EPA 510-K-92-812
Report No. KT-88-003(R)
FUEL VAPOR BACKGROUND
CONCENTRATION MEASUREMENT
AND TRACER TESTING
IN UNDERGROUND
STORAGE TANK BACKFILL
SAN JOSE, CALIFORNIA
Revised March 1988
Printed on Recycled Paper
-------
f
-------
: Report No. KT-88-003(R)
REPORT
FUEL VAPOR BACKGROUND CONCENTRATION MEASUREMENT
AND TRACER TESTING
IN UNDERGROUND STORAGE TANK BACKFILL
San Jose, California
Revised March 1988
Subcontract 216-8900-9
Submitted to:
Midwest Research Institute
5201 Leesburg Pike
Suite 209
Falls Church, Virginia
22041
By:
Kaman Tempo
816 State Street
Santa Barbara, California
93101
(805)963-6479
-------
-------
EXECUTIVE SUMMARY
This study was performed in two parts: 1) measurement of the fuel vapor
concentrations in underground tank backfill, and 2) development of tracer
testing methods for determing the efficacy of vapor monitoring systems. All
sites were located in the City of San Jose, California.
Background Concentration Measurement
Total fuel hydrocarbon vapor concentration was measured in 75 existing
vapor wells at 11 service stations and one emergency power generation site.
The wells were located within tank and pipeline backfill. Tank contents were
limited to gasoline and diesel.
Concentration was measured downhole, one foot from the bottom of the
well, using a Photovac TIP-I connected to a 25-foot length of 1/16" ID PTFE
tubing. Readings ranged from 0 to 2,000 ppm, with a mean value of 300 ppm.
Concentrations were reported by volume as isobutylene.
The following conclusions can be made from the background measurements:
• Background concentrations are commonly greater in tank backfill than
in pipeline backfill. Kaman Tempo believes that this is ascribable
to the common occurance of tank overfills.
t Stratification of soil vapor within tank backfill can occur. If a
well is perforated over a short interval, leakage in an unscreened
stratum may be undetected. Therefore, each vapor well should be
perforated over the entire backfilled depth.
Tracer Testing
All sites selected for tracer tests had very low fuel vapor background
concentrations. Measured volumes of the tracers were injected into a "source"
well after which concentration was monitored for several hours in "observa-
tion" wells within the pipeline and tank backfill.
-------
Three tracers were used: sulfur hexafluoride (SFg), fluorocarbon-11
(F-ll) and propane. SFg was detected by a thermal conductivity meter, whereas
F-ll and propane were detected using a dual-column gas chromatograph equipped
with a photoionization detector. The most conclusive data obtained was for
propane. Propane travel velocities were observed to be on the order of 0.5
feet per hour. SFg travelled more rapidly at approximately 7 feet per hour.
F-ll proved to be a poor tracer due to its poor detectability.
A method was developed for predicting a tracer's concentration at any
distance from the source well at any time after its release using a mathemat-
ical model. Such a model can aid the design of well spacings and the detector
selection for future monitoring systems.
The following conclusions can be made from the tracer tests:
• Vapor wells can be effectively used to monitor underground product
releases from both tanks and pipelines.
• Because tracers were not detected on the side of the tank opposite
the source well, in multiple tank installations, shorter well
spacings are recommended in the direction transverse to the tank
axis than in the longitudinal direction.
• Additional testing is needed to relate the diffusion rates (and
concommittant detection time) of petroleum products and surrogate.
compounds used as tracers.
-n-
-------
TABLE OF CONTENTS
Section
1 INTRODUCTION ........................................... i'1
2 INVESTIGATIVE METHODS ....................... . .......... 2'1
21 DESCRIPTION OF VAPOR MONITORING WELLS ............ 2-1
22 BACKGROUND CONCENTRATION TESTING METHODS ......... 2-2
2.3 TRACER TESTING ............................... ----- 2-4
2.3.1 Field Procedures, Sites #1 and #2 ......... 2-9
2.3.2 Field Procedures, Site #3 ................. 2-12
3 RESULTS ................................................ 3'1
3 . 1 BACKGROUND MEASUREMENT RESULTS ................... 3-1
3.2 TRACER TEST RESULTS. ......................... .... 3-6
3.3 GAS DIFFUSION MODEL DEVELOPMENT AND FIELD TESTING. 3-8
4 CONCLUSIONS .............. . ............................. 4~1
Appendix
A SELECTED CITY OF SAN JOSE HAZARDOUS MATERIALS DIVISION
SITE INFORMATION SHEETS.. .............................. A~l
B LETTER OF AUTHORIZATION ................................ B-l
C ADDITIONAL SITE SKETCHES .............................. C-l
D VAPOR MONITORING EQUIPMENT SPECIFICATIONS .............. D-l
E SITE #3 TRACER TEST DATA ............................. . • E-l
F CURVE MATCHING SOLUTION OF INSTANTANEOUS LINE SOURCE
GAS DIFFUSION MODEL .................................... F-!
-------
-------
LIST.OF TABLES
Table Page
2-1 Physical Properties of Selected Gases 2-8
2-2 Detector Sensitivity for Selected Compounds 2-10
3-1 Background Fuel Vapor Concentrations at Underground
Tank' Sites 3-2
-------
-------
LIST OF FIGURES
Figure Page
2-1 Sketch of Typical Underground Storage Tank Site 2-3
2-2 Sketch of Sites #1 and #2 2-5
-2-3 Sketch of Site #3 2-6
3-1 Frequency Plots of Concentration Measured in Pipeline
and Tank Backfill 3-7
3-2 Time Versus SFg Concentration Plots 3-9
3-3 Time Versus Propane Concentration Plots 3-10
3-4 Selected Tracer Test Chromatograms 3-11
3-5 Well V-2, Test A - Solution of D1 and A by Curve
Matching Method 3-13
3-6 Well V-5, Test A - Solution of D1 and A by Curve
Matching Method 3-14
3-7 Well V-5, Test B - Solution of D1 and A by Curve
Matching Method 3-15
-------
-------
SECTION 1
INTRODUCTION
This report describes the methodology and results of background fuel
vapor measurements and gaseous tracer tests in Underground Storage Tank (UST)
backfill. The twofold purpose of the study was: 1) to expand the data base
of fuel vapor concentration in tank backfill, and 2) to develop methods for
determining the efficacy of vapor monitoring systems. Kaman Tempo performed
this study during November and December 1987 under subcontract to Midwest
Research Institute as a contribution to the EPA UST Federal Guidelines and
Regulations.
Tracer tests are commonly used to construct gas transport models and are
easily applied to fuel vapor applications. Such models can provide an approx-
imation of the volume and rate of a theoretical product leak detectable by the
lower detection threshold of the monitoring system. Ultimately, this informa-
tion can be used for system design and operation optimization and for deter-
mining regulatory performance standards.
In a previous study (Kaman, September 1987), Kaman Tempo measured back-
ground levels in tank backfill at known leak sites and demonstrated that these
concentrations were statistically higher than those measured at sites with no
leak history. These tests did not address vapor migration or timeliness of
leak detection. Bench scale tracer tests using actual field products have
been conducted by other workers to study migration rates; however, the appli-
cation of their results must assume boundary conditions that may differ from
actual tank conditions. Therefore, Kaman Tempo believes that on-site tracer
testing is the most valuable tool available for evaluating vapor migration in
tank backfill, excluding monitoring of actual leaks in progress, both for
testing of individual installations and for design purposes.
The field work for this study was conducted in the City of San Jose,
California, where vapor monitoring is required in the backfill of all existing
USTs. The San Jose Fire Department Hazardous Materials Program maintains
information sheets for each site which lists vapor well and tank construction
data. The sites visited and tested in this study were selected from these
1-1
-------
sheets. Seventy-five vapor wells at twelve service stations were measured for
fuel vapor concentration. Tracer tests were conducted at three of these
sites. All chemical analyses and measurements were performed with field-
portable equipment.
Kaman Tempo would like to thank Charles J. Wilhelm and Joseph C. Afong of
the City of San Jose Hazardous Materials Program for their cooperation and
assistance in the field portion of this study.
1-2
-------
SECTION.2
INVESTIGATIVE METHODS
Twelve test sites were selected from 58 site information sheets provided
by the City of San Jose according to the following criteria:
Sites
• distributed within a four mile radius
• minimum number of fuel company brand names
Wells
• uniform construction and design
• not instrumented with continuous monitoring devices.
These criteria minimized logistics, reduced the number of variables affecting
results, and minimized disturbance of the existing monitoring systems. Site
accessibility and operator cooperation were also optimized. Information
sheets for the selected sites are presented in Appendix A. Site access was
further aided by a Letter of Authorization from the City of San Jose which was
presented to the attendant of each station (see Appendix B).
The site information sheets contain entry spaces for facility address and
name, number of tanks, tank and piping construction data, well construction
data, soil analysis data, depth to groundwater, and vapor monitoring data.
Because the information sheet program is still being developed, many items on
the sheets, including age of tanks, depth to groundwater, and site leak
history, were not available for most of the sites.
2.1 DESCRIPTION OF VAPOR MONITORING WELLS
Total fuel hydrocarbon vapor concentration measurements were taken at 11
non-instrumented service stations and one instrumented emergency power genera-
tion site. In all, 75 wells were tested. All service stations were operated
either by Arco or Unocal. Vapor well configurations at these sites typically
included two or more wells within the tank backfill area, wells at the end of
pump islands in the pipeline backfill, a well in the vent line backfill, and a
2-1
-------
background well at some distance from pipelines or tanks (Figure 2-1). The
configuration of wells for Site Numbers 1, 2, and 3 are shown in Figures 2-2
and 2-3. Configurations of the remaining sites are presented in Appendix C.
Tank contents were primarily gasoline, although some diesel and waste oil
tanks were encountered.
The wells were constructed of either one- or two-inch nominal diameter
PVC pipe, slotted to within two to five feet below the surface and fitted at
the top with a threaded coupling and threaded plug. The majority of the wells
had a one-inch diameter. An uncapped i-inch hose fitting was threaded into a
hole drilled into either the side of the threaded coupling or top of the plug.
