SOIL VAPOUR EXTRACTION AS AN INVESTIGACION, MONITORING AND REMEDIATION

DECIDING TOOL
Maite Garcia1, Juan Fernndez1 and Laia Juncosa1
1

TUBKAL INGENIERIA, S.L., Joan Gamper 25, E-08014 Barcelona (www.tubkal.com)

Key words: VOC, XCOV, SVE, UST, gas stations

Summary
After more than 300 vacuum assays performed by TUBKAL INGENIERIA, this paper describes SVE
tests applied for site quality investigation and control purposes and it includes some practical
simplified examples that show the capabilities of the technique.
1 Introduction
Conventional sampling techniques often do not access to most
polluted areas or contamination sources when these are located
below tanks, pipelines or other underground facilities; a typical
case in gas stations. XVOC pollution is also difficult to hit when
sampling, as they percolate with a narrow, fingering profile. The
result is that, even when soil samples from a site investigation
comply with soil quality criteria, there may be an undetected
contamination problem to cope with.
These limitations in site investigation are most worrying when soil
quality evaluation is critical, such as in a property transfer, when
there is no shallow groundwater to detect potential soil
contamination or in sensible residential use of land.
To overcome these limitations, TUBKAL INGENIERIA started to
use the Soil Vapour Extraction (SVE) technique, traditionally
applied for soil remediation, to detect VOC contamination leaks or
sources in unsaturated soils. The grounds are that if SVE
mobilises VOC present in soil through the application of vacuum,
this same technique could also be used to evaluate VOC soil
quality trough the quality of soil gas extracted, giving information
for a bigger volume of soil (10-20 m radius from the extraction
point) than punctual and discrete soil sampling.
2 Method
An SVE test consists of extracting soil-air from a piezometer (or vapour detector) and periodically
measuring air flowrate, vacuum in extraction and monitoring points, and quality of the air extracted
(VOC, O2, CO2, T, etc). Before the test ends, an air sample should be taken for VOC laboratory
analysis. That is, a SVE test is performed in a similar way than SVE remediation technique but it only
lasts a few hours and therefore more frequent data needs to be compiled.
Before executing SVE tests on the field, some aspects must be taken into consideration:
- Piezometers (or vapour detectors) should be adequately constructed to apply vacuum (mainly,
that there is a good surface seal and enough open filter to air, that is unsaturated soil) and
they should be located near all potential VOC sources (UST, pipelines, etc.).
- Nearby underground infrastructures may cause air preferential pathways, altering results and
finally disabling the detection of potential contamination sources. This can also be the case
depending on the soil layers geometry with different permeability and the depths of sources.
- A previous theoretical performance estimation is recommended. SVE flowrate may be
evaluated using Johnson equation (1991) which takes into account the site-specific
permeability. Therefore, time for the SVE test can be predicted so that it reaches the required
distance or radius of control to cover nearby potential contamination sources. On the other
hand, a vacuum pump should be selected according to the site-specific permeability.

Interpretation of results considers that soil-air is extracted from a homogenous cylinder in the subsoil,
with increasing radius as times elapses. On this base, the relation between time and radius of control
depends on air flowrate, height of soil layer where air moves and its porosity.
rcontrol = [ tcontrol x Q / ( x bsoil x nsoil)] ;
Where

rcontrol
tcontrol
Q
bsoil
nsoil

tcontrol = x r2control x bsoil x nsoil / Q

radius of control (m)

time of test (h)
3
air flowrate (m /h)
thickness of soil layer where air moves (m)
effective porosity (-)

For each extraction point VOC data (usually from PID, ppmv) should be plotted versus time and
versus distance. Interpretation of VOC evolution versus distance over the actual site plan will confirm
the control of all potential contamination sources and it may indicate leaking points or polluted areas.
See examples included at the end of this paper. Focus should also be placed on VOC data evolution
in terms of time and distance, as it may give more information about contamination profile.
VOC mobilization rate (kg/day) is also evaluated and plotted as described for VOC data. This
parameter would help overcome misinterpretation due to dilution effects in high permeability cases.
The mobilization rate is also a remediation deciding criteria. Mobilisation rate may be estimated on the
basis of in-situ VOC measures (ppmv) but it should be recalculated with laboratory analysis (mg/m3),
as in some cases there may be some relevant differences between both data.
T = Q x CVOC x 24 x 10-6
Where

T
Q
CVOC

mobilisation rate (kg/day)

air flowrate (m3/h)
VOC concentration (mg/m3)

CVOC = PID x FC x PMcomp / VM

On

CVOC
PID
FC
PMVOC
VM
VOC

or

CVOC = PID x FC x 10-6 x VOC

VOC concentration (mg/m3)

PID measurement, usually based on isobutylene (ppmv)
PID correction factor for specific VOC (-)
VOC molecular weight (g/mol)
gas molar volume (22,4 l/mol at 1 atm and 25C)
VOC gas density at specific pressure and temperature (mg/m3)

High VOC measures in extracted air and/or high VOC mobilisation rates indicate a VOC polluted area
nearby the extraction point or, if tests are repeated periodically, a potential recent leak.

