National Renewable Energy Laboratory
1617 Cole Boulevard
Golden, CO 80401-3393

NREL Subcontract TZ-2-11226-1

Thomas R. Miles, P.E.
Thomas R. Miles, Jr.
Consulting Design Engineer
5475 SW Arrowwood Lane
Portland, OR 97225

Larry L. Baxter 
Combustion Research Facility
Sandia National Laboratories, Livermore, CA

Richard W. Bryers emeritus
Foster Wheeler Development Corporation 
Livingston, NJ

Bryan M. Jenkins 
Biological and Agricultural Engineering
University of California, Davis, CA

Laurance L. Oden 
Albany Research Facility, Bureau of Mines, 
U.S. Department of the Interior, Albany, OR

APRIL 15, 1995

The complete report can be obtained from: Alkali Deposits Found in Biomass Power Plants (4 MB)

Volume 1
Volume 2
If those links fail search for "Miles" and "Alkali" at


Alkali in the ash of annual crop biomass fuels creates serious fouling and slagging in conventional boilers. Even with the use of sorbents and other additives, power plants can only fire limited amounts of these fuels in combination with wood. The National Renewable Energy Laboratory (NREL), U. S. Department of Energy, and the biomass power industry carried out eight full-scale firing tests and several laboratory experiments to study the nature and occurrence of deposits with the goal of increasing the quantities of these biofuels that can be used. This report describes the results of the laboratory and power plant tests that included: tracking and analyzing fuels and deposits by various methods; recording operating conditions; and extensive laboratory testing.

Occurrence of Deposits

Sintered or fused deposits were found on grates and in agglomerates in fluidized beds. Potassium sulfates and chlorides were found condensed on upper furnace walls where it mixed with flyash. Convection tubes were coated with alkali chlorides, carbonates and sulfates mixed with silica, alumina and complex silicates from flyash or fluidized bed media.

Deposits were evaluated using elemental analysis, X-ray diffraction and other mineralogical techniques. These analyses have advanced the understanding of the role of minerals in the combustion of biomass, and their occurrence in biofuels. Deposits occur as a result of the boiler design, fuel properties and boiler operation. The limited furnace volume and high furnace exit gas temperatures of most biomass boilers promote slagging or deposits from those biofuels that contain significant amounts of potassium or sodium, sulfur, chlorine and silica. All annual growth, whether from urban tree trimmings, annual crops or their residues, or from energy crops, contains sufficient volatile alkali (0.34 kg/GJ or 0.8 lb./MMBtu) to sufficiently lower the fusion temperature of the ash so that it melts in combustion, or the elements vaporize and condense on boiler tubes and refractories.

Deposits observed in this project are consistent with all known mechanisms for deposit formation: particle impaction, condensation, thermophoresis and chemical reaction. Particle impaction was the dominant mechanism, especially on cross flow convection tubes, followed by condensation on waterwalls and chemical reaction. Analysis by scanning electron microscopy (SEM) showed that compounds containing potassium, sulfur and chlorine were the principal bonding agents in most deposits and were usually associated with fuel blends containing annual growth materials such as agricultural crops and residues. Most deposits occur during post-combustion and cannot be predicted solely by analysis of the fuel composition.

Fuel and Deposit Analyses

More than 30 methods of analysis were identified for fuel characterization. Fuel sampling at biomass plants typically does not include enough information to evaluate potential sources of deposits. Routine fuel characterization should include: proximate analysis, ultimate analysis, heating value, chlorine, a direct measure of oxygen and elemental ash analysis. Samples should be ashed at 600 C to minimize the loss of volatile alkali. Microwave digestion in acids followed by atomic adsorption best accounted for fuel elemental composition. A successive leaching method called chemical fractionation was used to determine the reactivity of inorganic constituents in biomass fuels as measured by their solubility in water and acids (1 M ammonium acetate, and 1 M HCl).

Several techniques were used to analyze deposits. Scanning electron microscopy (SEM) helped to determine the composition and mineralogy of deposits. DTA (differential thermal analysis) and TGA (thermogravimetric analysis) were used to evaluate volatilization and fusion of ash constituents. Ash fusion temperatures using the pyrometric cone tests were of little value in predicting deposits, the alkali having been lost during ashing and calcining.

Pilot scale combustor simulations demonstrated deposit formation on combustor refractory and on heat transfer surfaces, confirming mechanisms observed in the field. Alkali volatilization, condensation and enrichment on flyash and in deposits were observed in the Sandia Multi-Fuel Combustor (MFC).

Boiler Design and Operation

Conventionally designed boilers are not suitable for burning high alkali fuels. Special boiler designs with low furnace exit gas temperatures, <1500 F, are required for annual crops or residues, including grasses and straws. Designs should include: adequate waterwall surface area or parallel heat exchange surfaces, and combustion air control to control gas temperatures, grates suitable for removing large quantities of ash, and sootblowing to clean tenacious deposits.

Limestone was the principal additive used in test boilers to maintain bed fluidization. While limestone improved operation the calcium appears as a constituent of deposits on convection surfaces (as CaCO3, CaSO4) and may reduce but does not prevent deposition. High alumina sand also reduced agglomeration in a circulating fluidized bed (CFB) but did not change the composition of deposits on the superheater tubes.

