- Combustible Gas Sensors
- Catalytic Sensor
- Sensor Output
- Speed of Response
- Semiconductor Sensor
- Thermal Conductivity
- Infrared Gas Detector
- Open Path Flammable Infrared Gas Detector
- Electrochemical Sensor
- Chemcassette® Sensor
- Comparison of Gas Detection Techniques
Combustible Gas Sensors
Many people have probably seen a flame safety lamp at some time and know something about its use as an early form of ‘firedamp’ gas detector in underground coal mines and sewers.
Although originally intended as a source of light, the device could also be used to estimate the level of combustible gases- to an accuracy of about 25-50%, depending on the user’s experience, training, age, colour perception etc. Modern combustible gas detectors have to be much more accurate, reliable and repeatable than this and although various attempts were made to overcome the safety lamp’s subjectiveness of measurement (by using a flame temperature sensor for instance), it has now been almost entirely superseded by more modern, electronic devices.
Nevertheless, today’s most commonly used device, the catalytic detector, is in some respects a modern development of the early flame safety lamp, since it also relies for its operation on the combustion of a gas and its conversion to Carbon Dioxide and water.
(Back to top)
Nearly all modern, low-cost, combustible gas detection sensors are of the electro-catalytic type. They consist of a very small sensing element sometimes called a ‘bead’, a ‘Pellistor’, or a ‘Siegistor’- the last two being registered trade names for commercial devices. They are made of an electrically heated platinum wire coil, covered first with a ceramic base such as Alumina and then with a final outer coating of Palladium or Rhodium catalyst dispersed in a substrate of Thoria.
This type of sensor operates on the principle that when a combustible gas/air mixture passes over the hot catalyst surface, combustion occurs and the heat evolved increases the temperature of the ‘bead’. This in turn alters the resistance of the platinum coil and can be measured by using the coil as a temperature thermometer in a standard electrical bridge circuit. The resistance change is then directly related to the gas concentration in the surrounding atmosphere and can be displayed on a meter or some similar indicating device.
(Back to top)
To ensure temperature stability under varying ambient conditions, the best catalytic sensors use thermally matched beads. They are located in opposing arms of a Wheatstone bridge electrical circuit, where the ‘sensitive’ sensor (usually known as the ‘s’ sensor) will react to any combustible gases present, whilst a balancing, ‘inactive’ or ‘non-sensitive’ (n-s) sensor will not. Inactive operation is achieved by either coating the bead with a film of glass or de-activating the catalyst so that it will act only as a compensator for any external temperature or humidity changes.
A further improvement in stable operation can be achieved by the use of poison resistant sensors. These have better resistance to degradation by substances such as Silicones, Sulphur and Lead compounds which can rapidly de-activate (or ‘poison’) other types of catalytic sensor.
(Back to top)
Speed of Response
To achieve the necessary requirements of design safety, the catalytic type of sensor has to be mounted in a strong metal housing behind a flame arrestor. This allows the gas/air mixture to diffuse into the housing and on to the hot sensor element, but will prevent the propagation of any flame to the outside atmosphere. The flame arrestor slightly reduces the speed of response of the sensor but, in most cases the electrical output will give a reading in a matter of seconds after gas has been detected. However, because the response curve is considerably flattened as it approaches the final reading, the response time is often specified in terms of the time to reach 90 percent of its final reading and is therefore known as the T90 value. T90 values for catalytic sensors are typically between 20 and 30 seconds.
(N.B. In the USA and some other countries, this value is often quoted as the lower T60 reading and care should therefore be taken when comparing the performance of different sensors).
(Back to top)
The most common failure in catalytic sensors is performance degradation caused by exposure to certain poisons’. It is therefore essential that any gas monitoring system should not only be calibrated at the time of installation, but also checked regularly and re-calibrated as necessary. Checks must be made using an accurately calibrated standard gas mixture so that the zero and ‘span’ levels can be set correctly on the controller.
Codes of practice such as EN50073:1999 can provide some guidance about the calibration checking frequency and the alarm level settings. Typically, checks should initially be made at weekly intervals but the periods can be extended as operational experience is gained. Where two alarm levels are required, these are normally set at 20-25% LEL for the lower level and 50-55% LEL for the upper level.
Older (and lower cost) systems require two people to check and calibrate, one to expose the sensor to a flow of gas and the other to check the reading shown on the scale of its control unit. Adjustments are then made at the controller to the zero and span potentiometers until the reading exactly matches that of the gas mixture concentration.
Remember that where adjustments have to be made within a flameproof enclosure, the power must first be disconnected and a permit obtained to open the enclosure.
Today, there are a number of ‘one-man’ calibration systems available which allow the calibration procedures to be carried out at the sensor itself. This considerably reduces the time and cost of maintenance, particularly where the sensors are in difficult to get to locations, such as an offshore oil or gas platform. Alternatively, there are now some sensors available which are designed to intrinsically safe standards, and with these it is possible to calibrate the sensors at a convenient place away from the site (in a maintenance depot for instance). Because they are intrinsically safe, it is allowed to freely exchange them with the sensors needing replacement on site, without first shutting down the system for safety.
Maintenance can therefore be carried out on a ‘hot’ system and is very much faster and cheaper than early, conventional systems.
(Back to top)
Sensors made from semiconducting materials gained considerably in popularity during the late 1980s and at one time appeared to offer the possibility of a universal, low cost gas detector. In the same way as catalytic sensors, they operate by virtue of gas absorption at the surface of a heated oxide. In fact, this is a thin metal-oxide film (usually oxides of the transition metals or heavy metals, such as tin) deposited on a silicon slice by much the same process as is used in the manufacture of computer ‘chips’. Absorption of the sample gas on the oxide surface, followed by catalytic oxidation, results in a change of electrical resistance of the oxide material and can be related to the sample gas concentration. The surface of the sensor is heated to a constant temperature of about 200-250°C, to speed up the rate of reaction and to reduce the effects of ambient temperature changes.
Semiconductor sensors are simple, fairly robust and can be highly sensitive. They have been used with some success in the detection of Hydrogen Sulphide gas, and they are also widely used in the manufacture of inexpensive domestic gas detectors. However, they have been found to be rather unreliable for industrial applications, since they are not very specific to a particular gas and they can be affected by atmospheric temperature and humidity variations. They probably need to be checked more often than other types of sensor, because they have been known to ‘go to sleep’ (i.e. lose sensitivity) unless regularly checked with a gas mixture and they are slow to respond and recover after exposure to an outburst of gas.
(Back to top)
This technique for detecting gas is suitable for the measurement of high (%V/V) concentrations of binary gas mixes. It is mainly used for detecting gases with a thermal conductivity much greater than air e.g. Methane and Hydrogen. Gases with thermal conductivities close to air cannot be detected E.g. Ammonia and Carbon Monoxide. Gases with thermal conductivities less than air are more difficult to detect as water vapour can cause interference E.g. Carbon Dioxide and Butane. Mixtures of two gases in the absence of air can also be measured using this technique.
The heated sensing element is exposed to the sample and the reference element is enclosed in a sealed compartment. If the thermal conductivity of the sample gas is higher than that of the reference, then the temperature of the sensing element decreases. If the thermal conductivity of the sample gas is less than that of the reference then the temperature of the sample element increases. These temperature changes are proportional to the concentration of gas present at the sample element.
(Back to top)
Infrared Gas Detector
Many combustible gases have absorption bands in the infrared region of the electromagnetic spectrum of light and the principle of infrared absorption has been used as a laboratory analytical tool for many years. Since the 1980s, however, electronic and optical advances have made it possible to design equipment of sufficiently low power and smaller size to make this technique available for industrial gas detection products as well.
These sensors have a number of important advantages over the catalytic type. They include a very fast speed of response (typically less than 10 seconds), low maintenance and greatly simplified checking, using the self-checking facility of modern micro-processor controlled equipment. They can also be designed to be unaffected by any known ‘poisons’, they are failsafe and they will operate successfully in inert atmospheres, and under a wide range of ambient temperature, pressure and humidity conditions.
The technique operates on the principle of dual wavelength IR absorption, whereby light passes through the sample mixture at two wavelengths, one of which is set at the absorption peak of the gas to be detected, whilst the other is not. The two light sources are pulsed alternatively and guided along a common optical path to emerge via a flameproof ‘window’ and then through the sample gas. The beams are subsequently reflected back again by a retro-reflector, returning once more through the sample and into the unit. Here a detector compares the signal strengths of sample and reference beams and, by subtraction, can give a measure of the gas concentration.
This type of detector can only detect diatomic gas molecules and is therefore unsuitable for the detection of Hydrogen.
(Back to top)
Open Path Flammable Infrared Gas Detector
Traditionally, the conventional method of detecting gas leaks was by point detection, using a number of individual sensors to cover an area or perimeter. More recently, however, instruments have become available which make use of infrared and laser technology in the form of a broad beam (or open path) which can cover a distance of several hundred metres. Early open path designs were typically used to complement point detection, however the latest 3rd generation instruments are now often being used as the primary method of detection. Typical applications where they have had considerable success include FPSOs, add jettys, loading/unloading terminals, pipelines, perimeter monitoring, off-shore platforms and LNG (Liquid Natural Gas) storage areas.