Well head assemblies were enclosed with steel traffic-rated monuments having
12-inch threaded steel tops. Some sites where measurements were not taken
included a lock box with a keyed padlock within the monument. Wells inter-
cepted the entire backfill depth interval. Wells within pipeline backfill
tended to be less than five feet deep, whereas tank backfill wells were as
deep as 18 feet.
2.2 BACKGROUND CONCENTRATION TESTING METHODS
Fuel vapor concentrations were measured downhole using a Photovac Model
TIP-I connected to a 20-foot, probe constructed of 1/16-inch I.D. PTFE tubing.
The intake end of the probe was weighted with a i-inch brass bell reducer
which acted as a centralizer to avoid aspirating water condensation from the
walls of the well casing into the instrument. TIP-I specifications are given
in Appendix D.
Prior to well measurement at each site, the instrument was zeroed to
ambient air and calibrated to 100 ppmV isobutylene. The same probe was used
for downhole measurement and instrument calibration in order to correct for
possible progressive contamination of the probe.
Each well was opened by removing the threaded PVC plug and then sounded
for water with an electric water level sounder. No downhole measurement was
made if the well was flooded. The total depth of the well was also recorded
and the well immediately recapped until ready for concentration measurement.
2-2
-------
Product Lines
i
/
Background Vapor Well
Vapor Well
Vapor Recovery Line
Backfill
Vapor Well In
Product Line Backfill
.Vapor Recovery
Lines
Vapor Wei!
In Tank Backfill
•Vent Lines
VENTS
Figure 2-1. Sketch of typical underground storage tank site.
2-3
-------
The concentration was first measured with the TIP probe held two inches
below the top of the casing. The probe was then slowly lowered at a rate of
about two feet per second to within one foot of the bottom of the well and the
concentration recorded at that depth. Concentration versus depth was recorded
in rare cases where unusual stratification was observed. In most cases,
concentration generally increased with depth as anticipated and concentration
versus depth was not recorded.
No attempt was made to aspirate the sample from the hose fitting on the
well head assembly due to the long pumping times anticipated to reach a stable
concentration level. Previous field experience has shown that more than ten
well casing volumes of soil vapor must be removed to achieve a stable pump-
discharge reading that suitably approximates a downhole reading obtained
without pumping (Kaman Tempo, March 1988).
2.3 TRACER TESTING
The two main criteria used to select sites for tracer testing were low
background concentration level and the presence of at least two vapor wells
within either the pipeline or tank backfill. Six of the twelve sites tested
for background concentration qualified as tracer candidate sites, and three
sites (Sites #1, #2, and #3) were actually tested. Sites #1 and #2 are
non-instrumented sites, whereas Site #3 is instrumented with an aspirating
system with a central control panel manufactured by Soil Sentry. The well.
configurations at the sites are shown in Figures 2-2 and 2-3.
Site #3 well head assemblies differed somewhat from the non-instrumented
sites. The instrumented wells were constructed of one inch PVC fitted with a
slip-cap connected to a filter and restricter valve, which was in turn fitted
with i-inch tubing to the aspirator pump and detector in the control panel.
Well monument boxes at Site #3 were sealed with a silicone sealant. In gen-
eral, these well monuments were drier than those at the non-instrumented
sites.
An ideal tracer compound for fuel leak studies should have transport
properties identical to one of the principal fuel components of environmental
concern (such as benzene, toluene, or xylene), be non-reactive, non-hazardous,..
2-4
-------
9-2
Bfanham Lane
Pump Island ]
Pump Island ]
c
-s-
n>
ro
i
ro
CO
rt-
n
o
-h
CO
CD
=«=
=tfc
ro
O
CO
3=>
-------
ro
7500 gallon |
Underground \/
Diesel Tank
2000 gallon
Above-
Ground
Water Tank
Fuel Lines
V-6
Concrete Pad
Covering Two
10,000 gallon
Diesel Tanks
V-5
Hospital Boiler Room,
Shops.Offices, etc.
Fuel Lines
Doors ', ! .
I
Soil Sentry
Control Panel
and Sensor
10 20 30 40 50 Feet
Figure 2-3. Sketch of Site #3.
-------
readily available and inexpensive, and detectable with field portable equip-
ment. For this study, the following compounds were used: sulfur hexafluoride
(SF6), trichlorof1uoromethane (F-ll), and propane, each of which satisfies a
unique set of the ideal criteria. The transport properties in soils of all
three tested tracers were unknown at the onset of this study. All are-non-
reactive with fuel components, comrnercially available and reasonably afford-
able in the needed quantities. In its highest commercially available purity
(99.99 percent) SFg is the most expensive of the tracers used. Purities of
the other tracers used were greater than 99.0 percent. Although detection
limits varied between compounds, all were detectable with field instruments.
Propane was selected largely because it is already in common use in evaluating
soil gas monitoring systems.
Because of significant disadvantages of propane as a tracer, the current
study attempts to evaluate the two other compounds as possible substitutes.
One of propane's drawbacks is that it is detectable by most monitoring sys-
tems. A tested system's alarm threshold level must be raised after propane is
used unless all residual propane is evacuated from the backfill after testing.
It is more desirable to use a tracer that does not degrade the effectiveness
of the monitoring system because threshold adjustments of this type can result
in false negative alarms. In addition, propane is a hazardous substance on
the basis of its flammability.
A comparison of physical properties of the selected tracers and compon-
ents of gasoline vapor is presented in Table 2-1. All three tracers would be
expected to diffuse more rapidly through tank backfill than actual fuel
components on the basis of their higher vapor pressures and Henry's Law
Constants and lower boiling points. The molecular weight and density of
propane approaches that of benzene, whereas F-ll and SFg more closely approach
the density and molecular weight of xylenes. The vapor pressures and Henry's
Law Constants for all three tracers are higher than those of fuel components
by a factor of about 100. The boiling points of the tracers are generally
much lower than those of the fuel components, F-ll having the closest approxi-
mation. Because both gasoline and F-ll are liquids at the expected soil
temperatures between 20°C and 24°C, F-ll's transport is of special interest.
2-7
-------
Table 2-1.
Physical Properties of Selected Gases*
Compound
TRACERS
Propane(2)
F-11®(2)
(Trichlorofluor
Photo-
ionization
Potential
MW fEW1} bp.-C
44.11 11.07 -42.1
137.37 11.77 23.8
•omethane)
Henry's Law Constant
Vapor
Pressure
p. kPa
941
88.5
(20°)
Solu-
bility
S. 9/m3
62.4
1100
(20°)
Density
a/I
1.969
(calc. 20°)
6.133
(calc. 20°)
kPa m3/mol
Experi-
Calculated mental
71.6
11.05 81.20
(20°) (20°)
Recom-
mended
71.6±2.4
..
SF6(3) 146.05 >12.0 -63.8* 753(3)
(Sulfurhexafluoride) 'Sublimates (-63.5°)
SELECTED COMPONENTS OF GASOLINE VAPOR
Sparingly 6.602
in water (3)
Benzene (2) 78.11 9.25
Toluene (2) 92.13 8.82
m-Xylene 106.2 8.56
o-Xylene 106.2 8.56
(2)
Afr.ffl = =
80.1 12.7 1780 3.487 0.557 0.562
(calc. 20°)
110.6 3.8 515 4.113 0.68 0.67
(calc. 20°)
139 1.1 162 4.741 0.721
(calc. 20°)
144.4 0.882 175 4.741 0.535
(calc. 20°)
1.200
0.550
±.025
0.670
±.035
0.700
±.1
0.500
±0.06
* Properties at 25°C except where noted.
1) From Photovac Technical Bulletin No. 11.
2) From: Dev5tt,D.A., Evans.R.B, Jury,W.A., and Starks.T.H.. 1986, Soil Gas Sensing for
Detection and Mapping of Volatile Organics, Draft, EPA Cooprative Agreement
CR 812189-01.
3) From: Handbook of Chemistry and Physics, 1973-1974, 54th edition.Chemical Rubber
Company Press, p B-143.
4) ibid, p D-187.
5) m-Xylene is the predominant xylene isomer in fuel products, whereas o-Xylene is of greater
concern to health.
6) From: Cummins, A.B.,and Given.LA., 1973, SME Mining Engineering Handbook, Society of
Mining Engineers, p 16-13.
2-8
-------
All three tracers and gasoline vapor :have Densities greater than air. SFg is
one of the heaviest known gases.
Tracer concentrations were measured with a Photovac Model 10S70 gas
chromatograph with photoionization detection (PID) and a Gowmac Model 20-600
gas analyzer with thermal conductivity detection (TCD). Both instruments are
described in detail in Appendix D. Table 2-2 shows the relative detector
sensitivity to these compounds. The PID detector is sensitive to propane,
F-ll and fuel components, whereas the TCD detector is sensitive to all of
these compounds. Therefore, the tracer tests with SFg had to be run before
the propane and F-ll tests, in order to avoid shifts in background level
during the duration of the test.
It should be noted that although the TIP and 10S70 have 10.6 eV detector
lamps, propane, whose photoionization potential is 11.0 eV, is reasonably
detectable (see Table 2-1). F-ll, which has an 11.7 eV photoionization
potential is much less detectable. The 12.0 eV ionization potential of SFg
renders it undetectable by photoionization.
The most sensitive detector of choice for the fluorinated compounds F-ll
and SF, would be an electron capture detector (ECD); however this detector
requires a degree of temperature control and a power supply not currently
available in portable equipment. The metal oxide semiconductor (MOS) type of
detector commonly used in most continuously monitoring equipment, including
that at Site #3, is sensitive to fuel components, propane and F-ll, but not to
SF6.
The gas chromatograph has the added feature of reporting tracer and
background fuel vapor concentrations as separate peaks. Therefore, any
possible diurnal changes in fuel vapor background levels do not influence
tracer test results.
2.3.1 Field Procedures, Sites #1 and #2
Preliminary tracer tests were conducted simultaneously at noninstrumented
Sites #1 and #2 in order to gauge the volume of tracer and test duration that
2-9
-------
TABLE 2-2. DETECTOR SENSITIVITY FOR SELECTED COMPOUNDS
Criteria
MOS detector sensitivity
PID detector sensitivity
TCD detector sensitivity
ECD detector sensitivity
Flamiability Hazard
Propane
Fll
BTX
0
Yes
+
+
u
0
0
+
+
No
No
0
Yes
Benzene, Toluene, Xylenes.
Explanation:
0 = Not detectable
+ = Detectable
-H- * Highly detectable
2-10
-------
would be required for a full scale test. Full scale tests were not conducted
at these sites because of automobile traffic through work areas during busi-
ness hours. SF-6 and F-ll were used, however, propane was not used so as not
to disturb the City's weekly monitoring of the sites.
In general, tests were performed by placing measured volumes of tracers
into a designated source well while monitoring concentration for several hours
in observation wells within the line or tank backfill. In order to avoid
convection effects, injection volumes were minimized and in no test was more
than one tracer injected at a time. Background concentrations were measured
in the observation wells prior to tracer injection in order to establish a
concentration base line.
At Site #1, a single test was performed using 100 ml of liquid F-ll.
Well V-4 was the source well and V-2 the observation well. Four readings,
including the background measurement were taken over a 115 hour interval. The
tracer was poured from the supply cylinder into a beaker and drawn into a 100
ml syringe. These preparations were made at a distance of 100 feet from the
test wells, in order to avoid surface contamination of the observation well.
The syringe was then connected to a 10 foot length of teflon tubing. The
tubing was lowered into the source well and the tracer injected opposite the
perforated section of well casing. The source well was then quickly capped
and the steel monument covered tightly.
In the observation well, measurements were made with the Photovac 10S70
probe at a depth of 14 feet. Using the internal pump on the 10S70, two tubing
volumes (30 ml) were removed to insure that the sample was representative of
downhole conditions rather than residual ambient air in the tubing. The
sample was then injected by the automated probe method (Appendix D) with an
injection volume of 1.0 ml and gain setting of 50 to 500. Immediately after
injection, the probe was removed from the well and the PVC plug and monument
cover replaced. The observation well was uncapped for a maximum of four
minutes at a time.
A pipeline and a tank backfill test were performed at Site .#2. The
pipeline backfill was tested by injecting 500 ml of liquid F-ll, using the
2-11
-------
same method as described for Site #1, into source well V-7. Observations were
made in Well V-8 for a period of approximately 24 hours. Wells V-7 and V-8
are located in the vapor recovery line backfill. Only two measurements,
including the background measurements, were made with the 10S70 for each well.
Measurements were taken in the same way described for Site #1.
The tank backfill at Site #2 was tested by injecting 10 liters of SFg gas
into well V-3 with three measurements taken in observation wells V-2 and V-4
over a 92 hour period. The SFg measurements required more sample preparation
in comparison to the F-ll measurements, because the Gowmac gas analyzer's
internal pump was not strong enough to draw a sample through the total length
of the sampling probe. Therefore, samples were collected in 5-layer bonded
sample bags using an oilless vacuum pump. Between each sample, bags were
flushed five times with ambient air and filled to capacity with ambient air.
The ambient air sample was then used to zero the Gowmac instrument. The probe
was then lowered to within one foot of the bottom of each well and connected
to the vacuum pump. The sample bag was then evacuated, the vacuum pump intake
connected to the probe, and the bag filled with sample gas. The sample bag
was then immediately connected to the instrument and the concentration
recorded.
2.3.2 Field Procedures, Site #3
Before tracer tests were conducted at Site #3, all wells were suction
tested for plugged perforations. An in-line vacuum gauge was placed between
the well head fitting and a vacuum pump. No response was observed on the
gauge during pumping, indicating good communication with the backfill. Strong
vacuum readings would indicate saturated or clayey soils having very low
permeabilities to gas flow. Had strong vacuums been observed, tracer testing
would not have been conducted at site #3.
Two complete sets of tracer tests were run, Test A during November 18
through November 20, 1987, and Test B during November 30 through December 2,
1987. In both tests, Wells V-l and V-4 were used as source wells, and Wells
V-2 and V-5 used as observation wells. Wells V-3 and V-6 were monitored as
2-12
-------
observation wells only during the first test set of tests. The major diffe-
rence in field conditions between sets was that the site's aspirating-type
automated monitoring system, manufactured by Soil Sentry, was in operation
during the first set. Each well was aspirated three times daily during the
first set, for 7 minutes at 1.8L/min, with cycles starting at midnight, 8:00
a.m. and 4:00 p.m. Therefore, 13 liters of soil gas were aspirated per cycle
at each well, or 5 to 20 well casing volumes, depending on depth. The second
set of tests was conducted to determine if the site's automated monitoring
system significantly affected the tracer test results.
During Test A, 500 ml of liquid F-ll and 7.5 liters of propane gas were
injected into Well V-4, whereas 5.6 liters of SFg and 3.75 liters of propane
were injected into Well V-l. During Test B, 7.5 liters of SFg and 15 liters
of propane were injected into Well V-4, whereas 3.75 liters of SFg and 3.75
liters of propane were injected into Well V-l.
As with Sites #1 and #2, tracers were not mixed in any test due to the
large injection volumes required and positive interference by propane to SFg
detection.
Test A measurements were taken according to the methodologies described
for Site #1. During Test B dedicated probes were placed in each well for the
duration of testing in order to minimize dilution by ambient surface air. The
dedicated probes also avoided cross contamination between observation wells.
During both sets of tests, the annular space between the well casing and probe
was sealed with a one-hole stopper. It was decided not to use F-ll in Test B
due to F-ll's poor detectability as shown by Test A.
2-13
-------
-------
SECTION 3
RESULTS
3.1 BACKGROUND MEASUREMENT RESULTS
Background fuel vapor concentrations were measured in 75 wells at 12
sites. The measured concentrations are reported with well and site obser-
vations in Table 3-1.
Downhole concentrations, regarded as representative of the concentration
in the backfill, range from 0 to 2,000 ppm, with a mean value of 300 ppm.
Concentrations are reported by volume as isobutylene.
Top of casing readings proved to be poor indicators of true soil condi-
tions as they were greatly influenced by dilution with surface air. For
example, concentrations less than 10 ppm were measured at Site #2 in wells
with downhole concentrations as high as 1,600 ppm. Conversely, the greatest
top of casing readings were obtained in wells giving the lowest downhole
readings at that same site.
Concentration in each .well generally increased with depth, with the
greatest concentration being measured in the bottom one-quarter of the per-
forated casing. The inverse condition was encountered only in Wells V-3, V-4
and V-5 at Site V-10. These three wells were among the deepest encountered in
the study. The inverse concentration gradient may be ascribable to the
partitioning and stratification of individual fuel components having different
TIP detection sensitivities within the well casing. Because the well casing
is screened throughout almost its entire length, the inverse concentration
gradient in the well may also represent actual stratification of backfill
contamination.
Measurements were unobtainable or questionable at six wells due to either
inaccessibility, broken casings or flooded conditions. Bentonite was encoun-
tered in several wells, indicating that the sandpack did not cover the entire
length of perforations.
3-1
-------
Tab!? 3-1. Background fuel vapor concentrations at
underground tank sites.
CO
ro
Site Well
# Address No.
1 858 W. Branham 1
2
(Tracer test site) 3
4
5
6
7
2 4610 Pearl Ave 1
2
(Tracer test site) 3
4
5
6
7
8
3 2105 Forest Ave 1
2
(Tracer test site) 3
4
5
6
7
4 1501 Parkmoor 1
2
3
4
5
6
Weil
Depth
fm
8
15
8
15
5
5
5
8
18
18
15
5
18
5
5
3
13
4
14
14
14
14
8
14.5
5
15
15
5
TIP
(Top of
Casing)
(ppm) ,
1
1
10
3
4.5
2
1
10
2
118
8
160
0
3
6
0
0
0
0
0
0
0
1
1
3
5
9
1
TIP
at
Depth
fpprrrt
30
15
9
15
6
3
1.5
9
33
160
15
-
-
22
16
0
0
0
0
0
0
0
1
3
5
9
300
240
350
8
Depth of
Meas-
urement
ffn
7
12
7
12
4
4
4
7
17
17
14
4
17
4
4
2
12
3
13
13
13
13
7
13
4
14
13
12
10
4
Well
Diameter
(Inches)
1
2
1
2
1
1
1
1
1
2
1
1
1
1
1
1
1
1
1
1
1
1
1
2
1
2
2
1
Product
Type
G
G
WO
G
G
G
G
R
R
R
SU
-
-
-
.
D
D
D
D
D
D
D
G
G
G
G
G
G
Remarks Ml
Well on supply line.
Well on supoly line.
Well on vent line.
Well on vent line.
Well on supply line.
Well on supply line.
Well on supply line.
Well on vent line.
Well on supply line.
Well on supply line.
Background well.
Well on supplv line.
Well
Condition
Vent line leak history.
Wet bentonite in hole.
Wet bentonite in hole. Flooded.
All wells fitted with Soil-
Sentry/ Genelco aspirated
monitoring system instru-
mentation.
(1) Wells are within tank backfill unless otherwise stated.
-------
Table 3-1 (Continued).
co
Site Well
# Address No.
5 1708Tully 1
2
3
4
5
6 2690 Union Ave 1
2
3
4
5
6
7 3010 Almaden Expr. 1
2
3
4
5
6
8 1152Tully 1
2
3
4
5
6
7
8
9 1405 Branham Ln. 1
2
Well
Depth
fftt
14
14
14
-
.
8
4.5
4.5
14.5
14.5
8
14
15
15
8
5
5
6
15
5
5
5
-
15
15
8
8
TIP
(Top of
Casing)
(ppm)
23
400
40
-
30
0
0
20
4
0
0
2
0
0
0
0
0
425
10
10
4
160
-
1
0
107
9
TIP
at
Depth
(ppm)
441
1808
1835
-
.
0
0
0
6
87
33
18
0
0
0
0
0
0
0
0
2000
250
10
6
800
-
1370
1050
270
30
Depth of
Meas-
urement
rm
12
12
12
-
.
7
3
3
2
13
10
5
13
7
10
14
14
7
4
4
5
12
4
4
4
-
12
12
7
7
Well
Diameter
finches)
2
2
2
1
1
1
1
1
2
2
1
2
2
2
1
1
1
1
2
1
1
1
-
2
2
1
1
Product
Type
G
G
G
G
G
G
G
G
G
G
G
G
G
G
G
G
G
G
G
G
G
G
G
G
G
WO
WO
Well
Remarks Condition
Well on supply line. PVC bushing stuck.
Well on supply line. Well flooded.
Background well.
Well on supply line. Pipe broken.
Well on supply line.
Concentrations at this site
verified w/ 40 ppm isobutyl-
ene standard.
Background well.
Well on supply line.
Well on supply line.
Pipe broken.
Well on supply line.
Well on supply line.
Steel lid stuck.
-------
Table 3-1 (Continued).
to
Site Well
# Address No.
10 4995 Almaden Expr. 1
2
3
4
5
6
7
11 1722 Meridian 1
2
3
4
5
6
Well
Depth
ifrt
8
15
15
15
15
5
8
8
15
14.5
15.5
5
5
TIP
(Top of
Casing)
(ppm)
82
110
950
116
225
23
10
2
4
0.6
Q
Q
56
TIP
at
Depth
(ppm)
176
400
600
700
750
900
500
780
800
800
240
48
62
85
105
120
158
130
15
0
4
6.5
60
85
105
60
6
0
0
110
Depth of
Meas-
urement
rm
7
14
13
12
9
6
3
12
10
9
5
13
12
11
10
9
6
3
4
7
7
14
13
12
10
7.5
5
14
4
4
Well
Diameter
(Inches)
1
1
1
1
2
1
1
1
1
2
1
1
1
Product
Type
G
G
G
G
G
G
G
G
G
G
G
G
G
Well
Remarks Condition
Background well.
Note inverse concentration
stratification. (500 ppm in
monument)
(40 ppm in monument)
Strong odor.
Note inverse concentration
stratification.
Well on supply line.
Well on supply line.
Ambient air =11.0 ppm.
Background Well.
Greatest reading 4 feet
from bottom of well.
No Stratification observed.
Well on supply line.
Well on supply line. 6" bentonite on bottom.
-------
CO
I
en
Table 3-1 (Cont'd)
Site
# Address
12 3147Senter
Well
No.
1
2
3
4
5
6
7_
Well
Depth
(ft)
15
15
5
15
5
14
7
TIP
(Top of
Casing)
(ppm)
1
0
4
0
10
1
2
TIP
at
Depth
Ippm)
1600
1300
8
1000
50
1500
16
Depth of
Meas-
urement
(ftt
12
12
4
12
4
12
6
Well
Diameter
(inches)
1
1
1
2
1
1
1
Product
Type Remarks
G
G
G Well on supply line.
G
G Well on supply line.
G
G
Well
Conditbn
-------
Fuel throughput data for each site are not available at this writing,
however, those sites having the greatest background concentrations were
self-service stations with heavy business traffic during our site visits. The
lowest concentrations measured in the study were at Site #3, whose tanks are
only rarely used and certainly have the lowest throughput of sites visited.
Leak history information is also incomplete at this time. According to
the Site #2 station operator, the highest background concentration was mea-
sured in Well V-3 adjacent to a tank having a vent line leak history. The
lowest concentrations measured in the study were at Site #3, the newest of the
tank installations where a leak was highly unlikely.
A comparison of concentrations measured in pipeline and tank backfill is
presented in Figure 3-1. Product, vent and vapor recovery pipeline backfill
were considered equally in the comparison. Most concentrations were less than
20 ppm in both pipeline and tank backfill. The tank backfill concentration
data exhibited the highest overall values and most uneven frequency
distribution.
3.2 TRACER TEST RESULTS
Of the three compounds tested, propane yielded the most conclusive data.
Propane was used extensively at Site #3. The greater effectiveness of propane
was largely due to its detectability with the Photovac 10S70. With the
methods used in this study, the detection limits for propane, SFg, and F-ll'
were 0.01 ppmV, 0.5 percent and 1 to 10 ppmV respectively. The detection
limit for F-ll increases greatly with lowering temperature below 65°F (18°C).
Ambient temperatures varied between 62° and 72°F during the tests. As a
result, F-ll vapor was never detected in the backfill, including measurements
in source wells.
Sites n and 12
In the preliminary tracer test at Site #2, SFg was observed to travel
from Well V-3 to V-4 in less than 17 hours, exhibiting a 4% concentration
increase. The distance between wells was 15 feet. There were not enough data
to define tracer breakthrough in the other wells at Site #2.
3-6
-------
500-1000
PIPELINE BACKFILL DATA
100-200
E
f
.o
1 50-100
1
o 20-50
O '
0-20
02468 1012141618202224262830
Frequency
TANK BACKFILL DATA
E
a
_Q.
I
1
I
5
02468 1012141618202224262830
Frequency
Figure 3-1. Frequency plots of concentration measured in pipeline and
tank backfill.
3-7
-------
Site #3
Conclusive tracer test data were obtained for SFg and propane in tests at
Site #3 (see Appendix E). Time versus concentration plots for SFg and propane
for these tests are presented in Figures 3-2 and 3-3. Selected chromatograms
from Site #3 tracer tests are presented in Figure 3-4. A rise in SFg concen-
tration was observed in approximately two hours in both Well V-2 during Test A
and Well V-5 during Test B. During Test A, SFg was not used at observation
well V-5/source well V-4. No distinct concentration rise could be interpreted
from the Test B SFg well V-2 data.
Over a distance of 15 feet, propane exhibited travel times between 24 and
30 hours in all tests except in Test B at Well V-2. This propane data is
generally more conclusive due to the increased sampling frequency made pos-
sible by using the 10S70 in an automatic cyclic sampling mode.
The right-hand sides of the field data plots have steep declining limbs
because the sources were instantaneous. Had a constant source been used, the
declining limb would be missing and concentration would remain relatively
constant after breakthrough. The other ramifications of the use of instantan-
eous versus constant sources upon data interpretation are discussed in the
following section.
There was no breakthrough in Wells V-3 or V-6 from either SFg or propane.
Apparently the density of the tracer gas prevented movement into V-3, whose
total depth is 12 feet above source well V-4. Lack of breakthrough in V-6 is
likely due to the tanks themselves, which may have provided a barrier to flow.
The tanks are installed parallel to a line drawn between V-4 and V-5. There-
fore, the flow path to V-6 from source well V-4 is more tortuous than to V-5.
3.3 GAS DIFFUSION MODEL DEVELOPMENT AND FIELD TESTING
Several continuous point source models, such as that of Weeks and others
(1982), have been tested. However, these models do not satisfy the boundary
conditions imposed by the subject field tests. It is therefore necessary to
develop a new practical mathematical model to predict the concentration of a
gas in backfill at any time and radius from an instantaneous line source of
3-8
-------
CO
I
10
CQ
tz
n>
to
ro
to
TI
en
o
o
3
O
-J
OJ
O
rt
CO
CO
I
O)
ro
o .
ro .
d°"
3 co
3 o
CO
00
IO
o
10
to
ro
*.
ro
o>
to
oo
09
o
ro
i
SF6 (%)
*> OJ
SF6 Concentration (%)
00
I
a
DO
3!
I
Ul
O
« s
s
o
I
00
I
en
I
*.
10
O
lO
*>
01
at
CO
ro
o
10
10
10
CD
10
00
CO
o
Geneico System aspirated
7 min every 8 hrs.
T=5.42 hrs
a
>
Tl
O)
2.
I
a
w
-------
Test A, Propane Well V-2 Data
0.4'
0.3-
ss
£
e
n.
0.2-
0.1 -
T=23.25 hrs
Genelco System Aspirated
7 min every 8 hrs.
T=3.00
T 1 1 1 1 1 1 1 1 1 1 1 1 T
-202468 10121416182022242628303234363840424446
Tlme(Hrs)
Test A, Propane Well V-5 Data
•20246 810121416182022242628303234363840424446
Tlme(hrs)
Test B, Propane Well V-5 Data
>
2.3.0
§ 2.0
I
-•- Propane Concentration
* Time Probe Disconnected
x Time Probe Removed From Well
-20 24 6 810121416182022242628303234363840424446
Time (hrs)
Figure 3-3. Time vs. Propane concentration Plots.
3-10
-------
j» • :•"
-. , r
v • v -
» 2
I a j
' :n
i 5*
WML-Sti * . rs-I 5C.-PL£a
rE""ESl»T'-FE 21 IBi'EC
G*If* 1000 "J'RSCP'VIK
* *
ig
i
.* « JJd.J
WV1Pt.E RUT* OEC 1 1387 8:
IWW.TSI3 • 7 23-5 SWPLES
tE"PERl»rUfi£ 30 iBSEC
(*:N iaaa OWBOPKK
1 u
SIWLt RUN DEC 1 138; 3- 6
KKOLTSIS * 3 23-S SWPLE3
TBrESuriJKE 31 IBSEC
3«1M leee CIWBOPKCK
•JNKNCUM
'JWNOUM
•JNKNCi»<
i IS.' 392.= «oS
3 3«.8 58 = «JS
5 12.3 31.0 -^9
3 163.3 7.« ..•:
'••:.' < '•SB.?
< OEC
"SIS • « 23-5
IE 35 13SE
iaaa
fl;t 1
t
j \ '-'=•
\ 5**
!h
i CTP
• •» ^ •••^a 31 _ .
u
: «
;5
> 16
« tf
a 21
V 9 ^?8.8 ^''C
!i
/P
5 il
« 23
~~
-J.i i3^.3 -JS
VHTI.E RUN PEC I 1387 77:32 Sdnr-E RUN PEC 1 138: ~n V
«NM.TS;S < 3 2!-=. Sf
-------
known length and mass. The model's development and application procedures are
described in Appendix F. The intent of the model is to aid in the design of
adequate well spacings and detector sensitivities in tank and pipeline moni-
toring systems, and to test the efficacy of existing systems.
A line source model was required for this study because of the great
thickness over which the source wells are perforated. The main drawback of
point source tests is the very low emission rate and concomitant low detection
limits (high detector sensitivity) required.
For additional simplicity, it was required that the source be instan-
taneous. In continuous source tests, controlling the source release rate can
be a complex task. However, an instantaneous source is easily approximated by
injecting a volume of tracer into a vapor well.
By applying the instantaneous line source model to the tracer concentra-
tion versus time data, the effective diffusion coefficient, D', and the
sorbtion corrected porosity, A, are estimated. These parameters characterize
both the tracer and particular backfill material.
Average values of D1 and A for propane calculated from the test results
in the pea-gravel backfill at Site #3 are 0.02 cm2/s and 30% respectively.
The field data plots for Site #3, Wells V-2 and V-5, showing type curve match
points, are presented in Figures 3-5, 3-6 and 3-7. The two curve fits shown
in each field data plot represent the range of possible solutions.
The Test B well V-5 test (Figure 3-7) yields the most reliable solution
because of the plentiful data on the rising limb of the curve. The model's
type curve only describes the rising portion of field data.
The identification of rising-limb data during curve-fitting was greatly
aided by referring to the arithmetic plots (Figure 3-3). Because only two
rising points existed for both wells in Test A, off-plot coordinates using
zero concentrations had to be used to achieve a solution. The V-2 solution
required off-plot coordinates (2.87, 0). The V-5 solution required off-plot
coordinates (5.69, 0).
3-12
-------
CO
I
10
0
O)
E
10
IU
'2
*
O 0
10 3
10
-4
10
0
E
Cuve
SOLLfTION
u«1Q2, V(u)«10-6, Q= 77.5mg/cm
EiLl: r2/t=i8 cm2/sec
C*t«=0.01 mg sec/cm3
D' « 0.014 cm2/sec
A * 0.043
Fit 2: r2/t = 20 cm2/sec
C*t • 0.03 mg sec/cm^
D' = 0.016 cm2/s*c
A « 0.013
10'
10'
rz/t (cm2/sec)
Figure 3-5. Well V-2, Test A - Solution of D1 and A by the curve matching method.
-------
C*t (mg sec/cm )
_L CO
o
o
o
I
ro
o
ro
GO
I
cn
i
un
-H
n>
in
00
O
C
<-+
o
3
O
-h
O
01
=5
D.
o —k
3 °
io ro
o
CO
o
-5
S
ft-
o_
IQ
i
•
1
•
c
•
•
•
,
•
i
_—- -~
B
j
--"
\
\
-
••"V
i
\.
•o
o
o
> q c
••
„ H H
Pb'-
0 g I>
S <" =
o *2
1.?
« c
a i
o (
Ef B
E -
' To ?
: S <
> *• T K
* jk. " *i
i _ -«. —
, i ° i
'1!? -* i
r «
L w
* CO
3
o"
3
o
o.
-------
CO
i
en
O
O
0.023 cm2/«ec
A a 0.191
Fit 2: r2/t « 0.32 cm2/sec
CM » 0.058 mg sec/cm3
D' * 0.026 cm2/««c
A x 0.359
um m
m
10
10
0
.10
1
r2/t (cm2/sec)
10'
10'
Figure 3-7. Well V-5, Test B - Solution of D1 and A by the curve matching method.
-------
Once obtained, D1 and A can be used to predict the concentration of vapor
in backfill at any time in distance from a point source typical of a fuel leak
using a point source model. The relationship between fuel and tracer diffu-
sion parameters must first be known in order to apply the model to fuel leaks.
This relationship would be obtainable by relatively simply benchscale column
tests. However, such tests were not within the scope of this project and were
not performed.
3-16
-------
SECTION 4
CONCLUSIONS
1. Vapor wells can be effectively used to monitor underground product
releases from both tanks and pipelines.
2. Background concentrations are commonly greater in tank backfill than in
pipeline backfill. Kaman Tempo believes that this is ascribable to tank
overfill. Such overfills are common and would affect only the immediate
vicinity of the tanks in moderate overfills.
3. Because stratification of soil vapor within tank backfill is possible,
perforating the entire backfill thickness is desirable. Shorter perfora-
ted intervals might cause leakage in an unscreened stratum to be
undetected.
4. Propane is the best suited compound of those tested for use with portable
gas detectors; however, its flammability and interference with continuous
site monitoring devices are undesirable properties. Therefore a surro-
gate compound with the detectability of propane and a low flammability is
desired.
5. The convective flow induced by the cyclic aspiration of sample wells in
the normal operation of the Genelco type system tested was not signifi-
cant enough to marketly effect the solution of diffusion parameters
calculated from the time versus concentration data. The arithmetic
time/concentration plots reveal a certain amount of intermittent noise
that may be associated with aspiration of the wells, however, where an
adequate number of data are available to filter out such noise, it is
revealed that cyclic aspiration does not affect the general trend of the
curve.
6. Vapor wells in shallow pipeline backfill may not be effective in detect-
ing tank leaks even when in close proximity to the tank backfill. This
4-1
-------
conclusion is supported by the observations made in pipeline Well V-3 and
tank Well V-5 at Site #3. Although Well V-3 was closer to source well
V-4 within the tank backfill, no increase in concentration was observed
in Well V-3. This observation may be ascribed to gas density effects
where denser-than-air tracers remain within the tank backfill, and
therefore did not enter the pipeline backfill. Conversely, when the same
tracers were injected in source well V-l in pipeline backfill, they were
detected in tank backfill (well V-2) at a lower elevation.
7. In multiple tank installations, shorter well spacings are suggested in
the direction transverse to the tank axis than in the longitudinal
direction due to the no-flow boundary effects caused by the tank walls
themselves.
8. The need for additional tracer testing is indicated whereby a relation-
ship between the diffusion parameters of tracers and fuel products can be
determined. These tests can be performed in relatively simple benchscale
soil column tests. Once this relationship has been defined, the diffu-
sion parameters, D1 and A, for a particular underground storage tank site
can be evaluated using surrogate compounds as tracers. These diffusion
parameters can be used to aid the selection of well spacings and detector
sensitivities in tank monitoring systems.
9. In order to simplify the procedures for field tracer tests, it would be
advantageous to evaluate a surrogate compound with sorptive properties
similar to fuel products. This would minimize the need for the relation
described in Item 8.
4-2
-------
SECTION 5
REFERENCES
1. Freeze, R. Allen, and J.A. Cherry, Groundwater, pages 317-345, 1979,
Prentice-Hall, Inc., Englewood Cliffs, New Jersey.
2. "Evaluation of U-Tube Underground Tank Monitoring Systems for Soil Vapor
Testing, Suffolk County, New York," Kaman Tempo, March 1988, Report to
Midwest Research Institute for EPA Prime Contract Number 68-01-7383.
3. Kreamer, D.K, Personal Communication, January 1988.
4. Weeks, E.P. D.E. Earp, and G.M. Thompson, "Use of Atmospheric Fluorocar-
bons F-ll and F-12 to Determine the Diffusion Parameters of the Unsatu-
rated Zone in the Southern High Plains of Texas," Water Resources Res.,
Volume 18, pages 1365-1378, 1982.
-------
-------
APPENDIX A
SELECTED CITY OF SAN JOSE HAZARDOUS
MATERIALS DIVISION SITE INFORMATION SHEETS
A-l
-------
-------
' t
Facility Address:
Facility Name:
File Number:
UNDERGROUND TANK MONITORING SYSTEM
INFORMATION SHEET
SITE
rOv\
K
Uv\Qgig.£L
T. 3
CONSTRUCTION INFORMATION
Number of Underground Tanks:
Tank
Size (Gallons)
Material Stored
Construction Material
Single/Double Wall
Year of Installation
Piping
Piping Construction Material:
Secondary Contained Piping? _
Method of Secondary Containment:
Year of Installation:
-------
Page 2
MONITORING INFORMATION
Vadose or Groundwater Well
Number of Wells: [
Well Number: 1 2
Vadose or Water Well \S
_ -
Depth of Well fS"" \ ^~ % $ *5~ £"
Depth of Groundwater :
Is Well in backfill?
Soil Anal. (Installation)
Soil Anal. Result (ppm)
Frequency of Monitoring: Daily, Weekly, JtonthjVor Other (specify)
Brand Name of Monitoring Device: L^>(^A p
Principle of Detection:
Person performing sampling: Owner or Consultant:
Periodic Calibrations?
Alarm Hook-Up? :
Vadose Well Only
Aspirated or Static:
Alarm Setpoint (ppm):
Initial Calibration Date:
Calibration Chemical:
Interstitial Space Monitoring
Frequency of Monitoring: Daily, Weekly, Monthly or Other (specify)
Brand Name of Monitoring Device:
Principle of Detection:
-------
UNDERGROUND TANK MONITORING SYSTEM
INFORMATION SHEET
SITE
Size (Gallons)
Material Stored
Construction Material
Single/Double Wall
Year of Installation
Piping
Piping Construction Material:
Secondary Contained Piping? _
Method of Secondary Containment:
Year of Installation:
Facility Address: _
Facility Name: fKLc 0 Ht\? .2.3 N
-------
Page 2
MONITORING INFORMATION
Vadose or Groundwater Well
Number of Wells:
Well Number; 1 2345 67
Vadose or Water Well ^ v ^ ^ ^ ^ V
Depth of Well (8 _&_ JJL J* ^- -£- _®L
Depth of Groundwater
Is Well in backfill?
Soil Anal. (Installation) T Y Y Y Y
Soil Anal. Result (ppm) ^O r\ o ^Q ^\ o . T> ^ \
Principle of Detection:
Person performing sampling: Owner or Consultant:
Periodic Calibrations?
Alarm Hook-Up? ;
Vadose Well Only
Aspirated or Static:
Alarm Setpoint (ppm):
Initial Calibration Date:
Calibration Chemical:
Interstitial Space Monitoring
Frequency of Monitoring: Daily, Weekly, Monthly or Other (specify)
Brand Name of Monitoring Device:
Principle of Detection: ; '.
-------
UNDERGROUND TANK MONITORING SYSTEM
INFORMATION SHEET
Facility Address:
Facility Name: _ OCTov^v>G.r VW-&r\v ^~a.J»L t*fr'? it ( 3 •/£>£> "
File Number: _ __
CONSTRUCTION INFORMATION
Number of Underground Tanks:
Tank 1 2 - 3
Size (Gallons)
Material Stored t> T> O
Construction Material _ _
Single/Double Wall _ _ _ _
Year of Installation
Piping
Piping Construction Material:
Secondary Contained Piping? _
Method of Secondary Containment:
Year of Installation:
-------
Page 2
MONITORING INFORMATION
_ Vadose or Groundwater Well
Number of Wells: \ \ >r w
Well' Number: 36_ 1 2345 67
Vadose or Water Well ^ ^ ___!_ _ ¥ _zL _^ __ ^L __±1
Depth of Well ^ \Q T£~ _M_ _U_ _V3 __ 1B_
Depth of Groundwater _ ____
Is Well in backfill? _ __ ; __ _
Soil Anal. (Installation) _ ____
Soil Anal. Result (ppm) _ ____
Frequency of Monitoring: Daily, Weekly, Monthly or Other (specify)
Brand Name of Monitoring Device:
Principle of Detection: _ _
Person performing sampling: Owner or Consultant:
Periodic Calibrations?
Alarm Hook-Up? ,
Vadose Well Only
Aspirated or Static:
Alarm Setpoint (ppm):
Initial Calibration Date:
Calibt-ution Chemical:
. Interstitial Space Monitoring
Frequency of Monitoring: Daily, Weekly, Monthly or Other (specify)
Brand Name of Monitoring Device:
Principle of Detection:
-------
UNDERGROUND TANK MONITORING SYSTEM
INFORMATION SHEET
svre
Facility Address: ^O I
Facility Name: UKO
File Number:
CONSTRUCTION INFORMATION
Number of Underground Tanks:
Tank
Size (Gallons)
Material Stored
Construction Material
Single/Double Wall
Year of Installation
Piping
Piping Construction Material:
Secondary Contained Piping? _
Method of Secondary Containment:
Year of Installation:
-------
Page 2
MONITORING INFORMATION
Number of Wells:
Vadose or Groundwater Well
^-^
Well Number; 1 2345 67 8
Vadose or Water Well ^ ^ ^ ^ ^ _A/
Depth of Well ^ ST (C \T ( *T S" _§
Depth of Groundwater
Is Well in backfill? ^ ^\ ^ \A CA u_
Soil Anal. (Installation) y ; 1_ 1
Soil Anal. Result (ppm)
Frequency of Monitoring: Daily, Weekly^onthjpor Other (specify)
Brand Name of Monitoring Device: ^ ^
Principle of Detection: _ _
Person performing sampling: Owner or Consultant:
Periodic Calibrations? _
Alarm Hook-Up? _
Vadose Well Only
Aspirated or Static:
Alarm Setpoint (ppm):
Initial Calibration Date:
Calibration Chemical:
. Interstitial Space Monitoring
Frequency of Monitoring: Daily, Weekly, Monthly or Other (specify)
Brand Name of Monitoring Device:
Principle of Detection:
-------
UNDERGROUND TANK MONITORING SYSTEM
INFORMATION SHEET
Facility Address:
Facility Name:
File Number:
CONSTRUCTION INFORMATION
Number of Underground Tanks:
Tank
Size (Gallons)
Material Stored
Construction Material
Single/Double Wall
Year of Installation
Piping
Piping Construction Material:
Secondary Contained Piping? _
Method of Secondary Containment:
Year of Installation:
-------
Page 2
MONITORING INFORMATION
\/ Vadose or Groundwater Well
Number of Wells: ^
Well Number; 1 2345 67 8
Vadose or Water Well _ ___X ^
Depth of Well \ C" f !T ^ C" ^~ _§
Depth of Groundwater
Is Well in backfill? O\ ^) (/) U La -^
Soil Anal. (Installation) ^
Soil Anal. Result (ppm)
Frequency of Monitoring: Daily, Weekly.(Konthiy or Other (specify)
Brand Name of Monitoring Device: V— ^ f^
Principle of Detection:
Person performing sampling: Owner or Consultant:
Periodic Calibrations?
Alarm Hook-Up?
Vadose Well Only
Aspirated or Static:
Alarm Setpoint (ppm):
Initial Calibration Date:
Calibration Chemical:
Interstitial Space Monitoring
Frequency of Monitoring: Daily, Weekly, Monthly or Other (specify)
Brand Name of Monitoring Device:
Principle of Detection:
-------
UNDERGROUND TANK MONITORING SYSTEM
INFORMATION SHEET
SITE:
Facility Address:
Facility Name:
File Number:
CONSTRUCTION INFORMATION
Number of Underground Tanks:
Tank
Size (Gallons)
Material Stored
Construction Material
Single/Double Wall
Year of Installation
Piping
Piping Construction Material:.
Secondary Contained Piping?
Method of Secondary Containment:
Year of Installation:
-------
Page 2
MONITORING INFORMATION
Vadose or Groundwater Well
Number of Wells:
^)o&L -
Well Number:
Vadose or Water Well V/. ^ V
Depth of Well \5> < !T ft *>
Depth of Groundwater ,
Is Well in backfill? VA ^)
-------
UNDERGROUND TANK MONITORING SYSTEM
INFORMATION SHEET
Facility Address: 3O \ O
Size (Gallons)
Material Stored
Construction Material
Single/Double Wall
Year of Installation
Piping
Piping Construction Material:
Secondary Contained Piping? _
Method of Secondary Containment:
Year of Installation:
Facility Name: OrxOCaJ? l£> _ MA? 23
File Number: _ _ T*.f>c&L
CONSTRUCTION INFORMATION
Number of Underground Tanks:
Tank
-------
Page 2
MONITORING INFORMATION
Vadose or Groundwater Well
Number of Wells:
•» t
Well Number: 1 2345
Vadose or Water Well \/ V V w V/
Depth of Well
Depth of Groundwater
Is Well in backfill?
Soil Anal. (Installation)
Soil Anal. Result (ppm)
Frequency of Monitoring: Daily, Weekly, (M^ntMyjor Other (specify)
Brand Name of Monitoring Device:
Principle of Detection:
Person performing sampling: Owner or Consultant:
Periodic Calibrations?
Alarm Hook-Up?
Vadose Well Only
Aspirated or Static:
Alarm Setpoint (ppm):
Initial Calibration Date:
Calibration Chemical:
Interstitial Space Monitoring
Frequency of Monitoring: Daily, Weekly, Monthly or Other (specify)
Brand Name of Monitoring Device:
Principle of Detection:
-------
UNDERGROUND TANK MONITORING SYSTEM
INFORMATION SHEET
Facility Address: \ t
Facility Name: U A * <:_ gg T
File Number:
CONSTRUCTION INFORMATION
Number of Underground Tanks:
Tank
Size (Gallons)
Material Stored
Construction Material
Single/Double Wall
Year of Installation
Piping
Piping Construction Material:
Secondary Contained Piping?
Method of Secondary Containment:
Year of Installation:
-------
I
Page 2
MONITORING INFORMATION
Vadose or Groundwater Well
Number of Wells:
Person performing sampling: Owner or Consultant:
Periodic Calibrations?
Alarm Hook-Up?
Vadose Well Only
Aspirated or Static:
Alarm Setpoint (ppm):
Initial Calibration Date:
Calibration Chemical:
. Interstitial Space Monitoring
Frequency of Monitoring: Daily, Weekly, Monthly or Other (specify)
Brand Name of Monitoring Device: _ _
Principle of Detection: . _ —
Well Number: 12345678
/ / / ^
Vadose or Water Well _Y_ _J£_ __^_ ^ v —*- ^- -^— ' '
Depth of Well \ ^~ \ ^ —2—
Depth of Groundwater
Is Well in backfill? _^ _S£ ^,
Soil Anal. (Installation)
Soil Anal. Result (ppm)
Frequency of Monitoring: Daily, Weekly^MontMyJr Other (specify)
Brand Name of Monitoring Device:
Principle of Detection:
-------
UNDERGROUND TANK MONITORING SYSTEM
INFORMATION SHEET
SITE
Facility Address: f A OST
Facility Name: A~^C O
File Number:
CONSTRUCTION INFORMATION
Number of Underground Tanks:
Tank
Size (Gallons)
Material Stored
Construction Material
Single/Double Wall
Year of Installation
Piping
Piping Construction Material:
Secondary Contained Piping? _
Method of Secondary Containment:
Year of Installation:
-------
Page 2
MONITORING INFORMATION
^ Vadose or Groundwater Well
Number of Wells:
Well Number: 1 2
Vadose or Water Well \/
Depth of Well ^
Depth of Groundwater _
Is Well in backfill? _
Soil Anal. (Installation)
Soil Anal. Result (ppm) <^(<-> *M C
Frequency of Monitoring: Daily, Weekly, Monthly or Other (specify)
Brand Name of Monitoring Device:
Principle of Detection:
Person performing sampling: Owner or Consultant:
Periodic Calibrations?
Alarm Hook-Up? ;
Vadose Well Only
Aspirated or Static:
Alarm Setpoint (ppm):
Initial Calibration Date:
Calibration Chemical:
Interstitial Space Monitoring
Frequency of Monitoring: Daily, Weekly, Monthly or Other (specify)
Brand Name of Monitoring Device:
Principle of Detection:
-------
UNDERGROUND TANK MONITORING SYSTEM
INFORMATION SHEET
. 10
x7^a /o t— l/\ \
Facility Address:
Facility Nair
File Number:
Facility Name: Afl^ O
CONSTRUCTION INFORMATION
Number of Underground Tanks:
Tank
Size (Gallons)
Material Stored
Construction Material
Single/Double Wall
Year of Installation
Piping
Piping Construction Material:
Secondary Contained Piping? _
Method of Secondary Containment:
Year of Installation:
-------
Page 2
MONITORING INFORMATION
\J Vadose or Groundwater Well
7
Number of Wells: _ / _
Hell Number; 1 2
Vadose or Water Well ^ _^_ _ __ _ _ _-
Depth of Well _l£ _^_ _l£ j£_ j£l _£L .!_
Depth of Groundwater _ ____ ___
Is Well in backfill? _ ____ __
Soil Anal. (Installation) ^ ^ VA __ ^- J^ _ M
Soil Anal. Result (ppm) ^a <\ ^ c ( o 4 Q
-------
UNDERGROUND TANK MONITORING SYSTEM
INFORMATION SHEET
S\TE:
.. •»
I T 2- ^- MS '~ i tfu^
Facility Address:
Facility Name: frvlC Q
File Number:
CONSTRUCTION INFORMATION
Number of Underground Tanks:
Tank
Size (Gallons)
Material Stored
Construction Material
Single/Double Wall
Year of Installation
Piping
Piping Construction Material:
Secondary Contained Piping? _
Method of Secondary Containment:
Year of Installation:
-------
Page 2
MONITORING INFORMATION
Vadose or Groundwater Well
Number of Wells:
Well Number:
Vadose or Water Well __ __ _ _
Depth of Well '3 ^ \ S"7 IS" ST
Depth of Groundwater
Is Well in backfill? ,
Soil Anal. (Installation)
Soil Anal. Result (ppm)
Frequency of Monitoring: Daily, Weekly, Monthly or Other (specify)
Brand Name of Monitoring Device: *— T> f^V y
Principle of Detection:
Person performing sampling: Owner or Consultant:
Periodic Calibrations?
Alarm Hook-Up? ;
Vadose Well Only
Aspirated or Static:
Alarm Setpoint (ppm):
Initial Calibration Date:
Calibration Chemical:
Interstitial Space Monitoring
Frequency of Monitoring: Daily, Weekly, Monthly or Other (specify)
Brand Name of Monitoring Device:
Principle of Detection:
-------
UNDERGROUND TANK MONITORING SYSTEM
INFORMATION SHEET
Facility Address: ^_
Facility Name: fenj^C. Q nAv9 2 "3
File Number:
CONSTRUCTION INFORMATION
Number of Underground Tanks:
Tank
Size (Gallons)
Material Stored
Construction Material
Single/Double Wall
Year of Installation
Piping
Piping Construction Material:
Secondary Contained Piping? _
Method of Secondary Containment:
Year of Installation:
-------
Page 2
MONITORING INFORMATION
\/ Vadose or Groundwater Well
-7
Number of Wells: '
Well Number:
Vadose or Water Well _ _ _
Depth of Well ST b" (^ 1 \ f ^S"
Depth of Groundwater _ ____
Is Well in backfill? _ ____
Soil Anal. (Installation) V ^ Y Y Y
Soil Anal. Result (ppm) ^Q ^(c ^O ^o^ ^U <[O_ _£U:
Frequency of Monitoring: Daily, Weekly, Monthly or Other (specify)
Brand Name of Monitoring Device: _ ^- S> t^\ V
Principle of Detection: ___^
Person performing sampling: Owner or Consultant:
Periodic Calibrations?
Alarm Hook-Up? ;
Vadose Well Only
Aspirated or Static:
Alarm Setpoint (ppm):
Initial Calibration Date:
Calibration Chemical:
Interstitial Space Monitoring
Frequency of Monitoring: Daily, Weekly, Monthly or Other (specify)
Brand Name of Monitoring Device: _____
Principle of Detection:
-------
APPENDIX B
LETTER OF AUTHORIZATION
B-l
-------
-------
ROBERT E. OSBY
Fire Chief
CITY OF SAN JOSE, CALIFORNIA
FOUR NORTH SECOND STREET, SUITE 1100 SAN JOSE FIRE DEPARTMENT
SAN JOSE, CALIFORNIA 95113-1305
(408) 277-4444
November 10, 1987
TO WHOM IT MAY CONCERN:
The city of San Jose Fire Department, in cooperation with the
United States Environmental Protection Agency ("EPA"), is
conducting a research study underground storage tank vapor
monitoring wells in the San Jose area. Mr. Bill Bergmann of
Kaman Tempo, a consultant to the EPA, is performing this study at
service stations and other underground storage tank locations
such as yours which are equipped with vapor monitoring wells in
tank backfill.
Please give your cooperation to this study. The information
being gathered is for research purposes. Please do not consider
this an enforcement action. Every effort will be made to
complete the work in a rapid manner so as not to significantly
interfere with your normal flow of business.
Should a question or problem arise related to this project,
contact Joseph Afong of the Hazardous Materials Program at
277-4659.
sincerely,
CHARLES J. WILHELM, Manager
Hazardous Materials Program
San Jose Fire Department
CJW:jwt
CJW/SURVYLTR/1
-------
-------
APPENDIX C
ADDITIONAL SITE SKETCHES
C-l
-------
-------
\U7WA
£©'
©
-5 ©
A
"
31IS
P3
U
m
31IS
-------
311S
\-
so
31IS
-------
£
o
P>
T®
0
C
->
6# 31IS
I
r
c
S © C ^wtrv )
H(7) (-
8# 31IS
-------
T©
Wcf/Wtf
c
r
%
©
I
0 !-# 31IS
-------
0
2
-I
0
r
A, f\
\J
^ (&
or
X
m
-------
-------
APPENDIX D
VAPOR MONITORING EQUIPMENT SPECIFICATIONS
D-l
-------
-------
APPENDIX.D
VAPOR MONITORING EQUIPMENT SPECIFICATIONS
TIP-I:
The TIP-I is a device that measures total hydrocarbon concentration in
air by photoionization. In operation, sample gas is pumped at approximately
275 ml/min into a 10.6 eV detection cell. Concentrations between 0 and 2,000
ppm by volume are read from a digital LCD display. The intake port is ter-
minated with a swagelok fitting which simplifies attachment of sample probes,
and a 15 micron sintered dust filter.
Photovac 10S70 Gas Chromatograph;
The 10S70 is a dual-column portable gas chromatograph with internal power
and carrier gas supplies using the same detection cell as in the TIP-I de-
scribed above. The pre-column is backflushed during analysis, allowing short
cycle times between samples. A carbopack column was selected for this project
to adequately separate propane and F-ll. Carrier flow rates were maintained
at 10 ml/min.
Sample injections are made either by direct syringe injections into a
septum or by an internal sampling loop that is loaded by an internal pump.
The pump intake port is terminated with a 1/8" swagelok connector identical to
that of the TIP-I. Loop injection volumes are controlled by varying the
duration that the sample loop is in series with the carrier gas flow into the
column. Loop injections can be made automatically at any time cycle interval
enabling the 10S70 to be operated while unattended.
Individual sample components are reported as separate peaks on a strip-
chart recorder. Concentrations are proportional to the integrated area under
each peak, printed in volt-seconds (VS) by the internal software at the base
of each chromatogram. By calibration with known standards, concentrations in
ppm can be calculated by hand or by the instrument's internal circuitry.
Kaman has detected benzene at levels as low as 1 ppb with the 10S70.
-------
Gowmac Model 20-600 Gas Analyzer:
The Model 20-600 is a device which measures total concentration of any
gas that has a thermal conductivity difference from air. The 4-inch analog
readout dial has 0.2% divisions on the 0 to 10% scale and 2% divisions on the
0 to 100% scale. Span potentiometers for the two scales were independently
calibrated for use with SFg in this project. Readings are susceptible to
interferences from other gases and do not necessarily represent SFg concen-
tration. However, for SFg tracer tests one may assume that changes in concen-
tration in a particular vapor well are due to SFg.
The Model 20-600 has an internal power supply, 200 ml/min pump, and a
1/8" swagelok intake-port termination like the TIP-I and 10S70. However, the
pump is not strong enough to maintain an adequate flow rate through more than
5 feet of tubing. Therefore, sample bags must be used.
-------
APPENDIX E
SITE #3 TRACER TEST DATA
E-l
-------
-------
Test A Tracer Tests: Wells V-2 and V-5, Site #3
Hours
after source
initialized
Well #2 0.00
0.02
0.67
1.70
2.15
3.68
5.02
6.23
8.68
8.87
9.62
10.37
11.12
11.87
12.62
13.37
14.12
14.87
15.62
16.37
23.25
23.47
24.17
24.67
25.55
Hours
after source
initialized
Well #5 0.33
0.93
1.28
2.70
3.33
4.63
6.17
8.27
23.25
24.07
28.00
Concentration
(VS)
0
0
0.01
0.006
0
0.023
0.01
0
0.009
0
0
0.007
0
0
0
0.009
0.006
0
0.007
0
0.357
0.123
0.256
0.117
0.157
Concentration
(VS)
0.398
0.118
0.129
0.154
0.023
0.008
0
0
6.6
13.7
2.9
Propane
Concentration
trna/rr\\}
OE+00 PROPANE
OE+00
2.29E-07
1.38E-07
OE+00
5.27E-07
2.29E-07
OE+00
2.06E-07
OE+00
OE+00
1.60E-07
OE+00
OE+00
OE+00
2.06E-07
1.38E-07
OE+00
1.60E-07
OE+00
8.18E-06
2.82E-06
5.87E-06
2.68E-06
3.60E-06
Propane
Concentration
(mo/rnn
9.12E-06 PROPANE
2.70E-06
2.96E-06
3.53E-06
5.27E-07
1.83E-07
OE+00
OE+00
1.51E-04
3.14E-04
6.65E-05
-------
Test A Tracer Tests: Wells V-2 and V-5, Site #3
Well #2
Hours
after source
initialized
-0.28
0.74
1.89
6.17
6.25
26.00
SF6
Concentration
(%}
3.0
2.5
5.0
0.8
0.7
2.0
SF6
-------
Test B Tracer Tests: Wells V-2 and V-5, Site #3
11/30 Well#2 SF6
Hours
after source
Date initialized
Injection Volume = 2 Bags = 3.6L
Concentration
11/30
-0.1
1.1
1.6
1.9
4.2
11/30 Well #5 SF6
Hours
after source
Date initialized
11/30 -0.9
-0.1
0.8
1.0
1.6
2.0
2.4
4.5
12/1 Well #2 Propane
Hours
after source
Date initialized
12/1 -0.4
-0.3
3.6
12/1 4.6
12/2 16.2
16.6
17.9
18.4
19.4
20.9
21.9
22.6
23.7
24.4
25.0
25.8
26.3
7.4
1.6
5.2
4.5
65
Concentration
(%}
0.5
0.8
3.4
1.2
1.5
8.5
8.0
7.3
Propane
Concentration
(VS)
0.008
0.020
0.000
0.000
0.230
0.006
0.010
0.005
0.008
0.009
0.000
0.008
0.008
0.006
0.008
0.012
0.005
SF6 INJECTED INTO WELL 1 AT T=0.
70°F Temperature In hole T=0.08.
End
Injection Volume=4 BAGS=7.2 L
Temp 63°F T=0.
End
Injection Volume=2 bags = 3.6L Injected Into well 1.
