Analysis of Heat Transfer Fouling by Dry-Grind Maize Thin Stillage Using an Annular Fouling Apparatus

Cereal Chem. 83(2):121–126 In dry-grind processing to produce ethanol from corn, unfermented solids are removed from ethanol by distillation and dried to produce distillers dried grains with solubles (DDGS), an animal food. Fouling of thin stillage evaporators has been identified as an important energy consumption issue in dry-grind facilities. Using an annular fouling apparatus, four batches of thin stillage were analyzed to determine repeatability of fouling rate and induction period measurements. Dry solids, protein and ash concentrations, and pH were correlated to fouling rate and induction period to determine how variation in thin stillage from the same dry-grind facility affects these fouling parameters. Effects of increasing Reynolds number (Re) in the laminar region on fouling rate, induction period, and fouling deposit protein and ash concentrations were also determined. Repeatability of fouling rate measurements was similar to other studies (CV < 7.0%) but repeatability of induction period measurements was high relative to other studies (CV < 88.7%). Fouling rate increased with increasing dry solids concentration. Thin stillage at Re = 440 had shorter induction periods and greater fouling rates than at Re = 880. Fouling deposits collected from Re = 440 tests had similar protein concentrations and lower ash concentrations compared with deposits from Re = 880 tests. The U.S. Clean Air Act mandates that certain areas of the country with air pollution problems use reformulated gasoline containing 2% oxygen (Lyons 2003). Two additives are used to increase oxygen levels in gasoline: methyl-tertiary butyl ether (MTBE), a petroleum derivative, and ethanol. In recent years, MTBE has been phased out due to its possible carcinogenic properties and its discovery in ground water. As a result, U.S. ethanol demand has increased from ≈670 million liters (200 million gallons) in 1980 to 13.6 billion liters (3.6 billion gallons) in 2004, most of which was produced from maize (Shapouri and Gallagher 2005). Fuel ethanol is produced from maize using two methods: wet-milling and dry-grind (DG) maize processing. In 2004, DG facilities accounted for 70% of current U.S. ethanol production. The DG process consists of grinding whole maize, using heat and enzymes to convert starch to glucose, fermenting glucose to ethanol with Saccharomyces cerevisiae yeast, separating ethanol from unfermented material, and drying the unfermented material to produce distillers dried grains with solubles (DDGS). Evaporation of water from the soluble nonfermentable fraction, known as thin stillage, consumes much of the energy in the DG process (Meredith 2003). Thin stillage is composed of proteins, ash, lipids, and other kernel constituents that were not fermented. Thin stillage (4–6% dry solids, w/w) is concentrated (25–30% dry solids, w/w) in multi-effect evaporators (Singh et al 1999) before being mixed with wet grains and dried to produce DDGS. Thin stillage evaporators foul rapidly, requiring plant shutdowns and cleaning every few weeks. Evaporator fouling increases heat transfer resistance, energy use, cleaning costs, downtime to restore evaporators to optimal performance, and capital costs to install extra evaporation capacity to compensate for cleaning. One strategy to improve long-term stability of the DG industry is to increase efficiency of unit operations in the process, particularly energy use. Understanding the fouling of heat transfer surfaces and the tendencies of process materials to foul these surfaces would provide information to increase energy efficiency (i.e., less energy to produce a given volume of ethanol) of the DG industry. Heat transfer fouling, the deposition of material on a heated surface, is found in heat transfer operations and is prevalent in processing of foods and biological materials (Changani et al 1997). Molecules found in foods, such as proteins, carbohydrates, and lipids, are heat sensitive and attach to heated surfaces in heat transfer equipment. A limited number of fouling studies have been published for maize processing (Singh et al 1999; Agbisit et al 2003). Instruments with annular flow configurations have been used to study fouling in other applications, including petrochemicals (Asomaning and Watkinson 1992; Panchal and Watkinson 1993; Wilson and Watkinson 1996). In maize processing, Singh et al (1999) observed that thin stillage from maize wet-milling fouled an annular fouling apparatus at a slower rate than thin stillage from a DG facility. They attributed this to higher oil content in DG thin stillage than in thin stillage from maize wet-milling. Additionally, Agbisit et al (2003) subjected steepwater from maize wet-milling to microfiltration. Steepwater permeates from microfiltration fouled the heated surface of the apparatus at a rate 80% slower than unfiltered steepwater. Many variables affecting fouling behavior of maize processing streams are poorly understood. Relationships between thin stillage composition, fluid temperature, pH, flow rate, and the fouling of heated surfaces have not been quantified. For example, flow rate directly influences the shear stress at the evaporator heating surface and tends to disrupt formation of fouling deposits. Shear stress on evaporator surfaces has been increased by increasing flow velocities and creating flow geometries that promote turbulence, such as spiral heat exchangers and the rippled surfaces of plate heat exchangers (Young and Sloan 2003). Belmar-Beiny et al (1993) found the amount of whey protein fouling deposited in a tubular fouling apparatus decreased with increasing Reynolds number (Re), a dimensionless parameter indicating flow turbulence. Effect of Re on the amount of fouling deposit formed was asymptotic; a larger deposition decrease was observed when Re increased from 1,800 to 4,000 than when increased from 4,000 to 9,000. Karabelas et al (1997) used a CaCO3 solution to observe resistance to heat transfer in plate heat exchangers. They observed an asymptotic effect of increased flow velocity on heat transfer resistance: as flow velocity increased the rate of change in the fouling rate, the rate of change of the overall heat transfer coefficient decreased. Effect of Re on fouling measurements observed with thin stillage has not been reported. The repeatability of the fouling measurement technique was unknown for maize processing 1 Department of Biosystems and Agricultural Engineering, Oklahoma State University, Stillwater, OK 74078. 2 Department of Animal Sciences, University of Missouri, Columbia, MO 65211. 3 Department of Agricultural and Biological Engineering, University of Illinois, Urbana, IL 61801. 4 Department of Veterinary Pathobiology, University of Illinois, Urbana, IL 61801. 5 Department of Veterinary Biosciences, University of Illinois, Urbana, IL 61801. 6 Corresponding author. Phone: 217-265-0697. Fax: 217-244-0323. E-mail: