United States
                    Environmental Protection
                    Agency
 Environmental Monitoring
 Systems Laboratory
 Las Vegas NV89114
                    Research and Development
 EPA/600/S2-87/027  June 1987
v>EPA          Project Summary

                    Soil-Gas  Measurement  for
                    Detection  of  Subsurface
                    Organic Contamination
                    Henry B. Kerfoot and Larry J. Barrows
                     Two techniques for soil-gas measure-
                   ments were used at a site in  Pittman,
                   Nevada. Two distinct organic contam-
                   inant plumes  were  investigated.
                   Twenty-three monitor wells had been
                   drilled across the plumes to provide
                   ground-water concentrations of ben-
                   zene/chlorobenzene  and chloroform.
                   One  soil-gas sampling  technique (a
                   commercial ' 'passive" sampler)
                   involved the placement of activated-
                   charcoal-coated wires approximately 1
                   foot  below the surface of the ground
                   for nine days. Analysis of the samplers
                   was made at a commercial laboratory
                   using pyrolysis and mass spectrometer.
                   The  other technique (an ''active"
                   sampler) consisted of a pipe  being
                   pounded into the ground and  sampled
                   using a small hand pump and gas-tight
                   syringes. Analysis of the samples was
                   made using a field  portable gas
                   chromatograph
                     The accuracy, precision and  repre-
                   sentativeness of the  two soil gas
                   techniques were evaluated.  Closely-
                   spaced, repetitive measurements were
                   made in the vicinity of the  monitor
                   wells. Frequent calibrations were made
                   on the field  portable gas chromato-
                   graph to  analyze the  accuracy and
                   precision of the analytical method. The
                   data  from  both  soil-gas techniques
                   were compared  to the data from  the
                   monitor wells to assess the represen-
                   tativeness of the soil-gas techniques.
                     The data  from  the commercially-
                   marketed  "passive" samplers  exhi-
                   bited a large degree of variability.  It was
                   not  possible  to map  either  ground-
                   water plume using this method at  the
Pittman  site.  The data from  the
 "active" sampler permitted the chlo-
roform plume to be accurately mapped;
however, this technique, as well as the
 ' 'passive" technique, were unable to
map the benzene/chloroform plume.
Biodegradation of the benzene vapors
in the vadose zone is believed to  be
responsible for the lack of detectable
concentrations.  Further studies  are
recommended.
  This Project Summary was  devel-
oped by EPA's Environmental Monitor-
ing Systems Laboratory, Las Vegas,
NV, to announce key findings of the
research project that is fully  docu-
mented in a separate report of the same
title (see Project Report ordering
information at back).

Introduction
  Efforts  to map the extent of ground-
water contamination have relied on the
drilling of monitor wells. The proper
location and number of wells is an issue
in site investigations with cost being  an
important factor  Geophysical  tech-
niques have proven to  be  useful  in
locating where monitor wells should  be
drilled particularly when  inorganic,
electrically-conductive contaminants are
involved.  In those instances where
organic contaminants are  involved,
particularly when concentrations are
relatively low, traditional geophysical
techniques are less helpful in mapping
the ground-water contamination prior to
the drilling of monitor wells. The sam-
pling of  vapors  in  the  vadose  zone,
commonly referred to  as  "soil-gas"
monitoring, has  become increasingly

-------
attractive to those who seek additional
techniques and information on the extent
of organic contamination at a site.
   The strengths and weaknesses of the
soil-gas  technique  are  still  being
researched. The number of practitioners
using the technique at sites during the
past few years have increased dramat-
ically. This together with the growing
number of reported instances where the
technique  has been  used with some
success would suggest that the soil-gas
technique  is  a  credible technique  to
consider in a site investigation along with
more tested,  older approaches such  as
those found under the general category
of geophysics.
   The soil-gas technique, being a rela-
tively new technique, cannot be consi-
dered to be a standard technique  with
well-defined  quality assurance/quality
control practices being  prescribed. Soil
gas is typically collected with adsorbents,
pipes, canisters,  and  bags, and the
sample  is generally  analyzed  with
organic vapor analyzers, field-portable
gas  chromatographs,   laboratory gas
chromatographs and mass spectrome-
ters. As with the placement of monitor
wells, the number and placement of the
less-expensive soil-gas  measurement
devices is an issue also. Neither standard
sampling and  analytical procedures nor
prescribed quality assurance precisions
of the method  defines the  extent  of
ground-water contamination.
   The representativeness of the method
of collecting a sample is also an item  of
research  interest.  While the method
appears to be capable  in many investi-
gations  of being able to map contami-
nants in the ground water, there is some
question as to whether  the contaminant
concentrations measured in the vadose
zone can reliably measure the extent  of
contamination of the underlying ground
water. In  some cases,  a soil-gas  mea-
surement may simply measure contam-
inants within  the vadose zone or con-
taminants that have migrated downward
from a surface spill.
   The  objectives  of  the field  study
described in this report were to  deter-
mine the accuracy, precision and repre-
sentativeness  of two common soil-gas
measurementtechniques in mapping the
contamination at a previously studied
site.

Procedure
  The site is located at Pittman, Nevada
(Figure 1).  Measurements made  along
the Pittman Lateral (a large pipe used to
supply water from Lake  Mead to the Las
                 Pittman
Figure 1.     Location of the Pittman Site.
Vegas metropolitan  area)  determined
that two distinct organic and inorganic
plumes  were  in  the ground water.
Twenty-three monitor wells had  been
drilled at 200 foot intervals and sampling
of those wells  over  a period  of years
indicated  that  benzene/chlorobenzene
and chloroform plumes were present in
the shallow ground water aquifer (Figure
2). Soil-gas measurements were made
in the vicinity of several of the wells.
  Two  techniques   were  used.  A
commercially-marketed,   activated-
charcoal-coated wire  (Figure  3) was

-------
      West
                 Benzene/Chlorobenzene
                                                                          East
        655
               650
                      645
                             640
                       0      500
                     Scale in Feet
  635    630
  Stations


 0       200
Scale in Meters
                                                   625    620
                                                     rest Well
          615
                 610
T Water Table
 Figure 2.    Hydrogeologic cross section along the Pittman Lateral.
 buried approximately one foot below the
 surface of the  ground and left for nine
 days. The samplers were returned to the
 firm for analysis by Curie-point pyrolysis
 and mass spectrometer (MS). The wires
 were heated  to approximately  400°C
 under vacuum in a 1.1 MHz,  1.5  kW
 Fisher  Curie-point pyrolyzer, and  the
 desorbed  gases were flushed  into an
 Extranuclear  Laboratories  Spectra  El
 quadrapole  mass  spectrometer  where
 low-energy ionization (15 eV) was used
 to  minimize  fragmentation. The  other
 technique used a custom-designed and
 manufactured steel pipe (Lockheed Gas
 Analysis  System-"LGAS") that was
 driven into the ground to a typical  depth
 of approximately four feet, and samples
 were obtained using a small hand  pump
 to evacuate the deadspace.  A gas-tight
 syringe was used to inject a small sample
 of  the  gas into a field  portable gas
 chromatograph. The  total  volume of
 sample  obtained from  the  ground was
 approximately 75 mL.
  A series of tests were conducted to
 develop quality assurance procedures for
 both  soil-gas  sampling techniques in
 addition to assessing the general ability
 of the techniques to map ground-water
 contamination. The spatial variability of
 soil-gas measurements  in  a small area
 was assessed, the change in concentra-
 tion over depth was evaluated, and the
 concentration of  contaminants  over a
 series of sample cycles was also  eval-
 uated at  one  point for the soil-gas
 measurements using the LGAS. Spatial-
variability of soil-gas  measurements
 made with the activated-charcoal adsor-
bent was also investigated.
      Results

        Soil-gas measurements made with the
      pipe and pump were able to map the
      chloroform but not  the benzene/chlor-
      obenzene contaminant plume (Figure 4).
      Measurements made with the activated-
      charcoal-coated wire were highly vari-
      able and  were unable  to map either
      ground water plume (Figure 5). Although
      no  measurements of the  biological
      activity in  the vadose zone were made
      above  the  benzene/chlorobenzene
      plume, it is believed that the inability to
      measure detectable levels of benzene in
      the  soil gas with either  method, any-
      where from 1   1 ,/2 feet above the water
      table to just below the ground surface,
      is due  to biological  degradation of the
      benzene  in  an aerobic  environment.
      Further  studies   in  this  area  are
      recommended
        Concentrations of chloroform and
      carbon tetrachloride increased linearly
      with depth above  the  contaminated
      ground water,  m agreement with a model
      for vertical transport of volatile organic
      compounds in  the vadose  zone  by gas
      diffusion.
        Detailed investigations of the sampling
      and  analytical procedures for the LGAS
      were helpful  in obtaining  reproducible
      results. Teflon components were elim-
      inated from the- probe and sampling train
      to avoid a "memory" effect in sampling
      low  levels of organics. A vacuum gauge
      was attached  10 the  sampling manifold
      on the probe to monitor the rate at which
      the vacuum diminished to  help ensure
      that  there was  a good seal between the
      probe and the  soil. The gauge was also
 helpful in determining  when the pores
 in the sample  tip  were  completely
 clogged. Clogging of the probe and leaks
 were  found to significantly affect the
 measurements that were made to assess
 spatial variability in the measurements.
 The length of time gas was present in
 the syringe also  was found to  have an
 effect on  the precision  of the measure-
 ment. Blanks and frequent use of  cal-
 ibration gas standards  were helpful in
 reducing and assessing the precision of
 the method.

 Conclusions
   The ability  of  the  soil-gas  probe to
 discern not only the center but edges of
 the chloroform plume  in the ground
 water indicates that soil-gas measure-
 ments, with proper safeguards, can be
 of value in mapping ground-water con-
 tamination from some organic contam-
 inants. The absence of  benzene in  the
 vadose zone indicates that caution must
 be exercised  in the use of soil-gas
 measurements to  map ground-water
 contamination. The  inability  of one
 "passive"  sampling  method (using
 activated-charcoal and pyrolysis to map
 organic contaminants)  should  not be
 interpreted as a failure of the  general
 approach.   After this  study was  com
 pleted, another "passive" method was
 used and the results were good. Further
 efforts at  developing  standard test
 procedures  for the  evaluation of  the
 adsorption, desorption, and analysis
 phases of passive sampling are planned.
 Further research  is  also  planned  in
 evaluating  the representativeness  of
 soil-gas measurements with any method
to contaminant levels in the soil, ground-
water,  or both.

-------
           -Oo3
                                      S^\
                                                             Ground Surface
, \J,iJ r  rJ  ° J ' Yl/ * l  -.   •,
*£ rJr>o£ •  .o>,( ^  ;=? vy
'^f 0 "'   "„ ,'f 0°°»°^ "''".
p^^j:^;, „:•
>r>°ooM>3.,c:  '^
^^o^'f^
Sfjcr"?', V
•C,  0. ^-.. olj'
S..5?.rf;^
                                                                Charcoal
                                                               Adsorbent
                                                               Inverted
                                                                Glass
                                                                 Tube
                                                             Ferromagnetic
                                                                  Wire
                                                             Threads
                           l°''^^•O^^^^o^^^' '

Figure 3.    Schematic drawing of a PETREX® sampler in place

-------
        woo-.
         800-
        600-
        400-
        200-
   !!
                631
                          629
                                   627
                                            625
                                                     623
                                                "HT-
                                                627
                                     -f
                                     619
             West
                                                                   East
        600-1
        500-
        400-
        300-
       200-
        100-

WN'ES
  631
WN'ES
  629
                                WN
                                          WN
                          WN'ES    WN1ES
627      625      623
Borehole Identification
                                                             621
                                                                       619
Figure 4.    Soil-gas and ground-water chloroform plumes.

-------
                        Benzene, M/Z = 78
                                                       Chlorobenzene, M/Z- 112
                                                                                         Chloroform, M/Z - 83
    2000 -,
     1500-\
©   7000H
P
£   500 -I
                Borehole Identification
               10 •<» "fr
               Co ~o>
                                                     Co CQ CQ
                                           4000
                                           3000
                                           2000
                                                                                    Borehole Identification

                                                                                    CQ co CQ CQCO co   COCQCQ
                                                                                 20000i
                                                                                  70000
                                                                                   5000
                                                                                      0
                                                                              -Non-detect	
> to
-* V.
ID c
C Ct)
b c
•T o


\ 5-
*3) -
n

n\\
}
\
n
1
} I

!"
k.
C
01
«J
0

-------

-------
United States                          Center for EnvironiTTent.il Research
Environmental Protection                Information
Agency                                Cincinnati OH 45268
Official Business
Penalty for Private Use $300
EPA/600/S2-87/027

-------