Gasification at low temperatures, <1400 F, with additives, may be necessary to inhibit alkali volatilization in order to burn large quantities of these biofuels. A review of straw experiences in Europe shows some success with straw pyrolysis. Additional research is needed.

Fouling and Slagging Indicators

Several methods were tested for anticipating slagging or fouling based on fuel composition. These included: concentration of alkali, water soluble alkali, tendency to form sulfates or chlorides, and direct observation on heating. Fuel elemental composition and the concentration of alkali, sulfur, chlorine and silica in the fuels appear to be the best indicators of the tendency of fuels to slag. Project data should be used to develop practical algorithms for industry. Observations of fuel samples heated to sticky temperatures during ashing showed differences between wood and agricultural residues but did not clearly identify problem fuels.

Project Participants and Reporting

Seven power plant sponsors joined the project representing nine biomass power plants: Delano Energy Company, Inc., Woodland Biomass Power Ltd., and Mendota Biomass Power, Inc., (representing Thermo Electron Energy Systems); Hydra-Co Operations; Sithe Energies, Inc.; Wheelabrator Environmental Systems, Inc.; and Elkraft Power Company, Ltd. (Denmark). Electric Power Research Institute, Foster Wheeler Development Corporation and the National Bioenergy Industries Association (formerly National Wood Energy Association) also contributed to the investigation.

The Bureau of Mines Research Laboratory in Albany, Oregon provided X-ray diffraction (XRD), scanning electron microscopy (SEM) and mineralogical analysis courtesy of Larry Oden. Sandia National Laboratories contributed pilot scale combustion tests and analyzed fuels and deposits as part of a parallel NREL project conducted by Larry Baxter. Bryan Jenkins, professor of Biological and Agricultural Engineering, participated through the courtesy of the University of California, Davis.

Results of the project were reported in presentations to nine conferences and will be published in forthcoming journals. Four project meetings were held with participants at Sandia National Laboratories. The suite of fuels and fuel samples developed during the tests were archived at Hazen Laboratories and were used in independent investigations by NREL and by Richard Bryers at Foster Wheeler Development Corporation. The conclusions from the project have changed the way participants evaluate deposits, and the way they buy and prepare fuels. Project data helped in the startup of a new biomass power plant in Florida. Through more than twenty five inquiries from biomass plants around the country it has been possible to assist other interested parties including: an industrial boiler manufacturer in the northeast, a power plant boiler manufacturer, other power plants, including a 20 MW power plant in New York and a plant cofiring wood and refuse derived fuel, many of the state and regional biomass energy agencies, and developers of short rotation woody crops and non-wood crops for fuel.

Future Work

Tube erosion and corrosion associated with deposits are of prime concern to industry participants. Development of a computer algorithm for predicting deposit formation combining information about fuels, deposits, boiler design and boiler operation would be useful. Further fuel characterization is needed. Deposit properties such as reflectivity, emissivity, porosity and tenacity also need to be characterized. Several other pilot and operational projects are suggested.


In summary and in answer to questions from the biomass fueled plant operators concerning how to proceed regarding alkali and deposits, the following steps are recommended:



A. Project Participants A.1 Investigators and Advisors A.2 Industry Sponsors and Participants B. Methods of Sampling and Analysis B.1 Analytical Request Form B.2 Standards for Biomass and Coal B.3 Summary of Analytical Methods Used B.4 Microwave Dissolution and Atomic Absorption Method B.5 Comparison of Microwave and Thermal Ashing B.6 Determination of Water-Soluble Alkali B.7 Chemical Fractionation Procedure B.8 Fuel Sampling Procedure C. Fuel Characteristics: Data Summary C.1 Wood Fuel Blends: Alkali Deposit Investigation C.2 Wood Fuels: Oak, Fir, Pine, Poplar, Forest Residues, Christmas Trees C.3 Urban Waste Fuels and Residues: Demolition, Land Clearing, Waste Paper, RDF C.4 Wood Fuels - Energy Crops: Willow C.5 Grasses and Straws: Bagasse, Bana Grass, Switchgrass C.6 Grasses and Straws: Energy Crops, Miscanthus and Reeds C.7 Grasses and Straws: Residues, Alfalfa, Mint, Wheat Straw, Rice Straw C.8 Nuts, Pits and Shells: Almond, Pistachio, Walnut, Olive, Prune Pits D. Deposits D1. Melting Temperatures of Potential Low-Melting Minerals Found in Biomass (Bryers, 1994) D2. Grate-1. Wood and 95% Wood/5% Imperial Wheat Straw Blend. D3. Grate-1. 80% Wood/20% Oregon Wheat Straw D4. Grate-1 type boiler. Wood and Landscape Residues. D5. Low Temperature Straw Boilers: Grate-2 (Bale), Grate-3 (Stoker) D6. FBC-1 Wood and Agricultural Prunings D7. CFB-1 and CFB-2 Wood and Agricultural Residues. D8. CFB-3 Wood and Pits