August 1989 EPA-700/8-88-053
Hazardous Waste Ground-Water
Task Force
Evaluation of
Amoco Oil
Yorktown, Virginia
f/EPA
UNITED STATES ENVIRONMENTAL PROTECTION AGENCY
-------
United States Environmental Protection Agency
Hazardous Waste Ground Water Task Force
Ground Water IVbnitoring Evaluation
Amoco Oil Company
Yorktown, Virginia
August 1989
Region III
-------
August 1989
UPDATE OF HAZARDOUS WASTE GROUND WATER TASK FORCE EVALUATION OF
AMOCO OIL COMPANY YORKTOWN, VIRGINIA
The United States Environmental Protection Agency's Hazardous
Waste Ground Water Task Force (Task Force) conducted an evaluation of
the ground water monitoring program at the"'Amoco Refinery (Amoco)
located in Yorktown, Virginia. The field inspection was conducted
during the period from January 20 to 24, 1987. Amoco is 1 of 58
facilities evaluated by the Task Force. The purpose of the Task Force
evaluation was to determine the adequacy of the facility's ground water
monitoring program in regard to applicable State and Federal requirements.
The Task Force effort came about in light of recent concerns as to
whether operators of hazardous waste treatment, storage and disposal
facilities were complying with State and Federal ground water monitoring
regulations.
The evaluation of Amoco focused on determining: (1) if the facility
was in compliance with applicable interim status ground water monitoring
requirements, and (2) if hazardous waste constituents were present in
the ground water. The inspection revealed that Amoco was not fully
complying with applicable interim status ground water monitoring require-
ments and that ground water samples from a number of on-site wells
contained hazardous waste constituents. This update provides informa-
tion on ground water related activities conducted by Amoco, EPA and the
Virginia Department of Waste Management (VDWM) since the Task Force
Inspection.
As of November 8, 1988, Amoco ceased all management of hazardous
wastes in landfarm #10, the last remaining land disposal unit in opera-
tion. Wastes are presently stored in above ground tank(s) and are
ultimately fed to the coking unit from which petroleum coke is produced.
According to 40 C.F.R. Part 261.6 (a)3(IX), "Petroleum coke produced from
petroleum refinery hazardous wastes containing oil at the same facility
at which such wastes are generated" is considered a recyclable material
and, therefore, not subject to regulation under Parts 261 through Part
266 and Parts 268 and 270 of RCRA.
On this basis, the coke production unit is exempt from regulation
under RCRA. However, the tank(s) in which the hazardous waste are
stored require a RCRA Part B permit. The application has been submitted
to VDWM and is currently under review.
Amoco remains an interim status facility and has modified its Part
A application to reflect the storage of hazardous wastes in tank(s).
The Part B permit application for the land treatment unit was officially
withdrawn in July 1987. Post closure permit applications for landfarms
-------
-2-
#10 and #12 nave been submitted to VDWM for review and approval.
According to the VDWM, the approval of these applications is still
under consideration. In addition, closure plans for the two landfarms
have been submitted to VDWM and have undergone extensive review. The
plans are currently undergoing revision and will be approved by VDWM in
the 3rd Quarter FY 1989.
Normal operations and maintenance of the landfarms is continuing,
except for the addition of new waste. Until final closure is achieved,
the landfarms are disked on a regular basis consistent with standard
operating procedures. This allows for greater incorporation of sludge
into the soil.
The ground water monitoring program is still in an assessment mode
as has been the case since December 1984 when statistically significant
increases were observed in downgradient monitoring wells.
Since the Task Force evaluation, determinations of the rate and
extent of contaminant migration have been made as reflected in annual
reports but not to the satisfaction of VDWM. According to Amoco, the
most recent annual report dated February, 1989, which discusses the
status of the ground water assessment investigation, presents the final
determination of vertical and horizontal extent of contamination.
Amoco has taken the position that this latest determination accu-
rately represents conditions at the facility and, therefore, additional
hydrogeologic studies in this regard are unnecessary at this time.
This report is currently under review by the VDWM. VDWM and EPA do not
agree that ground water conditions have beeen adequately characterized.
Better horizontal and vertical deliniation of ground water contamination
is required before permits can be issued. Ultimately, post closure
permits will be issued by VDWM and a compliance monitoring program will
be instituted superceding the assessment program. Post closure permits
have a corrective action component under which additional studies
directed at corrective measures evaluation and implementation can be
required.
Since the Task Force evaluation, several developments are worth
mention which reflect an effort on the part of Amoco to address defi-
ciencies identified by VDWM and EPA. and documented in this report.
The following developments have occurred:
1. Submission of a revised Sampling and Analysis Plan.
This plan was approved by VDWM in September 1987. In conjunction
with this approval, Amoco instituted new policies with regard to sample
collection and QA/QC protocols which enhance the quality of analytical
data. As of the first quarter 1988 sampling event, ground water samples
are no longer collected using a peristaltic (vacuum) pump but rather
by bailer. Monitoring wells, however, continue to be evacuated using
-------
-3-
peristaltic pumps. It is now standard practice to collect equipment
blanks during each day of sanpling and to carry travel blanks into the
field during sanpling events.
2. Performance of a supplemental subsurface investigation
(February 1988) which resulted in the following:
(a) Presentation of separate ground water contour maps for
the 8 foot, 18 foot, 38 foot, and 58 foot monitoring
wells;
(b) Drilling of three stratigraphic test borings to provide
continuous lithologic characterization to a depth of 58
feet;
(c) Analysis of the impact of seasonal and York River water
level fluctuations on ground water flow;
(d) Geophysical logging of 17, 58 foot CQ series wells;
(e) Split spoon sampling of soils within landfarms #10 and
#12 for chemical analysis and geotechnical evaluation;
and
(f) Performance of slug tests on several CQ series wells to
evaluate horizontal hydraulic conductivity and estimates
of vertical rates of ground water flow.
This study was performed at the request of VDWM in an attempt to
address deficiencies in the geologic and hydrogeologic characterization
of the facility. At this time, a more definitive characterization of
the horizontal and vertical extent of contamination is required.
The VDWM performed a Comprehensive Ground Water Monitoring Evaluation
(CME) at Amoco during January 1989. Major findings and reconmendations
resulting from the evaluation are indicated as follows:
1. Monitoring well CQ 1O-8 was improperly installed within land-
farm #11 and should be relocated downgradient and outside the
unit. All other compliance point monitoring wells located
within a hazardous waste management unit should be relocated
outside of the unit along the downgradient perimeter.
2. The monitoring wells comprising cluster CQ-10 had measured
casing heights of only 2 to 3 inches and should be 1.5 feet
above the ground surface. Sediment surrounding well casings
should be removed to expose the concrete protective pad for
the well cluster.
3. Ground water samples collected from monitoring well CQ-14-58
were turbid. Well design and construction specifications
-------
-4-
should be reviewed to determine whether the well can be rede-
veloped to obtain adequate samples. All monitoring wells
yielding turbid samples during the CME should be identified
by Amoco and reevaluated. If these wells cannot be redeveloped,
then replacement may be required.
During the OVE, VDWM and Amoco split samples collected from the
following monitoring wells surrounding landfarm #10.
CQ-12-18
CQ-14-58
CQ-13-58
CQ-10-8
CQ-18-68 (Trip Blank)
VDWM samples were analyzed for a modified list of priority pollutant
compounds and significant results are presented on the following table:
CO-12-18
2,4,5-Trichlorophenol 63.3 ppb
Bis (2-Ethylnexyl) phthalate 20.0 ppb
Chromium (unfiltered) 116.0 ppb
CO-18-68
Bis (2-Ethylhexyl) phthalate 137.0 ppb
2,4,5-Trichlorophenol 123.9 ppb
CO-14-58
2,4,5-Trichlorophenol 33.09 ppb
Di-N-Butyl Fhthalate 36.0 ppb
1,1,1 Trichloroethane 3.6 ppb
CO-13-58
1,1,1 Trichloroethane 3.2 ppb
CO-10-8
Bis (2-Ethylhexyl phthalate 290.0 ppb
2,4,5-Trichlorophenol 32.03 ppb
2,3,5-Trichlorophenol 19.82 ppb
2,3,6-Trichlorophenol 33.12 ppb
Results of unfiltered and filtered metals analysis were below
established IVbximum Concentration Limits (MIL'S) with the exception of
chromium at 116 ppb for well CQ-12-18.
-------
-1-
TABLE OF CONTENTS
I. EXECUTIVE SIMftRY
A.
1. Task Force Objectives 1
2. Facility Background/Location 2
B. SUMMARY OF FINDINGS AND (XNCLUSIONS 4
1. Ground Water MDnitoring Program During Interim
Status 4
2. Proposed Ground Water Monitoring Program for
RCRA Permit 13
3. Task Force Sampling Data 14
4. Conclusions 15
II. TECHNICAL REPORT 18
A. BACKGROUND 18
1. Site History 18
2. Enforcement Actions 18
3. Adjacent Land Use 19
B. INVESTIGATIVE METHODS 19
l. Record/Document Review 20
2. Facility Inspection 20
3. Laboratory Audits and Inspections 21
4. Sampling Audits 21
5. Sampling Program 22
C. WASTE MANAGEMENT UNITS AND OPERATIONS 23
1. Waste Management Units 23
2. Facility Operations 28
D. SITE rHTO^/HYDROGECfrOGY 32
1. Topography 32
2. Geology 32
3. Hydrogeology 36
(continued)
-------
-11-
II. (Continued)
E. GROUND WATER MCNITCRING SYSTEM 43
1. Monitoring Requirements/Interim Status 43
2. Current Monitoring Well Network 44
3. Well Construction Specifications 44
4. Site Characterization 49
5. Sampling Audit 50
6. Sampling and Analysis Plan and Field Procedures 55
7. Facility Water Quality Analysis and Data
Quality Assessment 56
8. Interim Status Ground Water Monitoring Data 57
F. TASK FCKCE DATA CX3LLEETICN/RESULTS 58
l. Sample Collection Methods 58
2. Task Force Sampling Results 65
3. Conclusions 70
REFERENCES
APPENDICES
Appendix A Laboratory Audit Reports
Appendix B Task Force Field Measurements
Appendix C Task Force Analytical Parameters
Appendix D Task Force Sample Results
Appendix E Data Quality Control Report
-------
Ill
LIST OF FIGURES
Fig. 1 Refinery Plot Plan 3
Fig. 2 C Series Monitoring Wells 5
Fig. 3 B Series Monitoring Wells 8
Fig. 4 C Amd D Series Monitoring Wells " 9
Fig. 5 Contaniinant Plume Configurations 10
Fig. 6 Ground Water Elevation Contours 11
Fig. 7 Solid Waste Management Units, 1 of 2 24
Fig. 8 Solid Waste Management Units, 2 of 2 25
Fig. 9 Topographic Map 33
Fig. 10 Geomorphology 34
Fig. 11 Fill Areas 35
Fig. 12 Stratigraphic Column 37
Fig. 13 Geologic Map 38
Fig. 14 Legend Geologic Map 39
Fig. 15 Geologic Cross Section 40
Fig. 16 Ground Water Contour Map 42
Fig. 17 RCRA Monitoring Wells 45
Fig. 18 Well Cluster Construction Diagram 47
Fig. 19 Peristaltic Pump in Operation 52
Fig. 20 Areas Impacted by Organic Hazardous Constituents 59
Fig. 21 Areas Impacted Inorganic Hazardous Constituents 60
Fig. 22 Monitoring Well Locations 62
-------
IV
LIST OF TABLES
Table 1 Waste Quantities 30
Table 2 Well Construction Details 46
Table 3 Well Purge Sample Sequence 63
*
j j
Table 4 MDnitoring Well-Purge/Sample Equipment 64
Table 5 Preferred Order of Sample Collection,
Bottle Type, and Preservation List 66
Table 6 Selected Inorganic Data, Landfarm #11 71
Table 7 Selected Inorganic Data, Landfarm #12 72
Table 8 Selected Inorganic Data, Background Wells 73
Table 9 Selected Inorganic Data, Landfarm #10 74
Table 10 Selected Organic Data, Landfarms #11 and #12 75
Table 11 Selected Organic Data, Landfarm #10 76
-------
I. EXECUTIVE SUMMARY
A. INIRCDUCnaN
l. Task Force Objectives
This report sumnarizes the results of the Ground Water Task Force
Inspection conducted during January 26 through January 30, 1987, at the
Amoco Oil Company Refinery (Amoco) located'in Yorktown, Virginia.
The Administrator of the Environmental Protection Agency (EEA)
established a Hazardous Waste Ground Water Task Force (Task Force) to
evaluate the level of compliance with ground water monitoring require-
ments at hazardous waste treatment, storage and disposal facilities.
The ground water requirements were promulgated under the Resource
Conservation and Recovery Act (RCRA). In question is the ability of
existing or proposed ground water monitoring systems to detect contam-
inant releases from hazardous waste management units. The Task Force
comprises personnel from EPA Headquarters' core team, Regional EPA
offices and the states. This investigation was conducted on behalf of
the Task Force by EPA Region III.
The principal objectives of the inspection at Amoco were to deter-
mine the level of compliance with the requirements of 40 C.F.R. Part 265,
Subpart F - Ground Water Monitoring; determine if the ground water
monitoring program described in Amoco's RCRA Part B permit application
is in compliance with 40 C.F.R. Part 270.14(c); and determine if haz-
ardous waste constituents have migrated to ground water around the
facility.
The specific objectives of the Task Force investigation of Amoco,
were to determine if:
(a) Amoco has in place a ground water monitoring system
capable of meeting RCRA 40 C.F.R. Part 265 ground water
monitoring requirements, (b) The wells in place at
Amoco have detected any contamination.
(c) Designated RCRA monitoring wells are properly located
and constructed.
(d) Amoco has developed and is following an adequate plan
and procedures for ground water sampling and analysis.
(e) Amocofs analytical laboratories are producing accurate
and precise results.
(f) The ground water quality assessment program outline is
adequate.
-------
-2-
(g) Record keeping and reporting procedures for ground water
monitoring are adequate.
A complete Project Plan which incorporates a Sampling Plan for
facility monitoring wells, a Laboratory Audit Plan for the audit of
both consulting and corporate laboratories and a Sampling Audit Plan
for the audit of facility sampling procedures is incorporated into the
report by reference and not included in whole.
v *
2. Facility Background/Location
Amoco is a 600 acre site located on the York River near Yorktown,
Virginia. The facility is located in York County. The latitude and
longitude for the facility are 37' 12' 45" N and 76* 26' 45" W. Figure
1 presents a plot plan of the refinery.
The refinery receives foreign crude oil with a medium to high
sulfur content. The facility processes approximately 53,000 barrels
per day of crude oil by altering the crude to create gasoline and home
heating oil. This facility also produces liquified petroleum gas,
chemical stocks, furnace oil, petroleum coke, and sulfur coke. The
refinery operates 24 hours a day, 7 days a week and employs 210 people.
The refinery has been in operation since the completion of construction
in 1957. The primary RCRA regulated activity at Amoco is the landfarming
of hazardous wastes. There are three landfarms at the facility desig-
nated landf arms #10, #11, and #12. Landf arms #10 and #12 were granted
RCRA interim status by EPA in November 1980, and were listed on the
RCRA Part A application sutmitted by Amoco to EPA in November 1980.
Amoco submitted a Part B permit application to EPA on July 25,
1983, for landf arm #11. Amoco plans to secure a RCRA permit for landf arm
#11 and sutmit RCRA closure plans for landf arms #10 and #12. Currently,
only landfarm #10 is an active treatment unit and has been in operation
since early 1980. Landfarm #12 has not been in operation since late
1982.
Amoco wishes to pursue a land treatment demonstration permit for
landfarm #11 prior to receiving a final RCRA permit for this unit.
Amoco must formally sutmit closure and post-closure plans for landfarms
#10 and #12 no later than February 9, 1988. The post-closure permit
will include provisions for post-closure ground water monitoring.
Closure of the land treatment demonstration area is included in
the Part B permit application should the demonstration fail. However,
according to the Virginia Department of Waste Management (VEWM), closure
in the event the demonstration fails has not been addressed.
Amoco's RCRA Permit strategy is to divide landfarm #11 into two
treatment areas. One treatment area will be constructed and operated
for one year as a demonstration facility. The second unit will be
-------
-3-
2 T 30" '71
> R K
37°1.V45"
Latitude
Flood
Ebb
I
«_*
• H
00
Outfall Outfall
001 ^ 00? ;....
71
"IS
"17
Fig. 1 Refinery Plot Plan •
-------
-4-
constructed, if warranted, after the completion of the demonstration.
Hie demonstration project cannot, however, conmence until the RCRA
Permit is issued to Amoco by the VDWM. It appears that a final permit
determination will not occur until July 1988.
B. SUTWfcRY OF FINDINGS AND CONCLUSIONS
1. Ground Water Monitoring Program During Interim Status
j j
The RCRA interim status ground water monitoring program for Amoco
conducted since January 1983, was investigated by the Task Force.
Amoco's ground water monitoring well network has been modified and
expanded on several occasions since 1980. Amoco had installed a series
of observation wells in 1980 and 1981 to monitor ground water conditions
across the entire facility. RCRA monitoring wells were installed in
1981 and 1984 in order to assess the ground water conditions at the
three landfarms. These wells are shown in Figure 2.
Amoco installed a series of 21 cluster wells in 1985 to further
address the vertical movement of organic and inorganic contaminants in
ground water. Seventeen clusters contain 4 wells each while 4 clusters
contain 2 wells each for a total of 76 wells. The wells installed,
date of installation and type are indicated as follows:
Wells Date Installed Type
Bl-12 1980 Observation
B13-24 1981 Observation
Cl-16 1981 RCRA
C17-20 1984 RCRA
CQ-CP wells 1985 RCRA
Amoco obtained background concentrations for all parameters listed
in 40 C.F.R. Parts 265.92(b)(1), 265.92(b)(2), and 265.92(b)(3) during
four quarterly sampling episodes in 1982. All 1982 ground water
samples were collected from the Cl - CIS RCRA monitoring wells. Amoco
collected their first series of semi-annual indicator parameter samples
as required by 40 C.F.R. Part 265.92(b)(3) on January 24, 1983. The
first semiannual indicator parameter results were obtained on March 22,
1983. The results showed statistically significant differences in the
indicator parameters, pH, total organic carbon (TOC), specific
conductance, and total organic halogens (TOX). To verify these
results, the wells were resampled as required. Samples were split
between Amoco's laboratory in Naperville, Illinois, and a contract
laboratory. The analytical results did not conform to previous results
due to possible laboratory error. However, the results of resampling
confirmed the existence, based upon the Student's t-test, of statisti-
cally significant differences between the 1983 data and the background
data generated in 1982.
-------
-5-
C18
X
C19
X
X
C6
LANDFARM 10
C5
C8
X
Cll
C9
X
LANDFARM 12
CIO C12
X X
C17
X
C13
Cl
C3
LANDFARM 11
C2
X
C4
X
Fig. 2 C Series Monitoring Wells
-------
-6-
All three landfarms exhibited statistically significant differences
for TOC and TQX. On August 31, 1983, Amoco notified the VDWM of statis-
tically significant differences in the indicator parameters. Written
confirmation of the significant increase was received by the VDWM on
September 14, 1983. Amoco also submitted a Ground Water Quality
Assessment Plan to the Department on the above date. After the initial
semi-annual sampling episide, Amoco believed the upgradient wells were
too close to the landfarms to be unaffected by them and that diffusion
and dispersion processes could cause constituents to migrate upgradient
over short distances. Three new upgradient monitoring wells were
installed during 1984 and were designated wells C-17, C18, and C-19.
These wells were sampled during 1984 for the same analytes as the other
RCRA wells. Amoco did not compare downgradient results with upgradient
results as required by 40 C.F.R. Part 265.94(2)(ii).
The Ground Water Quality Assessment Plan of September 14, 1983,
was not sufficient to characterize the rate and extent of contamination.
Several meetings between Amoco, EPA, and the VDWM took place throughout
1984 to discuss the direction of Amoco's ground water assessment program.
On November 1, 1984, Amoco submitted a ground water Quality Assessment
Program Report which presented the following data:
The ground water at landfarms #10 and #11 had lead concentra-
tions in excess of the MZL of 0.05 mg/l. However, previous
analyses (February 1982, May 1982, July 1982, and October
1982) have not shown any concentrations above this limitation.
Well C-9 had a concentration of arsenic above the M3L of
0.01 mg/l, specifically 0.16 mg/l. All other wells are below
the MCL. Well C-6 had a concentration of cadmium above
0.01 mg/l (0.021 mg/l) and a concentration of chromium above
0.05 mg/l (0.06 mg/l). There have been no previous concentra-
tions of cadmium above 0.01 mg/l (sampling dates February
1982, May 1982, July 1982, and October 1982) and reportedly,
one occurrence of chromium above 0.05 mg/l. The sample
collected in July 1982, reportedly had a chromium concentra-
tion of 0.24 mg/l. However, no samples before or since
have been near this value so Amoco considered the data to be
suspect as to its accuracy. Finally, wells C-6 and C-9 had
nitrate concentrations in excess of 10 mg/l (27 and 1560 mg/l
respectively). Here again, neither of these wells has any
previous history of concentrations above 10 mg/l.
Well C-6 detected benzene, toluene, and methylethylJcetone
(MEX). Wells C-9 and C-ll detected only benzene. The
presence of these constituents in the ground water may be
attributed to the overloading of soil in specific areas of
landfarms #10 and #12. Due to the clay content in the soil,
the constituents are not expected to be widely distributed.
These constituents were not detected in wells C-7 and C-8.
-------
-7-
This report includes the first sunmary of statistical differences
between upgradient (background) wells and downgradient wells. In addi-
tion, comparison between upgradient and downgradient background para-
meters and current (1984) indicator parameter analysis was included in
this report.
Amoco also proposed modifications to their ground water monitoring
program to make it more responsive to the conditions at the refinery.
The monitoring well network was expanded and the frequency of analyses
for some parameters was increased.
During 1985, Amoco continued to collect quarterly samples from the
C wells and also sampled the B and D observation wells. The B and D
observation wells are not RCRA wells but were installed by Amoco to
monitor ground water quality beneath the facility. The B wells were
installed in 1980 and 1981 and are shown on Figure 3. The D wells were
installed in 1980 to monitor ground water quality at Amoco's two sanitary
landfills and are shown in Figure 4. The B and D well samples were
analyzed for the indicator parameters and the parameters to establish
ground water quality. The results indicated high levels of chloride,
radium, and sulfate. The indicator parameter results were similar to
the results from the C well monitoring episodes. However, no background
data was available to perform any statistical comparisons.
On October 14, 1985, Amoco submitted a second Ground Water Quality
Assessment Program report to the VDWM. In this report Amoco stated
that they believed a solid background data base for comparison had been
established due to additional data analysis and sampling. Amoco had
identified two plumes of lead contamination in 1984 beneath landfarms
#10 and #11. These results were used to determine the rate and extent
of contamination, and looked very different from results reported in
November 1984. In summary, the constituents which appear above back-
ground are zinc and several aromatic hydrocarbons. The aromatics
detected .were benzene, toluene, ethylbenzene, and ortho, meta, and
para-xylene. Aromatics were detected in wells C-6 and C-ll. Lead
reported being detected in the ground water in the November 1, 1984,
report does not exceed background levels in this report. The majority
of the zinc concentrations were at or just above the analytical detection
limit. Figure 5 exhibits the approximate location and dimensions of
the zinc and aromatic hydrocarbon plumes.
As was discussed in the Hydrogeology Section of the October 1985
Report, Amoco believes the API separator area, to the east of landfarm
#10, is a ground water discharge area. As a result, Amoco believes
that any contaminant plume, resulting from the landfarming activity
will be drawn to the API separator thus reducing the potential for
radial migration beyond the landfarm boundaries. As can be seen by
Figure 6, a cone of depression in the immediate vicinity of the API
separator can be deduced.
-------
s
-------
Fig. 4 C and D Series Monitoring Wells
-------
-lO-
cs
e
o
®
U
CO
O
E
O
hi
U-l
0)
E
3
— t 0)
CU C
O
i-( J3
(0 hi
•H CB
•U U
C 0
0) hi
U -O
O >>
0..3S
U
C
^J
N
E
O
hi
VM
.
5
"<3
CO
?
(Q
(/)
k.
S
i<
•2 «
g 1
o 1
0.5
,9 =
f ^ LU
CJ i"
81
il
W H>
< §
2
k.
5
$
a
en
c
.g
*^
ca
3
.O)
O
CJ
V
1
E
75
c
0)
^^
£
a
C
J-l
CO
>-
5
M-
C
O
^J
0)
£=
(X
w
c
CO
c
•H
e
CO
w
C
o
CJ
ul
60
•H
fa
0
c
c
>
cc
^^
1
to
40
o
ii
i
-------
-11-
j*
Ol
>.
J5
To
OT
•Q
CO
-------
-12-
During October 1985, VDWM required Amoco to determine the vertical
extent of ground water contamination. Up until this time, Amoco had
only been assessing the rate and extent of contamination on a horizontal
plane. As a result of this demand, Amoco proposed to install 21 well
clusters throughout the refinery. The proposed cluster wells were
installed in the fall of 1985. There are 21 sets of stainless steel
observation wells for a total of 76 wells. Seventeen of the clusters
contain 4 wells, conpleted at the following depths: 8, 18, 38, and 58
feet. These wells are identified in all documents by the CQ prefix,
a well number, and the depth (e.g., CQ 5-18). Four of the clusters
consist of 2 wells completed at depths of 8 and 18 feet, respectively
(e.g. CP-l-8).
On September 22, 1986, Amoco submitted to the VDWM a Ground Water
Quality Assessment Program report summarizing the results of all the
quarterly sampling which took place in 1986 up to the report date.
Inorganic analysis during 1986 detected chromium and nitrate in wells
at landfarms #10 and #12. These results are consistent with earlier
data reported in October 1985. In reviewing all the assessment moni-
toring data for 1986, Amoco reached the following conclusions:
(a) The ground water elevation contours (and flow patterns)
are consistent with previous analyses.
(b) The extent of the organic hazardous constituents coincides
with the previous report, (October 1985).
(c) The horizontal extent of each of the inorganic hazardous
constituents is extremely limited.
(d) Horizontal ground water flow velocities were calculated
to range from 5 X 10~7 cm/sec to 4 X 10~8 cm/sec
(0.0014 to 0.01 ft/day).
(e) The horizontal rate of migration is negligibly small with
the possible exception of induced ground water flow into
the API separator.
(f) Vertical ground water gradients were found to range from
0.023 ft/ft upward to 0.033 ft/ft downward for the inter-
val between 8 and 58 ft.
(g) Vertical flow velocities were calculated to range from
5 X 10~7 on/sec to 4 X 10~4 cm/sec (0.0014 to 1.1 ft/day).
(h) There is no evidence of vertical migration of hazardous
constituents related to the operation of the landfarm
into any of the deeper observation wells.
(i) The vertical rate of migration is negligibly small.
-------
-13-
(j) No firm conclusions can be drawn regarding the presence
of elevated concentrations of selenium and arsenic.
The VDWM does not believe that Amoco performed sufficient sampling
and analysis for organic constituents to conclude that the vertical
rate of migration is negligibly small. Amoco also concludes that the
extent of organic contamination coincides with previous reports. The
VDWM has requested on several occasions that Amoco release all sampling
results for organic compounds. As of the Task Force evaluation, neither
VDWM nor EPA, has received these results. - *
EPA. has concluded, after studying all pertinent interim status
ground water reports, that Amoco has not deliniated the vertical and
horizontal limits of the uppermost aquifer. Amoco has not defined
potentially interconnected aquifers nor intervening confining zones.
The VDWM never officially approved of the Amoco ground water assessment
plan, however, Amoco is implementing the plan as written. Due to the
absence of well construction logs for the B, CP, and CQ series monitoring
wells, insufficient information exists to determine if the monitoring
wells are constructed in a manner consistent with the requirements of
40 C.F.R. Part 265.9KC).
In summary, Amoco suspended statistical analysis for the indicator
parameters in September 1983 after triggering assessment. Prior to
triggering assessment, Amoco conducted quarterly monitoring in 1982 for
the required background parameters. From January through September
1983, Amoco conducted the required semi-annual analysis and statistical
compraisons. During 1984, 1985, 1986 and continuing past January 1987,
Amoco continued to perform quarterly assessment monitoring. During the
second half of 1984, monthly sampling and analysis was conducted.
The assessment program implemented by Amoco, while intense in
scope, has not defined the critical hydrogeologic information needed to
determine the vertical and horizontal extent of ground water contami-
nation.
During sampling episodes in 1986, arsenic and selenium were detected
in concentrations above the analytical detection limits for these two
constituents.
2. Proposed Ground Water Monitoring Program for RCRA. Permit
Amoco submitted a Part B permit application for landfarm #11 to EPA. on
August 9, 1983. As previously stated, Amoco plans to close landfarms
#10 and #12 once a PCRA. permit is secured for landfarm #11. A post
closure permit for landfarms #10 and #12 is scheduled to be issued in
FT 1990 or 1991 by VDWM.
Amoco submitted with their Part B permit application, a description of
the detection monitoring program to be implemented for landfarm #11
-------
-14-
when the permit is secured. The detection monitoring program was
designed to monitor for indicator parameters, waste constituents, and
parameters to characterize the suitability of ground water as a drinking
water supply. The indicator parameters include: TOC, TQX, pH, and
specific conductance. The waste constituents include: oil and grease,
cadmium, lead, and chromium.
EPA and VDWM found the ground water detection monitoring plan to
be insufficient. On August 25, 1985, VDWM sent Amoco a letter indi-
cating that if landfarm #11 was to be permitted, EPA Appendix VIII
analysis, as found in 40 c.F.R. Part 264, must be performed for all
landfarm monitoring wells. Also, Amoco was to provide VDWM with informa-
tion defining the rate and extent of any ground water contamination
plume which has emanated from landfarm #11. Since the Part B permit
application has been submitted to VDWM and EPA, six Notice of Deficiencies
(NDD's) concerning the Part B permit have been issued by both agencies.
A NOD issued by VDWM on December 31, 1986, requested Amoco to submit
all construction specifications and boring logs for the 76 cluster
wells installed in November 1985. As a result of this request, it was
learned that boring logs did not exist for the 76 cluster wells.
3. Task Force Sampling Data
As part of the Task Force Investigation, ground water samples were
collected from 16 monitoring wells. A total of 21 field samples were
collected including a field blank, an equipment blank, a trip blank,
and a pair of duplicate samples. With few exceptions, samples were
analyzed for 194 organic and inorganic parameters through the EPA's
Contract Laboratory Program (CLP).
Laboratory error rendered a major portion of samples analyzed for
volatile organic compounds unreliable. The analytical laboratory
exceeded the allowable holding time of 7 days by 26 to 34 days for 12
of the 21 samples. Laboratory analytical problems also resulted in the
contamination of several method blanks with methylene chloride and
acetone. Of the volatile organic data that was validated, benzene and
total xylenes were identified in monitoring wells CQ 17-8 and CQ 17-18.
In well CQ 17-18, benzene was detected at 140 ppb and total xylenes at
650 ppb.
Semi-volatile organic results are considered semi-quantitative, at
best, considering that neither matrix spikes nor matrix spike duplicates
were analyzed by the laboratory. As with the volatile organic samples,
the maximum holding time for semi-volatile samples of 40 days was
exceeded for 12 of the 21 samples. Of the samples analyzed within the
maximum holding time, 2-Methyl naphthalene (110 ppb) and bis (2-Ethylhexyl)
phthalate at (2300 ppb) were detected in CQ 17-8.
According to data usability reports prepared by PPC Environmental
Management, Inc., the quality control for the graphite furnace metals
-------
-15-
was generally reliable. However, laboratory difficulties in initial
calibration verification rendered total antimony, arsenic, cadmium and
thallium results semi-quantitative. The only furnance metal detected
was arsenic at 12 ppb in well B-ll. This concentration does not exceed
the Maximum Concentration Limit (MX) for arsenic of 50 ppb.
Total and dissolved results for TCP metals can be considered
quantitative with the exception of total iron and dissolved barium and
potassium which are considered said-quantitative. Eighteen sanples
exhibited elevated concentrations of dissolved calcium, iron, magnesium,
potassium, and sodium. Dissolved iron was detected in all monitoring
wells at concentrations ranging from 70 ppb to 20,800 ppb.
Results for indicator parameters are considered quantitative with
the exception being Purgable Organic Carbon (POC) considered semiquanti-
tative. The vast majority of wells display elevated TOC concentrations.
Total phenols were detected in four monitoring wells at concentrations
ranging from 53 ppb to 94 ppb.
4. Conclusions
In summary, the Task Force, as a result of this investigation, has
found that the ground water monitoring program at Amoco does not fulfil
the following:
o 40 C.F.R. Part 265.92(a) - The Sampling and Analysis Plan
has not been revised in response to changes in the ground
water monitoring program, and as a result, does not accurately
describe procedures currently in use. In particular, tech-
niques and procedures for the following are not described in
sufficient detail.
- Field and laboratory Quality Assurance/Quality Control.
- Ground water evacuation and sampling
- Sample Analysis (Field and Laboratory)
o 40 C.F.R. Part 265.44 - A field log book which fully
describes all field procedures and observations including,
but not limited to, well evacuation, sampling methods, and
water level measurements has not been maintained.
o 40 C.F.R. Parts 265.90(a), (a)(l), (a) (2) and 270.14(c)(2)
Boring (Stratigraphic) and construction logs for the CQ and
CP series cluster wells were not prepared. Therefore, infor-
mation critical to the characterization of geologic and
hydrogeologic conditions is not available.
- The vertical extent of the uppermost aquifer and any
potentially interconnected aquifers has not been
sufficiently characterized.
-------
-16-
o 40 C.F.R. Parts 265.90(a)(1) and 265.91(a)(1) Upgradient/
Background wells may not provide results truly reflective of
ground water quality conditions unaffected by the facility.
40 C.F.R. Parts 265.90(a) and 265.92(a) - Use of
peristaltic pumps for purging and sampling of monitoring
wells are known to cause degassing of volatile organic
compounds (VOC's) due to sudden pressure changes.
o 40 C.F.R. Part 265.93(d) - The Ground Water Quality Assess-
ment Plan does not include a specific description of goals,
performance requirements and schedules for implementation,
and completion of required tasks. In addition, the following
deficiencies exist:
- A representative determination of aquifer properties
via performance of multiple well pump tests or, at a
minimum, slug tests has not been conducted to characterize
nonhomogeneous and anisotropic conditions
- Intake and discharge tubing associated with sample
collection pumps are constructed of tygon and polyethylene,
materials which can interfere with sample integrity.
- The assessment monitoring program is not complete in
that the rate and extent of contamination has not been
adequately characterized.
- The rationale for selection of individual well depths
within each cluster is based upon modeled projections as
to the maximum vertical migration of contaminants although
baseline determination of uppermost aquifer thickness
through direct investigation methods is lacking.
- Seasonal and temporal variations in ground water eleva-
tion and flow patterns in response to the York River have
not been adequately assessed.
In addition, the following findings are offered.
1, Several facility wells sampled by the Task Force have
• detected some levels of organic and inorganic constituents.
In addition, the Task Force noted the presence of light
phase irrmiscibles in well CQ 17-8. The source and/or
sources have not been positively identified.
2. The quality and accuracy of sample results is of concern
given difficiencies in Quality Control procedures noted
in audit reports for analytical laboratories utilized by
the facility.
-------
-17-
Ihe ongoing Ground Water Quality Assessment should place increased
emphasis on the determination of rate and extent of contaminant migra-
tion, both vertically and horizontally, and the identification of
discrete source(s) of contamination within the facility.
-------
-18-
II. TECHNICAL REPORT
A. EftCXGRQUND
1. Site History
Amoco Oil Company currently operates a 600 acre refinery on the
York River near Yorktown, Virginia. The refinery is located in York
County, Virginia. Construction of the refinery was completed in 1957.
The land on which the refinery was built was rural prior to construction.
The primary RCRA activity occurring at Amoco is the landfarming of
hazardous waste produced in the refining of crude oil. The refinery
processes approximately 53,000 barrels per day of crude oil. The crude
oil is altered through the refining process to create premium gasoline
and home heating oil. Also produced at the facility is liquified petro-
leum gas, chemical stocks, furnace oil, petroleum coke, and sulfer
coke. The refining process at this facility operates 24 hours a day, 7
days a week and employes approximately 210 people.
The Amoco facility contains two RCRA interim status landfarms.
Amoco was granted interim status for landfarms #10 and #12 in November
1980. Currently, only landfarm #10 is active. Landfarm #11 has not
been used since 1980 and landfarm #12 since late 1982. Amoco is cur-
rently pursuing a land treatment demonstration permit for landfarm #11
only. Amoco submitted a Part B permit application to EPA on July 25, 1983,
for landfarm #11. Amoco will cease operation of landfarms #10 and #12
prior to the November 1988 land based unit permit deadline.
In April 1984, Amoco submitted to EPA, through their consultant,
Stone & Webster Engineering Corporation, the Yorktown Refinery Land
Treatment Manual for landfarm #11. This manual included the land
treatment demonstration plan, waste characterization plans, operating
procedures, a ground water monitoring plan and a closure plan if the
land treatment demonstration is unsuccessful.
2. Enforcement Actions
As of January 31, 1987, no formal enforcement actions have been
taken against Amoco, Yorktown, by either EPA or the VDWM. Prior to the
Task Force Inspection, Amoco had received 4 Compliance Evaluation
Inspections by either EPA or VDWM. Three of the four inspections
detected some Class I and Class II violations.
The original Amoco Part A application listed the "Sour Water
Stripper" as being a regulated unit requiring a RCRA permit. Subsequent
documents submitted to EPA and VDWM by Amoco eliminate the "Sour Water
Stripper" as a RCRA unit since the process is fully enclosed.
-------
-19-
After Amoco submitted their Part B permit application to EPA for
landfarm #11 on July 25, 1983, a series of NOD'S were issued by EPA and
the VDVM. The following list contains the issue dates of the NOD's and
dates of Part B permit application submissions.
7/25/83 - Part B permit application sutmitted to EPA for landfarm
#11 and Sour Water Stripper
10/7/84 - Notice of Deficiency (EPA)
2/28/84 - Notice of Deficiency (EPA)
3/6/84 - Notice of Deficiency (EPA)^
7/27/84 - Notice of Deficiency (EPA)
11/6/85 - Part B permit application sutmitted to EPA for landfarms
#10 and #12
2/26/86 - Notice of Deficiency (VDWM) for landfarms #10 and #12
12/31/86 - Notice of Deficiency (VDWM) for landfarm #11
The deficiencies noted in the NOD's concerned mainly issues such
as waste application rates, soil information, waste analysis, clean-up
activities to be employed after the treatment demonstration, and land-
farm #11 construction specifications. As of the Task Force evaluation,
the Part B permit application for landfarm #11 remains deficient in
certain areas.
3. Adjacent Land Use
The Amoco Refinery was constructed on the York River in a section
of the Newport News area known as Goodwin Neck. The refinery was built
on the Goodwin Neck Peninsula which is bounded by the York River to the
north, Wbrmley Creek to the west, the Chesapeake Bay thoroughfare to
the east, and Back Creek and Seaford to the south. Amoco shares the
eastern edge of the peninsula with a Virginia Power Company fossil fuel
electrical generating station. The small town of Seaford is approxi-
mately three quarters of a mile southeast of Amoco. For
facility location refer to Figure 1.
B. INVESTIGAnVE METHODS
The Task Force investigation of the Amoco Refinery involved:
(a) Reviewing all pertinent facility records and conducting
facility personnel interviews;
(b) conducting on-site ground water sampling and analysis
during January 26 thru January 30, 1987;
(c) Auditing Amoco's analytical laboratory and contractor
laboratory;
(d) Auditing Amoco's ground water sampling procedures;
-------
-20-
1. Record/Document Review
All pertinent Amoco records from EPA Region III and the VDWM were
reviewed prior to and during the Task Force inspection. During
December 1986, Planning Research Corporation (PRC), Chicago, Illinois,
under contract to EPA, carpi led an information/document package for
Amoco. The PRC package consists of two volumes containing copies of
all pertinent documents and correspondence related to the refinery from
EPA and VDWM files.
j *
The PRC package was used as a comprehensive review and reference
document to aid in the Task Force investigation. Important documents
were also reviewed and copied at the facility. These documents were
copied by EPA and VDWM to fill in data gaps in the respective files.
A total of 33 documents were reviewed, copied, and collected at Amoco.
All these documents are on file with EPA Region III.
Specific documents reviewed included the Ground Water Sampling and
Analysis Plan, Ground Water Quality Assessment Plans, analytical results
from previous ground water sampling episodes, geologic reports, Part A
and Part B permit applications, facility operating records, and records
showing the quantities and source of hazardous wastes which are managed
in the RCRA landfarms.
Interviews with facility personnel and their consultants were
conducted throughout the investigation. Three Amoco representatives
and two consultants were interviewed during the Task Force inspection.
All discussions, interviews, and observations were recorded by Task
Force personnel in logbooks. All log books were collected at the
conclusion of the investigation and are currently on file at EPA Region
III.
Photographs of the RCRA landfarms, Task Force operations, and
Amoco ground water sampling procedures were taken during the inspection.
A total of 74 photographs were taken and are currently on file at EPA
Region III. A duplicate set of photographs was sent to Amoco at their
request.
2. Facility Inspection
The on-site inspection was conducted to collect in-situ data and
ground water samples from a select number of RCRA monitoring wells
surrounding the three landfarms and facility background wells. A
visual inspection of the refining process area was also conducted. In
addition, Solid Waste Management ifliits (SWMU's) were identified during
the inspection.
The ground water sampling techniques of J.R. Reed & Associates
(J.R. Reed) were also observed by the Task Force to ensure proper quality
-------
-21-
assurance during ground water sampling episodes. As was previously
mentioned, photo-documentation of the refinery waste handling operations
and Task Force sampling procedures took place.
3. Laboratory Audits and Inspections
J.R. Reed located in Newport News, Virginia, conducts all the
ground water monitoring activities at Amoco. J.R. Reed performs all
the in-situ and indicator parameter analyses and occasionally performs
analysis for organic parameters.
The Amoco corporate laboratory in Naperville, Illinois, performs
the organic analysis and the vast majority of inorganic analysis ground
water samples obtained at Amoco. Both the Naperville laboratory and
the J. R. Reed laboratory were audited by EPA's Central Regional
Laboratory in Annapolis, Maryland. The audits included a review of
analytical equipment and methods, quality control procedures and documen-
tation, and a completeness and accuracy review of pertinent laboratory
records. Essentially the audits are conducted to ensure that the labor-
atories posess the ability to produce high quality data.
The Laboratory Audit Plan is attached to the Final Project Plan of
January 22, 1987. This plan contains the specific details of the
laboratory audits.
The J.R. Reed Laboratory audit took place on January 30, 1987.
The Amoco laboratory in Naperville, Illinois, was audited on February 4-
5, 1987. Results of the laboratory audits are attached to this report
as Appendix A.
4. Sampling Audit
In order to assess the facility's sampling procedures, an audit
was conducted prior to the start of Task Force sampling activities. A
demonstration of sampling procedures was provided by J. R. Reed on
January 27, 1987. Well CQ 1-18 was selected for this demonstration.
Actual samples were not collected due to extreme winter weather condi-
tions. A walk through of sampling procedures was provided including
the actual purging of CQ 1-18 to demonstrate the operation of the peris-
taltic (vacuum) pumps employed on a routine basis by J.R. Reed field
staff.
The evaluation included the observation of J.R. Reed's well purging
and sampling procedures, interviewing sampling personnel, collection of
documents, review of the sampling plan and photo documentation of
procedures.
An outline of sampling audit procedures is provided in the Task
Force Project Plan and were followed during the demonstration. All
pertinent information and observations were recorded in field log books
and later transferred to Comprehensive Monitoring Evaluation (CME)
-------
-22-
checklists. Concurrent with the Task Force Evaluation, the VDWM scheduled
a CME of Amoco; thus, conpletion of the checklist was required. A CME
is an in-depth examination of all facets of a facility's RCRA ground
water monitoring program including characterization of geologic and
hydrogeologic conditions. The checklist also includes details of general
field methods, quality assurance and quality control (QA/QC), and chain
of custody.
Findings of the CME are incorporated in this report. Copies of
the checklist are contained in Task Force'files.
5. Sampling Program
The major objective of the Task Force investigation at Amoco was
to determine the impact the three PCRA landfarms may be having on local
ground water quality.
The sampling portion of the Task Force investigation involved
three activities: (1) The measurement of water levels in all monitoring
wells, (2) sampling a pre-selected number of RCRA wells surrounding the
landfarms and background wells, and (3) installing automatic water
level records on two monitoring wells to determine the extent of hydraulic
head change in response to cyclical tidal fluctuations. Water level
measurements were taken in an attempt to determine the direction of
ground water flow. Between January 26 and January 30, 1987, the Task
Force, with its contractor, Versar Inc., collected 21 field samples
including field blanks, equipment blanks, and a pair of duplicate samples.
The ground water samples were drawn from CQ Series wells, CQ 2-8 CQ 2-58,
CQ 6-38, CQ 9-58, CQ 10-18, CQ 11-8, CQ 12-8, CQ 12-38, CQ13-8, CQ 14-18,
CQ 14-58, CQ 15-8, CQ 17-8, CQ 17-18, CQ 17-38, CQ 17-58, and well B-ll.
All samples collected were split with Amoco.
All samples were shipped to the appropriate laboratory the same
day samples were collected or were shipped early the next day. All
inorganic samples were shipped to ENTBC Laboratories in Salem, Virginia;
all organic samples were shipped to EMSI Laboratories in Camarillo,
California; and all dioxin samples were shipped to Compu-Chem Labora-
tories in Research Triangle Park, North Carolina.
All sampling activities were conducted following the January 22,
1987 Project Plan. Sampling procedures are described in detail in the
Project Plan. Also described in detail in the Project Plan are:
(a) description of sampling protocol;
(b) proposed sampling schedule;
(c) container and preservative details;
(d) shipping requirements; and
(e) QA and QC procedures.
-------
-23-
The actual sampling procedures used during the sampling episodes
followed the guidelines set forth in the Project Plan and the Hazardous
Waste Ground Water Task Force Protocol for Ground Water Evaluations,
September 1986.
C. WASTE MANAGEMENT UNITS AND OPERATIONS
1. Waste Management Units
Amoco has identified 7 Solid Waste Management Units (SWMUs) including
the two RCRA landfarms. Currently, three of the units are active. The
following descriptions were taken from Amoco's response to an EPA letter
dated April 26, 1985, concerning SWMLT's. SWMU's located at the Amoco
facility are shown on Figures 7 and 8 and include:
Industrial Waste Landfill (Active)
Amoco was issued a permit by the Commonwealth of Virginia to
operate an industrial waste landfill in York County, Virginia on June 21,
1983. It consists of two separate, but adjacent areas known as Site A
and Site B. The area designated as Site A is 260 feet long and 260 feet
wide and is used for disposal of asbestos waste. This site is surrounded
by a chain link fence and signs are posted on all sides.
Asbestos was disposed of by digging a trench approximately 10 feet
wide and 20 feet long and not more than 1 foot deep. The material was
placed in plastic bags and placed in the trenches. At the end of each
day the asbestos was covered with at least 6 inches of soil. No asbestos
has been placed in Site A since February 4, 1983. Since that time, all
asbestos has been sent off-site to a VDWM approved landfill.
The area designated as Site B is 750 feet wide and 1050 feet long
and is used for disposal of silica-alumina catalyst, bauxite catalyst,
merox catalyst, inert waste, waste sulfur, and S-201 alumina. This
area is designed to manage approximately 1.8 tons/day of solid waste
for disposal. Signs are posted on all sides. The landfill is divided
into sections and all waste is placed in the designated section. The
waste is spread evenly across the sections and covered within one week
after disposal.
There is a chain across all entrances and these entrances are
locked. The Oil Movements mechanical supervisor is responsible for
management of both Site A and Site B and he has been trained in solid
and hazardous waste management. There is no closure plan at present
for these sites.
Hazardous Land Treatment Facility (Active)
The primary RCRA-regulated activity at this facility is landfarming
of hazardous wastes. These wastes consist of API separator sludge,
heat exchanger sludge, slop oil emulsions, leaded tank bottoms, oily
wastes, monoethanolamine, and ASP sludge.
-------
CO
4J
•H
C
C
-------
-25-
-------
-26-
Amoco lias interim status to operate two landfarms on its facility.
They are designed as landfarms #10 and #12, all within the refinery's
boundaries and within close proximity to each other.
Presently, landfarm #10 receives all the refinery's wastes. The
waste is distributed over the landfarm by a "Big A" vacuum truck. The
waste is disked into the soil using a skidder disking plow. This
aerates the soil and aids in the biological activity.
Landfarm #10 is 1450 feet long and 550 feet wide. It is surrounded
by a 2 foot earthen berm to control run-on and run-off. The run-off is
pumped to the Stormwater Retention Pond for processing in the refinery's
water treatment plant. No engineering drawings are available. Closure
plans were submitted with the Part B permit application in August, 1983,
to the EPA and the VDWM.
Landfarm #12 is 1050 feet long and 500 feet wide. No waste has
been applied to this landfarm since late 1982. It is surrounded by a 2
foot earthen berm to control run-on and run-off. No engineering drawings
are available. The closure plan was submitted with the Part B permit
application in August 1983 to the EPA and the VDWM.
Landfarm #11 is 600 feet long and 340 feet wide. This site has
not been used since early 1980 and is the proposed site for the treat-
ment of refinery wastes. The demonstration plan, waste characterization,
operating procedures, ground water monitoring plan and closure plan for
this site were submitted to the EPA and the VDWVl on April 16, 1984.
Stormwater Retention Pond (Active)
Oily water from storage tank water draws and unit sewer boxes
gravitate through underground concrete piping into the API separator.
The API separator is made of two parallel cells, each with a capacity
of 2500 gallons per minute (gpm). The two cells are provided so that
continuity of operation can be maintained while one channel is being
cleaned or repaired. Four 6 inch by 6 inch openings in the downstream
wall of the inlet box prevent oil from accumulating in this section of
the separator. A picket fence type distribution baffle is installed at
the inlet of each cell to prevent channeling of the water and thereby
ensuring distribution of the oily water over the full width of the
cell. The picket fence baffle extends to the operating water level so
that in the event of excessive clogging, flow over the baffle will occur.
The water level in the API separator is controlled by an adjustable
outfall weir downstream of the oil retention baffle. Oil impounded by
the oil retention baffle will be skimmed into the collecting trough by
operation of the Harding skimmer equipment. This mechanism, when
moving forward, will skim oil off the surface of the water and on the
return will scrape the sludge from the bottom of each cell to a sump
for removal.
-------
-27-
The average flow rate to the API separator is 1100 gpm of waste-
water per day. Hie dry oil sump is capable of holding approximately
2500 gallons of oil, but is seldom allowed to become that full. The
reason for this is the dry oil sump, when full, will flow into the API
separator. This oil is pumped off to storage as slop and reused as
charge to the Delayed Coking unit. Water overflowing the outfall weir
is pumped from the API spearator sump into the equalization basin
depending on the conditions of the wastewater, i.e., pH and ammonia.
If an upset condition occurs, the water is diverted to the Storm-
water Retention Pond (SWRP) until it clears up. Normal flow to the
SWRP is equalization pond overflow, backwash water from the Neptune
filter and final filter. The SWRP has a capacity of approximately 10
million gallons. The height of the dikes around the SWRP and the
equalization pond is approximately 9 feet. The normal operating
level in the equalization pond is 10 feet and the SWRP is 5 feet.
The sludge in the SWRP is approximately 2 to 4 inches in thickness.
The API separator and the SWRP sludge have been analyzed and the
results are included in the waste characterization section of the
Yorktown Refinery Land Treatment Manual prepared by Stone and Webster
Engineering Corporation, Inc., in April, 1984.
The equalization basin discharges into the MicroFloc filter to
collect suspended solids and then into the Activated Sludge Plant (ASP).
The ASP consists of a micro-organism aeration tank, clarifier thickener,
final filter, backwash clarifier, and aerobic digester.
API Separator Sludge Pits (Inactive)
Located east of the API separator are two sludge pits. These pits
are 80 feet wide and 130 feet long. Sludge from the API separator was
placed in these pits starting in 1965 and was discontinued in 1971.
The pits have one foot earthen berms which control run-on and run-off.
Plans are to remove the sludge with a front-end loader, place it in a
dump truck, and then spread it evenly over Landfarm #10. The sludge
cannot be placed on landfarm #10 after November 8, 1988. The sludge
will be disked into the soil. After all sludge is removed, the pits
will be backfilled with suitable soil and a clay cap placed over the
pits. An 18 inch layer of topsoil will be placed on the cap and seeded
with Kentucky Fescue 31 grass seed. Kentucky Fescue 31 is recognized
by the U.S. Department of Agriculture and the U.S. Soil Conservation
Service for use in the Yorktown area.
Nbn—Hg^-arrinn.c! Land Treatment Facility (Inactive)
In August 1981, non-leaded tank bottoms from tank 405 were distri-
buted evenly over an area 525 feet by 525 feet inside the firewall at
tank 110. This sludge was disked into the ground. No sludge has been
placed in this area since 1981. The area is surrounded by a 6 foot
-------
-28-
high earthen berm to control tank 110's contents and run-on and run-off.
No engineering drawings are available, nor has a closure plan been
developed for this area.
Leaded Tank Bottoms Pits (Inactive)
Tanks 608 and 609 are used for leaded gasoline storage. Tank 608
was out of service in June 1979, for inspection and repairs. During
this period a 4 foot by 8 foot by 2 foot deep pit was dug outside the
tank manway inside the firewall. Sludge and water which had accumulated
at the bottom of the tank was washed into the pit that had a small weir
located in the middle. This allowed the sludge to settle in one section
of the pit and the water to enter the other section. The same procedure
was used for Tank 609 in February 1978. No sludge from either tank has
been buried since 1979.
Industrial Waste Landfill (Inactive)
In 1966 a fire training area and a trash burning area were put
into operation. These areas were also used as a storage area for old
exchanger bundles and tube sheets. In 1972, the fire training area was
moved to another location within the refinery. The fire training pit
and related equipment was removed as were the fuel storage tanks.
The exchanger bundles and tube sheets were also removed and placed
in a storage area east of the warehouse storage yard.
The foundations for the fuel storage tanks were dismantled and the
area was backfilled and graded. This landfill has not been used since
1972.
2. Facility Operations
Amoco operates 24 hours a day, 7 days per week. The major product
of the refinery is Amoco Premium lead-free gasoline. Also manufactured
is furnace oil for home-heating, petroleum coke for industry, butane,
propane, and sulfur. The primary RCRA regulated activity at the facility
is the landfarming of hazardous wastes. The refinery process creates a
number of hazardous wastes which are landfarmed. The RCRA landfarms
accept a variety of refinery process sludges for the primary purpose of
disposal. The treatment methods which take place in land treatment
units include the following:
(a) Aerobic/anaerobic microbial decomposition of organic
constituents of the waste sludges into carbon dioxide
gas (C02) and water (H2O) ;
(b) Chemical oxidation and/or hydrolysis;
(c) Ion exchange - inorganics such as heavy metals are bound
up in the soil; and
(d) Chemical precipitation.
-------
-29-
During the Task Force investigation it was noticed that Amoco
continued to apply sludges to landfarm #10 even though the ground was
frozen. Under frozen conditions, it is doubtful whether any of the
processes mentioned above were effective in treating the hazardous
constituents of the sludges.
The landfarmed wastes generated by the refining operations include:
API Separator Sludge
' ,
Oily water from the process unit sewers and refinery tank farm
sewers and water from chemical cleaning processes are routed to the API
separator for gravity separation of oil, water, and solids. A traveling
rake system skims oil off the top into a wet oil sump and rakes the
bottom sludge into a sludge sump. The oil is pumped to the slop oil
tanks and the bottom sludge is collected for treatment in landfarm #10.
Slop Oil Dnulsion
Slop oil emulsion is generated from the intimate mixing of oil and
water. Sources of slop oil emulsion include the slop oil tanks, a
settling basin, the equalization basin and the storm water retention
ponds.
Leaded Tank Bottoms
Leaded gasoline is stored in several storage tanks. Leaded tank
bottoms are generated from the settling of water and particulates such
as rust, scale, and dirt. Bottom sediment is removed from these tanks
approximately once every ten years and is collected for treatment in
landfarm #10.
Heat Exchanger Sludge
The refinery uses York River water for once-through process cooling.
Water is routed through all the water-cooled heat exchangers and then
to an effluent tank before returning to the York River. Heat exchanger
sludge consists of solids which are generated when the heat exchangers
are taken out of service and cleaned by washing with high pressure jets.
Oilv Wastes
Oil-soaked materials such as soil, catalysts, and oil spill cleanup
materials are classified as oily wastes. Oily soil, generated from
leaks and spills, consists of soil from the dredging of refinery ditches,
the settling basin, the equalization pond, and the stormwater retention
pond. Oily catalyst is generated during the shutdown of the catalytic
cracking reactor and the regenerator. The catalyst in these units that
cannot be regenerated is disposed. Oily spill cleanup material consists
of biodegradable straw, hay, and oil absorbent blanket.
-------
-30-
Moraetjianolamine (MEA)
Hie refinery is designed to process high sulfur content crude oil.
The sulfur content of self-produced fuel gas must be reduced in order
to control emissions of sulfur dioxide. Sulfur in the form of hydrogen
sulfide (H2S) is removed from the gas by scrubbing with MEA. The MEA
is regenerated by stripping with steam to remove the absorbed H2S.
The continuous absorption-stripping process results in an accumulation
of impurities in this system that eventually requires the disposal of a
portion of the circulating MEA. The discarded material, consisting of
water, rust, sulfides, and MEA, is collected for treatment in landfarm
#10.
Storage Tank Bottoms
Non-hazardous tank bottoms are generated in tanks that contain non-
leaded fuel by the settling of water and emulsion mixed with particulates
such as rust, scale, and dirt. Bottom sediments from these tanks are
collected annually for treatment in landfarm #10.
ASP Sludge
Activated sludge is generated from the wastewater treatment plant's
activated sludge unit. Activated sludge is comprised of a microbial
biomass that feeds on organic matter in the wastewater. To maintain a
constant biomass, a certain amount of the activated sludge is removed
from the activated sludge unit and subsequently treated in landfarm #10.
Table 1 summarizes the total, annual quantities of waste produced
by the above sources which are treated in the KCRA landfarms.
Table 1. waste Quantities
Waste Type Annual Q|liant'ityr Tons/year
API Separator Sludge 275.0
Slop Oil Elmlsions 2.5
Leaded Tank Bottoms 1.0
Heat Exchanger Sludge 1.0
Oily Wastes 10.0
Monoethanolamine (MEA) 18.0
Storage Tank Bottoms 256.0
ASP Sludge 975.0
-------
-31-
In March 1984, Amoco completed a waste characterization study of
the wastes which are landfarmed. This study updated the waste characteri-
zation study conducted by Amoco in 1981.
The major differences between the two studies are listed below:
In general, the concentrations of most of the heavy metals decreased
in the recent study. Cadmium, for example/ was detected in the API
separator sludge in the 1984 study at 0.17 mg/kg compared to 0.7 mg/kg
in the 1981 study and was not detected in any of the other waste samples.
The concentrations of three metals, i.e., chromium, copper, and nickel,
showed a net increase in the 1984 study.
Oil and Grease
The net concentration of oil and grease was significantly lower
for the 1984 waste samples, compared with the 1981 data. For example,
the oil and grease content in API separator sludge decreased from 25
percent in 1981 to 1.5 percent in the recent study.
Amoco's waste characterization studies concluded that the following
hazardous constituents identified in Appendix VIII of 40 C.F.R, Part
261 are most likely to be present in the landfarmed process wastes:
Inorganic Constituents Organic Constituents
Arsenic Analine
Cadmium Benzene
Chromium Benzo (a) Pyrene
Cyanide Chloroform
Lead Chrysene
Mercury 1,1-Dichloroethylene
Nickel 2,4 Dimethylphenol
Selenium Fluoranthene
Naphthalene
Phenols
Toluene
An additional activity taking place at the Amoco Refinery is the
treatment of "sour water." Sour water is defined as water which has
become contaminated with hydrogen sulfide (H2S) and ammonia. The "sour
water stripper" meets the requirements of a totally enclosed treatment
system; thus, is not regulated under RCRA.
The sour water strippers were built to enable Amoco to meet the
requirements of the Clean Water Act for process water being discharged
to the York River. The sour water strippers process sour water collected
-------
-32-
fron the crude unit, the FCCU, the coker, and the ultraformer. The
stripping steam removes the H2S and aimonia. After being stripped, the
resultant process stream termed "sweet water" is sent to the crude unit
desalter for phenol removal. The stripper off gas is sent to the flare
and the balance of the sweet water goes to the refinery gland water
system. The sour water strippers are designed to process 200-250 GPM
of sour water with an ammonia removal of 98% and H2S removal of 99.8%.
D. SITE
1 . Topography
The Amoco Refinery is located within the lower York- James Peninsula
and is shown in Figure 9. Topography is characterized by a series of
plains whose surfaces dip slightly eastward towards major estuaries.
The plains are separated from each other by scarps or cliffs that face
eastward or parallel the James and York Rivers. The facility is situated
on Hampton Flat between Big Bethel Scarp and the Plumtree Island Trough
and Ridge area along the Chesapeake Bay and is displayed in Figure 10.
The Hampton Flat, a morphological subdivision of the Virginia Coastal
Plain consists of Goodwin Neck, the western part of Crab Neck, Fish
Neck and the Poquoson area. The refinery is located on Goodwin Neck, a
peninsula bounded by the Chesapeake Bay to the east, the York River to
the north, Back Creek to the south, and Wbrmley Creek to the west. The
flat is drained by Back Creek, Wormley Creek, and a number of smaller
unnamed streams. Within the confines of the refinery, there is less
than a 5 foot difference in surface elevation due in part to the place-
ment of hydraulic fill to raise the grade of low lying areas. Fill
thicknesses range from 1 foot to approximately 7 feet. Fill areas are
shown in Figure 11. The two low areas on the eastern side of the site
appear to have been tributaries of Bull Creek which flows north to the
York River. Present surface elevations range from approximately 8 feet
mean sea level (msl) to 13 feet msl.
2. Geology
The refinery is underlain by the marine clayey sand facies of the
Norfolk Formation of late Pleistocene age. The Norfolk formation is
the surficial stratigraphic unit overlying the Hampton Flat. Formation
thicknesses range from 8 feet to 15 feet below ground surface. Drilling
records for C series wells installed in 1981 reveal a compositional
variation from brown/gray sandy clay to a fine to medium clayey sand.
Beneath the Norfolk Formation is the Yorktown Formation of Miocene
Age. The Yorktown Formation is the uppermost member of the Chesapeake
Group which also includes the Calvert and Saint Marys Formations , in
decending order. According to Cederstrom (1957), formation thicknesses
within the Chesapeake Group, subdivided on the basis of paleontological
evidence, are as follows:
-------
-33-
$£*• *^j£\ jl
SCALE
o
1000 0 1000 2000 3000
MOO
I
-M-
Fig. 9 Topographic Map
LOCATION
-------
-34-
4)
C
0)
oi
oo
o
i—i
o
JS
o.
M
0
o
0)
O
u
O
I
-------
-35-
Fig. 11 Fill Areas
-------
-36-
Yorktown Formation: 100 feet
Saint Marys Formation: 140 feet
Calvert Formation: 100 feet
A stratigraphic column with descriptions of geologic formations in
the region is displayed in Figure 12. The Yorktown Formation is
described as a poorly sorted, silty to clayey, fine sand with sporadic
zones of shell fragments based upon records from a well five miles
west of the refinery. Johnson (1972) further indicates that the
maximum measured thickness of the Yorktown Formation in the general
vicinity of the refinery is approximately 60 feet. A site map display-
ing surficial geologic conditions is shown in Figures 13 and 14.
The Yorktown Formation has been further subdivided by Johnson (1973)
into six facies based upon texture, composition, and bedding features
of each unit. Available stratigraphic information suggests that at
least one of these facies, the silty sand facies underlines the Norfolk
Formation, discomformably, near the facility as shown on Figure 15.
The silty sand facies is composed of bluish-grey to greenish-grey
fine sand with interbedded clayey silt. Major components of the sedi-
ment include quartz grains and whole or broken calcareous fossils. In
some areas, the shelly sand facies may be present. This sediment is
predominantly a grey calcareous and quartzose sand with clay rich zones
and whole or broken shells to a lesser degree. The Yorktown Formation
has been distiguished from the Norfolk formation on the basis of grain
size and mineral content. In general terms, the Yorktown is charac-
terized by cleaner fine sands whereas the Norfolk formation has a higher
clay and silt fraction. Another distinguishing feature of the Yorktown
formation is its glauconite content which imparts a greenish or brownish
green color to the sediment.
Deep exploratory borings penetrating the upper 100 feet of the
Yorktown Formation have failed to reveal the presence of a laterally
continuous confining unit. Materials encountered consisted generally
of silty fine sand with shell fragments. This determination is based
upon borings drilled at Virginia Power Company (VEPCO) Generating
Station located immediately west of the refinery. In addition, several
deep borings were drilled to a depth of 87 feet within the refinery in
conjunction with the Part B permit application process and yielded
similar materials as encountered in the VEPCO borings.
3. Hydrogeology
The Norfolk and Yorktown Formations comprise the water table
aquifer east of Suffolk Scarp and are considered to be one hydrogeo-
logic unit since no intervening confining unit or low permiability zone
has been identified during previous subsurface investigations. The
saturated thickness of this unit has not been determined.
-------
-37-
G«olO(ic age .nd description
Origin
Otatrl
Recent beach aanda
Oepoauad by ocean
Th,ck depoaita preaent oojy
Wawr-beuue; p-^»n'
May yield a little water
. f
- ---J^-^--":
-_TTL.~ * - — 1=J~=- ~=- ~-
S"Svfe=-?
- _ .
Tmcn 1
— - _~i_"*ir^- -- w
•^_ __ , ~
^y^l;.:
~r— ""•" ~ ~ ' -TV"**-
US?
~*T. " — 1 "~ *-*" — 4""
-Sf-^-— i.
SiOlg
SSfiSSSsB:
^5t0-
r
\
i
' " — — — ~ *
^
\
\
N
\
\
-V
j
-^
..-„... mrrent* ai Hampton ,j *t*l*oc*no
age Cooalata of mottled
clay. glaucoaU|acept at Hampton and
lower Warwick, where
t ic concealed by younger
U** at water t*
walla along ar near
tho Pall Una
Fig. 12 Stratigraphic Column
-------
-38-
Fig. 13 Geologic Map
-------
-39-
EXPLANATrON
tr
z
u
3
0
al
m
s
Alluvium; Marsh Sediment; and Sand
al, organic material; silt and clay: some sand;
m, peal, silt and clay with organic material; s,
beach and dune sand
Unconformity
1 «.-•••
m
Qn
s
sc
ss
Norfolk Formation
Facies-cs, fluvial and estuarine clayey sand; m
(in cross section only), peat, silt and clay with
organic material; s, beach and near-shore marine
sand; sc, marine sandy clay; and ss, marine $iUy
sand
Unconformity
Windsor Formation
Marine and estuarine silt and sand
ill
Z
Id
U
.0
(ft
u
0.
Unconformity
IT
U
Yorktown Formation
Marine tilt, sand, and coquina
ui
2
UI
•u
o
i
Fig. 14 Legend-Geologic Map
-------
-40-
f
I
SE»
-20
• LEVEL
-20
Schematic cross-section showing the relationship between the
lithofacies of the Yorktown Formation in the Yorktown and Poquoson
West quadrangles: Norfolk Formation, Qn. Windsor Formation, Qw.
Yorktown Formation, Ty—weathered zone, w; silty sand facies, sis; shelly
sand facies, shs; cross-bedded coquina facies, cbc; coquina facies, coq;
sandy silt facies, ssi; and Chama facies, chm.
** Approximate location - Amoco Yorktown Refinery
Fig. 15 Geologic Cross Section
-------
-41-
Estimates of hydraulic properties of the aquifer are based upon
literature review or extracted from field studies performed in surround-
ing areas of similar geology. No field studies involving aquifer
tests, i.e., pump and slug tests, have been performed; therefore,
estimates of permeability used in the calculation of linear flow veloc-
ities (hydraulic conductivity) may not be totally representative.
Vertical permeability testing has been conducted on split spoon
samples from the Norfolk Formation to a maximum depth of approximately
ten feet. Vertical permeabilities were determined using either permea-
meter or consolidation test methods under laboratory controlled condi-
tions. Calculated permeabilities ranged from 2.6 x 1CT5 centimeters
per second (cm/s) to 7.8 x 10~^ an/s, a difference of three orders of
magnitude. Estimates of permiability for the upper 90 feet of the
Yorktown Formation based upon soil sample discriptions and gradation
data range from 1.2 x 10~5 cm/s to 9.3 x 10~8 cm/s.
Potentiometric surface maps prepared by the facility (Figure 16)
indicate that ground water flow is predominantly to the east-northeast
towards the York River and Bull Creek. However, distinct anomolies in
the flow system are apparent, suggesting man made influences at work.
The present interpretation reflects three primary features: (1) a ground
water flow divide oriented NE-SW through the central portion of the
site, (2) a ground water mound in the vicinity of landfarm #11, and (3)
a ground water sink near the API separator and landfarm #12. The ground
water low appears to be related to the API separator which is allowing
shallow ground water infiltration. The ground water high over landfarm
#11 could result from enhanced recharge caused by decreased run-off and
evapotranspiration, a consequence of sparse vegetative cover.
With the completion of 76 stainless steel well clusters in the fall
of 1985, it became possible to more accurately determine horizontal and
vertical hydraulic gradients across the site. Seventeen of the well
clusters contain 4 wells completed at depths of 8, 18, 38, and 58 feet.
Over the 8 to 58 foot interval, vertical gradients ranged from 0.033
ft/ft downward to 0.023 ft/ft upward. This apparent reversal in gradient
direction is explained by seasonal and temporal variations in ground
water recharge. Horizontal ground water velocities ranging from 4 x 10~6
cm/s to 5 x 10~7 cm/s have been calculated, using 1 x 10~4 cm/s as a
representative permeability value derived from previous studies and not
through direct testing.
Based upon vertical gradient determinations, vertical seepage
velocities were calculated. Assuming a vertical permeability of
5 x 10~5 cm/s for the interval from 8 to 38 feet, hydraulic conductiv-
ities ranging from 2.6 x 10~5 cm/s to 1 x 10~7 cm/s were generated.
For the interval from 38 to 58 feet, a vertical permeability of 5 x 10~3
cm/s was assigned, producing values ranging for 1.4 x 10~4 cm/s to
4 x 20~4 cm/s. The prevailing assumption is that the shallow subsurface
consists of a lower permeability layer, the Norfolk Formation, overlying
a higher permeability layer, the Yorktown Formation.
-------
-------
-43-
As a means of determining whether cyclical tidal changes in the
York River were causing short term fluctuations in ground water levels,
continuous strip chart recorders were placed in wells B-24 and CP 1-8.
Both wells are located in the northeast corner of the site, B-24 being
600 feet and CP 1-8, 1000 feet from the banks of the York River. Review
of strip chart recordings indicate fluctuations in water levels of 0.05
inches or less over the 24 hour observation period. Although not conclu-
sive, results suggest that tidal stage has little impact on the rise
and fall of the potentiometric surface in the vicinity of active and
inactive landfarms.
E. GROUND WATER M3SETQRING SYSTEMS
1. Monitoring Requirements/Interim Status
Amoco installed a series of observation wells in 1980 and 1981, to
monitor the ground water conditions across the entire facility. RCRA
monitoring wells were installed in 1981 (wells C-l-16) and 1984 (wells
C-17-20), to monitor ground water conditions at the three landfarms.
To further assess ground water conditions beneath the facility, Amoco
installed a series of 21 well clusters in 1985. A total of 76 additional
wells were included in the 21 clusters. Seventeen clusters contain 4
wells (CQ series) while 4 clusters contain 2 wells (CP series).
RCRA background ground water sampling began in 1982. Initial
background concentrations were obtained during four quarterly sampling
episodes. Amoco began semi-annual sampling episodes in January 1983,
for the required indicator parameters. Based upon the results of the
Student's t-test, a statistically significant difference between back-
ground samples taken during 1982, and those samples taken in 1983, was
detected. On August 31, 1983, Amoco sent to the VDWM a Notice of
Increase in Indicator Parameters. All the landfarms showed statistical
increases in TOC and TOX. Amoco submitted a Ground Water Quality
Assessment Plan to the VDWM on September 14, 1983. Amoco installed
additional upgradient wells in 1984 (wells C-17, 18, 19), because it
was believed that the existing upgradient wells were located too close
to the landfarms. The Ground Water Quality Assessment Plan submitted
by Amoco was not sufficient to characterize the rate and extent of
ground water contamination. Discussions took place during 1984 between
Amoco, EPA, and VDWM to clarify the direction of Amoco's assessment
monitoring program.
Based upon sampling in 1984, lead was the only inorganic constituent
found to have entered the ground water. Benzene, toluene, and methethyl-
ketone (MEK) were detected in two monitoring wells. EPA and VDWM
believed that overloading of landfarms #10 and #12 with waste material
contributed to the leaching of constituents into the ground water.
In order to better define the contaminant plume and to better
define the impact the landfarms may be having on the ground water, Amoco
-------
-44-
installed the previously mentioned 76 cluster wells. Additional sampling
detected elevated levels of chromium, nitrate, and several organic
compounds at landfarms #10 and #12.
2. Current iVbnitoring Well Network
The current RCRA monitoring well network consists of 95 wells
specifically installed as RCRA monitoring wells for landfarms #10, #11,
and #12. Figure 17 shows the location of the RCRA monitoring wells.
As previously mentioned, no well construetdon or boring logs exist for
the C series wells or the CQ and CP cluster wells. Observation wells
B1-B24 have been sampled in the past but the reliability of the data
remains suspect due to well construction and maintenance practices.
The observation wells, while not normally included in the RCRA quarterly
monitoring program, are used for water level measurement.
3. Well Construction Specifications
A summary of well construction details for the 16 RCRA wells
sampled during the inspection are shown in Table 2. The following
description of well design and construction specifications is based
upon information provided by the facility and was, in part, confirmed
by EPA and VDWM observations.
The original RCRA system, wells C-l through C-16, were installed
in 1981, under the supervision of Dames and Mx>re. The system was
supplemented by the installation of wells C-16 through C-20 in 1984.
The C series wells were installed using hollow stem auger methods
producing a borehole approximately 6 inches in diameter. Inside the
borehole, 4 inch inside diameter (ID), schedule 40 PVC casing was placed.
The well intake consists of a 5 foot section of machine slotted PVC
casing with a slot size of 0.010 inches and is situated at the bottom
of the borehole.
It is unknown whether the screen and solid casing are connected
with adhesives, retainer screws, or threaded couplings. Filter pack
material consisted of clean quartz sand of unknown size. The length of
the filter pack is not specified; however, it is assumed to be consistent
with screen length. The pack was installed using a shovel and not by
tremmie pipe. The annular space above the filter pack was filled with
bentonite clay pellets reportedly producing a 6 inch seal. The annular
space above the seal was filled with concrete cement to the surface.
Wells are fitted with protective covers (caps); however, they are not
secured with locking devices. Typical well cluster construction specifi-
cations are presented in Figure 18.
The purpose of this design was to monitor the upper 5 feet of the
saturated zone within surficial Norfolk formation sediments considered
highly vulnerable to contaminant releases from landfarms #10, #11, and
#12. The physical and chemical properties of potential contaminants
and to their mobility characteristics were not given ample consideration
in formulating the well design.
-------
-45-
-------
en
a
4J
u
w
in
~~"
*
^-*
m
****
r\T
^
^—
Levation
' — I
a
I
I
3
CO
4-*
O
S1
w
rH
•H
J
?
to
S
*"O
35
1
^Q
Ml
^
Qj
•H
O
rH
rH
g
a
Q
rH
j>
M
JJ
C
M
T3
§
S-l
0
S
r~H
03
D
5
u
•H
to
o
M
1— 1
-46-
OrH^inooaoinr^inSoii^pooiH
CTiCT\OCTlCTi %
inmm m m in mmm
^^* -^* *^^ >^ ~_- ~_~ ^^ v_^ >^, i[ [
*-* 'sroooooooooooooo o
o incN
^-w
"S
c
o
g
I
p
5
Q
^
Q
^
C
•r-l
T3
1
§
to
rH
i
•
N
£
•2
•rH
g
03
jj
OJ
C
"S
c
§
u
to
5
Is
Sz
4-1
o
g
a
S1
T3
0 o
-U 3
flii
00
C CT 03
•H C > • 3 • 0
T3 T3 0 0 CO <4-l
35 c -u 35 03 >t
r-l CU C 4J CO 03
3 Qj-rH CO T3 in g
co 0 35
0 0 CO 03 C O
0 O 03 -H
rH -H J3
C -g co 3
^ooooooooooooooo
innvocNf— irH«aif—(
-------
-47-
M
00
tfl
C
O
o
3
CO
a
o
CJ
tw
11
4-1
CO
3
-------
-48-
In November 1985, VDWM requested the vertical and Horizontal
extent of the plume (ground water contamination) be characterized. In
response, Amoco installed 76 stainless steel monitoring wells. The
wells were installed without the prior knowledge or consent of EPA and
VDWM. A total of 17 stainless steel well clusters, identified as the
CQ series, consisting of 4 wells each and 4 sets of well pairs, the CP
series, were constructed. In total, 76 monitoring wells were constructed.
The CQ series were completed at depths of 8, 18, 38, and 58 feet and
the CP series at depths of 8 and 18 feet. The wells were installed
using hollow stem auger methods. Either 8w 11 inch outside diameter
(OD) augers were used depending upon the desired depth of the borehole.
As a general rule, 11 inch OD augers were used on the 38 foot and 58
foot wells to improve filter pack placement around the screens.
Each well is constructed of 2 inch ID blank stainless steel casing
with a 5 foot wire wrapped screen having an opening size of 0.010 inches.
Sections of casing and screen have threaded connections; therefore, no
adhesives were applied which could alter ground water quality. The
filter pack was placed in the annular space to a height approximately 2
feet above the top of the screen. The composition and size of filter
pack materials is not specified in available documentation. Above the
filter pack, a 2 foot thick bentonite clay seal was placed. The clay
was introduced by dropping pellets down the annular space followed by
the addition of clean water causing hydration. Each well was then
backfilled with auger returns (formation materials) to within 3 feet of
the ground surface. The remaining 3 feet was filled with concrete
cement.
At the surface, a concrete apron was formed, thereby preventing
the ponding of surface water around the casing. With respect to the 8
foot wells, a 3 foot screen was installed and the filter pack extended
1 foot above the screen. The annular space was filled with concrete
cement to the surface.
The rationale for selection of filter pack and screen slot size is
not presented by Amoco. Selection may not have been made on the basis
of grain size analysis. Neither stainless steel nor PVC well construc-
tion materials were steam cleaned prior to installation. Cluster wells
were developed using air surging techniques and surveyed for vertical
and horizontal control.
Reasons for selecting particular well depths within each cluster
are not given by Amoco; however, reference is made to a modeling study
contained within the Closure Plan for landfarm #11, which supposedly
demonstrates that contamination is not expected to occur below 50 feet.
The study employed the Finite Element Difference (FEDAR) Model. This
study is considered suspect given the number of unsupported assumptions
regarding stratigraphic setting which may run counter to actual field
conditions. Two cases are examined, one in which a 1 foot thick clay
layer is present near the base of the Norfolk formation and the other
-------
-49-
where the clay layer is absent. Under both scenarios, contaminant
migration was not expected below 50 feet; however, in the absence of
the clay layer, contaminant travel times to the Yorktown formation
were significantly shorter.
4. Site Characterization
As of the January 1987 Task Force Inspection, site characteriza-
tion has not been completed to the extent necessary to describe, in
detail, geologic and hydrogeologic conditions. A general understanding
of hydrogeologic conditions exists; however, it is not detailed enough
to ensure proper placement of monitoring wells from the standpoint of
aerial distribution and vertical spacing.
CQ ad CP series monitoring wells do not extend beyond a depth of
58 feet. This depth was selected on the basis of modeled projections
as to the maximum downward migration of contaminants. This model
relies on a overly simplified representation of geologic conditions,
that of a two layer system, without the necessary supporting evidence
derived from direct subsurface investigation. Borings have not been
drilled to depths sufficient to determine the full vertical extent of
the uppermost aquifer and any hydraulically interconnected aquifers.
Investigations performed to date, have failed to provide any evidence
of a laterally continuous confining unit or low permeability zone
representing the base of the uppermost aquifer. This implies that
potential pathways of contaminant migration have not been fully explored.
Drilling logs and construction diagrams for the CQ and CP series
wells were not prepared, further limiting available information regard-
ing subsurface geologic conditions.
A representative determination of aquifer properties has not been
made and in particular any evidence of anisotropic and nonhomogeneous
conditions. As a consequence, the vertical and horizontal hydraulic
conductivity distribution has not been adequatly defined for the Norfolk
and Yorktown Formations.
Boring logs were prepared for C series wells and split spoon
sampling of formation materials was conducted at the time of installa-
tion. However, analysis of grain size and other physical properties
was not performed.
The rationale for selection of individual well depths within each
cluster, screen length, and placement as it relates to hydrogeologic
conditions is not fully explained. The results of the modeling effort
provide the only link as it formed the basis for selection of maximum
well depth.
The vertical component of ground water flow as a whole, or as it
relates to each screened interval, has not been formally established.
-------
-50-
Although, Ground Water Quality Assessment Program reports include the
calculation of vertical gradients between screened intervals, minimal
interpretation is provided in a facility wide sense. In the derivation
of vertical flow velocities, assumed values for vertical permeability
estimated from gradation testing and visual examination/classification,
were used and not from field or laboratory mesurements.
The ability of select upgradient wells to produce results repre-
sentative of background water quality conditions not affected by the
facility is in question given analytical data from well clusters CQ 10,
CQ 6, and CQ 2 which suggests impaired quality.
5. Sanpling Audit
Field technicians from J.R. Reed of Newport News, Virginia,
conducted ground water sampling on a quarterly basis. Their sampling
procedures were demonstrated on well CQ 1-18 on January 27, 1987.
Upon opening the well cap, field technicians failed to monitor for
the presence of organic vapors. Water level measurements are routinely
taken from a surveyed reference point to determine the height of the
water column in the well from which purge volumes are calculated.
During this process, the dedicated tubing array remained in the well
which could result in displacement of the water column as reflected by
higher than expected water levels. This phenomenon has not been examined
in any detail. An electric water level probe and steel tape are used
to measure depth to water. The sounder consists of an electrode con-
nected to a two-wire cable which is slowly lowered to the water surface.
Upon making contact, the circuit is completed and a buzzer is activated.
The cable is marked in 5 foot increments necessitating the use of a
steel tape to record measurements to the nearest 0.01 feet. Measurements
were taken from the top of the well casing. Total well depth measure-
ments were not taken during the demonstration. The methodology used by
the sampling team to detect light phase inmiscible layers was not
adequate. As the cable was retrieved from the well, no decontamination
was performed. This deviation from accepted procedure introduces
another outside variable which may adversely impact sample integrity.
During this procedure, field technicians wore disposable plastic gloves,
which are supposedly replaced after sampling of each well is completed.
Purging is acconplished by means of a peristaltic pump connected
to a pair of tubes dedicated to each well. Two sets of tubing, one of
smaller and the other of larger diameter, are available depending upon
well depth. Larger tubing is used exclusively for purging deeper wells
(38 feet and 58 feet) and smaller tubing for purging and sampling
shallow wells and sampling deeper wells. Amoco's Sampling and Analysis
Plan specifies the removal of at least one well casing volume of water
or, at a minimum, evacuation to dryness once allowing for full or
partial recovery prior to sampling.
-------
-51-
The peristaltic pump rests on a metal stand and is powered by means
of a portable gasoline generator. A section of rigid polyethylene
tubing remains in the well bore while a section of flexible tygon
tubing is attached to it above the well head and is wrapped around the
rotor housing extending out the other side (Figure 19). Upon activation
of the pump, the rotor begins to turn and within a brief period of time
achieves a controlled speed. During this process, the tubing is squeezed
between a series of ball bearings (rollers) which revolve in a circular
pattern around the rotor creating vacuum pressure, thus displacing the
water from the well. It was observed thaf ^a constant flow of water was
not maintained, rather the discharge came in pulses or surges. This
surging action can result in excessive agitation and rapid pressure
fluctuations causing the degassing of volatile organic compounds.
Purge water was discharged directly to the ground with no plastic
sheeting being placed around the well casing to prevent ponding of
water. Wells such as CQ 1-18, which have an established record of
rapid recovery, are normally sampled immediately after purging. For
purposes of this demonstration, one pump/tubing array was utilized;
however, during routine sampling, two pumps are operated simultaneously
per well cluster thereby reducing purge times. The smaller capacity
pump already described, is rated at one gpm and a larger capacity pump
operates at 4 gpm. Both are powered by one-third HP motors. It is
common practice to purge a deeper well utilizing the higher capacity
pump in association with larger diameter tubing while simultaneously,
purging a shallow well using the smaller capacity pump. According to
the VDWM, during routine sampling events, only the upper portion of the
well column is actually purged, a practice which would be ineffective
for wells with high recharge rates.
Approximate purge volumes were calculated by first determining the
height of the standing water column in the casing by subtracting the
depth to water from the total well depth. The volume of water in the
casing is then calculated by multiplying the water column height by a
gallon per foot conversion factor. This volume is then multiplied by
three to produce the total purge volume required. During purging,
dedicated tubing is moved up and down to create a surging action in an
attempt to remove sediment accumulated at the bottom of the well. In
effect, this procedure represents a form of secondary development.
According to Amoco, another approach sometimes taken is to place the
tubing on the bottom of the well to remove entrapped sediment which is
then slowly elevated to the desired height. Documentation verifying
the effectiveness of this procedure was unavailable.
For purposes of the demonstration, it was assumed that samples
were being collected for laboratory analysis. Samples were collected
in order of the parameters volatilization sensitivity. In other words,
parameters having a greater sensitivity were sampled first followed by
those parameters less prone to such affects. Sampling order by parameter
is listed as follows:
-------
-52-
.\
Fig. 19 Peristaltic Pump in Operation
-------
-53-
volatile organic compounds (VOC's)
total organic carbon (TOG)
total organic halogen (TDK)
metals (total)
metals (dissolved)
nitrates
inorganics (pH, specific conductance)
Wells are sampled using the same peristaltic pump arrangement
described earlier. Prior to sample collection, field technicians put on
disposable rubber gloves. It is a routine practice to elevate tubing
to an approximate height of 3 feet above well bottom in preparation for
sampling. Reasons for collecting samples at this particular elevation
are unclear. Samples were first collected for VOC analysis. Duplicate
50 milliliter (ml.) vials were filled in a manner to prevent the forma-
tion of air bubbles. The discharge rate when filling the vials is
manually controlled by pinching the tubing thus creating a constricted
opening. Since flow cannot be controlled in any systematic manner, it
is possible that rates exceed the 100 ml/lnin. maximum recommended in
the Ground Water Technical Enforcement Guidance Document (TEGD); thereby,
creating a greater potential for degassing of VDC's. Samples for TOC
and TOX analysis were collected next in amber glass bottles and fixed
with sulfuric acid. This conflicts with the TEGD (page ill, Table 4-1)
entitled Sampling and Preservation Procedures for Detection Monitoring
which recommends the use of sodium sulfite as a preservative.
Sample containers remained in coolers until they were filled and
were then immediately returned once labels were completed. Due to
exceedingly cold weather, bottles were placed in plastic bags and then
immediately placed in coolers without ice. Samples for metals analysis
were field filtered through a large mesh prefilter followed by a .45
micron (geotech) filter. The filtering apparatus was placed on plastic
bags lying on the surface of the ground. Samples for metals analysis
were split in two, i.e., two separate containers were filled, one for
total and the other for dissolved metals. Preservation was accomplished
in the field by the addition of nitric acid. Field analysis for indi-
cator parameters (pH, specific conductance and temperature) was not
conducted given the close proximity of J.R. Reed Analytical Laboratory
to Amoco. Standard practice has been to perform indicator parameter
analysis at the laboratory shortly after sample collection. Due to the
chemical instability of these parameters, delaying analysis until samples
arrive at the lab may produce results not representative of in-situ
conditions. On this occasion, field conductivity mesurements alone
were made.
The J.R. Reed Analytical Laboratory performs analysis for major
inorganic parameters including total metals, dissolved metals, TOC,
TOX, and specific conductance. Organic constituent analysis is performed
at Amoco's laboratory in Naperville, Illinois. Samples are shipped by
overnight courier and arrive within 24 hours.
-------
-54-
A field logbook which documents sampling procedures was not in
use. Rather, information was recorded on standard forms provided by
J.R. Reed and retained at their office. Copies of these forms were not
available at the facility. Evaluation of a select number of forms
revealed discrepancies running the gamut from minor to major. Forms
failed to provide specific listings for the following basic information.
Several examples are indicated as follows:
- presence of immissible layers
- sample withdrawl procedures
- Date and time of collection
- Field observations (Equipment malfunctions, unusual well
recharge rates etc.)
- Parameters requested
- Type of sample containers and sample ID numbers
The file containing field data sheets was not well organized
reflecting the need to properly maintain files and keep information
current. Amoco recently instituted a policy of field and trip blank
collection concurrent with routine quarterly sampling events. Prior
to this inspection, field and trip blanks were not collected. Equip-
ment blanks were collected in the past but on an infrequent basis and
for organic parameters only. Records indicating the exact frequency
were not maintained.
Laboratory grade water provided by the J.R. Reed Laboratory was
used in preparing blanks. Documentation that it meets organic free
water standards was not provided. It is recommended that QA/QC samples
be collected during each sampling interval to ensure that laboratory
analysis is accurate and precise and that cross contamination does not
occur.
In comparing actual field procedures to those described in the
Sampling and Analysis Plan extracted from the Part B permit application,
it is apparent that the plan had not been updated since the installation
of CQ and CP series well clusters. The plan fails to discuss sampling
procedures for these wells. Provisions should be made to update the
plan on an as needed basis concurrent with changes in the ground water
monitoring program.
A potential drawback in the sampling program is the use of peris-
taltic punps which are known to subject samples to negative pressures
which can effect the concentrations of dissolved gases, cne advantage
of the system is that the sample does not come in direct contact with
the sampling mechanism, only the tubing. However, since the discharge
line is constructed of tygon and the influent line of polyethylene,
both known for their susceptibility to adsorb/desorb organic species
under certain conditions, sample chemistry may be altered during the
collection process.
-------
-55-
Chain of custody records were included with each sample shipment;
however, the form currently in use fails to document such necessary
information as parameters requested, sample location, sample type
(ground water/imnissible layer) and signature of collector. For the
sake of completeness and to fully document sample possession from the
time of collection to analysis, such information is essential.
6. Sampling and Analysis Plan and Field Procedures
The Sampling and Analysis Plan presented in the original Part B
permit application dated July 1983 and subsequent revisions including
documentation within the October 1985 Ground Water Quality Assessment
Plan were reviewed as part of the Task Force Inspection. The plans
were critiqued, observations of their implementation were made in the
field, and facility representatives were interviewed regarding field
procedures and rationales.
The plan was not written to guide field work, but as documentation
in support of procedures already in place. The plan needs to be a
stand-alone document with enough detail such that the qualified sampler
can apply this document at the facility and assure consistency in sample
collection. In particular, the plans failed to provide any discussion
of sample preservation and handling procedures and field and laboratory
CSVQC procedures.
The sampling plan does not adequately describe actual field proce-
dures implemented and observed during the January 1987 inspection.
Consequently, the ability to determine whether field procedures are
consistent with that described in the plan is greatly hampered. It is
unclear whether practices observed in the field were the norm or had
been instituted recently. Both the plan and field procedures in use
have several notable deficiences which must be corrected before analy-
tical data can be considered unbiased and representative of the upper-
most aquifer. Suspect or questionable sampling procedures can introduce
contaminants not present in the ground water (false positives) or
potentially mask parameters actually present in the ground water (false
negatives). Specific problems identified during the inspection are
indicated as follows:
(a) The use of peristaltic pump(s) for sampling can result
in violatilization of certain constituents due to rapid
pressure fluctuations (pressurization followed by depres-
surization). This can produce an unacceptable degree of
variability in the analysis of pH, specific conductance,
heavy metals and volatile organic samples.
(b) Intake and discharge tubing is constructed of polyethylene
and tygon, respectively. These materials can alter
sample chemistry during the collection process through
adsorption or desorption of constituents from the ground
water or the introduction of constituents not originally
present.
-------
-56-
(c) Purge water and excess water collected during sampling
is discharged directly on the ground in the vicinity of
the monitoring well(s).
(d) Given the proximity of the J.R. Reed Laboratory to the
facility, field measurements for pH, specific conductance
and temperature are not performed immediately after
sample collection. If samples are not promptly trans-
ported to the laboratory changes in in-situ chemistry can
occur given the relative instability of these parameters.
(e) The practice of equipment blank collection and transport
of trip blanks to the field was instituted shortly before
the Task Force inspection. According to facility repre-
sentatives, equipment blanks were collected in the past
but on an infrequent basis. The collection of field and
trip blanks is necessary to determine the reliability
and validity of field and laboratory data.
(f) Documentation regarding QA/QC procedures was not avail-
able at the time of inspection, Amoco contends that
these procedures would be described in the Sampling and
Analysis Plan undergoing revision at the time of the
Task Force inspection.
(g) A field logbook containing essential details of field
activities does not exist. However, J.R. Reed field
technicians prepare what are termed "field data sheets"
each time monitoring wells are sampled. Data sheets are
stored at the J. R. Reed Analytical Laboratory and not at
the facility. The sheets examined were incomplete in
that certain information such as the presence of immissible
layers, well evacuation and collection procedures, data
and time of collection, types of sample containers, and
parameters requested was not addressed.
(h) The measurement of total well depth and sampling of
immissible layers, if detected, is not conducted and is
not discussed in the current Sampling and Analysis Plan.
7. Facility Water Quality Analysis and Data Quality Assessment
An audit of the Amoco Laboratory and the J.R. Reed Laboratory was
conducted on January 30, and February 4 and 5, 1987, respectively.
Their purpose was to assess the ability of each laboratory to analyze
ground water samples.
The major objectives of the audit were:
(a) to assess Amoco's and J.R. Reed's capability to conduct
ground water analysis and their general ability to produce
data of acceptable quality, and;
-------
-57-
(b) to investigate and assess the quality of actual ground
water data generated specifically for Amoco.
The J.R. Reed Laboratory performs analysis of indicator parameters
on a routine basis and only occasionally inorganic parameters. Since
J.R. Reed performs the actual sampling, they do all the in-situ measure-
ments such as temperature, pH, immiscible layer measurements, and
specific conductance. The Amoco Laboratory performs the majority of
the inorganic analysis and all the organic analysis on ground water
samples.
With respect to the inspection of the J.R. Reed Laboratory, the
most critical areas of concern are data validation, corrective action
when QC limits are exceeded or calibration data indicates an out-of-
control situation, and instrument detection limit determinations. The
inspection of Amoco's Laboratory revealed that QC and SOP'S require
improvement. Some analytical methods need minor modifications to comply
with EPA. requirements and will be so modified upon receipt of SW-846,
3rd edition. The enhancement of QC procedures includes additional
matrix spike and duplicate analysis.
8. Interim Status Ground Water Monitoring Data
Amoco has performed an extensive ground water assessment program
however, the program remains unfocused.
The first three years of Amoco's ground water assessment program
(1984 through 1986) provide indications of ground water contamination
at the facility. A lead contaminant plume was tentatively identified
in 1984 downgradient of landfarms #10 and #11. Additional sampling
performed later in 1984 failed to verify the lead contamination; however,
concentrations of zinc and several aromatic hydrocarbons above background
appeared downgradient of landfarms #10 and #12. The lead contamination
did not reappear in future sampling episodes. Arsenic and selenium
appeared in quarterly results during 1986. There has not been a histor-
ical trend established for either parameter. Because the arsenic and
selenium results appeared slightly above the analytical detection limits,
Amoco believes that the detection of these two inorganic constituents
is an anomaly.
Selenium and mercury were reported, during 1986, to have been found
in. three wells above the Maximum Concentration Limits (MCLs). Amoco
concluded that there is no known basis for the presence of the concen-
trations. Elevated levels of chromium were reported in eight monitoring
wells around landfarms #10, #11, and #12. Chromium is present in some
refinery waste streams. Amoco will attempt in future sampling episodes
to determine if a trend is emerging. However, subsequent sampling
results did not verify the presence of chromium.
Many of the results for inorganic constituents from the assessment
monitoring program have been ambiguous. Constituents such as lead and
-------
-58-
chromium appear in one sampling episode but do not reappear in subse-
quent sanpling. Parameters such as arsenic, selenium and mercury either
appear just above the analytical detection limits where the analytical
reliability is suspect or the inorganics detected have no history of
being used on the site and could, therefore, be considered anoitiolous.
By the end of four quarters of sanpling in 1986, Amoco had reached
the following conlusions concerning the groundwater at the facility:
(a) Ground water elevation contours (and flow patterns) were
consistent with previous analysis.
(b) The calculated horizontal ground water flow velocities
ranged from 1.0 X 1(T6 cm/sec, to 5.0 X 10~6 cm/sec
(0.003 to 0.0011 ft/day).
(c) The calculated vertical flow velocities ranged from
2.7 X KT6 on/sec to 9.0 X ~4 on/sec (0.008 to
2.6 ft/day). (Figures from fourth quarter sampling
1986).
(d) The extent of the organic hazardous constituents, both
vertically and horizontally, coincides with previous
reports (see Figure 20).
(e) In evaluating data for 1986, there is only one inorganic
plume (chromium) which has remained constant in extent,
both vertically and horizontally (see Figure 21). In
the third and fourth quarters of 1986, new trends were
developing which indicated additional inorganic plumes
extending 18 feet below land surface.
(f) The horizontal rate of migration is negligibly small with
the possible exception of induced ground water flow into
the API Separator.
(g) The vertical rate of migration for the confirmed organic
and inorganic constituents is negligibly small.
EPA and VOrM are not in full concurrence with the above conclusions.
F. TASK FORCE EftIA CXULBCTICN/PESULTS
1. Sample Collection Methods
To determine whether monitoring wells at the Amoco Refinery had
detected any contamination, selected wells within the facility were
sampled. Between January 27-30, 1987, the Task Force sampling contractor,
Versar, Inc., sampled 16 monitoring wells. The locations of the monitor-
-------
-59-
I!
p 5 * * * J -
dSS!s:"
2 5 5 5 2 3 s r
^ o f 5 ?! 71!
C u"3aC5i;
J oaoo
-------
-------
-Sl-
ing wells are shown in Figure 22. The list of monitoring wells sanpled
during the evaluation including dates of purging and sampling is provided
in Table 3.
The monitoring wells were selected on a priority basis during a
preliminary meeting of State and Task Force personnel. Selection was
made based on well location, construction, screen interval, and previous
water quality data to yield the most representative array of sampling
points. All wells selected were RCRA monitoring wells with the excep-
tion of well B-ll, chosen on the basis of 'i'ts proximity to the York
River and because it was the furthest downgradient well from the land-
farms. Not all wells selected were sampled for the complete suite of
analytical perameters due to difficulties in extracting the required
volumes of water and/or equipment malfunctions. For example, wells CQ
17-8 and CQ 17-18 were not sampled for dioxin analysis. The list of
chemical parameters for analysis at Amoco was based upon a modified
list of Appendix IX parameters specified under the Agency's Contract
Laboratory Program (CLP). The purpose of the revised list was to provide
a broad spectrum of analytes to detect and identify chemical constituents
most likely to be found in the groundwater as a result of releases. A
complete list of parameters by category is provided in Appendix C.
The organic samples were analyzed by FJ\BI labs in Camarillo,
California. The inorganic samples were analyzed by Centec Labs in
Salem, Virginia. Dioxin samples were analyzed by Compu-Chem Labs in
Research Triangle Park, North Carolina.
Upon opening the well cap, organic vapors were monitored using a
properly calibrated Photovactip. A positive reading above background
was measured for well CQ 17-8 (l-5ppm). Depth to water table measure-
ments were made using an electronic water level probe prior to well
evacuation. For wells not being sampled, dedicated tubing was left in
place prior to measurement. For the 16 wells scheduled for sampling,
the field team decided to remove tubing prior to measurment to facilitate
detection of immissible layers utilizing an interface probe.
Wells were purged using either a peristaltic pump provided by Amoco,
or teflon bailer. As shown in Table 4, designated wells were purged
using peristaltic pumps on the first day of the Task Force evaluation
after which remaining wells were purged using bailers due to pump
malfunction caused by the extremely cold weather.
Every attempt was made to remove three well casing volumes of
water prior to sampling. In two instances, wells CQ 14-18 and CQ 17-18
went dry prior to removal of three casing volumes. In the case of well
CQ 14-18, half the required parameters were sampled on January 29, and
the remaining half the next day allowing for sufficient recovery. The
process differed with respect to well CQ 17-18 where dryness ocurred
after the removal of only three gallons of water. It was determined
that dioxin, a very volume intensive parameter, would be eliminated so
that sampling for less volume sensitive parameters could be completed
on a expedited basis.
-------
-62-
Landfarm #10
(2)(3)*CQ-11
*CQ-12
CQ-14 *(2)(3)
CQ-13 *(2)(3)
C010
* (1)
T^nrifarm #17
* B-ll (2)(3)
*CQ-17
CQ-15
Landfarm #11
CO-6
* (1)
Not to scale
*CQ-9 (1)
(1) Volatile Data-Quantitative
(2) Volatile Date-Unreliable
(3) Semivolatile Data
(semi quantitative
excess holding time)
Fig. 22 Monitoring Well Locations
-------
-63-
o
o
•a
M
0)
CCCCCCCCCCC'-CCGC
•H -H -H -H -H -H -H -H -H -cH -iH CJ -H -H -H -iH
CN rH CN rH
CN
cn
o
CO
m
(0
Vj
o
<
o
3
a
I
CNCNCNCNCNCNCNCNrOCNfOCNCNCNCNCN
00
m r~ 00 r- r^ CN rH
CN rH CN
CN
TD
S1
00
m
vo
CN
O
o
§
o
•a
CT
CN rH CN
in
§
O
•a
M
ft
a-
CN i— 1 CN
U}
CN
-o
en M
jj CD
C 4J
o o m < o o o
rH CN
2
a
(0
PI g>
1— ( <7l VO in C— | rHCN
CNCNCNCNCNCNCNCNrOCNnCNCNCNCNCN
Ps o
iH <3 rH
CL, co a
O (I) W
'SS 8
CO (0
(B CTi • 4H
J5 CN W U
--•. rH 0) «
a
— o g o
, SH rH
s rj 3 *ri
3 73 (i O
Q)
CO CO CO CO CO CO CO CO CO
cNCN^ck^rn'c^c^^-iin'r^itrLr^
,O 4J CT -
> C C 0)
0) -H CT
D) 5 C rJ
•H rH fQ Q4
W rH E _
< OQ CJ
-------
-64-
Table 4. Monitoring Well-Purge/Sample Equipment
Date
1/27/87
1/28/87
1/29/87
ft
It
1/30/87
Monitoring Well
CQ 2-8
CO 2-58
CO 6-38
CQ 9-58
CO 10-11
CQ 17-8
CQ 17-18
CQ 17-38
CQ 17-58
CQ 11-8
CQ 12-8
CQ 12-38
CQ 13-8
CQ 14-8
CQ 15-8
Purge Equipment
Peristaltic Pump
it
ii
11
'Bailer
*
Bailer
M
Peristaltic Pump
Bailer
Sample Equipment
Bailer
Peristaltic Pump
Bailer
* not purged - grab samples collected
-------
-65-
Well OQ 17-8 was unique in that a non aqueous oil phase was detected
on the surface of the water table. Benzene and total xylenes were
detected at 140 ppb and 650 ppb, respectively. The field team decided
not to purge this well, rather grab samples were collected using a
peristaltic pump so as not to contaminate field equipment. The oil
layer was so thin that collection of discrete samples was not possible.
Purge water was collected in plastic garbage cans and, when full, trans-
ported by truck to the API separator where the contents were disposed
by prior agreement with Amoco. Wells were sampled as soon after purging
as possible. Sampling was initiated within 20 minutes after purging
with the exception of well CQ 17-18 where the lag time was 134 minutes
due to very slow recovery. Sample parameters were collected in priority
order starting with volatile organic compounds in the event an insuffi-
cient volume of water remained in some slow recovery wells. The list
of sample aliquots collected and their containers is shown in Table 5.
Teflon bailers 3/4 inches in diameter and 36 inches in length with
double check valves that are bottom filling and emptying were used.
Bailers were raised and lowered by teflon coated, wire connected to a
pulley system. Sample containers were placed in plastic buckets lined
with garbage bags to contain any spillage during sample collection.
Bailers were drained by a bottom valve plug which has the distinct
advantage of controlling the rate of discharge. Dedicated bailers were
assigned to each well sampled. A complete suite of QA/QC samples were
collected including trip blanks, equipment blanks, and duplicates.
Samples were split between the Task Force and J.R. Reed sampling
teams for parameters including blanks. The J.R. Reed crew promptly
placed sample containers in coolers in preparation for shipment to the
appropriate analytical lab.
Samples were preserved and prepared for shipment as soon as poss-
ible after collection. Chain of custody records indicate two large
shipments, in coolers, on January 28 and 30, by Federal Express to the
respective labs. Complete details on sampling procedures are provided
in the Task Force Project Plan and field records.
2. Task Force Sampling Results
During the Task Force investigation, 21 field samples were collected.
Included in the 21 samples was a field blank, an equipment blank, a trip
blank, and a pair of duplicate samples. Sample M3V770 (Well CQ 17-8)
was designated by the sampling team as a medium concentration ground
water sample. All samples were analyzed for 194 organic and inorganic
parameters through the EPA CLP with the exception of samples M3\770
(Well CQ 178), M3V771 (Well CQ 17-18), and M3A783 (Well CQ 14-18) which
were not analyzed for dioxins and dibenzofurans due to difficulty in
obtaining the required sample volume. Data Summary charts displaying
select organic and inorganic data is shown in Tables 6 through 11
inclusive which follows this section.
-------
-66-
Table 5. Preferred Order of Sample Collection. Bottle Type, and
Preservative List
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
14.
15.
Parameter
Volatile Organic Analysis (VOA)
Purgable Organic Carton (POC)
Purgable Organic Halogens (POH)
Extractable Organics
Pest ic ide/Herbic ide*
Dioxin*
Total Metals
Dissolved Metals
Total Organic Carbon (TOC)
Total Organic Halogens (TOH)
Phenols
Cyanide
Anions-sul fate
Nitrate/Armenia
Sulfides
Bottle
2-,
2-
2-
4-
2-
2-
1-
1-
1-
1-
1-
1-
1-
1-
1-
40
40
40
1
1
1
1
1
4
1
1
1
1
1
4
ml
ml
ml
qt
qt
qt
qt
qt
oz
qt
qt
qt
qt
qt
oz
VGA vials
VOA vials
VOA vials
amber glass
amber glass
amber glass
plastic
plastic
glass
amber glass
amber glass
plastic
plastic
plastic
glass
Preservative
HN03
HNO3
H2S04
H2S04
NaOH
H?S04
(pH<2)
(pH<2)
(pH<2)
(pH<2)
(pH>12)
(pH<2)
Zinc Acetate
NaCH (pH>12)
* Pesticide/Herbicide and Dioxin designated low priority. One or
both deleted if volume of water becomes critical during sampling.
-------
-67-
Volatile Organic Compound's (VDC's)
Although 21 samples were taken during the Task Force investigation,
laboratory error has rendered a major portion of the analytical results
unreliable. The analytical laboratory exceeded the VDC holding time of
7 days for 12 of the 21 sanples. The results from this category
were considered crucial. Holding times were exceeded from 26 to 34
days in excess of the allowed 7 day holding time. VDC results
which should be considered unreliable are:
-i s
fCft Well* Em Well*
766 Trip Blank 783 CQ 14-18
775 CQ 12-38 784 Cq 15-8
776 CQ 12-38 785 CQ 13-8
(field duplicate) 786 B-ll
777 CQ 12-8
779 Field Blank
780 Equipment Blank
781 CQ 14-58
782 CQ 11-8
Also, laboratory method blanks MB-1, MB-2, and MB-5, were contami-
nated with methylene chloride. As a result, all positive methylene
chloride results should be considered unreliable.
Laboratory method blanks MB-3, MB-4, and MB-5 contained acetone
which renders all positive acetone results unusable. False negatives
for sample M3V770 (Well CQ 17-8) are unacceptable. VDC results should
be considered quantitative for sample M3V767 thru M3A774 and MJV778.
The wells corresponding with the samples are: CQ 2-8, CQ 1-58, CQ 6-38,
CQ 17-8, CQ 17-18, CQ 9-58, CQ 17-38, CQ 17-58, CQ 10-18.
Of the nine samples which are quantitated for voc's, only two wells
were found to contain traceable amounts for the following:
Well Well
CQ 17-8 CQ 17-18
Benzene 140 ppb 16
Total Xylene 650 ppb ND
The two wells are downgradient from landfarm #12 which has not
been in operation since 1982, but were found to contain an immisible
layer.
Send-Volatile Compounds
The laboratory did not analyze matrix spikes or matrix spike
duplicates for the semi-volatiles and, as a result, all semi-volatile
-------
-68-
analytical results should be considered semi-quantitative at best.
Also, the analytical laboratory exceeded the maximum holding time of
40 days between extraction and analysis for 12 of the 21 samples.
Semi-volatile results which should be considered semi-qualitative,
at best, include:
M3A • Well* MQA Well#
766 Trip Blank '781 CQ 14-58
767 CQ 2-8 782 CQ 11-8
771 CQ 17-18 783 CQ 14-18
777 CQ 12-8 784 CQ 15-8
779 Field Blank 785 CQ 13-8
780 Equipment Blank 786 B - 11
Although all the semi-volatile results are semi-quantitative at
best, the following results were recorded for sample M3A 770 which was
not altered because of excessive holding times.
wall*
CQ 17-8
2 - Methylnaphthalene 110 ppfo
bis(2-Ethylhexyl 2300 ppb
Hithalate)
Hie above result for bis(2-ethylhexyl) pthalate should, at a minimum,
be considered semi-quantitative because of laboratory (method blank)
contaminat ion.
Metals
According to the FRC and EFA data usability reports, the quality
control for the graphite furnace metals (antimony, arsenic, cadmium,
lead, selenium, and thallium) was generally acceptable. However, the
dissolved arsenic, cadmium, and thallium results for the well CQ 12-38
field duplicate sample are semi-quantitative, at best, because of out-
side DQO spike recoveries. Because an initial calibration verification
for dissolved arsenic had to be rerun, all dissolved arsenic data should
be considered semi-quantitative. All total antiminy, arsenic, cadmium,
and thallium results should also be considered semi-quantitative.
Dissolved arsenic was detected in well B-ll at a concentration of 12 ppb.
No other metals were detected.
ICP Metals
All total and dissolved results for aluminum, beryllium, calcium,
chromium, cobalt, copper, magnesium, manganese, nickel, potassium,
sodium, vanadium, and zinc can be considered quantitative. All total
-------
-69-
iron results should be considered semi-quantitative. All dissolved
barium and potassium results should be considered quantitative. Eighteen
monitoring wells exhibited elevated concentrations of calcium, iron,
magnesium, potassium, and sodium. Zinc and tin was also detected in a
number of wells. The following lists the monitoring well and the
concentration of dissolved ICP metals found in the corresponding samples.
well*
CQ 13-8
CQ 12-38
CQ 12-38 (dup)
CQ 12-8
CQ 14-18
CQ 14-58
CQ 11-8
CQ 10-18
CQ 2-58
CQ 2-8
B-ll
CQ 6-38
CQ 9-58
CQ 17-8
CQ 17-18
CQ 17-38
CQ 17-58
CQ 15-8
(concentrations in ppb)
Calcium Magnesium Magnanese Potassium Sodium
" 38
3,110
3,020
41
607
74
1,530
2,790
32
693
276
118
11,000
615
967
12
125
38
110,000
543,000
530,000
204,000
217,000
159,000
184,000
136,000
61,800
128,000
158,000
103,000
219,000
139,000
52,400
108,000
100,000
89,500
29,300
98,200
94,500
10,900
31,200
4,330
51,800
54,800
3,900
8,350
83,200
3,260
11,000
68,300
16,400
4,800
4,200
16,400
5,420
7,990
7,850
6,100
10,200
2,430
11,600
11,900
3,340
7,430
25,400
929
2,650
15,100
1,390
1,200
1,920
6,820
10,400
1,080,000
1,060,000
60,200
2,480,000
34,200
169,000
850,000
13,900
52,200
374,000
19,800
61,600
195,000
510,000
54,000
44,300
89,300
All of the above monitoring wells exhibited elevated concentrations
of dissolved iron except wells: CQ 15-8 and CQ 13-8 with concentrations
ranging from 70 ppb to 20,800 ppb. Dissolved tin was detected in 8 of
the above wells in concentrations ranging from 55 ppb to 135 ppb.
Zinc was detected in concentrations ranging from 25 ppb to 461 ppb
in 8 of the above wells.
Inorganic and Indicator Parameters
All cyanide, bromide, chloride, fluoride, sulfate, total phenol,
TOC, and PCX results and certain TOX results are considered quantitative.
All POC result are considered qualitative only.
The following table lists the monitoring well and the corresponding
concentration for the indicator parameters TOC and TOX.
-------
-70-
WBll* TOG FOG Well* ICC POC
CQ 13-8 32,000 26 CQ 2-8 4,300 50
CQ 12-38 22,000 222 B-ll 67,000 22,500
CQ 12-38(dap.) 22,000 253 CQ 17-8 390,000 11,600
CQ 12-8 22,000 37 CQ 17-18 57,000 1,020
CQ 14-18 28,000 7,190 CQ 17-38 4,200 111
CQ 14-58 2,100 38 CQ 17-58 4,100 34
CQ 11-8 18,000 33 CQ 15-8 12,000 29
CQ 10-18 3,400 56 CQ-'6-38 2,100 41
CQ 2-58 ND 17 CQ 9-58 8,100 99
(Concentrations in ppt>)
Total phenols were detected in 4 of the above wells in concentra-
tions ranging from 53 ppb to 94 ppb.
3. Conclusions
The following inorganics and organics were found at the facility
as a result of the investigation:
Given that the majority of volatile and semi-volatile organic
sample results are unreliable due to laboratory error only a very small
number of sample results were recovered. Benzene was detected in well
downgradient clusters of landfarm #12. Total xylene was also detected
in cluster CQ 17. In this same cluster 2-methylnaphthalene and
bis(2-Ethylhexyl) phthalate were also detected.
High levels of calcium, magnesium, manganese, potassium, and sodium
were detected in 18 downgradient and upgradient monitoring wells located
throughout the facility.
In general, TOC and POC were found at higher concentrations in
downgradient as compared to upgradient wells. This confirms Amoco's
data collected during their inital ground water assessment phase.
Due to the limited amount of quantitative and reliable data avail-
able, sample results are not sufficient to determine accurately if
volatile and semi-volatile organic compounds have impacted the ground
water. Landfarm #12 shows some indication of ground water contamination
because benzene, total xylene and bis(2-Ethylhexyl) phthalate were
detected in well cluster CQ 17, which is downgradient of this landfarm.
Poor quality control rendered the majority of sample results for
heavy metals semi-quantitative, and therefore, these results will not be
used in evaluating ground water quality conditions at Amoco. However,
monitoring well B-ll did detect arsenic at 12 ppb.
-------
-71-
Table 6. Selected Inorganic Data. Landfarm #11*
Monitoring Wells
Parameter CD 6-38 CO 9-58
Calcium 103,000 219,000
Iron 3,700,, 9,510
Magnesium 3,260 11,000
Manganese 118 324
Potassium 929 2,650
Sodium 19,800 61,600
Tin ND 55
Sulfate .0071 .140
Chloride .009 .078
Sulpher ND . 024
TOX ND 41
POX 12 ND
TOC 2,100 8,100
POC 41 99
Total Phenols ND ND
ND Nbn Detected
* All Concentrations in micrograms per liter
(ug/1)
-------
-72-
03
m
*.
oo
00 ro
CO
I—too
00
m
OOOOCNCNO22
O 00 CN rH
C3 ^5*
FI m
-I "*
o o o
o
~ ° "* Q
Q o m 2
w
CN
rH
=»*=
s1
• H
fi
00
n
OrHOCNOOQQCNOVOrHQOrHQ
O VO 00 rH CN O OOOr~ CNrH
o in
IH
•a
2
a
fO
S1
T3
3
o
(U
rH
«
r-
-------
-73-
Cfl
a
n
0)
Q Q Q a
z z s z
m Q Q Q Q Q CN
m z a z z z
CO
QQQQQcoaQ
z z,z z z CN a z
Q
Z
QQQcNQQQDQ
ZZZ-^ZSZZZ
QQQ
ZZZ
QQcoaQ
ZZCNZZ
m
•X
w
"8
g
a
CJ
CO
OJ
o CN CN
w
I—I
f— 1
$
S1
i
•H
C
5
2
r— 1
i— 1
00
1
CN
8
00
in
CN
8
oo
1— 1
1
o
8
u
2
2
ftl
00
in
,— i
0
0
o
00
CN
i— 1
O
o
00
1— 1
VO
o
o
o
V£>
CO
•—I
|
•r-l
O
3
p-j
00
o ro
in cr\
ro vo
CO
o CN
O C"0
°1
ro
O O
o cy>
oo r-
p» CN
in
§8
•H m
W C
O (0
in
CN
i-H O
oo n
^j< -^
^
Q o
m
n
Q o
Z o
i— i
i— i
•H
rH tfl
0) CO
33
Sfi
^4
r-
o Q
0 Z
CN
CN
in
o Q
o z
ro
i— i
0 Q
0 Z
O
O
m
00
•H
m Q
CN Z
Q Q
Z Z
Q Q
Z Z
U
•H
H§
XX
vo
rH
Q O oo
Z 00 •
n CN
i— i
131
CN m
Q r^ Q
2 r^ z
•H
P S 12
3 P S
-C l-i CD
U M 3
Q oo
Z VO
o
Q o
i-H
0
Q 0
Z ^O
CN
IS
c n
1-&
3 i-H
«3 w
1— 1
o
o
i-H
CN
i— 1
•
^H
0)
T3
•H
8
i-H
6
CN
CN
m m o o o Q
o fH o in o z
o m m
CN ^
oo oo Q r^ Q Q
•» • Z i— i Z Z
o in
m
-------
•K
O
03
rH
*
""
S1
-H
|
"c
00
i
rH
8
00
in
I
rH
8
00
rH
1
r— 1
8
00
1
CN
8
Q.
a
00
ro
I
CN
rH
CO
ro
CN
rH
8
CO
CO
1—1
8
o
o
o
*r
CO
rH
O
o
0
en
m
rH
0
O
«.
r-
rH
CN
O
0
O
*3"
o
CN
O
O
O
ro
in
0
0
o
rO
Tj*
m
O
0
o
o
rH
0
0
CO
rH
in
o
fV)
ro
•~0
O rH
O rH
O rH
0
O
^
rH
o m
O CO
0 rH
O
00
0
rH
O CN
O *&
o
"^
o
rH
-74-
C^ CD ^ft ^D Q Q lO CD ^^ CO Q C5
^frH OZZCNCNvOCNZO
00 • CO rH O
O 00
CN rH
QQQoQQf^voooOQO
IsO O O CN rH
rH ' CN
J '
oo ro ro • •— i "" O
«. • m «•
in oo
CN
2 2 r*** • s ro O1^ r*** vo is ^D
m -H o o o
* * • «»
CN
CN
inQQOQQooovDrHQo
rH r* m • o o
CM CN
CN
ro CM o r- r- CM Q
cNQQQo «r~r^ •m'*z
••erZZZOCNr^inrHO
P^ *
^^
rH
rHaQQQQinvoQCNQO
vOZZZZZCMroZroZO
T rH i— 1 O
CN
ro
n
ro
00
0
rH
^
r-
oo
ro
CN
O
O
0
CN
CN
CN
Q
jf j
Q
^
Q
Z
Q
CN
CN
CN
O
Z
a
o>
c
•H
03
0)
CO
-U
Q)
-------
00
ro
-75-
QQQ
222
1
^5*
r— '
8
*-*.
ro
*
00
1
1— 1
8
00
1— 1 *-•
i-H ro
m v
d.
en oo
C 1
•H CN
Ll rH
4J O*
C
*£ rH
1 ""* -~-
O
o a
rH Q
£ co
C ro
M-l CN
r3 o
CD
§
£
-^
1
Ll
4-1
T)
-U
O
•H
§
S
T3
o
d)
i— (
(ii
yj
rH
-*— •
CO
ro
c^
rH
8
m
*s^»
CO
rH
^
l-l
8
ro
r-H
""* 00
ro
rH
8
Ll
• d)
^D 1 *
i—i m
P
rH O
rH O
=t*I
1
rH CN Q 4-1
2 I
CD 0 Ll
DJ C -O 0}
H H CU -H ^j
rH Ll CD
in in >. o £
1 1 si I-H r3
•5T -^ 43 JS U
1 1 0) U <0
CN CN 2 di
3
W-l ^«
Q 'O ro
J CN
""" 00
rH
r-
rH
8
CN
rH
8
Li
3
Q Q2
a>
Q Q a a a a »*
222 2 22-
QQQ Q Q Q 00
222 2 22-
^^^
QQQ Q QQQ TJ
222 2 222 07
•rH P*
03 Li
••S 81
*J C
ooo o rocNin §-H
••a* in ^H o a) 3 £
I-H ^o rH ro > 51
«• o -H r c
CN rH 4J -H -H
-Q TS £
r3 -U O W
-H -H W C
rH JJ O
O C 1 -H
Ll 13 -U
rH C D "3 (T3
>i D O1 -U l-i
X 03 4J
0) 1 1 Q C
03 1 J2 OJ
CU O 4J fO fO O CJ
C Pi rH d) 4J 4J O C
CU ^3 ^"1 ^^ frt (B ^^ Q
rH C CD ^^ 03 t-S uj 1 o
^•1 rH C [ t i Qg lr_i
Q)X>iCDCp'o HH UCJ£rH
CD rH i3 ^3 CN p Q tn tn ^ ^ c/) 3
N (0 OJ ^2 *— £ 1 1 1
(DOT -H III rH CN ro
fflHCN .Q CNCNCN ^x«^^_-
-------
-76-
Table 11. Selected Organic Data from Task Force
Background and Ebwngradient Monitoring Vfells*
Blanks
Parameter CQ 2-8 CQ 2-58 B-ll Field Trip Equipment
2-4-D
2-4-5 TP
chlorobensilate
ND
ND
ND
ND
ND
ND
8 "
2
4
ND
ND
ND
(1) VDC Data - Unreliable
(2) VDC Data - quantitative
(3) Serni-VOC Data - Semi-quantitative (holding times exceeded)
* All concentrations in micrograms per liter (ug/1)
-------
References
Amoco Oil Company, Land Treatment Facility, Yorktown Refinery,
Virginia, Response to EPA/VDH Garments - Volume III, Stone and Webster
Engineering Corporation, 1984
Geology of the Yorktown, Poquoson West and Poquoson East Quadrangles,
Virginia, Report of Investigation No. 30, Commonwealth of Virginia,
Division of Mineral Resources, James L. Carver, 1972.
Geologic Studies - Coastal Plain of Virginia, Bulletin 83 (Parts
1 and 2), Commonwealth of Virginia, Division of Mineral Resources,
James L. Carver, 1973.
Geology and Ground Water Resources of the York James Peninsula,
Virginia, D. J. Cederstrom, U.S.G.S. Water Supply Paper No. 1361,
U.S.G.P.O., Washington, D.C., 1957.
Amoco Oil Company, RCRA Part B Permit Application, 1983, RCRA
Ground Water Monitoring Technical Enforcement Guidance Document,
U.S. EPA, 1986.
Comprehensive Monitoring Evaluation Checklist and Findings,
Ccmnonwealth of Virginia, Department of Waste Management, 1987.
Final Project Plan, Hazardous Waste Ground Water Task Force Amoco,
Yorktown, Virginia, U.S. EPA Region III, 1987.
Information and Documents for Compliance Assessment of Land Treat-
ment Facility, Amoco Oil Company, Yorktown, Virginia, PRC Environmental
Management Inc., 1986.
Amoco Oil Co. Land Treatment Facility, Yorktown Refinery, Yorktown,
Virginia, Response to EPA/VAD Comments - Vol. II, Stone & Webster,
Boston, Massachusetts, November 1984.
Geology of the Yorktown, Poquoson West and Poquoson, East
Quadrangles, Virginia, (Report Investigation), Commonwealth of Virginia,
Division of Mineral Resources, James L. Calver, 1972.
-------
APPENDIX A
LABORATORY AUDIT REPORTS
-------
? UNITED STATES ENVIRONMENTAL PROTECTION AGENCY
V -• — $ REGION III
<<• PRO^C CENTRAL REGIONAL LABORATORY
839 BESTGATE ROAD 301-224-2740
ANNAPOLIS, MARYLAND 21401 FTS-922-3752
DATE . February 23, 1987
SUBJECT: Lab Inspection of Amoco Analytical Services, Naperville, IL
,m,->n Jeanne Hankins (3ES2
FROM : Chemist
Chris Thomas (3HW15)
' DELMARVA/DC/WV RCRA Enf. Section
THRU : Patricia J. Krantz (3ES23) ^^
Chief, Quality Assurance Section
I inspected the Amoco laboratory at Naperville, IL on February 4-5, 1987.
Those present were:
Leo Duffy - Research Supervisor
Bob Babcock - Supervisor
Pat Beaulieu - AA
Don Becker - ICP
Jim Jarrett - Anions
Pat Blockinger - TOC
Bruce Keen - Mass spec.
Mark Bambacht - Mass spec.
The Amoco staff were very cooperative and provided all information
requested in an open and forthright manner. Discussions with staff
members revealed a high level of technical adequacy and a willingness
to comply with all RCRA QC requirements.
Deficiencies were noted in several areas, and are as follows:
Sample Receipt
- No Standard Operating Procedure (SOP) available.
- Daily temperature logs shall be maintained on all refrigerators, to
include an acceptable range.
- Sample receipt/analysis form has no provision for noting condition
upon receipt. I recommended this be added either to the form or
the chain-of-custody, with special attention to samples requiring ice.
AA Analysis
- SOPs for As, Se and Pb shall be updated to reflect actual practice,
including QC.
- SOP shall be written for Tl.
- Hoods over the AAs will be checked for flow in the
- All samples are to be digested. . ^
-------
ICP Analysis
- SOP shall be updated and include precise DC practice. Terms such
as "about, approximately, should, needs to be" must be eliminated
from SOPs.
Glassware Preparation
- SOP shall be written for both inorganic and organic areas and be
posted above all wash areas.
Anions & TOC
- SOP shall be updated and include all QC practices.
VOAs & BNAs (extractables)
- Daily calibration standards shall include all compounds of interest.
- Data on daily calibration runs is to be maintained in an easily
accessible manner. Current practice complicates the possibility
of checking response factors.
- SOPs shall be written which reflect current practice, including OC.
Referencing EPA Methods 624 and 625 is not an adequate substitute.
- Recommend purchase of an additional mass spectrometer based on a
minimum estimate of 150 samples per month. An additional operator
is also highly recommended. Based on recent downtime experience
in 1986, the potential exists that sample integrity may be in
question due to holding time violations. Currently Yorktown site
samples are being analyzed by ETC Labs, in NJ, as the Amoco lab does
not have sufficient capacity.
- All groundwater extractions are to be done using a volume of 1000 ml.
Data Validation
- Review of the organic data was incomplete due to the difficulty of
checking the calibration data.
- Hg, Tl and Pb data were acceptable.
- ICP data reflects lack of precise OC frequency.
QA Manual
- One of the most critical portions of the QA manual, section 10 on
lab-specific practices, is not currently available. This must be
completed and implemented immediately.
- Instrument logs are to be maintained as prescribed.
Section on corrective action in the case where the system is
out of control shall be expanded to include reanalysis of
samples and redigestion/reextraction, as indicated by the
specific QC problems encountered.
-------
In general, I noted that the lab was well organized and had the
necessary minimum space requirements. Housekeeping was good with
especially well maintained hoods.
The Amoco staff agreed they would comply with all QC requirements in
SW-849, 3rd edition, as soon as it is available.
In summary, actual practice reflects concern for QC and proper meth-
odology. Some methods need minor modifications to comply with EPA
requirements and will be so modified upon receipt of SW-846, 3rd
edition. At that time, additional QC, i.e.,.matrix spikes and
duplicates, will be implemented.
JHrwbg
-------
J
ri
8 UNITED STATES ENVIRONMENTAL PROTECTION AGENCY
f REGION III
CENTRAL REGIONAL LABORATORY
839 BESTGATE ROAD 301-224-2740
ANNAPOLIS, MARYLAND 21401 FTS-922-3752
DATE : March 5' 1987
SUBJECT: J. R. Reed Laboratory Inspection
FROM • Jeanne Hankins
Chemist
TO • Christopher Thomas (3HW15)
HELMARVA/DC/WV RCRA ENF. SEC.
THRU : Patricia J. Krantz (3ES23) ";
Chief, QA Section
J. R. Reed K Associates laboratory was inspected on 30 January 1987.
The following personnel were involved:
James R. Reed, President
Carol M. Isenhour, Laboratory Manager
Beverly Blanchard, Quality Control Officer
The staff was cooperative and provided all information requested. It
was agreed that they would meet all QC requirements in the SW846 3rd
edition, as soon as it is available (April, 1987).
The following deficiencies were noted:
Metals
- All samples for the RCRA program will be digested before analysis.
- For the instrument calibration standards documentation will be
provided stating the concentrations used for each element.
- nates and initials must be on each standard.
Phenol s
- SOP will be provided for the method of standardization.
TOX
- Frequency of checking the carbon blank must be specified. ^ «*f
0
-------
S04=
- SOP refers to Standard Methods for preparation of standard -
a specific edition will be cited.
- SOP does not Deflect required method, i.e. stirring during
measurement and the use of timed measurements.
- SOP needs a better description of the calculations.
Quality Assurance Manual
- QA manual must specify or make reference to methods which
specify the frequency of duplicates and spikes.
- Corrective action section must be expanded to include reanalysis
of samples and redigestion/reextraction, as indicated by the
specific QC problems encountered.
- Continuing calibration verification must be addressed and
frequency cited.
- Section must be included to cover data validation between the
raw data and the final report.
Data Validation
- TOC data of 11/25/86 indicates that the system is out of control
and that corrective action was not initiated.
- TOC bound notebook has pages torn out.
Items that were struck out were not properly dated and initialed.
- Inadequate identification of samples in TOX book.
General Comments
Instrument detection limits determinations must be established.
- Calibration procedures must be documented for all analyses.
- Routine maintenance check sheets or logbooks to be established
for each instrument.
- Class S weights will be checked each day that the balance is
used.
- Bench space is inadequate. A minimum of 6 feet (unencumbered)
per analyst should be available.
-------
In summary, the most critical areas of concern are data
validation, corrective action when QC limits are exceeded or
calibration data indicates an out-of-control situation, and
instrument detection limit determinations. The lab staff
expressed a willingness to correct all deficiencies.
JHrwbg
-------
APPENDIX B
TASK FORCE FIELD MEASUREMENTS
-------
January 1987 Task Force Measurements
Well Surveyed Elevations (ft) Total Well Depth to
Number Top of Well Casing Depth (ft) Water (ft)
CQ 1-8 101.67 3.04
18 101.67 3.06
38 101.57 ., 3.05
58 101.32 3.50
CQ 2-8* 100.02 10.22 2.42
18 99.71 18.77 2.50
38 100.22 40.54 3.65
58* 99.97 60.25 2.90
CQ 3-8 100.90 3.24
18 100.77 3.06
38 100.67 4.31
58 100.58 4.29
CQ 6-8 98.81 2.03
18 98.78 2.12
38* 99.15 40.54 2.56
58 99.54 3.34
CQ 8-8 98.59 1.88
18 98.97 2.38
38 100.79 4.54
58 101.11 4.86
CQ 9-8 98.55
18 98.40
38 100.08 3.70
58* 100.07 60.66 3.80
CQ 10-8 99.63 7.15
18 99.56 20.36 6.86
38 99.66 5.30
58 99.91 5.30
CQ 11-8 99.88 10.34 5.41
18 99.61 5.08
38 100.30 5.34
58 98.50 5.44
CQ 12-8* 99.88 10.38 6.23
18 100.00 5.88
38 99.74 40.54 4.92
58 99.50 5.85
Elevation of
Water (MSL)(ft)
98.63
98.61
93.52
97.82
97.60
97.21
96.57
97.07
97.66
97.71
96.36
96.29
96.78
96.66
96.59
96.20
96.71
96.59
96.25
96.25
96.38
96.27
92.48
92.70
94.36
94.61
94.47
94.53
94.96
94.06
93.65
94.12
94.82
93.65
-------
January 1987 Task Force Measurements (Cont'd)
Well Surveyed Elevations (ft) Total Well Depth to Elevation of
Number Top of Well Casing Depth (ft) Water (ft) Water (MSD(ft)
CQ 13-8* 99.79 10.56 2.95 96.84
18 100.17 6.54 93.63
38 100.58 6.65 93.93
58 100.71 ^ 6.79 93.92
CQ 14-8 98.98 2.68 96.30
18* 99.50 20.32 3.84 95.66
38 99.52 5.38 94.14
58* 99.91 60.42 6.0 93.91
CQ 15-8 100.29 10.33 5.66 94.63
18 100.14 6.79 94.35
38 100.22 5.88 94.34
58 99.95 5.63 94.32
CQ 16-8 101.44 5.89 95.55
18 101.44 6.43 95.01
38 101.44 7.13 94.31
58 99.55 8.03 91.52
CQ 17-8* 101.20 10.39
18* 101.34 20.35 8.54 92.80
38* 101.00 40.55 7.52 93.48
58* 101.16 60.70 7.67 93.49
CP 2-8 98.29 2.44 95.85
CP 2-18 98.56 2.09 96.47
B-01 100.67 ( 8.46) 2.81 97.86
B-ll 99.37 11.94 5.39 93.98
B-13 103.91 (13.51) 4.58 99.33
B-14 101.94 (13.33) 2.91 99.03
B-18 101.76 (12.63) 3.76 99.00
CQ 17-58 101.16 60.70 7.67 93.49
B-ll 99.37 11.94 5.67 93.70
* Wells sampled by HWGWTF.
1. Depth of water measured from TOC.
2. Water level measurements not obtained for CQ9-8, CQ9-18.
3. CQ17-8, 18 and 58 tubing pulled prior to water level measurement.
-------
CO
1
h~*
t— •
N;
O
o
o
ho
1— 1
i— '
NJ
^,
0
O O O O O OO
^•J ^J ^J ^-J U1 »& U^
I i i i 1 I i
tn CO I— ' CO 00 Ol (— '
00 CO CO CO 00
»^»&oSK
l_l l_ l_l l_l l_l 1 — I
COVOOOOOh-^
t— '
o
ON en i—4 to CA o
o ui en en *» o o
O O -J O (—• t—1 1— ' M
u> vo o co o i— • i— i
^J ^J ^J ^f\ ^fr ^J ^^
LxJ ^J ^^ CO ^^ ^O ^O
oooooooo
f Vr?TVVV
COUICOI-'LnOJOlCO
CO 00 CO 00 00
C*J tO C^ ^O ^J ^O ^J ^)
^5 VO ^^ CO CO *xj ^J **J
J J
h-> (— i K-» h- 1
OOCOUIOJOJNJ^^
4^ CO I—1
UlOOOOl— 'toO
OOOOOOU1O
en i— ' 4^ i— '
VOK>00~J^DUI-J
^3 «xj OJ O1
I—. )—. I-- |_.
cocncncocotoO^J
-jcr.-je^ONc^^erN
f-^OJfcVOCOkOUl
s
1— ^
s?
ff
DJ
CD
o
n
M
c w
3 H-
8 ^ S"
1 CD
— O l-h 1
gC^ 3
0 t2
-^ O O rr
O rr QI
1 Q)
(DO \
O O ho
rr ON
5> 1
O. ^o
00
-J
i-1- rr
3 (C
H- (D
? 2
«e
°s
rr
M (D
h-1 M
(D 0)
n?
[S-
k/J
!
ID
•s-
-------
APPENDIX C
TASK FORCE ANALYTICAL PARAMETERS
-------
LC'7-C2-3
67-64-1
1C7-L3-1
'1-^3-2
75-27-4
75-25-2
7—83-9
108-90-7
75-00-3
110-75-8
67-66-3
74-87-3
96-12-8
124-43-1
106-93-4
75-34-3
107-06-2
156-60-5
156-60-5
75-09-2
78-87-5
10061-01-6
10061-02-6
123-91-1
100-41-4
78-93-3
110-86-1
100-42-5
95-94-3
79-34-5
127-18-4
56-23-5
108-83-3
75-25-2
120-82-1
71-55-6
79-00-5
79-01-6
75-01-4
79-06-1
624-83-9
Ac role IT.
Ace cone
Aery Ionicrile
Benzene
Broraodichlcrone:hane
Broraofora
Broraoechane
Chlorobenzane
Chloroechane
2-Chloroechyl vinyl echer
Chlorofonn
Chlororaechane
1,2-Di jrorao-3-chloropropane
D ibrcjnoch lo rome chane
1,20-Dibronoechane
1,1-Dichloroechane
1,2-Dichloroechane
trans-l,2-Dichloroechene
1,2-Dichloroechene
D'chlorooechane
i,2-Dichloropropane
cis-l,3-Dichloropropene
crans-l,3-Dichloropropene
1,4-Oioxane
Echylbenzene
Methyl echyl kecone (MEK)
Pyridlne
Sytrene
1,2,4,5-Tecrachlorobenzene
1,7. ,3,4-Tetrachlorobenzene
1,1,2,2-Tetrachloroethane
Tecrachloroechene
Tecrachlororoechane
Toluene
Tribronomechane
1,2,4-Trichlorobenzene
1,1,1-Trichloroe thane
1,1,2-Tr ichloroechane
Trichloroethene
Vinyl chloride
Acrylamide
Isocyanic Acid
-------
A ;r~/lcn:rr: 1-
Acrvlaci-e
METHOD 32"
33-32-9 Acenapr.tr. er.e
106-96-3 AcenaphtdLer.e
62-53-3 Ar.iline
120-12-7 Anthracene
56-55-3 3enz[ a] anthracene
92-87-5 3enzidine
56-55-3 3enzo(a)anthracene
205-99-2 3enzo[b]fluoranthene
2Q7-08-9 3enzo[kjfluoranthene
50-32-8 3enzo[a]pyrene
191-24-2 3enzo[g,h,i]perylene
100-^-7 Benzyl chloride
111-91-1 Bts(2-chloroethyoxy)mechane
108-60-1 Bis(2-chlroisopropyl) echer
117-81-7 Bis(2-echylhexyl)phchalace
101-55-3 4-Broraophyenyl phenyl echer
85-68-7 Butyl benzyl phchalace
106-47-8 p-Chloroaniline
59-50-7 p-Chloro-m-cresol
91-58-7 2-Chloronaphchalene
95-57-8 2-Chlorophenol
7005-72-3 Chlorophenylphenyl et±ier
71R-01-9 Chrvsene
53.70-3 Dibenz[a,h]anthracene
132-64-9 Dibenzofuran
100-01-6 4-Nicroaniline
98-95-3 Nitrobenzene
88-75-5 2-Nitrophenol
100-02-7 4-Nitrophenol
62-75-9 N-Ni^
621-64-7 N-Nitrosodipropylaoine
608-93-5 Pentachlorobenzene
82-68-8 Pentachloronitrobenzene (PCNB)
87-86-5 Pentachlorophenol
120-12-7 Phcnanchrene
108-95-2 Phenol
129-00-0 Pyrene
95-94-3 1,2,4,5-Tecrachlorobenzene
1,2,3,4-Tetrachlorooenzene
-------
-._ ,-- . r u." - _ r r r er. ;: er e
'2,3," ,3-7ecr3chiorociberizo-p-dioxiri
*S carries for buc no standard available
34-~--2 Di-n-buryl phchalace
95-5C-1 1,2-Dich'lorobenzer.e
541-73-1 1,3-Dichlorobenzene
126-46-" 1 ,-k-Dichlorobenzene
91-94-1 3,3'-Dichlorobenizidine
120-83-2 2,4-Dichlorophenol
94-75-7 2,4-Dichlorophenoxyacecic acid
84-66-2 Diethyl phthalace
105-67-9 2,4-Diioechylphenol
131-11-3 DLTechyl phchalace
534-52-1 4.6-Dinicro-o-cresol
^l-?8-5 2,4-Dinicroohenol
121-14-2 2,4-Dinicrocoluene
606-20-2 2,6-Dinicrocoluene
117-84-0 Di-n-occyl phchalace
122-39-4 Diphenylanine
206-44-0 Fluoranchene
7782-41-4 Fluorene
87-68-3 Hexachlorobucadtene
77-47-4 Hexachlorocyclopencadiene
67-72-1 Hexachloroechane
193-39-5 Indeno(l,2,3-cd)pyrene
78-59-1 Isophorone
95-48-7 2-Methyl Phenol
106-44-5 4-Mechyl Phenol
91-20-3 Naphthalene
METHOD
8080
Aldrin
alpha BHC
Bee BHC
Delca BHC
Gamna BHC (Lindane)
Chlordane
4.4'-DDD
4.4'-DDE
4,4'DDT
Dieldrin
>i«.
. :-i'J *
..-j'41^
^
;4,,
-------
Er.dcsuifar. ::
Endosu.Lr.3r. 5_ljfa;e
Er.a'rin
Endnri aldehyde
Heocacr.ior
Heotrachlor epcxide
Methcxycnlor'
Toxaor.er.e
PCS-1016
PCB-1221
PCB-1232
PCS-1242
PCB-1248
PCB-12S4
PCS-1260
Appendix VIII METALS
METHOD 6010
Aluminum
Barium
Beryllium
Boron
Cactaium
Chromium
Iron
Lead
Nickel
Thallium
Vanadium
Zinc
Selenium*
Arsenic*
*Ther2 elements are noc approved for 6010 but they are approved for dP .-
mecals ICP mechol. The CLP mecals ICP method is identical co the
METHOD 7470 ,
Mercury
•> j
-------
->-70-30-3 Crcconaidehyde
764—1-0 1, — DichIoro-2-Dutene
75-71-3 Dichlorodifluoronethane
75-35-4 1,1-Dichloroethene
10C61-02-6 trans-1,3-Dichioropropene
57-14-7 cis-l.l-Dimethyihydrazine
591-78-6 Hexanone
74-88-4 lodornethane
Pentachloroethane
1,2,3,5-Tetrachlorobenz ane
630-20-6 1,1,1,2-Tecrachloroethane
75-70-7 Trichloromethanechiol
96-18-4 1,2,3-Trichloropropane
95-35-4 Trinitrobenzene(l,3,5-)
75-01-4 Vinyl Acetate
75-05-8 Acetonitrile
75-69-4 Fluorotrichlororaethane
542-75-6 1,3-Dichloropropene
METHOD 8270 (continued)
87-65-0 2,6-Dichlorophenol
60-11-7 p-Diraethylarainoazobenzene
57-97-6 7,12-Difflechylbenz[aJanthracene
122-09-8 alpha,alpha-Dunechylphenechylaoine
122-66-7 l,2-Diphenylhydra2ine
97-63-2 Ethyl raechacrylace
62-50-0 Ethyl mechanesulfonate
1888-71-7 Hexachloropropene
120-58-1 Isosafrole
148-82-3 Melphalan
91-80-5 Mechaperylene
79.99.1 M—•"—' L*
- ^
-------
6 -
--.e-}.!.—. p-rx^r.zoquinor.e
100-15-6 Benzyl alcohol
88-35-7 2-sec-bucyl-4.6-di.rucrophenol"
5^2-76-7 3-Chloropropi.oni.cri.le
131-39-5 2-Cyclohexyl-^,6-dinicrophenol
226-36-3 Diber.z[a,h]acridine
38-~i-4 2-Nitroaniline
99-09-2 3-Nicroaniline
92^-16-3 N-Nicrosodi-n-bucylanine
1116-54-7 N-Nicrosodiechano'lanine
55-18-5 N-Nicrosodiechylanine
10595-95-6 N-NiCrosoroechylechylaraine
615-53-2 N-Nitroso-N-aechylurechane
59-89-2 N-Nicrosoraorpholine
100-75-8 N-Nicrosopiperidine
930-55-2 N-Nicrosopyrrolidine
99-55-2 5-Nicro-o-toluidine
76-01-7 Pentachloroechane
62-44-2 Phenacetin
109-06-8 2-Picoline
94-59-7 Safrole
1,2,3,5-Tecrachlorobenzene
58-90-2 2,3,4,6-Tetrachlorophenol
636-21-5 o-Toluidine hydrochloride
73-70-7 Trichlororaechanechiol
95-35-4 Trinitrobenzene
126-72-7 Tris(2.3-dibroraopropyl) phosphate
61-82-5 Aoitrole
504-24-5 4-Aminopyridine
98-07-7 Benzocrichloride
357-57-3 Brucine
1338-23-4 2-Butanone peroxide
510-15-6 Chlrobenzilace
106-89-8 l-Chloro-2,3-epoxypropane
50-18-0 Cyclophosphanide
2303-16-4 Diallate
311-45-5 0,O-Diechylphosphoric acid
55-91-4 Di-isopropylfluorophosphace (DFP)
60-51-5 Dinechoace
119-90-4 3,3'-Dimechoxybenzidine
119-93-7 3,3'-Diaechylbenzidine
77-78-1 Dimethyl sulface
298-04-4 Disulfocon
541-53-7 2.4-Dichiobiurec
96-45-7 Ethylenechiourea
62-74-8 Fluoroacetic acid (Sale)
-------
16752-77-5
56-*9-5
70-25-7
56-57-5
684-93-5
145-73-3
123-63-7
108-4 -2
1120-71-4
108—6-3
57-24-9
3689-24-5
78-00-2
126-72-7
108-31-6
123-33-1
109-77-3
81-81-2
Xechooyl
Z-MechyLazindine
3-Mechylcholdr.cnrene
N'-Mechyl-N'-ni:ro-N'-ni:rosog^ianidine
— Nicroquinoline-l-oxi.de
Endochal
Paraldehyde
Phenylenediani-e (o.n.p)
1,3-Propane sulrone
Resorcinol
Scrychnine
Tetraechyldirhiopyrophosphace
Tecraechyl lead
Tris(2,3-dibronopropyl) phosphace
Maleic anhydride'
Maleic hydraz:de
Malononicrile
Warfarin
METHOD
107-18-6
100-51-6
75-87-6
460-19-5
75-21-8
765-34-4
302-01-2
78-83-1
L26-89-7
60-34-4
75-86-5
80-62-6
107-19-7
METHOD
Kepone
8240-DI
Ally! alcohol
Benzyl alcohol
Chloral
Chloroacecaldehyde
Cyanogen
Dichloropropanol
Echyl Cyanide
Ethylene Oxide
Glycidylaldehyde
Hydrazine
Isobutyl alcohol
Methacry Ionic rile
Mechyl hydrazine
2-Mechyllacconicrile
Mechyl raechacrylace
2-Propyn-l-ol
8080
-------
3 :r.cseo
TOC METHOD 9C60
TOX METHOD 9020
Chloride METHOD 9252
Total nhervls METHOD 9066
Sulface METHOD 9036 or 9038
Nitrate METHOD 9200
ror.ia "Methods for Chemical Analysis of Water and Waste"
'JSEPA - ETCL (Cincinnati, 3/33, Method 350.1 or 350.3
EPA 600/4-84-008
POC Ground Water, vol. 22, p. 18-23, 1984
Dissolved -erals Total -lecals, and
•>aride IFB-WA 3—T092
• - 'il
'•utfSf*
.'- - ?V^
_ -, -,J•.'\l^*1
• ;••'. s,'. •.'s*"7'
- '£&%>$£;
• VvA»S«
- -. A?*
- ' .Vn*
^j- ff,
• ',* *^
"*^f
'"^<
-------
m O
i i
z u
X
X
O.
a.
QC
<
Q-
O
•—•
I—
>-
x
1/1
1
•
f.
•«
*
i
u
<
«
$
s
^i
V
4i
>.
-c
I
c
tl
(J
<
•
f.
•»
4
i
u
<
^
rvi
M
!•»
10
o
L.
TJ
X
f.
^
«j
B
«
C
4f
X
£
i
I C
1 «l
1 i
c
M
•
u
•
£
a.
M
•*
••
(M
^
*o
^^
'x
. e
.. •
4:
&
x
o
£
*#
4f
•*
N^
V
«
1
i
«^
«>
u
<
«
c
«
1.
j
^^
§
I
°x
**
•
u
<
(M
0^
S
Kl
t^
,
"x
(M
C
•
S
at
<>•
M
*
«
^
i
*^
«
u
<
•
«rf
•
it
•
'x
>
•*
S
s
c
•
•«
•
0
"x
c
«
f.
t*
»
•fl
u
•
u
*rf
41
u
<
*•
m
•*
«M"
m
5
m
^K
X
^
1
e
0
2
u
b.
in
>^
•M
"*
T>"
U
•
u
•rf
V
u
<
•o
u
•
u
«
u
•
X
K
O
1
U
e
£
u
o
•*
IM
«
|^
C
i
^
^
X
W
o
c
II
4
o
c
0
£
•5
«M"
Tl"
u
•
u
*«
ti
u
«
•
k
I
M
•
u
<
9
m
e
JC
•
t_
i
*i
tl
U
<
I
u
•
e
g
-------
o
^
i
i i
in" o
x f
X 0
X
X
o
at
UJ
CL
*
J
0
I
U
•
*
u
a
ii
X
ISI
^
m
(M
u
o
1
^
S
o.
in
(M
i.
o
u
o
^
e
•o
(M
e
8
tf\
IV
1
*
S
01
o
8
t.
•
e
u
S~~
u
•oj
s
o
u
c
tl
01
t.
—
••*
o
•tf
s
£
e
•e
S
w
A
C
tl
U
u.
C
JJ
•
^
C
X
f.
I
o
(M
-O
in
_^
_*•
S
•5
(Nl
*™
sT
•cene.
(a)anthr
pi*
0
„
c
•»
^
]
o
u
X
X
m
•»
-0
m
•
"x
£
1
b
k
x
•5
«NI
- "
e
V
X
u
I
u
c
OB
41
C
tl
•f
C
•
0
3
e
M
c
IM
in
o
c
^J
X
k
1
tl
[elaceph
c
CO
.
1
§
*A
IO
§
M
0*
2
C
X
1
a
(M
J^
X
w
1
^
*"
*
b
o
•5
in
'i
M
C
•a
j j
§
u
u
g
•
e
«
•o
m
i
u
(a)anthr
M
«
•
C
*c
an
m
(Nl
•O
I
c
to
^1
c
^
2
b
o
L.
f
CO
•s
,
o
"c
«,
s
(
o
u
"c
4
M
C
*•
c
I
0
Jf
u
?
i
e
e
u
i
}
c
•
?
g
e
«
?
a
S
,
0
«.
"c
I
M
tl
ca
i
*
i
ti
c
—
c
0
o
£
u
rg
.2
0
c
"x
X
i
r
o
b
0
u
"J
1
•
"x
^
t
•r
S
8
«
*
-------
-
S
o
§"
S«
IU
t> —
f- <
S 8
> §
V ME
X O
x
o
x
u
OL
!
u
•t
VI
u
2
u
'~
tl
M
X
in
C
1
X
tl
trotodlph
^
•O
0
,
2
^
^.
c
i
,
o
M
o
b
•^
c
it
c
'g
•
tl
M
C
v
•
c
5
1
1
a
^
.
S
rvi
in
,
IM
K.
O
e
|
M
^
|
2
^^
^»,
|
5
0
S
U
*J"
*?
•?
^
1
e
c
M
1.
O
u
a
0
^
e
i
S
e
o
^
•5
«M
•^
•»
S
C
C
M
c
c.
o
u
b
-
^.
1
8
0
•M
0
U
b
ft
•^
%
•
c
2
^
(^
1
e
£
:!
**\
4
,
wt
o>
,
O
k
o
^
*«
4W
•
+4
•^
w
i
c
41
C
tl
M
chlorobcn
a
_
«
C
4^
tf%
O
e
«*
•C
j
*^
^
!
|
1
M
chlorobcn;
a
a
^
_,
5
.£
2
o
0
•^
^
3
*-
•
g
c
-------
^
VI
u
*
o
§
^
s
?
o
i
QB
u
X
x
••
o
UJ
*
1
1
u
*
tt
u
tl
u
i
tl
(A
X
l/l
„
c
tf
-Oinitroben
^
i
in
(M
o
o
^"
o
w
•W
c
tl
tl
c
CD
CI
c
tl
Oinitrototu
•0
f>7
1M
a
rsi
s
-O
o
k
,^
c "
•5
^
n
^
X
8
,
OJ
§
^
,
ethenyl
ti
c
e
CD
I benzene
X
Ul
^
i
g
^
"x
•
*<
c
V
CD
•i
c
M
O
f.
u
K
X
_
,
i
^
,
o
c.
o
u
«r
«
|
•«i
•
.
3
>•
Kl
s
e
e.
I
*
c
ti
c
09
C
M
i
k
Kl
co
o
k
"c
\
tl
CD
tl
C
••
•chlorobenzi
e
»
K
^^
X
5
•
OJ
•°
•o"
u
u
3^
X
p
^%
•
u
I
M
- tl
~
**
*
•
^J
^
1
X
i
"x
5
Is.
3
w\
CO
k
"x
£
s
X
c
*
X
^
u
u
^
X
o
X
tenedicarl
i^
•
CD
IM
*"
tl
_
"x
c
"T
f^
s
•
k
•
^
|l|
*
'x
|
tl
^
U
u
_*
X
K
^}
k
U
•
~
^"
-------
u>
o
S
§
M
Ul
*BB
<
3
S
m
u
x
o
Ul
a.
a.
,
j
•c
(A
U
U
41
^Jj
X
f.
*
^
^
•O
41
in
*
"x
41
'
U
(J
Benienedi car boxy (
(M
_*
41
^l phthalat
•*
S
m
z
j;
^
j.
^
n
•
—
£
1
'
V
a :
u
X .
K
o
b
•
u
•5
c
M
e .
ni
41
i
0
c
e
S
N,
••
k
^
•
*
"x
u
o
^
•9
a
u
K
JS
h.
u
M
C
<\l
„"
1
U
0
to
•o
s
^
Bentencdiol
^
„"
c
S
X
JC
c
^
X
f.
\
a
I
«
o>
•5
chlorobenz
a
•o
3
*
*
o
t
o
~
u
•5
m
•
c
1
T>
1' Blph*nyl) *.*'
.
-
-------
VI
o
m
I
I
X
25 tu
- 5
u-T o
X |
K ta
x
0
1
f
1
M
•C
vt
u
s
*
u
V
M
I/I
V
c
.2
M
C
1
•C
A*
1
a
Kl
„
§
z
X
O
*rf
1
.'
KT
*
c
•
•5
^
«
•*
~
"x
g
f
«
4f
c
•5
•4
|
^
*«
•f
i
o
Kl
^
S
^
X
1
J
s
.
I
i
•5
•* •
^
~
"x
e
r .
—
"x
c
41
|
5
•
^
•o
o>
*
5
-*
~
X
e
I
M
__
2
M
S
N.
«o
3!
c
8
•5
^
v
^
~
"x
e
?
•^
u
c
tl
•5
^
1
o
f
u
•
K
II
^
•O
N.
O
a
u
2
f
*
K»
(N
^
•
«T
g
•5
3
t-"
^
C
._
•o
1
^*
0
e
JS
o
«
s
•o
.
o
e
u
•
.
n
C
,3 Butadi
^
c
'i
^
X
I
c
i
e
=
Kl
«
1
»
b
e
u
'e
*
£
»
•
2
§
3
^
O
X
£
V
X
Kl
•>
Is.
e
3
(M
e
~ .
^
•
^
o
o
u
a
-"
i
^
K.
m
e
.
UJ
.
e
u
o
^
u
?
•*
^
•uttn«,
tM
'
~
•
0
§
•
o»
f^
o
••
1
u
-
•
e
i
u
"m
~
^3
f»
e
^
i
U
U
I
•»
3
•B
|
O
«•
e
2
•^
3
•5
i
u
r
*«
o
S
I
Kl
5
e
•*
r
tm
e.
u
|
^
^
o
M
1
X
b.
^
u
-•
0
—
1
f
3
o
s
1
u
—
o
^
1
•
o
o
£
L.
e
u
-------
in
i
m
ui
i
a
i
0.
o.
J
1
u
^f
•K
5
*
S
m
u
• M
£
tl
M
X
-S
'c
X
u
in
«•
^
ti
•o
c
X
u
tl
c
o
c
1
M
e
a
V»
IA
S
•
I:
^
^-
^
c
—
"3
K
O
U-
X
u
fM
U
X
i
,
^
*
•o
1^"
V
i
x"
f^
V
'**
«
o
o
£
U
f
IA
>r
*o
•,
f^
•
*
|
*
o
u
X
u
u
•
N.
m
.
*
•O
in
^
»
*
<£.
^
3
**
«
e
e
£
u
f
^^
•*
(^
»
^4
•
s
4
K
tl
•§
U
X
u
u
X
n
•o
0
^
1
2
•o
in
o-
a
•o
in
J
*
V
(M
8
«•
**
•
0
o
£
U
f
«
in
•*
m"
«
IM
•
C
4|
2
£
O
U
X
u
i
e
I
*•
|A
*
%^
*•
S
m
m"
•
§*
•
S
%
1
u
X
u
ni
ti
c
V
•
i
o
u
X
htoroc
u
K
«f
«f
i
O
0
u
0
K
*
.
m
m
•i
.
^
•
*
ti
>^
^
•«
o
u
X
u
^
*•
c
u
i.
£
e
£
•*
1
o
PO
1
o
K>
c
U
4
k
£
£
•
z
M
1
o
1
1
c
•c
o
•o
a
o
N
C
1
4.
O
£
U
4.
NT
*
fO
*
n
•o
«•
o
•o
•
o
o
J{
4.
•
,
•0
~
e"
•••
K
o
•S
^^
•^
^
7
A
o
c
^
a
W1
c
K
O
•5
a
rodibcnzo-
o
£
U
^
K
•
X
tA
C
o
•o
a
orodibcnzo
£
U
•
C
•
a
C
K
O
7
a
orodibcnzo
£
U
^
—
•.hlpyrcnc
o
C
j£
a
o
3
e>
ryscrt*
u
^
*
£
O
a
c
3
o
M
c
^
a
3
(V
S
u
3
O
M
a
i
c
4.
3
rodibenzofi
0
£
U
K
X
-------
5
19
S
|
g 2
* i
in o
x |
X <3
X
*™
a
Ul
Q.
Q.
*
*
8
s
c
1
u
•C
VI
u
I
m
u
•i
M
X
VI
c
<^
o
M
1
1
o
u
c
••
W
c
b.
3
<4-
O
M
c
1
O
£
u
i.
•-
c
c
2
o
N.
O
»
^
^
«
* *
•o •
.01
"^ •
^r^^*
**^
.1"
c —
1.
K •
O O
«• u
•?z
Kt £
«M* M
M u
A O
•^ *
t0
f^
^ *
C r»
e >
\ «•> •
* <5 "
rt""1"^
^J
^
CM
C
|
*
N
§
S
10
.
*^
in
^
•
"^
e
L.
e
f
u
K ^
^~^*"
O 0
o*i
•- «9
"*-l
W."S.
**.^
fm ^
^
81
« •*
ITS
jg ^
^^
•^ 4
g i
o o
*i
8 5
a jc
•9 •
^
,
•—
^
*
•1
-------
^
VI
<9
I
i
X
•c
Ul
<
.
1
o
X
X
UJ
CL
|
^
|
U
•c
VI
u
S
u
~
E
<
C
o
^^
X
K
0
£
chloroct
CM
w
,
C
£
Ul
c
X
£
s
0
£
u
o
^
^
ir>
JC
o
•dichlor
«
,
2
£
Ul
g
£
^
0
O
f
u
a
%
I
k
^
S
•o
UJ
«
1
o
•dichlor
1M
^
C
£
Ul
a§
k
O
^J
U
t
^
^
O
O
u
^
r
•>
£
Ul
^
e
£
««
•
o
k
o
1
*V
i^,
^
t
2
01
,
e
e
•
^
S
«f
£
Ul
£
£
s
k
£
U
k
•6
O
it
O
O
£
U
k
»
2
£
Ml
i
f
s
|
^f
o
•*
s
-------
_
I/I
u
i
;
o
*
•c
IU
•»
O
i
s
X
X
o
>
Ul
a.
X
|
9
u
*
ac
u
|
X
u
a
«i
X
O
3
-.
I
^
^»
N.
-O
uorene
~
X
s
nj
•
•0
•»
N.
y-i
g
O
•
z
•V
41
C
M
9
U
2 Oiphenylhyd
^
f«>
^
*
0.
•?
o»
e
X
a
u
r«
o
c
"g
—
o
*•
§
_
-o
2
^
m
^
e
o
w
O
•
j
^
•
3>
^,
£
•o
a
-<
^»
e
1
I
X
^
•
o>
o>
^
u
2
at
X
''
^^
ing*n«s« (tot*
X
^
•
§
£
fi
M
C
a
e
X
e
X
u
X
•o
^te
o>
m
K.
X
u
3
U
i
X
c
'i
^
NltrosodtMth
M
*
S
3
e
e
X
jg
M
•
i
S
5
«
X
j
^
^
S
s
!
*
£
%
X
^
I
•o«od
-------
i_
VI
«
*
•C
o
t-
i
m
Ul
i
o
1
•
X
X
g
&
^
|
•
|
u
X
•c
VI
u
*
X
u
i
tft
X
VI
c
^~
i
o
u
o
IM
S
K
o
k
O
U
•5
•*
c
•
£.
X
C
|
k
O
^
•5
o
0
u
o
^
"i
•
!C
o
o -
1
o
u
•5
«
.g
X
1
1
_
• ,,
•o
•
f»
i
•*
c
^
41
X
4
o
u
k
I
kl
ID
fM
•
IT!
O
k
o
1
k
*»
•rf
ft*
g*
ft
g
^
1
i
_
(M
— ^
u ft
K •
*? °x
2 1^
« 9
o
e
££
u •
K «
ft^v
£^
2^^
O
-------
in o
x o
&
a.
1
u
•c
v>
u
J
u
Systenal
|
e
e
2
£
trT
•".
Kl
tn
*
c
o
2
"•
e
u
UJ
^ •
*
s
^^L
f> O
-•O
•" ^
(M •
••0
ji
*• ^.
u •
•
if\
O
M
O
£
e
"o
o
„
c
'i
"x
i
K.
ru
2
C
1
I
w
c
•
'x
1
(VI
«
»
5
*
8
rg
j
o
-------
o
i
s s
X 8
VI
<
u
Sys
Q
e
c.
8
s
Phenol
o
c
I
e
•5
•o
3
U
.
s
s
,
o
1.
•1
•5
ropheno
U
0
•
IM
S
O
IM
0
0
f.
u
i
IM
Phenol,
j s
^0
\
I
0
•
•0
o
i
1
•5
•*
IM
Phenol ,
^
tylpheno
^
|
0
•
«
in
o
'x
|
•5
IM
Phenol ,
1
"c
0
-
-
S
in
e
1
•*
"*"
Phenol,
g
•
:hloroph
-
^
•
•
g;
o
o
«
r
•*
*•"
1
1
M
:hloroph
r
^
•
-
j
8
Ichioro
V.
*•
4
•#
*•
Phono I ,
^^
>ropheno
u
o
-
5
fe
o
u
•5
•A
*
1
"o
u
O
^
X
.
tft
t
•tf
X
1
fl
o
o
1
*
-------
»»
M
•
i
> I
1 I
a
x B
x
M
i
a.
a.
..
j
g
j
u
•t
VI
u
1
u
V)
"o
b
u
b
^
•i
i
_
t
"o
V
a.
|
^
|
^
e
o
o
•
.
b
~c
-r
*
"o
o
a.
"3
hlorophc
U
S
Ok.
m
2
IV.
•a
0
b
e
u
i
|
a.
M
b
e
a.
IM
8
s
IN
b
•
"
£
I
hylthio)i
J
^^
"
M
^
X
•5
b
o*
^
u
u
totphorodi thl<
a.
§
O
<^
"3
a
I
n
b
M
'x
5
'o
Y
1^
"
**
^
X
•5
b
o
•o
u
u
Ho»phorodithl<
a.
b
3
I
Ik
£
«M
m
„
e
5
—
)«ul forty
g
^
1
"5^
g
i
•5
•
2
° •
2 ;
U
Notprtorothioit
0.0-dlMtl
a.
S
^
*«
b
•.
At
2
<«
in
•
•
^^
I
^
C
^*
O
"x
•
•5
b
*
o
u
hosphorothlol<
a.
o
o
f
o
X
M
b
X
a.
IN
b
"x •
^ •
i °
a "~
b
o
IM
K.
IM
b
M
X
~
O
"x
•o
b
o"
u
e
o
o
a.
'
|
JC
S.
^«
X
x
a
S
i
b
z
•
2
tropritny
•J
,
•*
b
x
I
•5
b
o
•o
£A
hosprtorothloii
a.
•
•5
osoplper
^
*
S
<«
b
e
(
o
o
b
^
j
a.
^
O
i
^
o
£
K.
§
b
*"•
I
o
a.
c
5
S
ropylnit
a
e
a
IN
•O
1
&
.
e
S
b
"c
*
1
}.
••
1
a.
b
o
u
i»
i
^v
0
»-
;
*
o
b
e
2
u
i
i
b
•
1
chloropr
i^
a
*»
«
^
t
e
0
u
b
a.
c
1
e
Trichlor
t
-------
*•
VI
^
X
i
I
i i
s s
K 0
X
K
i
IU
a
.
j
o
u
*
at
VI
u
i
l
u
s
w
X
VI
b
•I
5
4^
1
b
oroitop
.c
u
•1
0
.-
s
•9
e
o
b
0
JE
o
,_
X
o
^
II
c
a
0
b
b
]
^
c
„
N.
O>
^
b
^m
_c
•5
c
a
o
b
;8
x
u
IK
««
Ul
o
o
^
.
b
*«
"c
a.
o
b
b
^
I
}
a
u
M
h>
JS
CM
•«
,
O
b
£
U
jt\
•
b
•
0
b
K
.^
VI
c
o-
X
1
1
b
o
u
b
.
^^
V
*^
ft
V
0
u
I
a
o
b
a.
u
4 •
0)
0
1
"x
a
o
a
I
ja
•5
CM
b
^
r.
K.
-------
i
VI
O
O
i
3
S
§
X
K
i
Ul
j
o
m
¥»
U
|
u
1
X
V»
*
X
k
u
dj
f
f
i*
•
•o
IM
S
•*
*
«••
X
I
*
"x
«
.
IM
.g"
U
U
'ropeno
Ok
IM
^
0
u
~~
•o
JO
e
"o
«l .
a
o
IM
•^
O
1
•
V.
*
f*.
o
"o
c
\
IM
b
X
e
o
e
5
c
a.
8
•a
b
x
-
2
e
\
•5
X
a.
2
u
,
S
o
J,
_*•
I
IM
*
C
•5
X
Ob
„
c
•5
"o
b
Mitrocopy
•
IM
m
O
o>
0
o
b
'c
-
»
g
•5
b
X
a
^
£
^
O
*4
OMC
!
]
7
IM
"f
IM
ff
K.
e
VI
4*1
^
^
0
—
r
•j*
e
•
«
^^
^
«j
0
§
o
?
ni
e
S
VI
|
«•*
"3
IM
S
!
-------
v>
^
O
i
m
§
oc
UJ
<
^
s
§
o
M
X
0
*
IW
&
&
«
1!
S
s
|
u
>
flE
«rt
<
j
U
**
^
;
«
X
(M
CM
•O
o
-*
~»
f^
i
•5
c
>
4
4*
e
M
•^
u
c
ISI
•O
-O
•o
0
^
P4
-------
APPENDIX D
TASK PCRCE SAMPLE RESULTS
-------
.
*«
R
s:
.1.
ii
I!s:E
til
ii
i
^11
= * |!
y. u ?• i
«i -.
,r IK
S
S
I!
i
s
'I
i
-------
CVS
!!:!;'
Ill
C> *•
. SB
lil
•<> IB •!
P * S
ia*
!
!r
1
• * r
E
s: f s s S; s; S: $• * • s- «• £! *• «• fc
J; R ?; £ s: s: K i 1; r 1.
C. £ 5! r R S: t - S
••"••£•»••> *•
. • r
— t
R K: 2 f:
1
— *
:===.- . ,.. T
Jj S* £ ** ^ ^* £ S! S* S
~* r* ^* OT f» ** |H m •• V
* fi
-. -i
mm ^« j-
iff |£g|
1 (liif ill
ini
|; j; ». J- j; «
*' c- *"' P»! I9»«« «
^
5
SI
I
-------
E'":5
ffcl1
*
E s-
liii
"I;
rr
f» n>
IS
•
*
5; =
Hi
** * ***
|
i
* Ii!
Vi
•^
kgf
«**
r*«
-
ii * i"ig
ITi *
I • PP
*•
« I -•«! « "'IN'
£ f: 1C 9 f: R |*f t
1 I - | Its
E Jt;
s *
LA
CM
^ :
I a -
E S? 5i W
ilil
ft *
ES
i *
I
S
i
i
-------
leii
III
PSS
'
•.1-.
<> -<* r^i
Sis;
It r
Si* 8j II
ft
2!
fr;
*
o
I
CM
5* iii
~ * ?. 5 5
!£i
5 s !
f
££
C
=1
t
i
I
).,
-------
f; r. f.
1
k*J
£
Jif
i. I
. «.
l-Ei
I'Ei
lip
FIB
fg£
Igt
r?s
i'gi
5--Z!
fg£
Ip
III
*§i
Z -:
£ R f : *•- s
li *"
«*»
JE; £ i: - £
ft
- | ««-
; £ «
i* s *
R
"» — i
«. «• «> t.. H
r **
7 •"•
._.
%-
iriilil
liiifii m
i (III!
if li
f: -«;-£•; t f:
| c. s- ?; t
|. 5 R J - 1; £ ;• 5 £ f;
? s£~c'~r ?
1- ' !'!« ^ |
I" IT Tl
|- I'H= sl I
|S 1 Is 1 |
r MM* i i
sl* s I"!*5 *^ 1
§** «•••*• «*l «• O
*». gt ^i - gt j»
Is i IB i i
iiiii i J!i ifiii
*- *
^•' 0 W^ ^i
• • ft ^ Wl
lT> » W>
""' "' ?'
= ='
fi »' »».
t
-
BB*
S
s
I i 1 |
^ i
CM
<
99^)?>»)>JI)>'J
-------
•a
i*l
Is i!
ill
i;r*
§ii
PIS
let
ill
Z* ** K
lit
R " *
li i
CR E
,!|
" r 1 1« i
* ii" * p "$
g; Si- IK y*
1; 2 *
K |:=! *"-fsi 1:
| ?: " a |
« r i P •!
» i § ix s
•• i- • ii i
• |- •»!• ••
- i ii* i
" fc s is i
ai
5 !
i
1 s: i:l fj M"
Pi SI *' "'
| a 1 | " | | "
Si f» i?! K -*
| •• | Mil"
|: » • » f R |: {: |: F,
1 i Ills
£ ? s Jl:*: ^
3 i * |: r.
••i i*; ^< ?.
i l| ' !•' K
I R * I * " III "
ft f s f i: 8: =
9* «• *• <• —
s £ = e -
1 1 • |H 2
111 isls ISiili i:ii^s?i
ii
CO
OJ
-------
/T\ GLOSSARY' OF DATA QUALIFIER CODES (OPGANIC)
^ CODES RELATING JO IDENTIFICATION
(confidence concerning presence or absence oi compound*) i
' v
U •» Not detected. The associated number indicate*
approximate sample concentration necessary to be
detected.
. ,' (NO CODE) « Con-firmed identi-fication.
9 m Not detected substantially above the level reported
in laboratory or -field blanks.
F = Unreliable result. Analyte may or may not be present
in the sample. Supporting data necessary to con-firm
result.
N = Tentative identification. Consider present. Special
methods may be needed to con-firm its presence or
absence in -future sampling e-f-forts.
3 CODES RELATED IP. QUANT I TAT I ON
(can be used -for both positive results and sample quantitation
limits)i
J » Analyte present. Reported value may not be accurate
or precise.
K s Analyte present. Reported value may be biased high.
Actual value is expected to be lower.
L - Analyte present. Reported value may be biased low.
Actual value is expected to be higher.
UJ = Not detected, quantitation limit may be inaccurate
or imprecise.
UL • Not detected, quantitation limit is probably higher.
OTHER CODES
0 • No analytical result,
-------
OF DM!A QUALIFIER CODES (INORGANIC)
:CD£5 DELATING TO IDENTIFICATION
tre»on f idenca concerning presence or absence ot compound^ M
u = Not detected. The assoclated^numoer indicates
approximate sample concentration necessary to oe
detected.
',NO CODE) = Con-firmed iden 11 -f icat ion.
B * Not detected substantially above the level repc»- tea
in laboratory or -field blanks.
R = Unreliable result. Analyte may or .r.a- net oe Present
in the sample. Supporting data necessary to confirm
result.
r-j PELATED TO QUANTITATIQN
I oe used tor botn oositiv* results and sample quant i tat ion
it*1i
J - Anal/te present. Reported value may not be accurate
or precise.
'- - Aral.ts prasenr. Reported value may be biased hiqh.
Actual -sl'js is e"peered to be lower.
L •= -inal.-rs pre=ent. Deported /i>lue ma*- oe biased low.
Actual /3lue 1= e-pscted to be nigher.
L] = Anal.C'3 cresent . As values approacn the IDL
the quantitation may not oe accurate.
uJ - Not detected, quantitstion limit may b* inaccurate
or imprecise.
UL = Not detected, quantitation limit is probably higner,
0 fHER CODES
Q - No analytical -- = = ult.
-------
APPENDIX E
EPflA QUALITY CdSfTROL REPORT
-------
pro
.1-ssearcn Corooraucn
EVALUATION OF QUALITY CONTROL ATTENDANT
TO THE ANALYSIS OF SAMPLES FROM THE
AMOCO OIL FACILITY, VIRGINIA
FINAL MEMORANDUM
Prepared for
U.S. ENVIRONMENTAL PROTECTION AGENCY
Office of Waste Programs Enforcement
Washington, D.C. 20460
Work Assignment No.
EPA Region
Site No.
Date Prepared
Contract No.
PRC No.
Prepared By
Telephone No.
EPA Primary Contact
Telephone No.
548
Headquarters
N/A
June 30, 1987
68-01-7037
015-05481603
PRC Environmental
Management, Inc.
(Ken Partymiller)
(713) 292-7568
Rich Steimle
(202) 382-7912
-------
MEMORANDUM
DATE: June 30, 1987
SUBJECT: Evaluation of Quality Control Attendant to the Analysis of Samples
from the Amoco, Virginia Facility
FROM: Ken Partymiller, Chemist
PRC Environmental Management
THRU: Paul H. Friedman, Chemist*
TO: HWGWTF: Richard Steimle, HWGWTF*
Gareth Pearson (EPA 8231)*
Pat FCrantz, Region III
Mark Fillipini, Region IX
Christopher Thomas, Region III
This memo summarizes the evaluation of the quality control data generated by
the Hazardous Waste Ground-Water Task Force (HWGWTF) contract analytical
laboratories (1). This evaluation and subsequent conclusions pertain to the data
from the Amoco, Virginia sampling effort by the Hazardous Waste Ground-Water
Task Force.
The objective of this evaluation is to give users of the analytical data a more
precise understanding of the limitations of the data as well as their appropriate use.
A second objective is to identify weaknesses in the data generation process for
correction. This correction may act on future analyses at this or other sites.
The evaluation was carried out on information provided in the accompanying
quality control reports (2-5) which contain raw data, statistically transformed data,
and graphically transformed data.
The evaluation process consisted of three steps. Step one consisted of
generation of a package which presents the results of quality control procedures,
including the generation of data quality indicators, synopses of statistical indicators,
and the results of technical qualifier inspections. A report on the results of the
performance evaluation standards analyzed by the laboratory was also generated.
Step two was an independent examination of the quality control package and the
HWGWTF Data Evaluation Committee Member
-------
performance evaluation sample results by members of the Data Evaluation
Committee. This was followed by a meeting (teleconference) of the Data Evaluation
Committee to discuss the foregoing data and data presentations. These discussions
were to come to a consensus, if possible, concerning the appropriate use of the data
within the context of the HWGWTF objectives. The discussions were also to detect
and discuss specific or general inadequacies of the data and to determine if these
are correctable or inherent in the analytical process.
Preface
j'
The data user should review the pertinent materials contained in the
accompanying reports (2-5). Questions generated in the interpretation of these data
relative to sampling and analysis should be referred to Rich Steimle of the
Hazardous Waste Ground-Water Task Force.
I. Site Overview
The Amoco Oil facility is a refinery located in Yorktown, Virginia which has
been in operation for approximately thirty years ago. The facility has three
landfarms of which only one is being used. The wastes which have been landfarmed
include API separator sludge, heat exchange sludge, slop oil emulsions, MEA
(monoethanolamine), leaded and unleaded tank bottoms, etc.
The facility monitoring wells monitor the two closed and one open landfarms.
The existing ground-water monitoring data indicates the presence of heavy metal
and organic contamination down gradient of at least one of the landfarms.
Twenty-one field samples including a field blank (MQA779), an equipment blank
(MQA780), a trip blank (MQA766), and a pair of duplicate samples (well CQ12-38,
samples MQA775 and MQA776) were collected at this facility. Sample MQA770 was
designated by the sampling team as a medium concentration ground-water sample.
All other samples were designated as low concentration ground-water samples. All
samples were analyzed for all HWGWTF Phase 3 analytes with the exception of
samples MQA770, 771, and 783 which were not analyzed for dioxins and
dibenzofurans.
II. Evaluation of Quality Control Data and Analytical Data
1.0 Metal?
1.1 Metals OC Evaluation
Total and dissolved metal spike recoveries were calculated for twenty-four
metals spiked into two low concentration samples (MQA768 and 776). Twenty total
metal average spike recoveries from these samples were within the data quality
objectives (DQOs) for this Program. The total barium, cadmium, iron, and silver
average spike recoveries were outside the DQO with values of 70, 63, 58, and 53
percent, respectively. One of the total aluminum spike recoveries and one of the
total iron spike recoveries were not calculated because the sample results were
greater than four times the amount of spike added. Seven individual total metal
spike recoveries from the low concentration matrix samples were also outside DQO.
This information is listed in Tables 3-la and 3-2a of Reference 2 as well as in the
following Sections. A listing of which samples were spiked for each analyte is also
available in Table 3-2a of Reference 2.
-------
All twenty-four dissolved metal average spike recoveries from the two low
concentration matrix spike samples (MQA768 and 776) were within the data quality
objectives (DQOs) for this Program. Four individual dissolved metal spike recoveries
from these samples were also outside DQO. One each of the dissolved calcium,
iron, magnesium, manganese, and sodium spike recoveries were not calculated
because the sample results were greater than four times the amount of spike added.
This information is listed in Tables 3-lb and 3-2b of Reference 2 as well as in the
following Sections. A listing of which samples were spiked for each analyte is also
available in Table 3-2b of Reference 2.
-r j
No medium concentration samples were spiked for either total or dissolved
metals.
The calculable average relative percent differences (RPDs) for all metallic
analytes were within Program DQOs. RPDs were not calculated for about two-thirds
of the metal analytes because the concentrations of many of the metals in the field
samples used for the RPD determination were less than the CRDL and thus were not
required, or in some cases, not possible to be calculated.
Required analyses were performed on all metals samples submitted to the
laboratory.
No metal analyte contamination was reported in the laboratory or field blanks.
1.2 Furnace Metals
The quality control for the graphite furnace metals (antimony, arsenic,
cadmium, lead, selenium, and thallium) was generally acceptable.
The total antimony, cadmium, and lead and the dissolved arsenic, cadmium, and
thallium spike recoveries for spiked sample MQA776 were outside DQO with values
of 55, 24, 48, 74, 66, and 158 percent, respectively. All results for these metals
should be considered semi-quantitative at best except for the total cadmium and
lead results which should be considered qualitative at best.
The correlation coefficient for the method of standard addition (MSA) analysis
of total lead in samples MQA770, 775, and 782 was below DQO. Dissolved lead
results for samples MQA770 and 775 should be considered qualitative and those for
sample MQA782 should not be used.
MSA analyses should have been performed on total cadmium in sample MQA775
and on dissolved lead in samples MQA773 and 783. Results for these metals in
these samples should be considered semi-quantitative except for dissolved lead in
sample MQA773 which should be considered qualitative.
The precision for the duplicate injection of total selenium in samples MQA775
and 783 was above DQO. Total selenium results for sample MQA775 should be
considered semi-quantitative while results for sample MQA783 should be considered
qualitative.
An initial calibration verification (ICV) and a continuing calibration
verification (CCV) for dissolved arsenic had to be rerun. All dissolved arsenic data
should be considered semi-quantitative.
Several CCVs were reported by the laboratory as failed in the analyses for
total and dissolved cadmium and dissolved lead. As a result, total cadmium results
-------
for samples MQA766, 767, 768, 769, 783, and 784; dissolved cadmium results for
samples MQA775, 780, and 781; and dissolved lead results for samples MQA768, 769, _
772, 775, 780, 781, 782, and 785 should be considered semi-quantitative. For these
and other reasons dissolved lead results for sample MQA773 should be considered
qualitative.
A continuing calibration blank (CCB) for total cadmium was out of DQO. Total
cadmium results for samples MQA782 and 786 were affected and should be
considered semi-quantitative.
j *
All total arsenic and thallium and dissolved antimony and selenium results
should be considered quantitative. Total cadmium and selenium and dissolved lead
results should be considered quantitative with the exceptions listed below. All total
antimony and dissolved arsenic, cadmium, and thallium results should be considered
semi-quantitative. Dissolved lead results for samples MQA768, 769, 772, 775, 780,
781, 782, and 785, total selenium results for sample MQA775, and total cadmium
results for samples MQA782 and 786 should be considered semi-quantitative. Total
lead results, with exceptions, total cadmium results for samples MQA770, 776, and
777, dissolved lead results for sample MQA773, and total selenium results for sample
MQA783 should be considered qualitative. Total lead results for sample MQA782
should not be used. The usability of all graphite furnace analytes is summarized in
Section 4.0 and 4.1 at the end of this Report.
1.3 ICP Metals
The matrix spike recoveries for dissolved potassium and total barium, silver,
and tin in sample MQA776 and total iron in sample MQA768 were outside DQO with
recoveries of 126, 40, 10, 44, and 58 percent, respectively. As a rule, high spike
recoveries indicate a high bias in the data and low recoveries indicate a low bias.
Dissolved potassium and total iron results should be considered semi-quantitative.
Total barium and tin results should be considered qualitative. Total silver results
should not be used.
The low level (twice CRDL) linear range check for all total and dissolved
chromium results exhibited high recoveries of 15 to 40 percent. The low level
linear range check is an analysis of a solution with elemental concentrations near
the detection limit. The range check analysis shows the accuracy which can be
expected by the method for results near the detection limits. The accuracy
reported for these metals is not unexpected.
The serial dilution percent differences for total and dissolved barium in sample
MQA776 were above DQO. Results for total and dissolved barium should be
considered semi-quantitative with a high bias.
High sulfate concentration in samples MQA775, 776, 778, 782, and 783 may
have caused a low bias in the barium results for these samples.
Precision results for dissolved zinc in field duplicate sample pair MQA775/776
were excessive. The comparative precision of field duplicate results is not used in
the usability evaluation of sample results. It is not possible to determine the
source of this imprecision. The poor precision may be reflective of sample to
sample variation rather than actual sampling variations. Therefore, field duplicate
precision is reported for informational purposes only.
All total and dissolved aluminum, beryllium, calcium, chromium, cobalt, copper,
magnesium, manganese, nickel, sodium, vanadium, and zinc results should be
-------
considered quantitative. Dissolved barium and potassium and total iron results
should be considered semi-quantitative. Total barium and tin results should be
considered qualitative. Total silver results should not be used. The usability of all
total and dissolved ICP metal analytes is summarized in Section 4.2 and 4.3 at the
end of this Report.
1.4 Mercury
Total and dissolved mercury results should be considered quantitative with an
acceptable probability of false negatives.
2.0 Inorganic and Indicator Analvtes
2.1 Inorganic and Indicator Analvte QC Evaluation
The average spike recoveries of all of the inorganic and indicator analytes,
except for POC and sulfate in the low concentration samples (only the low
concentration samples were spiked), were within the accuracy DQOs. Accuracy DQOs
have not been established for the bromide, fluoride, nitrite nitrogen, and sulfide
matrix spikes.
Average RPDs for all inorganic and indicator analytes were within Program
DQOs. The RPDs were not calculated if either one or both of the duplicate values
were less than the CRDL. Precision DQOs have not been established for bromide,
fluoride, nitrite nitrogen, and sulfide.
Requested analyses were performed on all samples for the inorganic and
indicator analytes.
No laboratory blank contamination was reported for any inorganic or indicator
analyte. The trip blank (MQA766) contained 15 ug/L of TOX and 61,000 ug/L of
sulfide contamination. The TOX CRDL is 5 ug/L and the sulfide CRDL is 1000
ug/L.
2.2 Inorganic and Indicator Analvte Data
All results for cyanide, bromide, chloride, fluoride, sulfate, total phenols, TOC,
and POX should be considered quantitative with an acceptable probability of false
negatives.
The holding times for the nitrate and nitrite nitrogen analyses ranged from 6
to 12 days from receipt of the samples which is longer than the recommended 48
hour holding time for unpreservcd samples. All nitrate and nitrite nitrogen results
should be considered semi-quantitative.
Sulfide matrix spike recoveries from samples MQA776 and 778 were low with
values of 10 and 81 percent. Sulfide contamination was found in the trip blank
(MQA766) at a concentration of 61,000 ug/L. The sulfide CRDL is 1000 ug/L. As a
HWGWTF convention, all positive sulfide results five times this amount or less
should not be used, all results five to ten times this amount should be considered
qualitative, and all positive results greater than ten times this amount, as well as
all negative results, should be considered quantitative. Therefore, sulfide results for
samples MQA766, 779, and 780 (the sampling blanks) and samples MQA769 and 785
(the negative sulfide results) should be considered quantitative and all other positive
sulfide results should not be used.
-------
Calibration verification standards for POC were not analyzed. A POC spike
solution was run during the analytical batch but the "true" value of the spike was
not provided by the laboratory. EPA needs to supply the inorganic laboratory with
a POC calibration verification solution. Until then, the instrument calibration can
not be assessed. Additionally, the laboratory failed to run final calibration blanks
at the end of all three of the analytical batches and continuing calibration blanks
during the first of the three analytical batches, as required. The matrix spike
recovery for POC from sample MQA776 was below DQO with a recovery of 46
percent. The POC results should be considered qualitative.
v J
TOX contamination was found in the trip blank, sample MQA727, at a
concentration of 15 ug/L. The TOX CRDL is 5 ug/L. As a HWGWTF convention,
all positive TOX results five times this concentration or less should not be used, all
TOX results between five and ten times the greatest of the values should be
considered qualitative, and all results ten times the level of contamination or
greater, as well as all negative results, should be considered quantitative.
Additionally, high chloride concentrations in samples MQA775, 776, 778, and 783 may
have enhanced the TOX concentration measured in those samples. Therefore, TOX
results for samples MQA766, 769, 779, 780, and 783 should be considered quantitative
and all other TOX results should not be used.
3.0 Organics and Pesticides
3.1 Organic OC Evaluation
All matrix spike average recoveries, with the exception of the semivolatile
matrix spikes which were not analyzed, were within established Program DQOs for
accuracy. Individual matrix spike recoveries which were outside the accuracy DQO
will be discussed in the appropriate Sections below.
All surrogate spike average recoveries, with the exception of the semivolatile
matrix spike and matrix spike duplicate samples which were not analyzed, were
within DQOs for accuracy. Individual surrogate spike recoveries which were outside
the accuracy DQO will be discussed in the appropriate Sections below.
All reported matrix spike/matrix spike duplicate average RPDs were within
Program DQOs for precision. Individual matrix spike RPDs which were outside the
precision DQO will be discussed in the appropriate Sections below.
All average surrogate spike RPDs, with the exception of that for the matrix
spike and matrix spike duplicate compounds which were not analyzed, were within
DQOs for precision. No surrogate standard was used or required for the herbicide
analyses.
No semivolatile matrix spike or matrix spike duplicates were analyzed.
Laboratory (method) and sampling blank contamination was reported for
organics and is discussed in Reference 4 as well as the appropriate Sections below.
Detection limits for the organic fractions are summarized in Reference 4 as
well as the appropriate Sections below.
3.2 Volatiles
Quality control data indicate that the volatile organics were determined
acceptably. The chromatograms appear acceptable. Initial and continuing
-------
calibrations, tunings and mass calibrations, matrix spikes and matrix spike
duplicates, and surrogate spikes were generally acceptable. Laboratory blank
contamination and excessive holding times were reported.
The analytical laboratory exceeded the volatile holding time of seven days for
12 of the 21 samples. Holding times for samples MQA766, 775, 776, 777, 779, 780,
781, 782, 783, 784, 785, and 786 ranged from 26 to 34 days in excess of the allowed
seven day holding time. Volatile results for these samples should be considered
unreliable.
j j
Laboratory (method) blanks MB-1, MB-2, and MB-5 contained methylene
chloride contamination. This common laboratory contaminant was present at
concentrations of 3 to 9 ug/L. The methylene chloride CRDL is 5 ug/L. All
positive methylene chloride results should not be used due to this laboratory blank
contamination. Laboratory (method) blanks MB-3, MB-4, and MB-5 contained
acetone at concentrations of 1 to 4 ug/L. The acetone CRDL is 10 ug/L. All
positive acetone results should not be used due to this laboratory blank
contamination.
The organic analytical laboratory did not use the contract specified primary
ions to quantitate results for many of the HSL compounds on one (Finnigan OWA)
of its instruments. This has no affect on the results. The laboratory has been
made aware of this discrepancy and is correcting it for future analyses.
Estimated method detection limits were CRDL for all samples except MQA770
which was 20 times CRDL. Dilution of this sample was required due to the high
concentration of organics. The volatile compound results should be considered
quantitative for samples MQA767 through 774 and 778. Results for all other samples
should be considered unreliable due to excessive holding times. AH positive
methylene chloride results and acetone results should be considered unusable. The
probability of false negative results is acceptable with the exceptions of sample
MQA770 and the samples analyzed after excessive holding times.
3.3 Semivolatiles
Initial and continuing calibrations, tuning and mass calibrations, surrogate
spikes, and chromatograms were acceptable for the semivolatiles.
No matrix spikes or matrix spike duplicates were analyzed for the semivolatile
samples. Because of this lack of precision verification, all semivolatile results
should be considered semi-quantitative at best.
The analytical laboratory exceeded the semivolatile 40 day maximum holding
time between extraction and analysis for J_2 of the 21 samples. Holding times for
samples MQA766, 767, 771, 777, 779, 780, 781, 782, 783, 784, 785, and 786 ranged
from 10 to 25 days in excess of the allowed 40 day holding time. Semivolatile
results for these samples should be considered semi-quantitative at best.
The acid surrogate spike recoveries of phenol-D5, 2-fluorophenol, and 2,4,6-
tribromophenol from samples MQA778 and 778RE (its reanalysis) ranged from no
recovery to 4 percent recovery. These results are outside DQO. The acid fraction
results for sample MQA778 should be considered unreliable.
Four of the semivolatile laboratory (method) blanks, MB-2, MB-3, MB-4, and
MB-5, contained bis(2-ethylhexyl)phthalate contamination at concentrations of 2 to 4
ug/L. The bis(2-ethylhexyl)phthalate CRDL is 10 ug/L. Positive bis(2-
-------
ethylhexyl)phthalate results for samples MQA766, 767, 768, 771, 773, 774, 775, 778,
780, and 786 should not be used. Bis(2-ethylhexyl)phthalate results for samples
MQA777, 779, 781, 782, 783, 784, and 785 should be considered unreliable/All other
positive (samples MQA770 and 776) and all negative bis(2-ethylhexyl)phthalate results
should be considered semi-quantitative. Phenol contamination was found in MB-3 at
a concentration of I ug/L. The phenol CRDL is 10 ug/L. All positive phenol
results (samples MQA766, 772, 781, 782, and 783) should not be used. Negative
phenol results should be considered semi-quantitative except for sample MQA778 for
which the entire acid fraction should be considered unreliable.
The organic analytical laboratory is not using the contract specified primary
ions to quantitate results for 2,4,6-tribromophenol and 4-nitrophenol.
Sample MQA770 was diluted by a factor of 20 prior to analysis. The extract
exhibited the general chromatographic characteristics of a petroleum oil. False
negatives for this sample are a possibility due to high sample dilution and mass
spectral interferences from the oil matrix.
Due to a dilution factor of 2.0 for all other samples, the estimated detection
limits for the semivolatiles were approximately twice the CRDL.
The semivolatile data are acceptable and the results should be considered semi-
quantitative for all samples with exceptions. Acid fraction results for sample
MQA778 should be considered unreliable as all three acid surrogate recoveries were
out of DQO. The positive bis(2-ethylhexyl)phthalate results, except for samples
MQA770, 776, 777, 779, 781, 782, 783, 784, and 785, and all positive phenol results
should not be used due to laboratory blank contamination.
3.4 Pesticides
The initial calibrations, matrix spike/matrix spike duplicates, surrogate spikes,
blanks, holding times, and chromatography for pesticides were acceptable.
Heptachlor was reported at a concentration of 0.06 ug/L in sample MQA767.
This value is very close to the heptachlor instrument detection limit of 0.05 ug/L.
The value may be a false positive as the peak specified by the laboratory to be
representative of heptachlor had a retention time very slightly outside the
laboratory established retention time window. Additionally, there is no historic
evidence of chlorinated pesticides being disposed of at this facility.
On confirmation column DB17 the dibutylchlorendate retention time shift was
outside DQO for samples MB-2 (a method blank), MQA773 and 778, and for standard
INDA II.
The estimated method detection limits for the pesticides analyses is the CRDL.
The pesticides results should be considered quantitative with the exception of the
heptachlor result in sample MQA767 which should be considered unreliable due to
possible misidentification.
3.5 Herbicides
Matrix spike/matrix spike duplicates, method blanks, and holding times were
acceptable for the herbicide analyses. No surrogate standard was used for the
chloroherbicides or organo-phosphorous herbicide analyses.
-------
The herbicides for which the laboratory analyzed include only 2,4-D, 2,4,5-T,
2,4,5-TP, chlorobenzilate, phorate, disulfoton, parathion, and famphur.
Numerous chloroherbicide peaks were observed in the method, field, and trip
blank chromatograms. Some of the blanks were reported to have chloroherbicides
present. The tentative identification and quantification of chloroherbicides in all
samples should be considered unreliable due to this blank contamination.
The laboratory failed to use the 3-point external standard calibration method
for the herbicides. A one point method was used.'
Method 8150 is not adequate for the determination of chlorobenzilate.
Chlorobenzilate could be more accurately determined by using the pesticide method.
Chloroherbicide standard chromatograms were specified by the laboratory to be
representative of the four chloroherbicides for which the laboratory analyzed.
However, five peaks were observed in the chromatograms. The fifth peak may have
arisen from the derivatization of chlorobenzilate. There is the possibility that both
the methyl and ethyl esters of chlorobenzilate are formed when using Method 615.
The organo-phosphorous herbicide results should be considered qualitative due
to the lack of herbicide surrogates and confirmation column analyses. The
chloroherbicide results should be considered unreliable due to blank contamination
and the absence of surrogate analyses. The estimated method detection limits were
the CRDL for the herbicides.
3.6 Dioxins and Dibenzofurans
Dioxin and dibenzofuran spike recovery from the fortified blank sample and the
spiked field sample ranged from 80 to 117 percent which is considered to be
acceptable accuracy. No performance evaluation standard was required or evaluated
for dioxins and dibenzofurans. No precision (RPD) information was available as no
dioxins were detected in the laboratory or field duplicate samples. Required
dioxin/dibenzofuran analyses were performed on all samples submitted to the
laboratory except for samples MQA770, 771, and 783. No contamination was found
in the laboratory (method) or field blanks.
Due to a method modification supplied to the laboratory by the Sample
Management Office, the column performance check solution was not analyzed by the
laboratory.
The laboratory analyzed one blank spiked sample and one spiked field sample
(MQA776). They failed, however, to analyze the field sample prior to spiking. In
this case, none of the target analytes were found in the duplicate field sample so
there was no impact on the data usability.
Results for sample MQA786 should be considered qualitative. The sample was
re-extracted due to severe matrix interference. The re-extracted sample did not
meet the carbon-13 ion ratio criteria.
Samples MQA767, 768, 768-D (duplicate), 769, 772, 774, 776, 777, 780, 781, 784,
785, and 786 did not meet the DQO requirement for resolution of the percent valley
being less than or equal to 25 percent. The results for these samples should be
considered semi-quantitative.
The dioxin and dibenzofuran results, with exceptions, should be considered to
be semi-quantitative because the method precision has not been established. Results
-------
for sample MQA786 should be considered qualitative as the re-extract for this
sample was out of DQO. Results for sample MQA766 should not be used as the
internal standards were obscured and matrix interferences were observed at various
masses. The probability of false negative results, with the exception of sample
MQA766, is acceptable. No dioxins or dibenzofurans were reported in any of the
field samples.
III. Data Usability Summary
4.0 Graphite Furnace Metals. Total
Quantitative: all arsenic and thallium results; cadmium and selenium
results with exceptions
Semi-quantitative: all antimony results; cadmium results for samples MQA782
and 786; selenium results for sample MQA775
Qualitative: lead results with exceptions; cadmium results for samples
MQA770, 776, and 777; selenium results for sample MQA783
Unusable: lead results for sample MQA782
4.1 Graphite Furnace Metals. Dissolved
Quantitative: all antimony and selenium results; lead results with
exceptions
Semi-quantitative: all arsenic, cadmium, and thallium results; lead results
for samples MQA768, 769, 772, 775, 780, 781, 782, and 785
Qualitative: lead results for sample MQA773
4.2 ICP Metals. Total
Quantitative: all aluminum, beryllium, calcium, chromium, cobalt, copper,
magnesium, manganese, nickel, potassium, sodium, vanadium,
and zinc results
Semi-quantitative: all iron results
Qualitative: all barium and tin results
Unusable: all silver results
4.3 ICP Metals. Dissolved
Quantitative: all aluminum, beryllium, calcium, chromium, cobalt, copper,
iron, magnesium, manganese, nickel, silver, sodium, tin,
vanadium, and zinc results
Semi-quantitative: all barium and potassium results
4.4 Mercury
Quantitative: all mercury results
-------
4.5 Inorganic and Indicator Analvtes
Quantitative:
Semi-quantitative:
Qualitative:
Unusable:
all cyanide, bromide, chloride, fluoride, sulfate, total
phenols, TOC, and POX results; sulfide results for samples
MQA766, 769, 779, 780, and 785; TOX results for samples
MQA766, 769, 779, 780, and 783
all nitrate and nitrite nitrogen results
all POC results
sulfide and TOX results wijh exceptions
4.6 Qreanics
Quantitative:
Semi-quantitative:
Qualitative:
Unreliable:
volatile results for samples MQA767 through 774 and MQA778
with the exceptions of positive methylene chloride and
acetone results; pesticide results with an exception
semivolatile results with exceptions
all organo-phosphorous herbicide results
volatile results for samples MQA766, 775 through 777, and
779 through 786; semivolatile acid fraction results for
sample MQA778; bis(2-ethylhexyl)phthalate (a semivolatile)
results for samples MQA777, 779, 781, 782, 783, 784, and
785; heptachlor (a pesticide) result for sample MQA767;
all chloroherbicide results
Unusable: all positive methylene chloride and acetone (volatiles)
results; all positive phenol (a semivolatile) results;
positive bis(2-ethylhexyl)phthalate (a semivolatile) results for
samples MQA766, 767, 768, 771, 773, 774, 775, 778, 780, and 786
4.7 Dioxins and Dibenzofurans
Semi-quantitative:
Qualitative:
Unusable:
dioxin and dibenzofuran results with exceptions
results for sample MQA786
results for sample MQA766
-------
IV. References
1. Organic Analyses: EMSI -^
2421 West Hillcrest Drive
Newbury Park, CA 91320
(805) 388-5700
Inorganic and Indicator Analyses:
Centec Laboratories
P.O. Box 956
2160 Industrial Drive
Salem, VA 24153
(703) 387-3995
Dioxin/Dibenzofuran Analyses:
CompuChem Laboratories, Inc.
P.O. Box 12652
3308 Chapel Hill/Nelson Highway
Research Triangle Park, NC 27709
(919) 549-8263
2. Draft Quality Control Data Evaluation Report (Assessment of the Usability of
the Data Generated) for Case E-2363HQ, Site 53A, Amoco, VA, 5/6/87, Prepared by
Lockheed Engineering and Management Services Company, Inc., for the US EPA
Hazardous Waste Ground-Water Task Force.
3. Draft Inorganic Data Usability Audit Report, for Case 2363HQ, Amoco, Virginia,
Prepared by Laboratory Performance Monitoring Group, Lockheed Engineering and
Management Services Co., Las Vegas, Nevada, for US EPA, EMSL/Las Vegas,
5/5/1987.
4. Draft Organic Data Usability Audit Report, for Case E-2363HQ, Amoco, Virginia,
Prepared by Laboratory Performance Monitoring Group, Lockheed Engineering and
Management Services Co., Las Vegas, Nevada, for US EPA, EMSL/Las Vegas,
5/5/1987.
5. Draft Dioxin/Dibenzofuran Usability Audit Report, for Case E-2363HQ, Amoco,
Virginia, Prepared by Laboratory Performance Monitoring Group, Lockheed
Engineering and Management Services Co., Las Vegas, Nevada, for US EPA,
EMSL/Las Vegas, 5/5/1987.
------- |