Other SVE test data to evaluate are:

-

Radius of influence for each SVE test, which is the maximum distance where the vacuum is
detected. Radius of influence should be higher than radius of control to guarantee the
objectives of the SVE test. Radius of influence should be measured on site trough monitoring
points and, ideally, for different SVE operating conditions (flowrates and vacuums). Although
not recommended, radius of influence may also be estimated by another Johnson equation
(1991).

Air permeability in the basis of Johnson equations (1991) to check the soil layer that allows air
mobilisation with vacuum.

Changes in SVE operation parameters according to flowrate and vacuum evolution in terms of
time (and distance). This sort of changes may be caused by sudden air intrusion through a
nearby slope or another relevant underground infrastructure, which may distort the ideal
circular situation. Oxygen and carbon dioxide data may give some extra hints about the cause
of a change in operating parameters.

The detailed methodology for SVE tests and their interpretation for gas stations are detailed in
chapters 2 and 3 of the document Methodological guide to perform vacuum tests in vapour detectors
in gas stations developed in 2012 by TUBKAL INGENIERIA under contract of the public Agncia de
Residus de Catalunya1.
3 Application Example: periodic monitoring of soil quality in a gas station
Yearly application of SVE tests in an urban gas station with 4 piezometers installed as vapour
detectors demonstrated a gasoline leak around one UST, later confirming that it was a result of
defective tightness in the supply pipe.
SVE tests performed in years
2009 and 2010 indicated
similar VOC background
levels in the subsoil of the
gas station (VOC around 100
ppmv and maximum of 400
ppmv and mobilisation rates
below or around 1 kg/day for
all tested points).

Date

2010

2011

Point
CV1
CV2
CV3
CV4
CV1
CV2
CV3
CV4

Duration
(hh:mm)
3:00
6:00
3:00
3:30
3:00
5:58
3:00
3:30

Flowrate
(m3/h)
35
22
32
45
33
24
31
44

Vacuum
(mbar)
185
185
187
185
181
191
186
185

VOC
(ppmv)
400
100
< 100
<100
3.600
500
< 100
900

Rate
(kg/day)
1,3
0,3
0,1
0,1
11,8
0,9
0,1
4,4

Nevertheless, SVE tests in 2011 showed an increase in values, mainly in vapour detector CV1 where
VOC measures in extracted air were above 3.000 ppmv and mobilisation rate was calculated of
around 12 kg/day through air samples analysis. Compiled data in 4 vapour detectors indicated that the
most polluted area was nearby CV1 and between CV1 and CV4; approximately at a distance of 2-3 m
from CV1 and 6-7 m from CV4. See graphics and site map interpretation.
1

http://www20.gencat.cat/docs/arc/Home/LAgencia/Publicacions/Sols%20Contaminats/sols_guia_carburants.pdf

GAS STATION - CV3

GAS STATION - CV1

500
2011
2009
2010

400

3.000
VOC (ppmv)

VOC (ppmv)

4.000

2011

2.000

2009
2010

1.000

300
200
100

3
4
Distance (m)

GAS STATION - CV2

GAS STATION - CV4

1.000

2011
2009

800

2010

VOC (ppmv)

VOC (ppmv)

Distance (m)

1.000
800

600
400

2011
2009
2010

600
400
200

200

0
0

4
5
Distance (m)

Distance (m)

After leak confirmation and repair, remediation was easily performed by SVE in the piezometer CV1
until VOC lecture in extracted air was reduced to 10 ppmv in less than 3 months; around 400 kg of
gasoline were removed. The subsequent SVE tests concluded effective remediation of soil and
indicated once again low background VOC levels, which allow continuing with periodic vapour
monitoring.

4 Application Example: site investigation at a chemical facility

PCE and TCE groundwater contamination in a deep aquifer was detected and a chemical facility
nearby was required to perform a first site investigation.
Soil and groundwater investigation included drilling and installation of 4 deep piezometers (45 m)
located considering groundwater flow direction and 9 shallow vapour detectors (5 to 8 m) all nearby
potential contamination sources. Conventional soil gas survey was not implemented as subsoil is very
heterogeneous with interbedded clay layers with fine sand layers.
Only very low levels of XVOC were detected in the analysed soil samples; concentrations ranged from
below detection limit to a maximum of 0,33 mg/kg of PCE. These low levels respond to vapour
migration but they show that XVOC source areas, if so, have not been identified nor prospected. On
the other hand, groundwater contamination was confirmed but these specific results did not provide
either sound information about potential sources.
Afterwards,
SVE
tests
were
performed and XVOC contamination
was detected in soil air quite in a
small area of the property; PID
measures were above 1.000 ppmv in
4 of the 13 points tested, reaching a
maximum of 5.800 ppmv in
piezometer
Pz4.
Air
samples
analysed showed high concentrations
of PCE and, in a lesser proportion,
TCE and 1,2-cis-dicloroethylene.
Mobilisation rates were as high as 54
kg/day in piezometer Pz4, indicating
a high polluted source area nearby.

Duration
(hh:mm)
5:45
4:25
4:05
5:50
3:30
4:00
2:00
3:00
3:00
1:30
1:55
3:00
1:55

Point
Pz1
Pz2
Pz3
Pz4
Cv1
Cv2
Cv3
Cv4
Cv5
Cv6
Cv7
Cv8
Cv9

Flowrate
(m3/h)
145
160
110
118
65
103
78
50
85
80
72
92
77

Distance
(m)
8,7
8,0
6,4
7,9
10,2
13,7
8,4
8,3
10,8
7,4
7,9
11,2
8,2

VOC
(ppmv)
50
110
100
5.800
150
25
402
270
1.200
20
2.500
110
1.800

Rate
(kg/day)
0,6
1,1
0,8
54,0
0,8
0,2
2,7
1,2
8,5
0,2
15,5
0,8
10,5

SVE tests results showed that contamination source was not related to different ASTs containing
XVOC solvents (either now or in the past); the most polluted area happened to coincide with an old pit
located in a corner of the site although not clearly defined in geometry (now all the facility is paved),
where several chemical products were probably dumped some years ago. But even the 2 soil samples
analysed from the borehole in this area (Pz4) did not show high XVOC concentrations.

13

2.500

13

2.000

10

2.000

10

1.500

1.500

1.000

1.000

500

500

0
0

5
6 7
8
Distance (m)

10 11 12 13

0
0

CHEM ICAL FACILITY - CV9

5
6 7
8
Distance (m)

10 11 12 13

CHEMICAL FACILITY - Pz4

3.000

15

7.000

70

2.500

13

6.000

60

5.000

50

4.000

40

3.000

30

2.000

20
10

10

1.500

1.000

500

1.000

4
5
Distance (m)

VOC (ppmv)

2.000

Rate (kg/dia)

VOC (ppmv)

15

Rate (kg/dia)

2.500

VOC (ppmv)

3.000

Rate (kg/dia)

VOC (ppmv)

CHEM ICAL FACILITY - CV7

15

0
0

4
5
Distance (m)

Rate (kg/dia)

CHEM ICAL FACILITY - CV5

3.000

Tests also showed that SVE was very effective as a remediation technique and installation was
optimised with only 4 SVE points. In 4 months more than 1.500 kg of XVOC (PCE and TCE mainly)
have been effectively removed from the unsaturated soil and remediation proceeds.
5 Summary and Conclusions
SVE tests can be used as a powerful investigation tool to detect VOC sources or hot spots in
unsaturated soil, either to further optimize soil investigation efforts or to decide whether it is advisable
to proceed with remediation.
Moreover, if applied periodically in an existing network of vapour piezometers, this technique can also
be used to detect future soil contamination from new VOC leaks from the facility; making SVE tests a
very useful and competitive environmental tool to survey soil quality in gas stations and in other
facilities with underground VOC tanks and pipelines.
References
Agncia de Residus de Catalunya (ARC), Guia metodolgica per a la realitzaci dassaigs de buit
(proves deficincia) en captadors de vapors en estacions de servei, July 2012
ARC, Guia de prevenci de la contaminaci del sl per a les activitats potencialment contaminants del
sl sota lepgraf CCAE 50500: venda al detall de carburants de lautomoci, March 2009.
US Army Corps of Engineers. Engineering and Design Soil Vapour Extraction and Bioventing, EM
1110-1-4001, June 2002.
US EPA, Detecting Leaks, Successful Methods Step-by-Step November 1989
US EPA, Tank Issues, Design and Placement of Vapour Monitoring Wells, October 1990
US EPA, Soil Vapour Extraction Technology - Reference Handbook, February 1991
US EPA. Standard Test Procedures For Evaluating Leak Detection Methods: Vapour-Phase Out-OfTank Product Detectors, March 1990.
US EPA, Background hydrocarbon vapour concentration for underground fuel storage tanks, 1992
US EPA, Decision-Support Software
HyperVentilate, February 1993.

for

Soil

US EPA. Guidelines for Vapour Monitoring, April 1998.

Vapour

Extraction

Technology

Application: