523% llHHlHlllNlIHIIWHHIHHHIWHIINIHIIHHIHWWI Lnivzrzity This is to certify that the thesis entitled SEPARATION OF CORN ETHANOL STILLAGE INTO ITS SOLID AND LIQUID PORTIONS presented by Barbara Elizabeth Goodrich has been accepted towards fulfillment of the requirements for M.S. degree in Chemical Engineering g 1‘. Major professor Date W3.— 0-7639 M50 is an Affirmative Action/Equal Opportunity Institution Illllllll lllllllllllllllfllllll \/ 3 1293 01079 7730 I MSU RETURNING MATERIALS: Place in book drop to remove this checkout from LlBRARlES w your record. FINES will be charged if book is returned after the date stamped beIow. APR IS 1995 i iii! @ 0 LI: SEPARATION OF CORN ETHANOL STILLAGE INTO ITS SOLID AND LIQUID PORTIONS By Barbara Elizabeth Goodrich A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of MASTER OF SCIENCE Department of Chemical Engineering 1983 ABSTRACT SEPARATION OF CORN ETHANOL STILLAGE INTO ITS SOLID AND LIQUID PORTIONS By Barbara Elizabeth Goodrich Farm scale ethanol production can only be economically feas- ible if the by-product, corn ethanol stillage, can be utilized. Some fractions of stillage can be used as livestock feed, while others have use as fertilizer. The limiting criteria is the moisture con- tent, which must be kept low if the material is to be used as feed. The protein content of the feed should be maximized if possible. The unsuitable material can be used as fertilizer because of its mineral content. Four simple separation devices were investigated for separa- tion devices were investigated for separation performance. The separated material was analyzed for protein and moisture content. It was then categorized as to whether it should be used as livestock feed or fertilizer. Finally, these products were analyzed in terms of their market value. Stillage made from coarser ground corn and separated on the gyratory device appears to be the best choice. ACKNOWLEDGMENTS I wish to thank my advisor, Dr. Bruce w. wilkinson, for all of his support during this project and also a very special thanks to Mr. Gary M. Webber of Animal Science for all the help and knowledge he so willingly provided. ii LIST OF LIST OF Chapter I. II. III. IV. VI. TABLE OF CONTENTS TABLES FIGURES INTRODUCTION NUTRITIONAL VALUE OF NET DISTILLERS GRAIN . STILLAGE PREPARATION . SEPARATION TESTS SNECO and CRIPPEN Separators . . . SNECO Circular Vibro-Energy Screen . CRIPPEN Inclined Screen . EIMCO Rotary Vacuum Filter Vacuum Filter RESULTS AND DISCUSSION M.S.U. Stillage . SWECO Separator . CRIPPEN Separator Commercial Stillage CRIPPEN Separator SNECO Separator . Comparison of SNECO Separated M. S. U. Stillage, CRIPPEN Separated M. S. U. Stillage, and CRIPPEN Separated Commercial Stillage . . Buchner Funnel with Nhatman Paper . EIMCO Filter . . Analysis of Alternative Uses of the "Net" Fraction . . . . . . CONCLUSIONS AND RECOMMENDATIONS APPENDICES . BIBLIOGRAPHY Page iv 12 12 14 16 18 20 20 20 26 27 29 29 36 38 38 43 45 63 Table 10. 11. A-l. A-2. A-4. A-S. LIST OF TABLES Effect of varying feed location for particle size X-Z 999p . . . . . . . . . . . Effect of varying feed location for particle size 995u>x3438p. Effect of varying feed location for particle size x < 438 . . . The moisture and protein content of SNECO separated M.S.U. stillage . . . . . . . The fraction yields of SNECO separated M. S. U. ostillage (per 100 lbs of feed) The moisture and protein content of CRIPPEN separated M.S.U. stillage . . . . . . . age The moisture and protein content of CRIPPEN separated commercial stillage . . . . . . . . The fraction yields of CRIPPEN separated commercial stillage . . . . . . . . . . Filtration results using Nhatman paper to further sep- arate M.S.U. stillage fine particle fractions . A market value breakdown of products per bushel of corn . . . . . . . . . . . . SNECO separator--M.S.U. Stillage . CRIPPEN separator--M.S.U. stillage CRIPPEN separator--commercial stillage . Fraction yields Moistures and proteins iv The fraction yields of CRIPPEN separated M. S. U. still- Page 23 23 23 24 24 26 27 27 28 37 41 47 48 49 50 52 Figure 001-pr 10. 11. 12. LIST OF FIGURES Distillation Column . Process Flowchart SNECO Circular Vibro-Energy Screen . CRIPPEN inclined screen EIMCO Rotary Vacuum Filter Vacuum Filter . _ Collected Samples Feed Point Locations Comparison of Protein Content in Separated Corn Ethanol Stillage . . . . . . . Comparison of Percent Protein by Particle Size in Separated Corn Ethanol Stillage . Comparison of Moisture Content by Particle Size in Separated Corn Ethanol Stillage . . Distribution of Total Protein and Total Moisture in Feed by Particle Size . . . . . . . Page 10 11 13 15 17 19 21 22 3O 31 33 35 CHAPTER I INTRODUCTION Over the past decade, beginning with the 1973 Arab Oil Embargo, the energy market has experienced gross fluctuations which have produced cost escalations and temporary shortages. The fear of future shortages has stimulated the search for alternative sources of liquid fuels. These instabilities have been of particular concern to the farmer whose livelihood depends on the availability and afford- ability of liquid fuels to operate his farm machinery. The possible on-site production of ethanol from corn poses a potentially attrac- tive source of liquid fuel. Since the typical farmer produces more corn than is required to feed his livestock, this surplus corn could either be sold, assuming that there is a buyer, or converted to fuel via fermentation. If the conversion process could be accomplished on the farm, it would eliminate certain transportation and handling costs plus give an additional sense of security to the farmer. The fermentation of corn results in a “beer" containing 8-10% ethanol. Recovery of ethanol through distillation producesa stillage (distillation bottoms) stream composed mainly of