United States
Environmental Protection
Agency
Environmental Monitoring
Systems Laboratory
Las Vegas, NV 89193-3478
Research and Development
EPA/600/SR-92/219 May 1993
Project Summary
Measurement and Analysis of
Adsistor and Figaro Gas Sensors
Used for Underground Storage
Tank Leak Detection
Marc Portnoff, Richard Grace, Alberto Guzman, and Jeff Hibner
Two different sensor technologies
and their properties were analyzed. The
analysis simulated a leak which occurs
from an underground storage tank.
Figaro gas sensors and the Adsistor
gas sensor were tested in simulated
underground storage tank environ-
ments using the Carnegie Mellon Re-
search Institute (CMRI) automated gas
testing facilities. This automated sys-
tem monitored the sensors' responses
while dynamically exposing them to
various mixtures of methane, butane,
and xylene. The sensors were also
tested to determine the effects of hu-
midity on their responses. Sensor re-
sponses were characterized by
sensitivity, selectivity, and speed of re-
sponse and recovery to select test con-
centrations of methane, butane, and
xylene The test results are presented
as a list of sensor specifications to
allow the potential end user a direct
comparison of these two different types
of sensors.
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 documented in a
separate report of the same title (see
Project Report ordering information at
back).
Introduction
Over two million underground storage
tanks (USTs) are currently being regu-
Mention of trade names or commercial products does
not constitute endorsement or recommendation for
use
lated by the EPA. By 1993, the vast ma-
jority of these tanks are to be equipped
with leak detection monitors to alert tank
owners of any problems. Vapor monitor-
ing equipment, housed in a monitoring
well surrounding the UST, is a common
choice for protecting the environment from
gasoline or product spills from a leaky
tank.
The concept behind vapor monitoring is
that a small liquid leak will generate a
large increase in product vapor concen-
tration. By proper placement of the moni-
tor wells, the product vapor will readily
migrate to the monitoring wells. There, the
vapor sensors will detect the increased
vapor concentrations and sound an alarm.
This study was initiated by the EPA
Office of Underground Storage Tanks to
help the regulators of UST, who use va-
por phase product leak detectors, to bet-
ter understand the capabilities and
limitations of commercial vapor sensors
used in continuous vapor phase product
leak detectors. The study was limited to
characterizing two types of commercial
vapor sensors: The Figaro sensor and the
Adsistor sensor
Procedures
Four types of Figaro gas sensors, mod-
els number 812, 813, 822, 823, and the
Adsistor gas sensor were tested in simu-
lated UST environments using the CMRI
automated gas testing facilities. The char-
acterization of these sensors resulted in a
set of specifications that allows direct com-
parison between the different sensor types.
The Figaro sensors are metal oxide semi-
conductor devices that operate at elevated
Printed on Recycled Paper
-------
temperature. The Adsistor sensor oper-
ates at ambient temperature, and it works
on the principle of gas adsorption in a
polymeric material.
The selection of test gases was based
on a study performed by Geoscience Con-
sultants, Ltd., in 1988. Based on this study,
methane was chosen as a potential inter-
ference that may cause false alarms for
UST monitors. Also iso-butane and m-
xylene were chosen as tags because they
represent major chemical constituents in
gasoline. The sensors were tested to de-
termine their sensitivity and cross sensi-
tivities to methane, butane, xylene, and
humidity.
Three test chambers were built to house
the sensors. One chamber was built to
test nine Adsistor sensors and two cham-
bers to house 12 Figaro sensors, six of
each type. The volume of each test cham-
ber was 1.2 liters. Test chamber tempera-
tures were monitored during testing. The
Adsistor test chamber temperature oper-
ated at room temperature, about 22 de-
grees Centigrade. The Figaro test
chambers ran hotter at about 33 degrees
Centigrade due to the local heating in-
duced by the Figaro sensor's operating
power requirements.
Results and Discussion
The Figaro 823 was chosen for illustrat-
ing the behavior of Figaro sensors for
several reasons. First, the test results
document that the Figaro 812, 822, and
823 sensors all have comparable re-
sponses, considering the statistical spread
in their respective responses. Second, the
Figaro 812 sensor has been discontinued,
being replaced by the 822 model. The
822 and 823 sensors are described by
Figaro as being the same sensor but pack-
aged differently. Finally, the Figaro 813
sensors are very sensitive to methane and
are of limited use for monitoring UST prod-
uct leaks.
The Adsistor sensors tested had model
parameters and sensor responses within
11% of each other.
All the Figaro sensors tested showed
wide variations in the sensor model pa-
rameters and measured responses. For
the Figaro 823 sensors, the spread in
percent standard deviation ranged from
15% to 100%. Similar variations in sensor
behavior were observed for the Figaro
822 and 812 sensors. The Figaro 813
sensors showed a more reproducible re-
sponse with the spread in percent stan-
dard deviation ranging from 3% to 30%.
Gas concentration ramp tests were used
to determine the test gas to which the
sensors were most sensitive. The sensors
were then modeled for this target gas.
The Adsistor sensor clearly responded
to xylene at concentrations over 100 ppm
as indicated by its increased resistance.
The sensor's resistance did not change
when exposed to methane and butane at
concentrations up to 5000 ppm. The
Adsistor sensor was not sufficiently sensi-
tive to the lower xylene concentration
range. Readings of 62 and 68 ppm xylene
in the presence of 10 and 30 ppm xylene,
respectively, reveal the baseline or zero
reading for these sensors. Also, the read-
ing of 233 ppm xylene in the presence of
300 ppm indicates the model is insuffi-
cient to truly characterize this sensor. How-
ever, the small spread of 3.7 ppm among
the nine Adsistor sensors indicates that
the sensors are responding similarly.
For the Figaro 823 sensor, the resis-
tance decreased with respect to all the
test gases. However, it was most sensi-
tive to xylene. The Figaro 823 sensors
are sensitive enough to measure 10 ppm
xylene and were therefore modeled and
calibrated for xylene. This sensitivity to
xylene was also observed for the Figaro
812 and 822 sensors. Thus, the Figaro
812, 822, and 823 sensors were all mod-
eled and calibrated for xylene. In the case
of the Figaro 813 sensors, they were more
sensitive to methane and therefore were
calibrated as methane sensors.
Adsistor sensor sensitivity is not affected
by changes in the level of humidity. For
the Figaro 823 sensor, changes in read-
ing of more than 50% were observed when
the humidity varied from wet to dry condi-
tions.
The Adsistor sensors are insensitive to
both methane and butane, and they are
selective to xylene even in the presence
of a mixture of methane and butane.
The Figaro 823 sensor cross sensitivity
to butane is larger in a mixture than would
be expected. At the background level (500
ppm methane, 500 ppm butane, 100 ppm
xylene), the Figaro 823 sensor reads over
300 ppm xylene. This error can be attrib-
uted mainly to the presence of 500 ppm
butane. The Figaro 823 sensor is insensi-
tive to methane by the slight increase in
the xylene level as the methane is in-
creased to 5000 ppm. When the butane
level is raised to 5000 ppm, the xylene
reading increases to over 1000 ppm, and
when the xylene level is raise to 1000
ppm, the xylene reading is increased to
1700 ppm.
Conclusion
Both the Adsistor and Figaro sensors
have properties that are useful for UST
leak detection. The two sensors respond
well to xylene and are relatively insensi-
tive to methane, which is the primary in-
terfering compound to sensor systems in
subsurface detection of leaking under-
ground storage tanks. The Figaro sensor
is more sensitive to lower levels of sol-
vents than the Adsistor. The Adsistor sen-
sors as a group had better reproducibility
and had a much smaller humidity interfer-
ence in comparison to the Figaro sensors.
These facts make the Adsistor easier to
deal with from an instrumentation and cali-
bration point of view. However, the Adsistor
sensors were observed to have longer
xylene recovery times than the Figaro sen-
sor, which is a factor one must take into
consideration when choosing a sensor
technology.
*U.S. Government Printing Office: 1993 — 750-071/60239
-------
-------
Marc Portnoff, Richard Grace, Alberto Guzman, and Jeff Hibner are with Carnegie
Mellon University, Pittsburgh, PA 15213-2183.
Katrina E. Varner is the EPA Project Officer (see below).
The complete report, entitled "Measurement and Analysis of Adsistor and Figaro
Gas Sensors Used for Underground Storage Tank Leak Detection," (Order No.
PB93-126589; Cost: $19.50; subject to change) will be available only from
National Technical Information Service
5285 Port Royal Road
Springfield, VA 22161
Telephone: 703-487-4650
The EPA Project Officer can be contacted at
Environmental Monitoring Systems Laboratory
U.S. Environmental Protection Agency
Las Vegas, NV 89193-3478
United States
Environmental Protection Agency
Center for Environmental Research Information
Cincinnati, OH 45268
Official Business
Penalty for Private Use
$300
BULK RATE
POSTAGE & FEES PAID
EPA
PERMIT No. G-35
EPA/600/SR-92/219
------- |