Early designs use dual wavelength beams, the first coinciding with the absorption band peak of the target gas and a second reference beam which lies nearby in an unabsorbed area. The instrument continually compares the two signals that are transmitted through the atmosphere, using either the back-scattered radiation from a retroreflector or more commonly in newer designs by means of a separate transmitter and receiver. Any changes in the ratio of the two signals is measured as gas. However, this design is susceptible to interference from fog as different types of fog can positively or negatively affect the ratio of the signals and thereby falsely indicate an upscale gas reading/alarm or downscale gas reading/fault. The latest 3rd generation design uses a double band pass filter that has two reference wavelengths (one either side of the sample) that fully compensates for interference from all types of fog and rain. Other problems associated with older designs have been overcome by the use of coaxial optical design to eliminate false alarms caused by partial obscuration of the beam and the use of xenon flash lamps and solid state detectors making the instruments totally immune to interference from sunlight or other sources of radiation such as flare stacks, arc welding or lightning.
Open path detectors actually measure the total number of gas molecules (i.e. the quantity of gas) within the beam. This value is different to the usual concentration of gas given at a single point and is therefore expressed in terms of LEL meters.
(Back to top)
Gas specific electrochemical sensors can be used to detect the majority of common toxic gases, including CO, H2S, Cl2, SO2 etc. in a wide variety of safety applications.
Electrochemical sensors are compact, require very little power, exhibit excellent linearity and repeatability and generally have a long life span, typically one to three years. Response times, denoted as T90, i.e. time to reach 90% of the final response, are typically 30-60 seconds and minimum detection limits range from 0.02 to 50ppm depending upon target gas type.
Commercial designs of electrochemical cell are numerous but share many of the common features described below:
Three active gas diffusion electrodes are immersed in a common electrolyte, frequently a concentrated aqueous acid or salt solution, for efficient conduction of ions between the working and counter electrodes.
Depending on the specific cell the target gas is either oxidised or reduced at the surface of the working electrode. This reaction alters the potential of the working electrode relative to the reference electrode. The primary function of the associated electronic driver circuit connected to the cell is to minimise this potential difference by passing current between the working and counter electrodes, the measured current being proportional to the target gas concentration. Gas enters the cell through an external diffusion barrier that is porous to gas but impermeable to liquid.
Many designs incorporate a capillary diffusion barrier to limit the amount of gas contacting the working electrode and thereby maintaining “amperometric” cell operation.
A minimum concentration of Oxygen is required for correct operation of all electrochemical cells, making them unsuitable for certain process monitoring applications. Although the electrolyte contains a certain amount of dissolved Oxygen, enabling short-term detection (minutes) of the target gas in an Oxygen-free environment, it is strongly advised that all calibration gas streams incorporate air as the major component or diluent.
Specificity to the target gas is achieved either by optimisation of the electrochemistry, i.e. choice of catalyst and electrolyte, or else by incorporating filters within the cell which physically absorb or chemically react with certain interferent gas molecules in order to increase target gas specificity. It is important that the appropriate product manual be consulted to understand the effects of potential interferent gases on the cell response.
The necessary inclusion of aqueous electrolytes within electrochemical cells results in a product that is sensitive to environmental conditions of both temperature and humidity. To address this, the patented Surecell™ design incorporates two electrolyte reservoirs that allows for the ‘take up’ and ‘loss’ of electrolyte that occurs in high temperature/high humidity and low temperature/low humidity environments.
Electrochemical sensor life is typically warranted for 2 years, but the actual lifetime frequently exceeds the quoted values. The exceptions to this are Oxygen, Ammonia and Hydrogen Cyanide sensors where components of the cell are necessarily consumed as part of the sensing reaction mechanism.
(Back to top)
Chemcassette® is based on the use of an absorbent strip of filter paper acting as a dry reaction substrate. This performs both as a gas collecting and gas analysing media and it can be used in a continuously operating mode. The system is based on classic colorimetry techniques and is capable of extremely low detection limits for a specific gas. It can be used very successfully for a wide variety of highly toxic substances, including Di-isocyanates, Phosgene, Chlorine, Fluorine and a number of the hydride gases employed in the manufacture of semiconductors.
Detection specificity and sensitivity are achieved through the use of specially formulated chemical reagents, which react only with the sample gas or gases. As sample gas molecules are drawn through the Chemcassette® with a vacuum pump, they react with the dry chemical reagents and form a coloured stain specific to that gas only. The intensity of this stain is proportionate to the concentration of the reactant gas, ie, the higher the gas concentration, the darker is the stain. By carefully regulating both the sampling interval and the flow rate at which the sample is presented to the Chemcassette®, detection levels as low as parts-per-billion (ie, 10 -9) can be readily achieved.
Stain intensity is measured with an electro-optical system which reflects light from the surface of the substrate to a photo cell located at an angle to the light source. Then, as a stain develops, this reflected light is attenuated and the reduction of intensity is sensed by the photo detector in the form of an analogue signal. This signal is, in turn, converted to a digital format and then presented as a gas concentration, using an internally-generated calibration curve and an appropriate software library. Chemcassette® formulations provide a unique detection medium that is not only fast, sensitive and specific, but it is also the only available system which leaves physical evidence (i.e. the stain on the cassette tape) that a gas leak or release has occurred.
(Back to top)
Comparison of Gas Detection Techniques
(Back to top)