An Overview of Hazardous Gas Detection
Robert B. Coulter, P.E.
Course Outline
This two hour online
course provides the student with a basic understanding of the current technology
of hazardous gas detection.
This course includes a multiple choice quiz at the end.
Learning Objective
At
the conclusion of this course, the student will:
About the Author
Robert B.Coulter, PE, is a provider of safety, process engineering and environmental consulting services including inhouse training on safety and environmental topics. For more information, visit his website at www.rbcoulter.com.
Course Introduction
When discussing hazardous gas detection, the first thing that may come to mind is the image of the canary going down into the mine. Fortunately, the current state-of-the-art is much more sophisticated; however, there is yet no "perfect" gas detector. Perfect is defined as responding immediately, extremely accurate, repeatable, specific or nonspecific (as needed), and inexpensive. In this course we will discuss first the theory of gas detection and then the practical application of gas detection.
Course Content
Definitions:
Detector
Technology (Theory):
The following
are the gas detectors that are generally available on the market. A description
of the theoretical mechanism behind each type of detector is also shown:
Flammable
Gas Detectors (FG)
Nearly all
of these detectors currently available on the market consists of a catalysis
material surrounding a RTD (Resistive-Thermo Device). The sensor basically "burns"
the flammable gas and measures the heat output to determine the amount of flammable
gas in the atmosphere. Because of this, these detectors WILL NOT work on nonflammable
gases. For example, many chlorinated organic compounds are nonflammable and
will not respond on this detector. Also, these detectors will not sense a smothering
(low oxygen) hazard. Air is typically required for detection. The sensor will
not work on a mixture of the flammable gas and an inert gas. These devices are
relatively small and inexpensive. The sensors can be fixed in one location or
incorporated into a battery powered handheld device. Some detectors have a small
pump that forces a small stream of air into the sample chamber.
Flammable
gas detectors are high range devices. Don't expect to resolve a reading better
then 2% LEL. For example, if Benzene is the gas (technically vapor) that is
to be detected, then a 2% LEL reading for this gas is about 300 ppmv. Benzene
is a health hazard above 10 ppmv. In other words, these detectors would have
little application for toxic gases. Also, if the detector is calibrated for
a particular gas then its readings will not be correct when detecting other
gases. The manufacturer often provides these "correction" factors for various
gases. Still it is very cumbersome to carry around, or program in, a list of
all the correction factors if you are dealing with multiple vapors or gases.
It is recommended to simply calibrate the device to the LEAST sensitive gas.
Specifically, this is the gas that produces the smallest detector reading at
a particular %LEL (say 20%). The other gases would then produce false high readings
which is much preferable to a false low reading.
To see an
example of a flammable gas detector, click
here. Remember to click the BACK button on your browser to return to the
course.
Electrochemical
Detectors (EC)
These detectors
have a small electrochemical cell with a membrane entrance for the gas. The
gas permeates the membrane and causes a current to flow in the cell. The amount
of gas present directly correlates to the current produced. Usually the cell
is water based. These detectors have a lagging (slow) response because of the
need for the gas to permeate the cell.
Many manufacturers
call these "toxic" gas detectors. This is somewhat confusing since there are
many toxic gases that it will not detect. The gases detected are generally "acid"
gases like carbon monoxide, hydrogen sulfide, and hydrogen chloride. Benzene,
which is toxic, would not be detected by this type of sensor.
These are
usually low range devices and relatively inexpensive. They are often combined
with flammable gas detectors into portable units.
Most oxygen
detectors are of this type. Oxygen detection is important when the hazardous
gas is not flammable nor toxic. A very large concentration (say greater than
4%) of any gas can reduce oxygen levels below a safe level. This should not
be overlooked as gases like carbon dioxide and nitrogen,which are often used
to suppress a fire hazard, can cause injury or death because of their ability
to reduce oxygen levels. Essentially, any gas can be hazardous under certain
conditions.
To see an
example of an electrochemical detector, click
here. Remember to click the BACK button on your browser to return to the
course.
Solid
State Detectors (SS)
These detectors
usually involve a thin film or matrix embedded on an electronic circuit. The
gas sample interacts with the matrix causing a resistance change in the circuit.
Information
about these detectors is sketchy. It appears as though they are mostly low cost,
high range devices. Depending on how the matrix and circuit is designed, these
devices tend to emulate the performance of EC and/or FG detectors.
Infra-red
Detectors (IR)
This type
of detector works by measuring the adsorption of a particular guess at preselected
wavelengths of Infrared (IR) light. Gases can be thought of having an IR "fingerprint"
based on it's molecular structure. Different IR wavelengths are adsorbed depending
on the types of bonds (C-H, C=H, etc.) present in the molecule. The amount of
IR radiation adsorb is directly correlated to the amount of the gas present.
A IR wavelength
is chosen that represents the type of gas that is being detected. For example,
if hydrocarbons are being monitored, a wavelength that is adsorbed by the C-H
bond may be selected. The device would then be configured to direct this wavelength
through the sample chamber. For example, methane has four C-H bonds. If methane
is in the sample chamber then a response directly related to the these four
bonds would be expected. However, many hydrocarbons have more C-H bonds per
volume of gas than methane. Butane has ten C-H bonds. The presence of butane
in the same sample chamber will cause a much greater response in the detector.
These detectors
have a distinct advantage over the previous two detectors discussed (flammable
and electrochemical) in that the IR detector can be configured to be much more
discerning for the target gas. For example, the flammable gas detector could
not distinguish between propane and ethylene. Both of these gases only appear
as "flammable" gases. However, an IR detector could be setup only to detect
the C=C bond in the ethylene. The propane could then be "invisible" to the detector
if one only wanted to know the ethylene concentration.
These detectors
can be low or high range and are moderately expensive. This detector also has
the unique ability to monitor "fence lines". This is done by setting up an IR
beam transmitter at one point and an IR detector as far away as 500 feet.
To see an
example of an IR detector, click
here.Remember to click the BACK button on your browser to return to the
course.
Photoionization
Detectors (PI)
Where the
IR detector makes use of the Infrared radiation, the photoionization utilizes
the ultraviolet spectrum. UV light of an energy less than 11.7 eV (electron
volts) is produced and directed into the sample chamber. If a compound is present
that is "excited" by the UV light, electrons will be produced and sensed by
the detector. The response of the detector is directly related to the number
of electrons produced which is related to the type and concentration of gas
present. Gases that are ionized above 11.7 eV cannot be detected because of
interference with oxygen.
The set
of gases that can be detected by this sensor is relatively small (compared to
FG, IR, and FID). The gases that are detected by this analyzer tend to be higher
molecular weight organic vapors. Methane, some chlorinated organic and many
cyanide compounds are not detected. These detectors; however, are not very complicated
to operate/maintain and are reliable.
These detectors
can be low or high range and are moderately expensive.
Flame
Ionization Detectors (FID)
This detector
sends the sample over a hydrogen flame where the sample is "ionized". The detector
measure the number of these ions as an electrical signal. For all practical
purposes, the signal is directly related to the number of carbon atoms present
in the sample. Carbon dioxide , CO2, is not detected, but this is actually an
advantage since this gas is not usually desired to be detected.
The big
advantage of this detector above almost all others is that it is generally not
affected by the presence of water. Many applications involve trying to discern
hazardous gases in almost 100% relative humidity environments. The disadvantage
is that these detectors are somewhat bulky (if portable) and involve burning
a hydrogen flame to run the detector.
These detectors
can be low or high range and are fairly expensive.
To see an
example of a flame ionization detector, click
here. Remember to click the BACK button on your browser to return to the
course.
Mass
Spectrometer (MS)
Because
gases have varying molecular weights (masses), this mass (or mass fragments)
can be detected to determine the amount and type of gas present. The basic technique
is that a gas sample enters into a fragmentation chamber where the gas is split
into a distribution of mass sizes. The magnitude and size distribution of these
masses is a function of the concentration and composition of the gas sample.
This magnitude and size distribution is determined by sending the fragments
through a vacuum chamber surrounding by a magnetic field. The magnetic field
is varied by electronics to only allow a particular mass fragment to reach the
detector at a particular moment. Usually the magnetic field is adjusted following
a preprogrammed"path" to detect certain "mass channels" that correspond to the
target gases to be detected. The detector can be a high range or low range detector.
Ultrahigh
vacuum is needed to run these detectors. Calibration is difficult. It goes without
saying that these are among the most expensive detectors. The distinct advantage
of these detectors is their discerning ability is very high and are equally
suitable for detecting organic and inorganic gases.
Gas
Detector Tubes / Paper Tape
Gas detector
tubes are small glass tubes containing a mixture of specific chemicals designed
to react with a particular target gas. The mix of chemicals is chosen in such
a way that a color stain develops in the tube to give a visual indication of
the presence of the target gas. In most cases there are markings on the glass
tube that indicate the concentration of the target gas. Usually the length of
the stain corresponds to the gas concentration.
This is
non-electronic technology depending solely on chemistry. The usually procedure
is two draw a specified volume of gas through a glass tube with a small hand
pump. The glass tubes can be provided to detect a wide variety of chemicals
throughout various ranges.
Paper tape
technology is similar but the stain occurs on a tape strip. This technology
is usually employed on isocyanate compounds.
These gas
detection methods are relatively inexpensive. However, they cannot be used as
a first warning device. They are almost always used after it is known (by some
other means) that a hazardous gas is present. Their use is intended to more
specifically identify and quantify the hazardous gas. Even at that, some knowledge
of the possible gases present must be known. Each tube is generally made for
detecting a specific gas. If the list of possible gases is large (say greater
than 10) then it could take over thirty minutes to sample/test for each individual
gas (3-4 minutes per gas compound). Tubes and the paper tape cannot be reused.
To see an
example of typical gas detector tubes, click
here. Remember to click the BACK button on your browser to return to the
course.
Gas
Chromatography (GC)
This is
really not a detector. It is a small tube (a column) with packing that the sample
gas is sent through first before being sent to a detector. The idea is to separate
different gases that may be present in the sample. Different gases would be
indicated by their varying arrival times at the detector. The detector could
be of any type but is usually an FID (see above). The GC enhances the ability
of a detector to discern the concentrations of individual compounds (species)
in a gas mixture. For example, a gas mixture of toluene and benzene sent directly
to an FID detector would show one peak. The response of the detector would simply
indicate the cumulative presence of the two species with no indication of the
"split". If the sample is sent through a properly configured GC then two peaks
may be seen at the detector where the magnitude of each peak would directly
correlate to the concentration of each species.
The disadvantage
of having a GC in a hazardous gas detection system is the longer response time
to detect the gas. The cost to add GC to a detection can be high. However, the
desire to know the "split" of the gas concentrations may outweigh these disadvantages.
For example, one gas in a sample could be very toxic while another gas is only
flammable.
Monitors,
Multiple Point Monitoring, and Alarms
As with
other input devices the output of most gas detectors can be routed to a remote
location for monitoring. This can be a computer or specially designed electronic
monitor.
Sometimes
multiple detection points are required. One can install detectors with transmitters
at each point,or install a solitary central detection system with sample tubes
running to all desired detection points. Multiple detectors have the advantage
of increased response time. A solitary detector is less expensive particularly
if it is an FID, IR, or MS; however, a switching sampling system and controls
must be added if one wants specific data about a particular sample location.
Usually
alarms are desired to indicate a hazardous gas level. Alarms are usually set
at some fraction of LEL (say 20% LEL) in the case of flammable gas monitoring,
or approximately one half of the permissible exposure limit (PEL) for toxic
gases. Oxygen monitoring is usually alarmed at 18% oxygen in air (21% is normal).
It is a good idea (in many cases it is required) to have a emergency action
plan in place that outlines the response to a hazardous gas alarms.
Examples
of Practical Application:
The following
are hypothetical examples of how gas detection technology can be applied:
Example
#1
A company
stores several nonspecific flammable liquids in storage tanks of various sizes
in a room adjacent to a work area. The company is interested in detecting leaks
and spills from the storage tanks so that a major spill and potential fire could
be avoided. Based on this information, what would be the optimal detector for
this system?
In this case the liquids are flammable and nonspecific. A PI detector is likely
to be inadequate because some flammable materials do not respond to it. An EC
detector is not likely to work on these organic compounds. It is likely that
FI, IR, FG, or MS detectors will work in this application but are expensive.
The FG detector is the least expensive and should be considered first.
Example
#2
In the example
above, where would the best placement of the detectors be -- above the tanks
near the ceiling or just above the floor?
Flammable liquids are most likely to generate vapors that will sink. A detector
placed nearer the floor will more likely detect a leak sooner.
Example
#3
In example
#1, In the example above, a company representative has indicated that toluene
is the main constituent in the storage tanks. After a review of the other constituents
in the tanks it is decided that the detectors can be calibrated with a span
gas consisting of air and toluene. If the span calibration bottle is desired
to contain toluene in air at 20 % LEL, what concentration of toluene (in %)
does the gas supply company put into the bottle? (Assume the LEL of toluene
in air is 1.3 %) Note: The span gas is used to set the upper calibration point
on the transmitter.
The answer is 0.26 % Toluene in air. This is obtained by taking 20 % of 1.3.
Example
#4
A company
rep is interested in purchasing a "toxic" detector with an alarm to warn of
unspecified gas hazards. The company rep does not have any knowledge of the
types of gases that could be toxic hazards. The company rep is considering the
purchase of a "toxic" detector from a safety catalog. A review of this detector
indicates that it is based on the electrochemical (EC) technology. Is the company
rep making a good decision in purchasing this detector?
Probably not. This "toxic" detector, as described in many vendor's publications,
is only suitable for a specific group (usually inorganic) of toxic gases. Many
toxic organic gases would not indicate on this type of detector. The company
rep would probably cause more harm than good by installing this detector because
of the false sense of security it could bring. Anyone interested in a generic
"toxic" sensor should consult a professional who is knowledgeable of multiple
gas detection technologies and is capable of understanding the customer's specific
needs.
Example
#5
A beverage
company is interested in a detector for ethanol vapors which can be flammable
and pose a health risk. The company currently has flammable gas (FG) detectors
installed but they complain that they are unreliable. A tour of the facility
indicates a large amount of water vapor present. What should the company consider
next to replace the FG detector?
It appears as though the companies FG detectors are affected by the water vapor.
Many inexpensive detectors have this problem. IR could also be negatively affected
by water. The best choice is an FID detector.
Example
#6
A refinery/chemical
complex is interested in purchasing portable gas detectors for finding leaks
from process equipment. The complex consists of a chemical plant that has specific
organic chemicals and the refinery that has a variety of nonspecific hydrocarbon
compounds. FI and PI detectors are being considered. The customer needs multiple
units for both plants. Water vapor is not an issue at either facility. Based
on this information, what should be considered first?
In general the PI detector is less expensive, easier to calibrate/operate and
to carry. The main problem with the PI detector is that some compounds cannot
be detected. A review of the chemical complex target gases indicate that all
gases have photoionization potentials less that 11.7 eV. So for the chemical
plant the PI would be the best choice. However, the refinery complex has many
nonspecific organic compounds that can't all be identified. In fact, methane
is present in the refinery which cannot be detected by photoionization (Methane's
ionizational potential = 12.98 eV). The best choice for the refinery is FID.
Example
#7
A firm cleans
tank trailers in a semi-closed garage. There are hazards from carbon monoxide
(CO) and hydrogen sulfide (H2S). The carbon monoxide comes from the engines
of the tractors. The hydrogen sulfide emanates from some tank trailers. The
firm is interested in an electrochemical detector. However, the sensor is five
times more sensitive to hydrogen sulfide than carbon monoxide. In other words,
if the sensor is calibrated to read ppms of CO, it will show a reading five
times higher than the actual PPM when detecting H2S. Let's assume the PEL for
H2S is 10 PPM, and the PEL for CO is 35 PPM. For the CO calibrated sensor, where
should the alarm be set?
It
would be a good idea to set the toxic alarm at half the PEL. Half of 35 PPM
is about 17 PPM. Let's assume we can set the alarm at 17 PPM. This would alarm
adequately for CO. H2S needs to be alarmed at 5 PPM of H2S. Because of the five
times sensitivity of H2S, this gas would set off the alarm at 17/5 or 3.4 PPM.
This is below 5 PPM, so the assumption of setting that alarm at 17 PPM is correct.
Example
#8
A safety
engineer is interested in knowing the PEL (permissible exposure limit), LEL
(lower explosive limit), and photoionization potential of ammonia in order to
configure a gas monitoring system. What are these values?
A
good source of these values is provided by NIOSH's Pocket Guide to Chemical
Hazards (see the link below). The ammonia entry lists a NIOSH REL (25 PPM) and
an OSHA PEL? (50 PPM). Let's use the smaller number, 25 PPM The LEL is listed
as 15 %. The ionization potential is 10.18 eV.
Current state
of hazardous gas detection employs technology that must be well understood to
make optimal use of a particular monitoring system. Gas detectors can be low
cost to very expensive, and all types have their advantages and disadvantages.
One should use caution when selecting a gas detector. If there is any doubt,
consult a professional familiar with gas detection for proper advice.
To conclude this course the student should read "Key Concepts in Gas Detection Systems" by clicking the first link below. The second link is NIOSH's guide to chemical hazards. It lists the exposure limits, LELs, photoionization potential and other useful information for many compounds. The student should examine a few chemicals in this database to be familiar with it. After reading the required course material the student should be ready for the test.
Related Links
For additional technical information related to this subject,
please visit the following websites or web pages:
Scott
Instruments - Key Concepts in Gas Detection Systems (PDF reader needed)
NIOSH
Pocket Guide to Chemical Hazards
Once you finish studying the above course content, you need to take a quiz to obtain the PDH credits.