Commercial PFBC clears development hurdles
Demonstration projects help foster PFBC`s commercial potential
By Timothy B. DeMoss
Pressure is on worldwide to reduce atmospheric pollutants, especially with regard to power generation. Plant owners, however, must maintain the lowest possible electricity cost while reducing emissions. To meet these demands, designers of advanced coal-combustion technologies have been concentrating on the coal-fired combined cycle as one solution. One technology that is now considered “commercial ready,” is pressurized fluidized-bed combustion (PFBC).
PFBC boilers offer higher efficiency (up to 39 percent in first-generation systems) and lower costs [about (US)$300/kW less] than a conventional pulverized-coal boiler with flue-gas desulfurization. These attractive attributes come with the added benefit of reduced emissions.
The PFBC power plant burns crushed coal in a pressurized combustor, providing steam and gas for a combined cycle. Heater tubes in the fluidized bed provide steam to the steam turbine, while the bed`s combustion gases drive a gas turbine. Before expanding through the turbine, combustion gases pass through a series of cyclones to remove about 98 percent of the ash in the gas. The latest designs use hot-gas filters to remove more particulate and other contaminants and to protect the turbine from erosion and corrosion damage. Finally, turbine exhaust gas is passed through an economizer and an electrostatic precipitator before exiting the plant through the stack. The gas turbine`s compressor provides the air to fluidize the coal bed.
Reducing plant emissions at low capital cost while being able to burn a variety of coal-based fuels drives the PFBC initiative. A PFBC combustor mixes a sorbent, usually limestone or dolomite, with the coal to capture sulfur in the fluidized bed. By cleaning as much as 95 percent of the sulfur pollutants in the combustion chamber, PFBC eliminates the need for post-combustion sulfur control. A relatively low combustion temperature–860 C is typical–helps prevent thermal NOx formation. NOx emissions generally measure below 0.2 lb/MBtu.
At the recent 13th International Conference on Fluidized Bed Combustion (FBC), many papers reflected the commercial viability of PFBC. The authors of two papers in particular provided details of similar experiences with PFBC on different parts of the globe. Hideki Goto, of Electric Power Development Co. Ltd. (EDPC) in Japan, presented “Operation Experience from the 71-MW Wakamatsu PFBC Demonstration Plant.” “An Analysis of Four Years of Operation of the 70-MW Tidd PFBC Demonstration Plant” was written by Mike Mudd and W.P. Reinhart of American Electric Power in the United States. Both Japan`s Wakamatsu demonstration plant and the US`s Tidd plant shared similarities in their plants` successes and problems.
In the United States, the Department of Energy`s (DOE) Clean Coal Technology (CCT) program gave rise to the country`s first commercial-scale PFBC plant. The CCT program chose the Tidd plant, located in Brilliant, Ohio, USA, as a repowering demonstration, with the goal to prove PFBC technology capable of commercial, base-load power generation. Construction began on the plant in 1988, and Tidd went on-line in November 1990.
EDPC built the Wakamatsu plant as part of Japan`s government-supported PFBC research and development, with the purpose of evaluating PFBC`s adaptability for utility plants and developing new PFBC technologies. EDPC began commissioning operation in September 1993.
In the 10 months following Tidd`s first coal firing in November 1990, the unit operated for only 818 hours. According to Mudd and Reinhart, numerous problems were responsible for this downtime, including ash blockage in the cyclone ash-removal system, fires at the cyclone gas inlets, blockage in the coal feed system, economizer fouling and boiler vertical separator-level control malfunctions. Goto cited similar problems with the Wakamatsu plant. EDPC experienced cyclone plugging (resulting in damage to the cyclone`s ceramic lining), plugged slurry pipes and nozzles, and cracks in the air distributor.
After a 12-week outage to correct problems at Tidd, the unit operated for 530 hours before plant operators discovered cracks at the root of a number of gas-turbine blades. Although the ruggedized gas turbine was a first-of-a-kind design specifically for PFBC, its failure did not result directly from PFBC operation. Testing prior to the turbine`s operation simply failed to predict the high-frequency fatigue responsible for the cracking.
After repairing the blades, Tidd operators spent the next nine months struggling to maintain high plant capacity, while dealing with numerous minor problems ranging from fuel nozzle and ash system fouling to fluidized-bed, air-distribution sparge duct cracking. Then, in February 1993, the gas turbine threw two low-pressure turbine blades. Although this caused extensive damage, it presented an opportunity to completely reconfigure the secondary ash-removal system and to replace the air-distribution sparge ducts.
While Tidd and Wakamatsu had numerous problems with early operation, solving the problems and continuing with successful operation has lent experience and confidence to PFBC commercialization. Both plants reported meeting or exceeding expected emissions and efficiency values.
Referring to the frustrating outages and failures that plagued Tidd early on, Mudd and Reinhart pointed out that Tidd lacked redundancy normally found in a commercial plant, but more importantly emphasized the philosophy behind the demonstration. “Operation and testing at the Tidd facility,” said the authors, “has not been done in a manner solely to maximize operating time, but rather to determine the unit flexibility and operating limits, and improve unit performance.”
This philosophy was fruitful. After the 21-week outage to repair the gas turbine, Tidd began to operate very successfully. In 1994, Tidd operated for 4,767 hours on coal with 55-percent availability. The plant`s 1994 record brought its first-three-year`s coal-fired total to 6,050 hours. Tidd`s coal-fired demonstration ended in late March 1995 after accumulating 11,445 operating hours. Wakamatsu is still in demonstration operation.
Perhaps the only hurdle still hampering commercial deployment of PFBC is the development of a reliable hot gas cleanup (HGCU) system. HGCU is necessary in PFBC plants because cyclones are not capable of cleaning 100 percent of the ash from the combustion gas before it encounters the gas turbine. The presence of ash in the gas requires ruggedized gas turbines, limiting the performance of the combined cycle. Furthermore, ash that does make it past the cyclones requires electrostatic precipitators (ESP) at the stack to meet particulate emissions standards. Capturing particulates early in the cycle would eliminate the need for secondary cyclones and stack gas ESPs, in addition to permitting a wider selection of gas turbines.
The basic design for HGCU is a ceramic barrier filter system. A filter pressure vessel houses hundreds of ceramic filter elements, called candles, which collect ash on their surface while allowing the gas to pass through. In the Westinghouse design used at the Tidd facility, ceramic filter element arrays were formed by attaching individual candle elements to a common plenum and discharge pipe. Periodic backpulses of compressed gas from a single-pulse nozzle source cleaned the fly ash collected on the surface of the filter elements, and the ash was discharged from the filter vessel`s hopper. When arranged vertically from a support structure, individual plenum assemblies formed a filter cluster.
The filter cluster represents the basic module needed for constructing a large filter system. The individual clusters are supported from a common, high-alloy tubesheet and expansion assembly that spans the pressure vessel and divides it into “clean-gas” and “dirty-gas” sides. The cluster approach also permits efficient maintenance and replacement of individual filter elements.
Engineers on the Tidd project determined that the filter`s basic design was structurally adequate. However, early tests revealed a problem with ash “bridging” between candles. Besides making ash removal and filter cleaning difficult, ash bridges broke several of the candles by producing bending-moment stresses at the candle bases. Testing of the Westinghouse system at Ahlstrom Pyropower in Finland and of a similar HGCU system at Wakamatsu indicated that severe thermal gradients can also break candles.
Engineers at Tidd noted that cohesiveness of the fine fly-ash particles, and the particles` tendency to sinter at temperatures above 760 C, were significant contributors to the bridging problem. Eliminating the primary cyclone increased the dust loading and particle size and solved the problem. This step also served to highly integrate the HGCU system into the PFBC cycle. However, at the Ahlstrom test facility, which was running simultaneously with the Tidd project, bridging was occurring in the Westinghouse system even though the PFBC had no primary cyclone.
According to Westinghouse, Ahlstrom, US DOE and Electric Power Research Institute (EPRI) authors in “Testing of the Westinghouse Hot Gas Filter at Ahlstrom Pyropower Corporation,” presented at the FBC conference, engineers at Ahlstrom believed that two factors contributed to ash bridging. First, severe bridging seemed limited to the top plenum section, which pointed to poor inlet flow distribution. Second, high residual dust cake on all the candle elements suggested poor cleaning. Reduced pulse-cleaning pressure for the test run that had bridging problems may have been the cause.
Following the test run, the project team implemented several facility and filter modifications to address process problems and thermal transient events. In this attempt to eliminate candle element damage, engineers improved main air-compressor surge control and pulse valve operation, increased control range to the feed pump, and improved gas flow distribution entering the filter unit by modifying the inlet baffle and shroud.
Although the changes made at Tidd and Ahlstrom eliminated ash bridging, candle damage remained a problem because of large thermal gradients, and long-term durability of ceramic candles at extreme temperatures remains a major hurdle for HGCU system engineers.
LLB`s HGCU design
Testing of advanced materials continues. However, LLB Lurgi Lentjes Babcock Energietechnik GmbH (LLB) has taken a unique approach to the candle problem at their 15-MWt PCFB test facility near Oberhausen in Germany. In their paper, “Hot Gas Filter Experience Around the LLB 15-MWt PCFB Test Facility,” presented at the FBC conference, George von Wedel, Udo Kalthoff and Horst Mollenhoff described an HGCU design concept aimed at side-stepping inadequate candle materials.
In LLB`s filter concept, instead of suspending candles in a tubesheet, which produces unfavorable tension in the ceramic material, designers arranged the candles upside down with the open end held in a seat arranged on a horizontal header. In addition to the compressive stress given by the weight of the candle itself, a weight on top of the candle is responsible for keeping the candle seated when operators backpulse the system (Figure 2). A special candle configuration ensures that the candle does not experience excessive horizontal forces and bending moments. The inverted candle arrangement was successful in eliminating candle damage once engineers isolated vibration from the fluidized bed`s operation.
LLB also experienced ash bridging in their early testing. However, similar to attempts at Ahlstrom, reconfiguring the gas flow avoided the problem. Although HGCU system testing has provided solutions to many of the problems facing PFBC engineers, more experience is needed with different fuel types before HGCU is ready for commercialization. Solving these problems is essential for PFBC to successfully enter its next advanced phase.
The next generation
With first-generation PFBC ready for commercial deployment, a new technology with its roots in PFBC is poised to emerge as a clean-coal technology leader. Second-generation or advanced PFBC (APFBC) will soon be demonstrated under the US DOE`s CCT program, with private interests conducting research and development as well. APFBC can be described loosely as a combination of first-generation PFBC and integrated gasification combined cycle (IGCC).
In an IGCC system, applying heat and pressure to coal synthesizes a fuel gas, which in turn is used to fuel a gas turbine. Steam for the bottoming cycle is generated using exhaust heat from the gas turbine and from heat rejected by water used to cool the syngas.
The marriage of PFBC and IGCC is made possible by converting a portion of the coal`s Btu energy into syngas. This partial gasification takes place in a carbonizer at a temperature below the coal`s slagging temperature, producing syngas and char. Limestone, fed into the carbonizer along with the coal, plays a vital role as sulfur remover in the APFBC process. The limestone sorbent captures sulfur as CaS, later converted to calcium sulfate in the fluidized bed by feeding excess air to the combustor. Early tests of this process indicate 95-percent sulfur-removal capability. Sulfur removal at this early stage eliminates the need for expensive sulfur-capturing bed cleanup systems currently used with IGCC.
By using staged combustion, the fluidized-bed combustor lowers NOx emissions as well. Estimates for NOx emissions are about 0.3 lb/MBtu; although an LLB APFBC pilot plant performed consistently below 0.003 lb/MBtu NOx, according to engineers from LLB, Foster Wheeler, Westinghouse, and Air Products and Chemicals Inc., in their paper, “Four Rivers Second Generation Pressurized Circulating Fluidized Bed Combustion Project.”
Cyclones and particulate control devices (PCD) remove light char and calcium sulfide from the syngas before it is introduced into the topping combustor for the gas turbine. All char and particulate matter is combusted in the fluidized bed, with calcium sulfate and ash as the only waste products. The hot flue gas from the FBC also passes through the gas cleanup system on its way to the topping combustor, where it acts as the oxidant for burning syngas from the carbonizer.
Like first-generation PFBC and IGCC, APFBC is a combined cycle. The evolution from first- to second-generation PFBC is shown in Figure 3. A fluidized-bed heat exchanger and an heat recovery steam generators provide the steam for the bottoming cycle. Using this design changes the ratio of steam-generated power to gas-generated power from 4-to-1 to 1-to-1 compared to first-generation PFBC.
By taking advantage of this shift from Rankine to Brayton cycle, engineers predict that commercial-scale advanced-PFBC combined cycles will enjoy a considerable efficiency boost to as high as 46 percent or more. Advances in gas-turbine technology could increase the efficiency level to more than 49 percent, according to authors of a paper, “Design and Operating Considerations for an APFBC Plant at Wilsonville,” presented at the FBC conference. Authors of the Four Rivers paper, also presented at the American Society of Mechanical Engineers Conference, point out that such efficiency increases will result in coal consumption that is 25-percent lower per unit than a pulverized-coal or atmospheric FBC plant. A comparison of advanced-PFBC efficiency to other advanced-coal technologies is shown in Figure 4.
Much of the increase in efficiency is due to the increased gas-turbine inlet temperature, made possible by burning the syngas in the topping combustor. Burning syngas does not, however, come without problems. The hot gas cleanup system cleans syngas particulates and alkali. However, the syngas still contains fuel-bound nitrogen, which would convert to NOx if burned in a standard combustor. The combustor must also be able to withstand the 870 C combustor inlet temperature. Most gas-turbine combustors are designed for 370 C.
Research and development programs conducted at various companies are producing valuable experience and data to further the overall development of advanced-PFBC systems. What is lacking so far are demonstration plants similar to the first-generation PFBC demonstrations mentioned earlier.
Commercialization of APFBC may lie in the success of the Four Rivers Energy Modernization Project (FREMP). Air Products and Chemicals Inc., principal contractor for the project, is using the advanced-PFBC technology to repower their chemicals manufacturing facility in Calvert City, Ky., USA.
Foster Wheeler will design, fabricate and erect the APFBC power island with LLB supplying the flue-gas PCD, fuel paste feed and ash-disposal system. Westinghouse will supply the carbonizer fuel-gas PCD.
In their paper, “Advanced Pressurized Circulating Fluidized Bed System: Beyond the Demonstration Stage,” presented at the FBC conference, I.F. Abdulally and I. Alkan of Foster Wheeler Energy Corp. described development of APFBC into a technology with “imminent” commercialization potential. The FREMP plant will provide data on system components and the integrated system at a commercial scale. Conservative estimates place APFBC commercialization around 2005. FREMP should provide needed stepping stones to achieve commercialization by this date.
Despite the success enjoyed with the programs to date, advanced-PFBC must still clear some formidable hurdles before it is ready for full commercialization. EPRI`s John Wheeldon discussed some of these problems in a recent interview and in a paper he co-authored with R.R. McKinsey and A.I McCone of Bechtel and G.S. Booras of EPRI.
Wheeldon said, “If you can get the coal in and the ash out, PFBC is a very simple technology.” Tidd operators can relate to this sentiment, considering their fouling problems. But Wheeldon did say that current advanced-PFBC designs incorporate circulating FBC technology as opposed to bubbling-bed technology, and that may alleviate some of the problems encountered at Tidd. A dry-feed coal-delivery system may also be an aid in this regard.
Besides the coal/char transfer problem, Wheeldon and his co-authors outlined several other development issues in “An Assessment of Advanced Pressurized Fluidized-Bed Combustion Power Plants,” presented at the FBC conference. Issues that require resolution are:
– high-temperature, high-pressure gas filters with commercially acceptable operating lives;
– a topping combustor which achieves the required turbine inlet temperature while restricting NOx emissions to less than 0.1 lb/MBtu;
– ash discharge material that is sufficiently low in sulfides, suitable for utilization applications and can be disposed of readily;
– a gas-turbine inlet gas that is sufficiently low in alkali vapor so as not to cause blade-corrosion problems;
– and for the full advanced-PFBC system, a fuel gas control valve that can operate at a high-temperature, reducing atmosphere.
Wheeldon also said none of the specialized components for APFBC have undergone long-term testing. It is this unknown that could determine the success of the APFBC projects. When asked when utilities should expect full-scale commercialization of APFBC, Wheeldon said accelerating advances in the hot gas filters is essential for commercialization to occur very soon.
“You can`t do anything without them. Commercialization of bubbling-bed PFBC is possible today,” said Wheeldon. Because circulating PFBC technology requires filters to operate at a higher temperature and efficiency, commercialization must wait until HGCU is proven as a long-term technology.
Tidd`s repowering was the first utility demonstration of PFBC in the United States.
Filter internals are lowered into the pressure vessel prior to initial operation.