quotes the EPA's internal
action memorandum for the site and is co-authored by the on-scene
coordinator):
"Because the BEST solvent extraction process required
neither major transportation costs (as in the case of off-
site incineration) nor an involved testing and permitting
process (as in the case of on-site incineration), it was
chosen as the most suitable and cost-effective disposal
options for this site. ... the BEST unit has consistently
separated the sludge into fractions that meet or exceed
contractual requ irements . "
The October 17, 1987 National Journal (including a quote from an
aide to a congressional committee):
Region IV 's on-scene coordinator "took a big risk . . . that
few of his fellow bureaucrats . . . are willing to take."
The on-scene coordinator "and his superiors in EPA's Atlanta
Regional office bet on a new and relatively untried method
to destroy both cancer causing PCBs and lead ... 'Everybody
talks a good game about using technology, but nobody is
doing it except in Region IV ... Our hat is off to them ' " .
78
-------
Report of Audit E5eH7-04-0181
General Refining
Oil and lead remain in
off-site drainage ditch.
Oil eroding off site
from the storage tanks.
79
-------
I
I
I
-------
Report of Audit E5eH7-04-0181
General Refining
Cluster of whole tanks and tank
sections left at the site
Storage tanks filled with oil
containing PCBs and lead.
81
-------
-------
Beport of Audit E5eH7-04~0181
General Refining
Hole in the site fence. Waste
pile eroding into drainage ditch.
Waste pile extending under
the site fence and eroding
into the drainage ditch.
83
-------
-------
Report of Audit E5eH7-04-0181
Peak Oil
Hazardous incinerator ash stored at the site.
Storage tanks containing waste oil,
solvents, and other contaminants.
85
-------
-------
Report of Audit E5eH7-04-0181
Peak Oil
Ash pile.
Site visitors.
87
-------
-------
Report of Audit E5EH7-04-0181-81927
Conclusion
As of March 8, 1988, Region IV had not officially completed the cleanups
at the two sites. However, lagoon solidification, conducted after BEST
was removed from the site, was completed in March 1987 at General
Refining, while incineration at Peak Oil was completed in October 1987.
The planned project cost ceilings stood at $2.7 million for General
Refining and $4.2 million for Peak Oil. The Region has spent over $6
million without removing the lead hazard and PCBs from the two sites.
EPA personnel relied on manufacturer product representations that did not
come to pass. Further, the EPA could not easily hold the manufacturers
to their claims because the contracts did not require the technologies to
contractually perform to the standards claimed by the manufacturers.
DRAFT RECOMMENDATION
The Regional Administrator should:
a. Require that meaningful contractual treatment requirements be in
effect and enforced.
b. Stress the importance of achieving cost-effective cleanups using
treatment technologies that will treat site wastes.
c. Take appropriate action to deal with hazardous materials remaining at
the General Refining and Peak Oil sites.
REGION IV RESPONSE
The Regional Administrator agreed waste piles at both sites required
further cleanup. However, the Regional Administrator disagreed that
testing the BEST unit and the SHIRCO incinerator was either inappropriate
or unsuccessful. Specific comments for each site are located in Appendix
VI. In addition to the specific comments, the Regional Administrator
also provided the following general comments (see Appendix VI):
"Although we have learned from our experiences at Peak Oil and
General Refining, we feel strongly that these two sites are
success stories for EPA. Threats to the environment and public
health have been mitigated. This work has dramatically helped
EPA materials handling at Superfund sites and provided in-sight
on permanent remedies for reducing toxicity and mobility of
pollutants... We will also continue to use innovative/alternative
technologies which are appropriate to the wastes present at
Superfund sites."
In the Region's specific comments, the Region states the BEST unit
operated successfully. The Region cites the treatment of 3,800 tons of
sludge which enabled the site lagoons to be closed. The Region also says
the SHIRCO incinerator operated successfully, citing the reduction of
site PCBs to acceptable levels.
89
-------
Report of Audit E5EH7-04-0161-81927
In addition to the comments on the finding, the Region provided the
following specific comments on the recommendation.
a. "This is a contracting problem that is constantly being
addressed. Region IV has looked at some of the contracting
problems at Peak Oil and General Refining and has resolved
these issues in the new fixed bid contracts on our other
incinerator sites.
b. "The Peak Oil site was cost-effective in that the unit price
for the incinerator was 1/3 of the ENSCO bid and 1/5 of the
cost of hauling to a fixed facility incinerator.
c. "The Region has been and is continuing to address final
removal action at these sites."
EVALUATION OF REGION'S RESPONSE
Besides the remaining cleanup, at issue in the finding is the fact the
Region elected, in the face of regulations to the contrary, to use two
unproven technologies at immediate removal sites. Had the two
technologies worked, the manufacturers would have proven technologies t.
market. However, the taxpayer would have paid to prove the two
technologies. This latter fact would open the government to complaint:
of unfair competition practices and possible bid protests from parties
that may have desired to compete for the opportunity to test and develc
their technologies.
In considering whether the technologies operated successfully, the Region
down-plays certain facts. Specifically:
a. The Region does not comment on the fact the on-scene coordinator
ordered the BEST unit removed from the site because of its processing
failures.
b. The Region says the BEST treated 3,800 tons of sludge which enabled
them to close the four lagoons. The BEST did processed 3,448 tons.
However, it was supposed to process 10,000 tons. After the on-scene
coordinator ordered the BEST off the site, the sludge in the lagoons was
solidified.
c. The BEST was to separate oil from the sludge which could be sold.
The Region did not include salable oil as a contract term. The Region
now says the oil could not be sold because of a drastic drop in the world
market. However, the project documents clearly state the lead content of
the oil produced by BEST prevented the planned sale.
d. The solids produced by the BEST unit remain at the site. The solids
were to have passed the EP Toxicity test for lead but did not. As a
result, the solids are a hazardous waste.
e. Both the solids produced by BEST and the 7,000 tons of site waste the
BEST unit did not process must now be treated.
90
-------
Report of Audit E5EH7-04-0181-81927
f. The SHIRCO incinerator did not destroy PCBs at the 99.9999 percent
destruction efficiency required by TSCA.
g. PCBs above the TSCA limit were released out the incinerator's stack
along with three tons of excess particulate matter containing lead.
h. The incinerator ash remains at the site. The incinerator was to have
fixed the lead in the ash but did not. The manufacturer's claim was not
included in the contract. The incinerator did not fix the lead to the
ash and the ash is a hazardous waste which remains to be dealt with.
The recommendations were revised for the final report. In regard to
SHIRCO's cost effectiveness, the incineration price for Peak Oil was $225
per ton. The indirect costs charged under the ERGS contract inflate that
figure. The EPA's SITE program, in a draft report dated December 1987,
evaluated the costs to use the SHIRCO incinerator considering the
technical aspects of the technology. They concluded the operating costs
for the SHIRCO incinerator ranged from $196 per ton for an ideal
operation, to $795 per ton "... for the highly inefficient overall
operation at the Peak Oil site".
RECOMMENDATION
The Regional Administrator should:
a. Establish and implement adequate controls to ensure that on-scene
coordinators effectively monitor contractors to achieve treatment
standards required by contractual terms.
b. Take appropriate measures to ensure that the hazards remaining at
General Refining and Peak Oil after prior removal activities are
adequately addressed.
91
-------
Report of Audit E5EH7-04-0181-81927
This page intentionally left blank.
92
-------
Appendix I
DISTRIBUTION
Office of Inspector General
Inspector General (A-109)
Region IV
Regional Administrator
Assistant Regional Administrator, Policy and Management
Director, Waste Management Division
Audit Followup Coordinator
Public Affairs Office
Associate Administrator for Regional Operations (A-101)
Director, Financial Management Division (PM-226)
Director, Office of Research and Development (RD-674)
Comptroller (PM-225)
Agency Followup Official (PM-225)
Office of Congressional Liaison (A-103)
Office of Public Affairs (A- 107)
Director, Procurement and Contracts Management Division (PM-214)
93
-------
------- |