ISSUING a specification for Continuous Emission Monitoring System (CEMS) requires planning and research. Consultation with environmental process and combustion engineers on-site will provide a list of components required for both environmental reporting and process improvements.
Environmental reporting can be a tricky one, as the company’s EPA licence may specify reporting only NOx. The following year, SO2 may be added to the licence. To incorporate an additional measurement after CEMS installation would be costly.
Thus, it is advisable to consult with others in the same industry and the EPA to find out what else may be required currently and in the future.
Confirm EPA licence requirements for reporting in ppm or mg/Nm3. Confirm if reporting requires measurement to be normalised for O2 or CO2. This is relevant for components after combustion processes.
If normalisation is specified, O2, CO2 will also need to be measured continuously. Industry also must report annual mass emissions for the National Pollutant Inventory (NPI) and National Greenhouse and Energy Reporting Act 2007.
These reports require tonnes per year of greenhouse gases, particulate and so on. Additional information is required to calculate reported data, for example stack gas flow and stack area.
Other important parameters for CEMS specifications would be stack temperature, required accuracies of analysers, zero and span drift tolerances and very importantly, an operational test period (without maintenance). CEMS equipment needs to have at least 95% availability.
Need for CEMS
Continuous emission monitoring provides valuable process information as well as environmental reporting data compared to periodic assessment. Periodic assessment by manual sampling provides data for an instant in time, which may not highlight intermittent process conditions.
Process variations could result in unknown emission exceedances and wasted fuel or product. If a Continuous Emission Monitoring System (CEMS) is installed as a process monitoring and control tool, then with process improvement, environmental issues will tend to look after themselves.
Gas Analyser Technologies
Hot Extractive Systems:
These require a stack sample probe with heated sample line to the analyser. The sample gas to the analyser needs to maintain a temperature above acid dewpoint to prevent water formation. Water formation at the gas analyser will cause severe cross sensitivity issues with other measured gases, plus expensive analyser maintenance.
Analysers for hot extractive systems need to handle typical flue gas temperatures around 150°C making them expensive. The main advantage is the analyser zero and span are easily calibrated. The main disadvantage is high temperature extraction and filtering creates increased maintenance costs.
Cold Extractive Systems:
These require a stack sample probe with heated sample line to a gas conditioning panel. Part of the sample preparation prior to entering the analyser is to remove the water vapour by rapidly chilling the sample.
A typical cold extractive analyser system.
This sample conditioning system requires a significant amount of maintenance to ensure all sample preparation components are operating correctly. The major drawback with this type of system is that a significant percentage of any soluble gas, such as NO2, SO2 and to a lesser extent, NO, is removed along with the water vapour.
To eliminate the need to remove the water vapour and to ensure condensation does not occur in the sample line or analyser components, a technique of diluting the sample at the stack take-off was developed.
Protection against condensation is achieved by diluting the sample to a level at which even the lowest ambient temperature would not cause any condensation to form. Dilution is generally provided with instrument air at typical ratios of 100 or 200 to 1.
The diluted sample is transported from the take-off point to the system housing, where additional sample preparation components and the analysers are mounted. The major drawback of this system is that, in addition to maintaining the dilution system, the analyser has to be significantly more sensitive to monitor the diluted gas.
Cross Stack and Reflective Cross Stack Analysers:
These were a significant step forward in Continuous Emission Monitoring analysers, negating the need for expensive, bulky and high maintenance sample systems.
A typical cross stack analyser.
The disadvantage soon became clear with the inability to directly zero and calibrate the instrument. There are also restrictions on the size of stacks which could be reliably monitored with this type of system.
These systems rely on the stack as the sample cell, sending pulses of infrared or ultraviolet light through the stack to either the receiver or a reflector, which is then returned to the stack mounted transmitter/ receiver unit. A benefit is that the analysed samples are unmodified.
In-Situ Open Path Analysers:
These have the reflector mounted on the probe, with a slot in the probe to allow the gas to pass between the in-stack window and the reflector.
To zero and calibrate this analyser, a second reflector is swung in front of the in-situ stack window and the system zeroed. In addition, test gas can be passed into the enclosed portion of the probe, enabling the instrument to verify calibration.
The major issue with this configuration is that the full system is not challenged – a second reflector is used and effects on the stack measuring reflector are not taken into account. Purge air blowers are required as measuring optics are exposed to the stack environment.
Typical In-Stack Open Path Analyser:
In-situ Enveloped Folded Beam Analysers have the transmitter and receiver mounted in one enclosure, and pulses of infrared or ultraviolet light pass out through a probe containing two lenses.
One lens is on the exit of the optical housing, the second, a process lens, is mounted in the stack end of the in-situ probe.
A typical in-stack open path analyser.
The pulses of IR or UV then pass through a second portion of the probe, which is fitted with sintered panels, allowing the flue gas to freely pass into the measuring cell. The pulses of IR/ UV light strike a retro-reflector and are returned through the same path to the transmitter receiver.
To comply with Environmental Authorities requirements, analysers must have calibration verified. Extractive analysers can be challenged by diverting zero and test gas into the sample cell, enabling the instrument to be recalibrated. This is also possible with Enveloped Folded Beam technology.
Normally, flue gas passes through the sintered panels, filling the in-situ measurement chamber, where the absorption of IR or UV light takes place. Periodically, either automatically or on demand, a solenoid valve can be activated, allowing instrument air to be discharged into the in-situ measuring chamber, forcing out the flue gas.
The analyser then checks zero and adjusts if necessary. In the same way, certified test gas, traceable to a National Standard, can be introduced into the measuring chamber, enabling the instrument to check and if necessary, adjust calibration.
Multi component analysis
Traditionally analysers were designed to measure a single gas species. If multiple gas analysis was required, then a series of analysers was used. With the requirement to monitor and report several pollutant gas emissions, the modern CEM System is capable of simultaneously monitoring, displaying and reporting concentrations of five or six species.
For many years the only way of reliably monitoring several flue gas components, such as NO, NO2, SO2 and HCl was to remove the water vapour from the stack gas sample prior to carrying out the analysis. This was due to the cross sensitivity between water vapour and components to be measured.
Infrared absorption spectra.
Water vapour absorption occurs at the wavelengths pollutant gases are measured. Two techniques have been applied to reduce cross sensitivity. By using gas filter correlation (GFC), the prime sensitivity is improved and cross sensitivity dramatically reduced.
In addition, by monitoring water vapour and applying a cross sensitivity correction, the effect of water vapour can be virtually eliminated, ensuring the instrument accuracy is within the 2% requirements.
These techniques can be used to remove the cross sensitivity of other species, for example CO2.
To ensure the analyser is within ±2% accuracy, it is necessary to carry out automatic correction for changes in sample temperature and pressure. This is achieved by continually monitoring the temperature and pressure within the sample cell and compensating for any changes.
The pressure compensation deals with changes in barometric and flue gas pressure. In addition, if the certified test gas applied for calibration causes a pressure increase, pressure compensation would remove the effect.
Selection of the Continous Emission Monitoring system will depend largely on the application.
In-situ analysis cannot claim to be the answer to all stack gas applications, however they can now compete very favourably with extractive systems.
The main benefits are maintenance requirements are significantly lower than extractive systems; installation costs can be lower as costly sample lines are eliminated. The need for analyser shelters is removed and because the sample cell is in the gas flow, response time is improved.
Typical enveloped folded beam analyser installation.
These measure obscuration or transmission of light through a medium such as smoke and report data in terms of opacity. The relationship between transmission and opacity is: % Transmission (light through the smoke) + % Opacity (light blocked by the smoke) equals 100%. Optical Density (OD) is calculated as OD = -log10(1/T) where T is transmission.
To relate % opacity to particulate, measured as mg/m3, is interesting as there is no direct relationship between the two units of measure. Therefore to relate an opacity monitor to mg/m3 particulate, numerous isokinetic manual samples are required to plot a trend of % opacity to mg/m3 particulate.
Even after this tedious, expensive exercise, variations in smoke colour, for example black, grey, white, greatly affect calibration accuracy. Opacity monitoring was about measuring smoke colour when industry smoke stack emissions were high, before efficient dust collection evolved.
For accurate % opacity measurement, double pass opacity monitors are required. All optical surfaces must be kept clean and purge air blowers are usually required.
This is a technology to measure light flicker caused by particulate passing through a light beam. The more particles pass through the light beam, the greater the AC variation. This AC variation is proportional to mg/m3 dust concentration.
Unlike opacity monitors, zero dust concentration is fixed, therefore no requirement for ongoing zero adjustment, because no dust, no flicker, no flicker equals zero mg/m3. Calibration is by comparison to isokinetic sampling with both results in mg/m3 particulate.
Optical Scintillation Particulate Monitors also measure the opacity signal, which is ratioed against the AC variations measurement of particulate. The ratiometric measurement provides accurate mg/m3 measurement to 90% soiled optics.
This feature and the absence of moving parts greatly reduces maintenance and downtime, which helps satisfy EPA requirements of >95% availability of emission monitoring equipment.
This technology relies on particulate in the air stream colliding with the sensor rod to create an electrostatic charge transfer from the particulate to the sensor rod. Particulate electrostatic charge is proportional to dust concentration.
This technology works well with dry, non conductive dust, typically in baghouses, to monitor bag rupture.
If particulate is moist, sticky or conductive, two issues occur, causing measurement error or failure.
Dust builds up on the sensor rod which insulates it from further charge transfer, and if the sensor insulator is coated with moist or conductive dust, electrostatic charge tracks to the duct wall to ground.
Both of these circumstances cause system failure and the monitor reads low emissions with no indication of filter condition.
This technology is a later innovation of the Triboelectric principle and measures the electrostatic charge induced to the sensor probe from particulate passing the probe. Charge transferred by collision is filtered out, as only the specific frequency of induced charge is measured.
This subtle innovation allows accurate measurement of dust concentration with the sensor probe coated with particulate and allows the manufacturer to provide fully PTFE insulated sensor rods for moist and conductive applications. Maintenance is minimal as dirty sensor probes do not affect measurement accuracy.
Light Scatter Systems:
This technology is based on ambient dust monitors, allowing very accurate measurement of low dust concentrations. Air purge blowers are always required to keep in-stack optics clean and cool.
Wet air applications, for example dryers and wet scrubbers, should be avoided, same as for all optical based systems.
There isn’t a single particulate monitoring technology to suit every application, so be sure you select the right technology for your specific application. Be aware of your current and future requirements plant wide, including legislative requirements.
Make sure your supplier is able to support you before and after purchase. Choosing the right technology for your application also depends on type of process, emission limits, arrestment type, air flow, duct or stack diameter, moisture content, flue gas temperature and as always, available budget.
[Frank Silberberg is Managing Director, Group Instrumentation.]