Measuring Unburned Carbon in Flyash with an Automated, On-line Monitor

Issue 3 and Volume 112.

By David Clifford, Thermo Fisher Scientific, Environmental Instruments Division, Air Quality Instruments

Increasingly stringent NOX regulations have required plant operators to modify traditional combustion processes. SOX regulations and deregulation have forced plant operators to diversify their coal sources. And new mercury control requirements have been instituted by the U.S. Environmental Protection Agency. These changes challenge plant operators who are trying to achieve the low carbon levels in flyash needed to allow it to be used in the concrete industry. Part of the solution lies with improved flyash monitoring.

On-line carbon-in-ash (CIA) monitors automate and increase the availability of combustion efficiency tracking data that can have several benefits, including:

  • Increased combustion efficiency
  • Increased quantity of low-carbon flyash available for sale
  • Reduced amount of flyash requiring disposal and landfilling
  • Increased consistency, availability of loss on ignition (LOI) type measurements
  • Eliminated or reduced manual LOI procedures
  • Improved mercury capture and control efficiency while balancing plant operating efficiency
  • Increased SCR catalyst life

For example, plant operators can use flyash residual carbon data to assess the effectiveness of plant optimization efforts. Flyash carbon content reduction is a direct result of improved combustion efficiency; therefore, plant operators employ the CIA measurement as a key component in their strategy to reduce fuel use. Also, by tracking changes in flyash carbon levels, plant operators can monitor coal mill performance, optimize burner configuration, track furnace performance and conduct other engineering studies. Furthermore, lower carbon levels in the flyash can increase the availability of low carbon flyash for sale and reduce or eliminate the need for off-site disposal.

The impacts of these benefits, either alone or collectively, can significantly contribute to a plant’s financial position. In fact, by using industry average prices for fuel and the average cost of ash disposal in combination with increased revenue from improved flyash sales, calculations suggest that the payback time for some CIA monitor installations, including the one discussed here, can be as little as a month or two.

Thermo Fisher Scientific Air Quality Instruments undertook a project to produce an automated, on-line, CIA monitor that incorporates recent advances in electronic devices and analysis technologies to provide reliable, automated, highly-accurate and precise measurement of unburned carbon in flyash samples at coal-fired power plants. The company tested the monitor and summarized the results of these tests to assess the monitor’s performance under actual plant operating conditions on coal-fired utility boiler exhaust ducts.

Operating Principle

In general, the monitor measures the amount of unburned carbon in the flyash sample using an inertial mass measurement/thermal oxidation technique. The sample analysis technique is similar to the LOI test performed in a laboratory and directly measures the CIA percentage. However, by measuring the amount of carbon dioxide (CO2) produced by the oxidation of elemental or organic carbon present in the sample, the analysis is not influenced by variation in coal types. The monitor can achieve a measurement accuracy of less than ± 0.5 percent CIA with a minimum detection limit of about 1.0 percent CIA. Additionally, the percent-CIA measurement was improved by integrating an optional dual sampling system into the system configuration to allow time-shared monitoring of two sample points in a duct or single sample points in two different ducts that are closely located.

The monitor’s onboard computer automatically controls five main steps that are included in the sampling/analysis cycle:

  1. 1. Isokinetic sample extraction and collection on a filter cartridge affixed to the mass transducer
  2. 2. Total sample mass measurement using the tapered element oscillating microbalance (TEOM) mass transducer
  3. 3. Oxidation sample in a high temperature furnace with combustion products collected and analyzed for CO2 using the onboard non-dispersive infrared (NDIR) CO2 detector
  4. 4. Data processing, calculation and reporting of percent CIA
  5. 5. Filter cleaning and repositioning to start new test

The entire sample collection and analysis cycle is completed in approximately 12 minutes; therefore, the monitor can provide up to five readings an hour.

The percent CIA is calculated as follows:

First, the collected sample mass is measured on an industrial grade TEOM mass transducer using the mass measurement principle (following Patashnick and Rupprecht, 1991) relating the change in mass to a change in the TEOM’s oscillation frequency as described by the following equation:

Eq. 1 ΔM = K0 ∗ ((1/f12) – (1/f02))

Where, ΔM is the change in collected mass,

f0 = the initial frequency

f1 = the final frequency

K0 = calibration constant of the TEOM’s tapered element.

In the next step, the sample is transferred into a high-temperature furnace and quickly heated to approximately 800 C. Clean, dry air pumped through the furnace flushes the combustion gases containing CO2 to a non-dispersive, NDIR CO2 detector. The mass of CO2 is determined using the following equation:

Eq. 2 CO2(kmol) = volume CO2(l)


The third step is where the mass of CO2 is converted to the mass of carbon (C):

Eq. 3 C(gmol) = CO2(gmol)

Eq. 4 C(g) = C(gmol) ∗ 12.011(g/gmol)

In the fourth step, the percent CIA (w/w) measurement is then calculated by dividing the carbon mass by the total original sample mass (ΔM):

Eq. 5 %CIA = C(g) ∗ 100%


Figure 1 (page 40) is a simple schematic diagram of the monitor sample and analysis components.

Click here to enlarge image


Performance Evaluation Overview

In re-engineering the monitor, Thermo Fisher Scientific’s goal was to preserve its measurement performance characteristics while addressing reliability and maintainability concerns. The company addressed these issues in the new monitor with an objective to ensure that the monitor can be maintained with a weekly maintenance visit taking less than 15 minutes and with preventive maintenance service every six months taking approximately one to four hours.

The test’s first objective was to determine how the new monitor operated under real-world conditions. Thermo Fisher Scientific installed monitors in several locations. One monitor was installed in a Midwest power plant that burns local sub-bituminous coals for fuel. A second monitor was installed at a utility station in the Northeast where eastern U.S. and South American bituminous coals were used for fuel. Both plants obtain fuel from multiple sources. After a short startup/conditioning period, Thermo Fisher Scientific tracked operating reliability, maintenance requirements and measurement performance over a three-month period. At both locations, the monitor was installed just downstream of the economizer. Plant staff performed routine operations and weekly maintenance while Thermo Fisher Scientific personnel performed non-routine maintenance and repairs.

The test program’s second objective was to qualify the monitor’s measurement accuracy during typical plant operating conditions by periodically collecting and analyzing samples with the monitor and time correlated samples taken from the flyash collection hoppers.

Operating Performance

The Midwestern plant is operated in a baseload supply mode so it typically runs at a high operating rate during the daytime hours but reduces load in the early morning hours. The plant blends incoming coal with stockpiled fuel.

Monitor installation requirements include a suitable sample port, power (120VAC, 20A) and instrument air (5.4 to 6.8 atm, 0.57 m3/min peak for approximately one minute). Installation, start-up and calibration require approximately four hours when power and air supplies are already installed.

The monitor was operated as it would be when part of normal plant operations. The plant engineer responsible for the environmental monitoring programs served as the monitor operator. The period shown in Figure 2 represents a period typical of plant and monitor operations.

Click here to enlarge image

Over the three-month evaluation period shown in Figure 2, the monitor operated properly and required only weekly maintenance (filter cartridge exchange, function checks), except for two minor malfunctions that resulted in unexpected downtime. The first malfunction was caused by the operator inadvertently positioning the mass transducer incorrectly, resulting in damage to the transducer mounting. When repairs were made, the monitor’s operating firmware was updated to prevent this from reoccurring. The other malfunction occurred when the instrument air supply pressure exceeded the safe operating range of the monitor’s linear slide mechanism. The monitor’s operating firmware properly and automatically put the monitor in suspend mode and issued an alarm warning to the operator who responded to reset the pressure regulator to the proper setting.

At the second test site in the Northeast, installation requirements were the same as for the Midwestern location. After a short start-up period, the monitor was allowed to operate as if under normal plant operations. Plant operations personnel operated and maintained the monitor.

Figure 3 presents a summary of percent CIA measurement data along with duct velocity and temperature data that can be used to help identify changes in plant load and operations.

Click here to enlarge image

The evaluation at this site proceeded without significant design, hardware or operator training issues. Operating reliability met the design goals.

Measurement Accuracy

For measurement accuracy evaluation, personnel installed the monitor in accordance with the system operating manual. The sampling point was located upstream of the plant’s flyash collection hoppers and downstream of the economizer. Manual samples were collected from the hopper. The sample collection times were consistent with the times the monitor’s automatic samples were taken. Manual samples were split into three aliquots. The plant’s main chemistry laboratory analyzed one of the splits. The generating unit plant laboratory analyzed one sample, and the monitor analyzed one sample using manual analysis mode. The plant laboratories used the standard LOI method (ASTM C311). In this method, a sample is weighed gravimetrically, introduced into a high temperature furnace wherein any unburned carbon present in the sample is oxidized. The sample is then removed from the furnace, allowed to cool and then reweighed. Oxidation results from a loss of the carbon in the flyash sample due to conversion of the carbon to CO2 that is lost to the atmosphere. Consequently, the analysis reflects the weight percent of carbon in the original sample.

Figure 4 presents the results of the analysis comparisons. On average, the difference between the plant lab analysis and the monitor’s analysis was less than 0.5 percent CIA. Secondly, even though the automatic sample taken by the monitor (was collected by extracting the sample from a single fixed point in a large duct, the automatic analysis tracked almost perfectly with the time-consistent, manually collected samples analyzed by the plant lab (shown in red).

Click here to enlarge image

The monitor’s accuracy as determined by comparison to manual samples clearly demonstrates that the monitor is capable of meeting the accuracy goals regardless of plant operating conditions or coal type being used. The accuracy results also show that the monitor’s electronics, linear slide mechanism and software platforms provide the needed reliability without negatively affecting measurement performance of the basic inertial mass measurement/thermal oxidation design.

Author: David Clifford has 18 years experience in the chemical, process control and instrumentation industries. He holds a bachelor of science degree in chemical engineering from Northeastern University and an MBA from Boston College.


Patashnick, H.; Rupprecht, E. “Continuous PM-10 Measurements Using the Tapered Element Oscillating Microbalance”, J. Air Waste Management Association 1991, 41, 1079-1083.

Dry Bottom Ash Handling Technology

In late January, Basin Electric Power Cooperative said it will use a Magaldi Ash Cooler (MAC) system from Allen-Sherman-Hoff as the bottom-ash handling system in its new 385 MW pulverized coal plant at the Dry Fork Generating Station in Wyoming. The plant will burn Powder River Basin coal, which tends to be a heavy slagging fuel. Because the plant is also in a water-constrained area, a low-water or completely dry system was favored.

View inside the Magaldi Ash Cooler system. Photo courtesy Diamond Power.
Click here to enlarge image

MAC technology removes, cools and conveys bottom ash from negative draft, pulverized coal-fired boilers. The system uses no water and utilizes the negative pressure inside the boiler to induce air flow through the MAC conveyor to cool the bottom ash.

The technology was first applied to bottom ash removal in 1985 at a 35 MW power plant in Italy. Since then, the technology has been installed in nearly 100 units worldwide, including retrofits and new construction.

In basic operation, hot bottom ash exits the boiler throat and passes through a refractory-lined hopper onto a belt. The belt conveys the ash to the head of the conveyor while atmospheric air is induced into the unit by negative pressure inside the boiler. This air results in further combustion and cooling of the ash. The air also cools the conveyor casing, belt and refractory-lined hopper returning some of the radiant heat losses back to the boiler.

The hopper is sealed to the boiler via a mechanical expansion joint to absorb boiler tube movement. This seal consist of several layers of special fabric and ceramic materials which provide for thermal insulation, mechanical strength and flexibility.

The hopper has refractory-lined, pneumatically-operated doors allowing the MAC conveyor to be isolated from the boiler. This allows the hopper to be used for temporary storage (typically, at least eight hours of storage) should maintenance be needed on downstream equipment.

Water losses on a PC fired boiler using a traditional wet type bottom ash system include water absorbed with the bottom ash disposal product, water evaporation into the boiler and water leakage throughout the system. For a once-through system, water usage is higher. Eliminating water from the system eliminates all water-related corrosion problems, energy required to handle the water, housekeeping and safety issues related to leakage and freezing.

The MAC conveyor is designed to use the boiler’s negative draft and ambient air for cooling. This means most of the heat loss is returned to the boiler. Heat loses include the internal energy of the ash as it exits the boiler. Typical bottom ash temperatures at the boiler exit are 1,600 F to 2,000 F. Typically, this heat is lost to the water in a wet system. With a MAC system, it is used to pre-heat air, returning it to the boiler for the combustion process.

Bottom ash leaving the boiler can have as much as 10 percent to 15 percent unburned carbon content. With a dry system the ash is allowed to continue burning, heating the air that will be returned to the boiler.—David Wagman