Propane
Concentration
(ma/mh
1.83E-07 73°F at T= -0.67.
4.58E-07
OE+00
OE+00
5.27E-06 ?: Residual from W#5 rdg: See
1.38E-07 W#5 T=37.95 rdg.
2.29E-07
1.15E-07
1.83E-07
2.06E-07
OE+00
1.83E-07
1.83E-07
1.38E-07
1.83E-07
2.75E-07
1.15E-07
-------
Test B Tracer Tests: Wells V-2 and V-5, Site #3
Hours
after source
Date initialized
1 1/30 0.0
0.8
1.6
1.6
11/30 3.1
12/1 3.1
3.1
15.5
16.3
17.1
17.9
18.5
19.0
19.5
20.0
20.4
21.0
21.4
22.4
22.9
23.2
23.4
23.7
23.9
24.2
24.4
24.7
25.4
25.9
26.2
27.0
27.2
27.5
27.7
28.0
28.2
28.5
28.7
29.0
Injection Volume :
Source stre
Propane Propane
Concentration Concentration
(VS) (ma/ml) ...
1.300
1.300
1.200
0.790
0.550
0.010
0.000
0.030
0.120
0.150
0.120
0.290
0.220
0.270
0.290
0.321
0.370
0.310
0.600
0.770
0.880
0.990
1.200
1.000
1.000
0.910
2.900
0.820
1.400
1.400
0.570
1.000
1.200
1.600
1.800
1.600
1.600
2.000
2.200
2.98E-05
2.98E-05
2.75E-05
1.81E-05
1.26E-05
2.29E-07
OE+00
6.88E-07
2.75E-06
3.44E-06
2.75E-06
6.65E-06
5.04E-06
6.19E-06
6.65E-06
7.36E-06
8.48E-06
7.10E-06
1.38E-05
1.76E-05
2.02E-05
2.27E-05
2.75E-05
2.29E-05
2.29E-05
2.09E-05
6.65E-05
1.88E-05
3.21 E-05
3.21 E-05
1.31 E-05
2.29E-05
2.75E-05
3.67E-05
4.13E-05
3.67E-05
3.67E-05
4.58E-05
5.04E-05
Page 1 of 2
8 bags*(1.8L/bag)*1969mg/L = 28400mg
Source strength = (28400mg)/14ft = 2029 mg/ft = 66.6mg/cm
Pumped 2 L from W5 to enhance flow
atT=14.59.
Temp =64°F @ T=16.59
Hiatus in rdgs to check instr. flow
rates. Adjusted for longer loop
loading time after 1=24.7 .
-------
Test B Tracer Tests: Wells V-2 and V-5, Site #3
11/30 Well 5 Pronane (continued)
Hours
after source
Date initialized
12/1 29.2
29.5
29.7
30.0
12/1 30.2
12/2 30.5
30.7
31.0
31.2
31.5
31.7
32.0
32.2
9 32.5
32.7
33.0
33.2
33.5
33.7
34.0
34.2
34.5
34.7
35.0
35.2
35.5
35.7
36.0
36.2
36.5
36.7
37.0
37.2
37.5
37.7
38.0
38.4
39.3
39.7
42.6
44.9
45.1
Propane
Concentration
/VS)
2.100
2.400
2.400
2.500
2.700
2.600
2.600
2.400
2.700
2.700
2.600
2.700
2.500
2.100
2.100
2.200
2.200
2.200
2.500
2.700
2.700
2.700
2.600
2.600
2.600
2.700
3.000
3.000
2.600
2.200
2.500
2.600
2.700
2.800
2.600
2.500
0.140
0.150
0.320
1.500
1.000
0.690
Propane
Concentration
(mg/ml)
4.81 E-05
5.50E-05
5.50E-05
5.73E-05
6.19E-05
5.96E-05
5.96E-05
5.50E-05
6.19E-05
6.19E-05
5.96E-05
6.19E-05
5.73E-05
4.81 E-05
4.81 E-05
5.04E-05
5.04E-05
5.04E-05
5.73E-05
6.19E-05
6.19E-05
6.19E-05
5.96E-05
5.96E-05
5.96E-05
6.19E-05
6.88E-05
6.88E-05
5.96E-05
5.04E-05
5.73E-05
5.96E-05
6.19E-05
6.42E-05
5.96E-05
5.73E-05
3.21 E-06
3.44E-06
7.33E-06
3.44E-05
2.29E-05
1.58E-05
Page 2 of 2
See W#2 T=16.18.
After sampling W 2. Dilution effect.
END
-------
Test B Tracer Tests: Wells V-2 and V-5, Site #3
WELL #2 DISCONNECT TIMES
Hours Disconnected
after source Duration
initialized (hrsL
Remarks
WELL
DATE
12/1
12/2
0.0
3.7
12.6
13.0
14.1
14.6
15.6
16.7
18.1
#5 DISCONNECT
Hours
after source
initialized
14.6
16.6
17.9
21.5
24.7
25.4
26.2
38.0
38.5
39.8
40.9
42.8
3.67
1.00
0.45
1.17
0.10
0.50
1.00
1.33
1.48
TIMES
Disconnected
Duration
fhrs>
0.17
0.17
0.17
1
0.67
0.75
0.75
0.25
0.75
1.25
1.85
2.23
Probe not removed
Probe not removed
Probe not removed
Probe not removed
Probe removed
Probe not removed
Probe not removed
Probe not removed
Probe not removed
WELL #5 Probe REMOVAL TIMES
Hours REMOVED
after source Duration
DATE initialized fhrsl
12/1
14.6
16.6
17.9
24.7
40.9
0.17
0.17
0.17
0.67
1.85
-------
APPENDIX F
CURVE MATCHING SOLUTION OF INSTANTANEOUS
LINE SOURCE GAS DIFFUSION MODEL
F-l
-------
-------
Curve Matching Solution
Of Instantaneous Line-Source Gas Flow Model
Development of Model
A general mathematical diffusion model for a sorbing gas in 3-phase porous
media has been developed from Pick's 2nd Law by Weeks et al.(1982). Pick's
2nd Law is stated mathematically as
rj2^« -jx>
i-> o O oLf
o^-TT (D
where
D = the general coefficient of diffusion of gas into air (L2/T)
C = the concentration of the gas (M/L3)
x= the distance from the source (L)
f= time (T).
In order to account for properties of 3-phase media, Eq(1) is modified to
where
T= tortuosity of the medium
fliAi,= gas-filled, liquid-filled and total porosities
p, s = densities of the liquid and solid phases of medium
Kw „ = partition ing coefficients for liquid and solid phases.
Let,
which is defined as the sorption-corrected porosity (dimensionless).
Dividing both sides of Eq (2) by A , the following diffusion model is obtained:
-------
Solving Eq(4) for instantaneous, infinite line-source boundary conditions, we
have the solution
-
c__ Q Um (5) (Kreamer,1988)
4nD'At
where
Q = mass per unit length of instantaneous source (M/L)
D' = effective diffusion coefficient in porous medium
(L2/T), defined by,
(6)
The solution in Eq(5) is analogous to the Theis solution for groundwater flow
into a well, where groundwater parameters of drawdown, transmissivity and
storage coefficient are analogous to gas tracer parameters of concentration,
effective diffusion coefficient and sorption-corrected porosity. The vapor
well function e-u , which we will abbreviate V(u) , is analogous to the Theis
well function W(u).
By further simplification, the solution is
C = —Q7—V(u) (7)
4nDAt * '
where
- (8)
Solution Procedure
By injecting a source of mass Q into the source well and measuring the
resulting increase in tracer concentration with time in an observation well,
D' and A can be evaluated by a curve-matching technique similar to the
Theis solution method (Freeze & Cherry, 1979). A type curve is generated by
plotting u versusV(u) on log-log paper (Figure F-l). The field data are
plotted on a separate,identical sheet of log-log paper. The field plot is then
superimposed upon the type curve, keeping the coordinate axes parallel, and
moved until most of the data points fall on the type curve (Figure F-2). Using
-------
10
ID
'2
§10-3
10
'4
Type Curve
•j Q~5 I i—I I i ii ill—»_i—. .... ..II -I
10
"1
10
10
u
1
10
10'
10
Field Data
1
10
10^
r2/t
10
10
ID
°
'1
10
~3
10'
Figure F-l. Type Curve and example field data.
-------
Type Curve .
,-1 * -
Field Data
1U '
10'2
10'3
io-4
10"5
10
-
-1
\
.10?
\
\
\ Mat
J
.1rf
ch Point: u=lC
r2/t
.10?
°, V(u)
=32,Ct=
/
•
=10-4 :
30 :
"
.10?.....'
10°
1C'1
10-25
ID'3
10
102 u 103
» 9
rz/t
10'
10'
Figure F-2e Curve-matching procedure. Maintaining parallel axes,
field data overlay is shifted to make the best visual fit with the
type curve. A match point is selected and the coordinates (u, V(u))
and (r /t» C t) read from both sets of axes.
-------
f
C
an arbitrary type curve match-point with coordinates (u, V(u) ), the
corresponding coordinates (r2/t, C*t) are read from the field data plot
overlay. For simplicity, type curve match point coordinates with unit
multiples of integral powers of ten (eg. 1*10-3, 1*10+6. etc.) are preferred.
Solving Eq(8) for D',
; D'=4^r) w
and solving Eq(7) for A ,
„_ 0
The match-point coordinates from both curves are subsituted into Eq's(9) and
(10), from which D' and A are caculated.
Once parameters D' ,A , and source mass O are known, the concentration, C
can be predicted in tank or pipeline backfill at any distance r from a line
source at any time / after the injection. It is simply necessary to calculate
u from Eq(8), take its natural exponential, and calculate concentration, C ,
from Eq(7).
Freeze, R. Allen, & J.A. Cherry. 1979. Groundwater. Prentice-Hall,Inc.,
Englewood Cliffs, N.J., pp317-345.
Kreamer, D.K.,January 1988, Personal Communication.
Weeks.E.P., D.E.Earp, & G.M.Thompson. 1982. Use of atmospheric
fluorocarbons F-11 and F-12 to determine the diffusion parameters of the
unsaturated zone in the southern high plains of Texas. Water Resources
Res, v.18, pp. 1365-1378.
-------
------- |