water, but which also contains unfermented grain parts in both solid and liquid form. This material is disposed of or utilized in the farm operation. The protein content of the stillage is adequate to make it attractive as a feed supplement for cattle. However, its high moisture content makes it unusable as a direct feed supplement. Cattle will only tolerate so much moisture in their diet or their dry matter intake will go down. This results in a reduced yield which translates into less meat on their carcasses at market time. Excessive moisture also causes manure handling problems. Some fractions of separated stillage are unsuitable for use as livestock feed because of a high moisture content. These fractions may have use as a fertilizer because of their nutrient content. The important nutrients contained in this material are nitrogen, potassium, and phosphorus. Commercial ethanol plants further process the stillage to recover the protein content in a dry concentrated form. The solid con- tent of stillage is separated out by screening and passed through a hot gas drier. The drying process requires a significant capital and energy investment. The liquid portion (containing some soluble pro- teins) is evaporated to remove the water and the resultant concen- trated soluble protein is added to the solids during the drying proc- ess. This operation produces a dry concentrated protein-rich "Distillers Dried Grains with Solubles (DDGS)" product which is suit- able for long-term storage and transportation to markets. The equip- ment requirements and the energy consumption incurred make this process attractive only in large commercial scale plants. Prospective owners of farm scale ethanol plants have limited funds available for investments and expenditures. The distillation column in itself is a major investment.1 Finding funds for an exten- sive stillage drying process is another matter. The potential exists, however, for direct feeding of separated distillers grain to cattle that reside on the farm. In this case, it would be unnecessary to dry the grain if storage can be minimized. The separation process should be as simple as possible to enable the farmer to minimize his investment in terms of capital and time. Furthermore, the protein content of the separated grain (which is known as "Net Distillers Grain (NDG)") should be maximized. This is complicated by the fact that the insoluble protein associated with small particle sizes and the soluble protein will probably not be recovered due to the expense associated with their separation from the water phase. A simple screening process which separates the wet grains from the water phase has the potential of arranging NDG into a form suit- able for feeding cattle. Furthermore, if the protein content can be established as a function of particle size, selective preparation of a particular particle range might be used to maximize the protein content. The purpose of this study was to demonstrate and evaluate several separation processes which do not require the drying of grains. The variance of the protein and moisture content in differ- ent fractions of a corn-fermented stillage stream was also determined. This study was part of a joint undertaking by the Departments of 1Joseph w. Geiger, “Design, Energy, and Economic Analysis of Small Scale Ethanol Fermentation Facilities,” (M.S. Thesis, Michigan State University, 1981). ' Chemical Engineering and Animal Science. Chemical Engineering was responsible for the separation of the stillage into NDG. Animal Science had the responsibility of investigating different methods of storing NDG on a long-term basis. The funding for this project came from the U.S. Department of Energy Grant DE—FGO7-BIID-12334. The feed stream used for this work was prepared at the M.S.U. Beef Cattle Research Center Ethanol Pilot Plant. This facility was built during the previous joint effort of the Departments of Chemical Engineering and Animal Science with a grant from the Michigan Depart- ment of Agriculture.2 Separation tests were conducted using three pieces of equipment: 1. CRIPPEN Mfg. Co. Inclinced Screen, Model KV 1236 2. SNECO, Inc. Circular Vibro-Energy Screen, Model L518C3333 3. EIMCO Rotary Vacuum Filter In addition, further tests using a buchner funnel with Whatman filter paper were run on the liquid residue of separations 1 and 2 above. The funnel employed a vacuum pulled by a water aspirator. A series of tests using the CRIPPEN unit was also con- ducted on commercial stillage supplied by a local ethanol plant. 21mm, p. 3. CHAPTER II NUTRITIONAL VALUE OF "NET” DISTILLERS GRAIN Fermentation and distillation processes remove only the starch or sugar in the feedstock, thus concentrating the remaining nutrients.3 When corn is used for feedstock, the concentration of nutrients is increased approximately 300% (dry weight basis) when compared with the original corn feedstock.4 A number of studies have been conducted concerning the suit- ability of NDG as a livestock feed source and supplement. In his article, "Net Distillers' Grains, An Excellent Substitute for Corn in Cattle Finishing Rations," Stanley D. Farlin5cites testing done at the University of Nebraska during the summer of 1980. Three dif- ferent percentages of N06 were substituted for corn in the feeding of three groups of cattle. A fourth group served as the control group. The control rations consisted of 85% corn, 5% dry supplement, and 10% hay. In addition, urea supplement was employed so that the total ration protein was 11%. The percentages substituted for corn were 3National Research Council, Feeding Value of Ethanol Produc- tion By-Products (Washington, D.C.: National Academy Press, 1981), p. 15. 4 Ibid. 5Stanley 0. Farlin, "Wet Distillers' Grains: An Excellent Substitute for Corn in Cattle Finishing Rations," Animal Nutrition and Health (April 1981): 35. 25%, 50%, and 75%. The NDG contained 75% moisture and from 27—29% protein on a dry basis. It was found that the carcasses of cattle fed at 50% were 23 pounds heavier than the groups fed at 25% and 75%. The carcasses of cattle fed at those latter percentages were similar in weight to that of the control group. Carcass character- istics such as quality grade, ribeye area, fat thickness, and dressing percentage were not affected by the use of NDG.6 It has been documented that the nutrient composition of NDG is comparable to DDGS when they are compared on a dry matter basis.7 Animals have been found to perform equal to or better on wet by- products than on the same by-product in a dried form.8 There is evidence that DDG, when used as a feed supplement, increases milk yield and percent milk fat in lactating cows.9 The amount of extra moisture that can be incorporated in a cattle's diet is limited by the amount which will not cause a reduc- tion in weight gain rate.10 If the animal becomes waterlogged, their dry matter intake will go down. 6 7J. C. Waller, et al., “Use of Fuel Ethanol By-Products in Livestock and Poultry Diets" (paper prepared for Michigan State University), p. 3. ‘ 81bid. Ibid. 9Distillers Feed Research Council, Distillers Feeds (Cin- cinnati, Ohio), pp. 48—53. 10National Research Council, Feeding Value of Ethanol, p. 40. High moisture by-products such as NDG are subject to micro- bial contamination.11 Therefore, it is important that they are stored properly. Lake (1976)12 said that the feeding of high mois- ture feeds reduced the bunker life of the feedlot diets. Because of the spoilage that might occur in the bunker, it is necessary to feed the cattle again within 12 hours before the feeds spoil.13 If NDG becomes spoiled, it loses its palatability. Cattle are likely to eat less of it and possibly refuse it altogether. This could result in a reduction of dry matter intake. 11d. C. Waller, et al., "Separation and Storage of High Moisture Distillers Feeds" (paper prepared for DOE, Michigan State University), p. 20. 12National Research Council, Feeding Value of Ethanol, p. 40. 13111111. CHAPTER III STILLAGE PREPARATION The M.S.U. Ethanol Pilot Plant operates on a 500 gallon batch fermentaion scale. Dry corn was ground to pass a U.S. No. 8 (0.093 inch) screen and mixed with water to form a fluid mash. Takatherm alpha-amylose enzyme was added to the mash following a pH adjustment to 6.5. (A 50% sodium hydroxide solution was used if pH needed to be raised; a 50% sulfuric acid solution was used if the pH needed to be lowered.) The mixture was boiled for at least an hour to ensurethe hydration of starch to the hexose sugars, maltose and glucose. The batch was cooled to 194°F and a final pH adjustment to 6.5 was made. Additional enzyme was added and the mixture was vigorously agitated to promote enzymatic activity. The solution was held at 194°F until an iodine test showed no starch intact in either the solid or liquid phases. Cool water was then added to lower the temperature to 135-140°F. The pH was adjusted to 4.2 and diazyme glucoamylose enzyme was added to produce at least 10% glucose. (If the corn starch is not completely hydrolyzed to glucose and maltose, the yeast will only be able to produce ethanol equal to the amounts of sugars present.) M.S.U.-produced stillage was overtreated with enzyme to guarantee complete hydrolysis. The resultant material was cooled to 90°F by the addition of cold water. Brewers yeast was then added. Fermentation proceeded for two to two-and-a-half days until carbon dioxide was no longer evolved or until glucose could no longer be detected in the mash. The resultant beer was stored in an agitated tank until dis- tillation. At that time it was pumped to the tap of a 12-inch glass sieve tray stripping column (see Figure 1). Reboiler heat was sup- plied by a steam coil (as contrasted to steam sparaging which is often done.) An ethanol stream was drawn off the condenser at about 120 proof. The stillage from the reboiler was essentially ethanol free and 90-93% water. See Figure 2 for a flowchart of the process. 10 Figure 1.--Distillation Column. 11 GRIND WATER it I ‘ L J. ALPHA- pH ADJUST— 1 LAMYLASE MENT COOKER WATER IL COOLING HADJUST— p MENT DIAZYME GLUCO- AMYLOSE MASH _ L . COOLING DRY 1 | '3$§VX§§SI 1 e , CARBON FERMlENTOR DIOXIDE ‘ STILL IL ETHANOL. STORAGE I i S'E‘PARAT— ING DEVICE Figure 2.--Process Flowchart. CHAPTER IV SEPARATION TESTS SNECO and CRIPPEN Separators For most tests, the stillage was pumped directly from the reboiler to a test separation unit. The pumping and separation rates were determined by the still operation. Screens were set up on these separators with two mesh sizes mounted such that three fractions of particle sizes were collected. These fractions were designated "Dry," "Intermediate," and "Net" in reference to their relative moisture content. The "Dry" fraction was material that stayed on the top screen; the "Net“ was that which passed through both screens. Each fraction was collected over a timed interval so that a flow rate could be assigned to it. The collected fractions were analyzed for protein via a modified Kjeldahl method (see Appendix C) for moisture content by oven drying at 60°C for 24 hours. SWECO Circular Vibro-Eneggy Screen (See Figure 3) This device consists of circular screens (device used in experimental work had two screens and three discharges) with an 18- inch diameter working surface and is powered by a 0.5 HP electric motor. It vibrates about its own center of mass. This vibration is 12 CUTAWAY. . .SWECO VIBRO—ENERGY SEPARATOR Figure 3.--SHECO Circular Vibro—Energy Screen. but mm M W muted hum! 14 caused by eccentric weights on the upper and lower ends of the motor shaft.14 The upper weight acts to create vibration in the horizontal plane which moves the material across the screen to the periphery.15 The lower weight tilts the machine creating vibration in the vertical and tangential planes.16 The SNECO separator is commercially used to separate stillage. Larger units are used to separate high moisture waste streams in vegetable processing plants on a commercial basis. The unit was supplied by a regional commercial supplier. Samples were collected on a timed basis by placing a container under the appropriate discharge. CRIPPEN Inclined Screen (See Figureg4) This unit consists of a rectangular screen with a 32" x 101" working area powered by a 0.5 HP electric motor. It was designed to separate dry particles rather than wet sludge such as stillage; thus its mechanical motion was not optimized for handling wet mate- rial. Many CRIPPEN units today are used in a certified seed busi- neSS'UJseparate weed or off-type seeds from certified seedstock.17 The screen angle of the device was varied to see if an optimum existed. If the screen angle was too near horizontal, material on 14SNECO, Inc., SNECO Vibro-Energy Separator (Los Angles, California), p. 8. 151m. 151m. 17Naller, "Use of Fuel Ethanol By-Products in Livestock and Poultry Diets," p. 6. 15 3.7. I" . .\ or. an ar- vuoov-pr‘r, . Figure 4.-CRIPPEN Inclined Screen. 16 the screen would clump together. 0n the other hand, if the angle was too steep, the residence time would be too short to permit an ade- quate separation. The tests that were conducted on the commercial product using this device were carried out on "cold" product which had been transported in barrels to M.S.U. from the commercial ethanol plant. The material was fed manually semi-continuously (small frequent batches) to the separator. Samples were collected on a timed basis by placing aluminum containers underneath the device to catch the various fractions. Losses occurred frequently due to the difficulty of catching all the material. EIMCO Rotary Vacuum Filter_(See Figure 5) This device consists of a cylindrical drum supported in an open-top vat which allows the drum to rotate about its own axis in the 18 The lower portion of the drum is confined within 19 horizontal plane. the tank walls and the upper portion is exposed. The drum shell contains a number of shallow compartments which are covered with a drainage grid and a filter cloth.20 A vacuum is applied to those compartments of the drum which pass through the material to be fil- tered. This, in turn, creates a vacuum within the compartments, 18R. H. Perry, and C. H. Chilton, Chemical Engjneers' Hand- book, 5th ed. (New York: McGraw Hill, 1973), pp. 19-76. 191m. ZOIbid. 17 cab_re E==Ca> scaboz oosz--.m manure 18 causing a flow through the filter medium, conduits, and automatic valve.21 A layer of cake solids is deposited upon the filter, covering the submerged particles of the drum.22 Stillage from the M.S.U. Ethanol Pilot Plant was transported in barrels to the Chemical Engineering Laboratory for separation using the EIMCO device. Consequently, the tests were conducted on cold stillage. Vacuum Filter (See Figuregg) The liquid and solid portions in the fraction previously designated as "Net" were further separated by vacuum filtration. A buchner funnel with an 11-centimeter diameter was used with various Whatman filter paper for the filtration. The funnel employed a vacuum pulled by a water aspirator. The filter paper ranged from "slow" to "fast" in filter speed depending on porosity. Protein tests via a modified Kjeldahl method were run on the filtrate to see how much protein remained in the liquid phase. 21Ibid.. pp. 19-77. 221bid.. pp. 19-78. 19 33:3 cmczuamv amp—I 5:33-16 83m: CHAPTER V RESULTS AND DISCUSSION M.S.U. Stillage Tests were conducted on the SNECO separator, the CRIPPEN separator, a buchner funnel, and the EIMCO filter. Samples were collectedin'plastic containers (see Figure 7). Both the SNECO and CRIPPEN devices gave reasonable separations, whereas the EIMCO device would not achieve a separation. For this latter system, the presence of a significant quantity of fines plugged the filter cloth and halted the separation of the product. Diatomaceous earth filter aid was added to the stillage at 10% of the stillage solids content to enhance filtration. Greater amounts were felt to make the material unusable as animal feed. Even with filter aid, the filtration process could not be carried out. The buchner funnel used in conjunction with Whatman paper allowed for further separation of the "Net" fraction. SNECO Separator The location of the feed point (see Figure 8 for location of feed points) was varied to see if an optimum existed. The results are listed in Tables 1, 2, and 3. The feed point location seemed to impact the moisture content and percent of the total flow in the particle size range x 3_995u 20 21 .mm—asmm umuumFFoo--.~ weaned '3”. .0533 W CUTAWAY. . . SWECO WORD-ENERGY SEPARATOR Figure 8.--Feed Point Locations. 23 TABLE 1.--Effect of varying feed location for particle size x.: 999p Position %,of Total Flow % Moisture % Protein (dry basis) 1 14.5 82.8 27.9 2 16.7 82.9 27.0 3 20.9 87.0 27.4 4 16.5 82.0 28.2 TABLE 2.--Effect of varying feed location for particle size 99511 > x 3 43811 Position % of Total Flow % Moisture % Protein (dry basis) 1 9.2 87.8 50.6 2 10.6 88.8 52.1 3 7.4 86.8 49.4 4 7.7 87.4 45.3 TABLE 3.--Effect of varying feed location for particle size x < 438 Position % of Total FLow % Moisture % Protein (dry basis) 1 76.3 96.4 33.5 2 72.8 96.5 38.2 3 71.7 96.4 35.5 4 75.9 96.4 31.6 24 more than it affected the particle size ranges 995p > x 3.438p and x < 438p. In the particle size range x Z 9950, the moisture content is highest when the feed point position is in position 3. This is probably due to a short residence time on the top screen resulting in a poor separation. The manufacturer recommends that the feed point be located in position 1 which is in the middle of the screen. Overall, the variation caused by feed point location was slight and the results were, in general, quite consistent. The moisture and protein content of the separated stillage varied with particle size. Fraction yields were also dependent on particle size. See Tables 4 and 5. TABLE 4.--The moisture and protein content of SNECO separated M.S.U. stillage Particle Size Moisture % Protein % (dry basis) (H) Average High Low Average High Low 995 §_x 82.9 87.0 80.0 27.7 29.1 26.8 995 > x 3_438 87.7 88.9 86.7 49.6 52.3 39.0 438 > x 96.4 96.6 95.9 34.3 39.3 27.1 TABLE 5.--The fraction yields of SNECO separated M.S.U. stillage (per 100 lbs of feed) Particle Size Average Yield lbs Water lbs Dry lbs Protein (p) lbs Solids (dry basis) 995 3.x 17.0 14.0 3.0 0.83 995 > X > 438 9.3 8.2 1.1 0.58 438 > x 73.7 71.0 2.7 0.91 25 By examining Table 4, several trends can be seen. The mois- ture content increases as particle size decreases. This is to be expected since most of the solids' weight is concentrated in the large particles as seen in Table 5. Furthermore, the dewatering of the large particles is more easily accomplished than with the fines. The protein content of the intermediate particle size is significantly higher than either the fine or coarse fractions. The solids' fractions produced are still extremely wet with more than 80% moisture. This is to be expected in light of the gravity separa- tion process utilized and the relative difficulty of water removal from fine solids. The average mass yield is greatest for the fine particles and lowest for the intermediate particles. The intermediate particles may have the highest protein content, but they have the lowest contained protein. The main advantage of screening the product is the reduction of the moisture content. The feed is calculated to have an overall moisture content of 93.2%, whereas the solid fractions collected have a moisture content which ranged from 80 to 87%. This reduction, although not great, represents a reduction of over a factor of two in the amount of water ingested by the animal for an equal weight of solids consumed. However, the separation process used to prepare this feed resulted in the loss of about one-third of the solids (and contained protein) into the liquid stream. Some portion of this material is in soluble form and cannot be recovered, except by evap- orative concentration. Another portion, finely divided suspended 26 solids, cannot be easily recovered due to the very fine particle size. See Appendix A for further data. CRIPPEN Separator The results obtained with this device were less reproducible since it was not meant to separate wet sludges such as stillage. Because of its design, it was also difficult to pull consistent sam- ples and losses occurred frequently. Nhile varying the screen mesh size, tests were conducted by altering the screen angle in an attempt to improve the separation. The best angle for the tests appeared to , be about 17° with respect to the horizontal. Separation results for the CRIPPEN separator using M.S.U. stillage are summarized in Tables 6 and 7. (Details are given in Appendix A under Moisture and Proteins and under Fraction Yields.) TABLE 6.--The moisture and protein content of CRIPPEN separated M.S.U. stillage Particle Size Moisture % Protein % (dry basis) 0) Average High Low Average High Low 940 §_x 82.4 82.6 81.8 29.1 32.0 24.2 940 > x :_622 94.1 95.1 93.0 30.7 32.1 29.3 622 > x 95.2 95.6 94.5 31.3 33.1 27.7 27 TABLE 7.--The fraction yields of CRIPPEN separated M.S.U. stillage Particle Size Average Yield lbs Dry lbs Protein (u) lbs Ibs Water Solids (dry basis) 940 §_x 14.7 12.1 2.6 0.76 940 > x > 622 2.7 2.5 0.2 0.06 622 > x 82.6 78.6 4.0 1.24 The higher protein composition occurs in the fine particle size. However, due to the poorer separation with the CRIPPEN unit, the variation in protein by particle size is less pronounced. Most of the flow is in the fine particle fraction. Commercial Stillage CRIPPEN Separator The stillage produced by a commercial ethanol plant was sep- arated using the CRIPPEN unit. The quantity of material was somewhat limited but the results are given in Tables 8 and 9. (See Appendix A for further data.) TABLE 8.--The moisture and protein content of CRIPPEN separated commercial stillage Particle Size Moisture % Protein % (dry basis) (u) Average High Low Average High Low 940 §_X 86.9 87.1 86.6 25.3 26.2 24.3 940 > x :_622 93.6 93.6 93.6 23.6 23.6 23.6 622 > x 94.5 94.8 94.3 23.2 26.8 20.4 28 TABLE 9.-—The fraction yields of CRIPPEN separated commercial stillage Particle Size Average Yield lbs Hater lbs Dry lbs Protein (p) lbs Solids (dry basis) 940 §_x 44.9 39.0 5.9 1.49 940 > x 3_622 9.2 8.6 0.6 0.14 622 > x 45.9 43.4 2.5 0.58 Little variation occurred in protein composition between frac- tion sizes with this material. Due to an initially higher water content, it was more difficult to remove water from this material (as indicated by its higher moisture content) than the M.S.U. still- age. A calculation of the percent solids and percent protein in the commercial feed shows 9.0% and 2.21% (wet basis), respectively. It is apparent, then, that this material had a higher solids content than the M.S.U. product. 0n the other hand, a much larger fraction exists as large particles. This is due to a coarser grind of the corn. It is noted that the moisture content of the coarsest frac- tion was decreased from 91.0% (feed) to only about 86.9% so that the amount of moisture ingested by cattle would still be quite high. A large fraction of protein can be recovered easily by coarse screen separation. The fine fraction, however, will contain a significant protein value. The variance in yield (see Appendix A, Fraction Yields) is due to the manual batch method used to transport the stillage to the separating device. 29 SNECO Separator The results using the SNECO device to separate commercial stillage would have been very helpful. However, at the time commer- cial stillage was available for testing, the SNECO device was not. Based on the M.S.U. results, SNECO separated commercial stillage would exhibit higher solids content and lower labor requirements. Comparison of SNECO Separated M.S.U. Stillage, CRIPPEN Separated M.S.U. Stillege, and CRIPPEN Separated Commercial Stillage. The total protein content of the commercial stillage is lower than the M.S.U. stillage. This could have occurred for several reasons. Either the corn used for the commercial product had a lower protein content to begin with and/or the starch was not com- pletely hydrolyzed during fermentation. The two M.S.U. stillages had similar protein contents; the variation is probably due to a differ- ence in the degree of hydrolyzation. 0f the intermediate fractions, the SNECO separated M.S.U. stillage had the highest contained protein. In Figures 9, 10, and 11 the top line in each bar represents the maximum value, the middle line is the average value, and the bottom line represents the minimum value. If only one line appears, then there was only one piece of data for that particular bar. In the large particle fraction range, the CRIPPEN separated M.S.U. stillage had the highest protein composition. The SNECD sep- arated M.S.U. intermediate fraction was much higher in protein com- position than either the CRIPPEN M.S.U. or CRIPPEN commercial separated intermediate fraction. It is quite possible that for the 30 I 40‘ M.S.U./SWECO M.S.U./CRIPPEN C. 3 I, com-1500 IAL/ 45' 30 CR1 PPEN (SL- 9955 I: ,— 9405 x .3 ,2 +3 .. 9405 x 5 20 940>X2622 g 995>t 2438 .1” 622:4 <3 433> i 940>X2622 622 >X Particle Size (p) Figure 9.--Comparison of Protein Content in Separated Corn Ethanol Stillage 31 Figure 10.--Comparison of Percent Protein by Particle Size in Separated Corn Ethanol Stillage. 32 522 2183.. NOA RA zmaaHmU\ (Hummzzou «mumuHMdmu _ _ ? zmaaHmU\.:.m.z QSAXAm oum3m\.:.m.z w v 2m aH U\4 x 3_438p. Note that the intermediate fraction in all three groups has the lowest mass of contained protein of all the fractions. Upon examination of the total moisture content (Figure 12), it can be seen that the CRIPPEN commercial large particle size has a higher moisture content than either the M.S.U. CRIPPEN or SWECO fraction. The commercial product also had a greater amount of mate- rial in the large particle fraction than either of the M.S.U. mate- rials. Because of the poor separation using the CRIPPEN device, water would sit on the large particles. Since the commercial product had such a large yield in this fraction, dewatering was more difficult. A longer residence time would be needed in order to dewater the com- mercial product to a dryness comparable to the M.S.U. product. The moisture content of the commercial CRIPPEN and the M.S.U. CRIPPEN intermediate fractions are similar, but the M.S.U. SWECO is much drier. In the fine particle range, all the separations have similar moisture contents. 23Corn Industries Research Foundation, Corn Gluten Feed and Corn Gluten Meal (Washington, D.C., 1959), p. 12. 35 TOTAL PROTEIN TOTAL MOISTURE M.S.U.-CRIPPEN COMMERCIAL-CRIPPEN COMMERCIAL-CRIPPEN Figure 12.--Distribution of Total Protein and Total Moisture in Feed by Particle Size (u) 36 The division of total protein mass was more evenly dis- tributed in the M.S.U. SWECO stillage than in either of the other two stillages. The commercial CRIPPEN separated stillage had its highest contained protein in its large particle fraction, whereas the M.S.U. CRIPPEN separated stillage's appeared in its fine particle fraction. The distribution of the total moisture in the commercial j CRIPPEN separated stillage is about the same for the large and fine ‘-"L‘ particle fractions. In the other two separations, the fine particle fraction contains most of the moisture. Buchner Funnel with Whatman Paper The fine particle fraction, previously designated as "Wet," was further separated using a buchner funnel. Tests were conducted both on SWECO and CRIPPEN separated stillage fine particle fractions. Four different types I”: Whatman filter paper were used. The results are summarized in Table 10. This separation did an excellent job capturing much of the remaining protein. The captured protein remained on the filter paper. Overall, there was little variation whether or not a higher or lower porosity filter paper was used in the capture of the pro- tein. After the filtration, the filtrate obtained was clear, but mold growth did occur despite refrigeration during the time lag between the filtration run and the protein analysis. This spoilage is probably responsible for the variation in protein content. 37 TABLE 10.--Filtration results using Whatman paper to further separate M.S.U. stillage fine particle fractions Paper Filter Speed Porositya No. of Paper of paper (p) in feed N Contentb N Contentb in Filtrate % Protein Passing Filter Particle Size x < 438p and moisture content = 96.4% 4 Fast 20-25 1.8 1 Medium Fast 11 1.8 2 Medium 8 1.8 5 Slow 2.5 1.8 0.20 0.25 0.25 0.20 11.1 13.9 i 13.9 11.1 Particle Size x < 724p and Moisture Content = 95.8% 1 Medium Fast 11 2.7 2 Medium 8 2.7 0.319 0.288 11.8 10.7 Particle Size x < 724p and Moisture Content = 96.6% 4 Fast 20-25 2.5 2 Medium Fast 8 2.5 5 Slow 2.5 2.5 0.238 0.331 0.219 9.5 13.2 8.8 aWhatman Paper Division, 1979 Laboratory Catalog Paper Products, Publication 800, Clifton, New Jersey. bN stands for nitrogen. Unit is mg N/ml. 38 Although the mold does not affect protein content, it does produce a tough rubbery growth. If this growth is not broken down suffi- ciently, the sample will not be representative. This procedure was very difficult and often unsuccessful. Since the filter paper is made of a cellulose material, there is no problem in feeding it along with the captured protein to cattle. Furthermore, they would be difficult to separate. However, on a large scale basis, this method of capture is unfeasible due to the expense of the materials involved and the labor and time required to separate large volumes of material. The filtration rate is very slow. EIMCO Filter No separation was possible. Analysis of Alternative Uses of the "Wet" Fraction The "Wet" fraction could be utilized as fertilizer because of the certain minerals contained in the solids portion of the stillage. The solids portion has the following mineral content: 1.68% phos- 24 The remainder of the phorus, 2.2% potassium, and 4.78% nitrogen. solids material is organic matter such as carbohydrates and very small amounts of other minerals such as calcium. An economic analysis can be performed on all the products basing it on a bushel of corn. 24Personal conversation with Gary M. Webber, Research Assistant, Animal Science, Michigan State University, April 1983. 39 The April, 1983,25 commodity prices are: corn 0 $3.09/bu, anhydrous ethanol @ $1.70/gal, DDGS 0 6.5¢/lb, phosphorus 0 22.8¢/lb, potas- sium 0 11.4¢/lb, nitrogen 0 30.0¢/lb, and lime 0 1.675¢/lb. Assume that 7.19 lbs of dry matter (DDGS) are produced per gallon of ethanol and 2.5 gallons of ethanol are produced per bushel of corn.26 There is an additional cost for neutralizing the acidity of the material to be used as fertilizer, but it is negligible. (See Appendix B). The material has a pH of 4.0. If it were applied directly to the soil without neutralization, it would be detrimental to plant life. In Scotland, this material has been used success— fully as a fertilizer after it was neutralized with lime.27 If this fine material is not utilized as fertilizer, it must be disposed of. The hauling expense would vary, depending on the number of miles the material is transported and the size of the hauling operation. Cost estimates range from 1.5¢/9al up to 28 6.0¢/gal. There might also be a septic tank charge once the material is at the disposal site. An upper bound on this cost is 25Chemical Marketing Reporter, April 1983, Schnell Pub. Co., Inc. Also The Drovers Journal, Shawnee Mission, KN, April 1983. 26d. Waller and Gary M. Webber, "Development of a 'Controll- able' Farm Scale Research Still and Assoc. Research Package" (Paper prepared for the Department of Animal Science, Michigan State University). 27Personal conversation with Gary M. Webber, Research Assistant, Animal Science, Michigan State University, April 1983. 28Personal conversation with Dr. Mackenzie L. Davis, professor of Civil and Sanitary Engineering, Michigan State University, July, 1983. 40 1.0¢/gal.29 Therefore, on the high end, disposal costs should run 7.0¢/gal. Table 11 tabulates the market value breakdown of the products produced under two headings. Under the first heading, the "Wet" fraction is utilized as fertilizer, while under the second heading it is sent to a disposal. For the CRIPPEN commercial stillage, the "Intermediate" fraction is added to the "Wet" fraction since its moisture content is so high. All figures are based on one bushel of corn and maximum disposal costs were used. (See Appendix B.) For a true economic assessment, investment and operational costs must be included. The SWECO device costs more than the CRIPPEN device, but operationally it is less labor intensive. Since labor costs constitute a major expense in any operation, the number of man-hours reQuired is important. Excluding investment and operational costs, it can be seen from Table 11 that the Crippen separated commercial product has a slightly better market value than the others, when the "Wet" fraction is used as fertilizer. However, it has a significantly better market value than the others when the "Wet" fraction is disposed of. It should be noted though that the DDGS of the CRIPPEN separated com- mercial product is slightly wetter than the other two. It is assumed that percent protein (dry basis) is not a limiting criteria. The CRIPPEN commercial product had the highest contained protein, but the lowest percent protein on a "dry" basis. 29Personal conversation with Dr. Mackenzie L. Davis, professor of Civil and Sanitary Engineering, Michigan State University, July11983. 41 U88: an on 383 cc» umsmmu 83 38338382 83;» mo cowpcoa a 33 .3o ummoamwu co 38~3pwpcme mm 888: mg 383338 38:5 33 .u883 mm .muwpom 33v umcwmucou co ummun 83 83383 383382 833 pan .muoo apmacu no: 83 33 om .3838: mcwmpcoo 38338385 8333 A.mFamcmq Isou 83 ucmucou acaumwos 833 .33 83383 38 3833833883 833 .83 38333833 .383. 833 83 88888 83 383338.8 .838388338333. 8.88833338 3838383383 3333333 833 3833 3388833 .8833 88 8883 mp :83 covuumce .382. 8;» :3 883383303 uo: 38338385 8;» 338 3833 8853888 mczmww moon gumm8 3-3 883.83 83383 383.83 33303 888.33 83383 383.83 33303 ~-v 88.3 3883 38883833 3383.o «3.333 + 8838833 8833 3383.8 .3.333 + 8888883 8833 8838.8 83383 8833333383 88.3 3838338 838.833=< m~.3 3838338 838883333 3-3 88.83 3.83 3-3 88.83 3.83 88833338 .3.m.z--3883333 388.33 83383 383.83 33383 38°.N3 83383 383.83 33383 3-3 No.0 3883 38883833 8883.8 388.8833 8833 8838.8 .38838883 8833 om~3.o 83383 8833333383 88.3 3838338 838383333 88.3 3838338 838833333 AIV mo.m» ccou AIV mo.mm ccou 88833338 _838.83383--3833333 3~N.O3 83383 383.83 33333 838.83 83383 383.83 33383 N-“ 38.3 3883 38888833 8383.8 .3.333 + 88.8833 8833 8383.8 .3.3=_ + 88.8833 8833 8333.0 83383 3833333883 .mN.3 3838338 838.8333< m~.3 3838333 838883333 AIV mo.mm cgoo ATV mo.mw ccou 88833338 .3.m.3--oum3m 38888833 83 3.88 38338838 .383. 3833333883 88 88333333 38333833 .383. ccou 30 383833 383 mpuanoca 3o czouxmmcn mapm> 383385 x 3_438 12.0 87.8 52.3 995 > x 3 438 8.7 87.9 48.8 995 > x 3_438 7.9 87.8 50.6 995 > x 3 438 8.1 87.8 50.6 995 > x 3_438 7.4 86.8 49.4 995 > x 3_438 8.1 86.7 51.6 995 > x.: 438 7.3 88.0 39.0 995 > x 3_438 9.8 88.7 52.1 995 > XI: 438 11.3 88.9 52.1 438 > x 72.0 95.9 27.1 438 > x 74.7 96.5 36.0 438 > x 79.5 96.6 37.0 438 > x 79.0 96.4 34.0 433 > x 71.7 96.4 35.5 433 > x 76.8 96.4 32.2 438 > x 74.9 96.4 31.0 433 > x 72.2 96.5 39.3 433 > x 73.4 96.5 37.0 47 48 TABLE A-2.--CRIPPEN Separator-M.S.U. Stillage Particle Size (H) % of Total Flow % Moisture %D:;o§§:?s) 940 §_x 16.1 81.8 24.2 940 5.x 16.1 82.6 28.3 940 §_x 11.8 82.6 31.7 940 §_x 13.7 82.6 32.0 843 §_x 11.6 81.2 32.3 843 §_x 33.1 81.6 34.2 814 §_x 23.4 81.6 29.1 814 §_x 19.4 82.4 29.8 814 §_x 11.9 82.1 29.4 814 §_x 15.1 82.9 30.5 940 > x 3_843 0.4 94.3 29.5 940 > x.: 724 14.6 92.8 32.8 940 > x 3 724 5.3 89.5 37.1 940 > x 3_622 2.6 95.1 32.1 940 > x 3_622 2.6 93.0 29.3 843 > x 3 724 4.4 93.5 32.2 843 > x _>_ 724 7.6 92.0 33.2 814 > x 3_724 2.4 90.7 37.8 814 > x 3_686 6.8 90.1 37.0 814 > x 3_622 2.0 90.0 37.5 814 > x 3_622 1.1 89.6 40.3 724 > x 83.9 95.0 31.2 724 > x 62.5 93.6 25.3 724 > x 80.8 94.6 32.0 724 > x 71.7 95.3 34.7 724 > x 82.7 94.1 29.4 535 > x 81.3 95.7 35.5 686 > x 83.8 95.2 31.5 622 > x 83.4 95.6 33.3 622 > x 76.5 95.4 32.3 622 > x 74.6 94.5 27.7 A? “firm” V 49 TABLE A-3.--CRIPPEN Separator-Commercial Stillage Particle Size (u) % of Total Flow % Moisture %D:;°§§:?S) 940 §_x 57.7 86.6 24.3 940 §_x 54.4 87.1 26.2 814 5.x 53.9 85.6 23.5 814 §_x 33.3 83.9 20.6 724 j'x 10.0 85.5 26.4 686 §_x 26.0 84.6 20.2 686 fbx 26.0 85.4 22.6 940 > x 3_843 3.8 94.2 25.4 940 > x 3_814 6.4 94.7 25.7 940 > x 3_622 11.5 93.6 23.6 814 > x.: 724 8.2 92.1 24.6 814 > x.: 724 8.2 94.8 24.8 814 > x 3_686 5.0 95.0 26.2 814 > x 3_622 4.8 94.5 25.6 814 > x 3_622 4.8 94.4 25.8 724 > x.: 686 10.1 95.2 28.9 724 > x 3_686 10.1 94.7 22.2 724 > X.3 622 3.9 93.8 25.8 724 > x 3_622 3.9 94.0 24.8 686 > x 3_622 7.5 95.5 30.3 686 > x 3_622 7.5 95.0 28.3 843 > x 36.8 94.8 25.5 814 > x 35.9 94.3 22.5 724 > x 58.5 94.6 24.8 686 > x 76.0 94.6 23.5 686 > x 41.1 94.5 24.4 622 > x 34.2 94.4 22.4 622 > x 51.6 94.3 20.4 622 > x 86.1 94.8 26.8 '_ .4. t“ “'u’,‘-.\.:W——_ 50 TABLE A-4.--Fraction Yields . . . . lbs Protein Particle Size (u) Avg Yield, lbs lbs H20 lbs Solids (Dry Basis) SWECO Separator-M.S.U. Stillage 995 5 x 17.0 14.2 3.0 0.83 995 > x.: 438 9.3 8.2 1.1 0.58 438 > x 73.7 71.0 2.7 0.91 CRIPPEN Separator-M.S.U. Stillage 940 < x 13.9 11.5 2.4 0.70 940 S'x > 724 10.0 9.1 0.9 0.31 724 > x _' 76.1 71.9 4.2 1.28 940 < x 14.7 12.1 2.6 0.76 940 S'x > 622 2.7 2.5 0.2 0.06 622 > x " 82.6 78.6 4.0 1.24 843 < x 21.4 17.4 4.0 1.33 843 3’x > 724 5.7 5.3 0.4 0.13 724 > x —' 72.9 68.9 4.0 1.22 814 < x 18.2 15.0 3.2 0.95 814 S'x > 724 2.5 2.3 0.2 0.18 724 > x " 79.3 74.9 4.4 1.34 814 < x 16.4 13.5 2.9 0.86 814'; x > 686 6.4 5.8 0.6 0.22 686 > x _' 77.2 73.7 3.5 1.17 814 < x 18.0 14.8 3.2 0.95 814'; x > 622 1.6 1.4 0.2 0.08 622 > x " 80.4 76.5 3.9 1.21 CRIPPEN Separator-Commercial Stillage 940 j_x 58.0 50.4 7.6 1.91 940 > x.: 843 3.9 3.7 0.2 0.05 843 > x 38.1 36.1 2.0 0.51 940 53x 57.0 49.5 6.5 1.89 940 > x.: 814 6.5 6.2 0.3 0.08 814 > x 36.5 34.4 2.1 0.47 940 < x 44.9 39.0 5.9 1.49 940'; x': 622 9.2 8.6 0.6 0.14 622 > x 45.9 43.4 2.5 0.58 814 < x 39.5 33.5 6.0 1.32 814 3 x_: 724 7.5 7.0 0.5 0.12 724 > x 53.0 50.1 2.9 0.72 51 TABLE A.4.--Continued . . . . lbs Protein Particle Size (u) Avg Yield, lbs lbs H20 lbs Solids (Dry Basis) 814 < x 40.7 34.5 6.2 1.37 814 3’x > 686 4.6 4.4 0.2 0.05 868 > x _' 54.7 51.7 3.0 0.72 814 < x 41.3 35.0 6.3 1.39 814 3 x > 622 4.5 4.3 0.2 0.05 622 > x -' 54.2 51.2 3.0 0.70 724 < x 12.7 10.9 1.8 0.48 724 3'x > 686 12.8 12.2 0.6 0.15 686 > x —' 74.5 70.4 4.5 0.98 724 < x 14.0 12.0 2.0 0.53 724 3'x > 622 5.5 5.2 0.3 0.08 622 > x _' 80.5 76.1 4.4 1.02 686 < x 28.6 24.3 4.3 0.92 686 3.x > 622 8.3 7.9 0.4 0.12 622 > x _' 63.1 59.6 3.5 0.81 u - _. ._ -- . L‘A.‘ . ' w 52 TABLE A.5.--Moistures and proteins Moisture, % Protein, % Particle Size (u) Avg. High Low Avg. High Low SWECO Separator-M.S.U. Stillage 995 §_x 82.9 87.0 80.0 27.7 29.1 26.8 995 > x.3 438 87.7 88.9 86.7 49.6 52.3 39.0 438 > x 96.4 96.6 95.9 34.3 39.3 27.1 CRIPPEN Separator—M.S.U. Stillage 940 §_x 82.4 82.6 81.8 29.1 32.0 24.2 940 > x > 843 94.3 94.3 94.3 29.5 29.5 29.5 940 > x 3 724 91.2 92.8 89.5 35.0 37.1 32.8 940 > x 3 622 94.1 95.1 93.0 30.7 32.1 29.3 843 > x _' 81.4 81.6 81.2 33.3 34.3 32.2 843 > x > 724 92.8 93.5 92.0 32.7 33.2 32.2 814 > x 82.4 82.9 82.1 29.7 30.5 29.1 814 > x > 724 90.7 90.7 90.7 37.8 37.8 37.8 814 > x 3 686 90.1 90.1 90.1 37.0 37.0 37.0 814 > x 3 622 89.8 90.0 89.6 38.9 40.3 37.5 724 > x 7' 94.5 95.3 94.1 30.5 37.4 25.3 686 > x 95.5 95.7 95.2 33.5 35.5 31.5 622 > x 95.2 95.6 94.5 31.1 33.3 27.7 CRIPPEN Separator-Commercial Stillage 940 < x 86.9 87.1 86.6 25.3 26.2 24.3 940 3'x > 843 94.2 94.2 94.2 25.4 25.4 25.4 940 > x 3 814 94.7 94.7 94.7 25.7 25.7 25.7 940 > x E 622 93.6 93.6 93.6 23.6 23.6 23.6 814 > x 84.8 85.6 83.9 22.1 23.5 20.6 814 > x > 724 93.5 94.8 92.1 24.7 24.8 24.6 814 > x 3.686 95.0 95.0 95.0 26.2 26.2 26.2 814 > x 3 622 94.5 94.5 94.4 25.7 25.8 25.6 724 > x —' 85.5 85.5 85.5 26.4 26.4 26.4 724 > x > 686 95.0 95.2 94.7 25.6 28.9 22.2 724 > x'g 622 93.9 94.0 93.8 25.3 25.8 24.8 686 > x'— 85.0 85.4 84.6 21.4 22.6 20.2 686 > x 3_622 95.3 95.5 95.0 29.3 30.3 28.3 843 > x 94.8 94.8 94.8 25.5 25.5 25.5 814 > x 94.3 94.3 94.3 22.5 22.5 22.5 724 > x 94.6 94.6 94.6 24.8 24.8 24.8 686 > x 94.6 94.6 94.6 24.0 24.4 23.5 622 > x 94.5 94.8 94 3 32.2 26.8 20.4 APPENDIX B SAMPLE CALCULATIONS 53 II. ETHANOL AND DDGS MARKET VALUE SAMPLE CALCULATIONS Ethanol Basis: 1 bushel of corn 2.5 gal ethanol/bu corn x $1.70/gal ethanol = $4.25/bu corn 413 DDGS 4 Basis: 1 bushel corn 5,000,000 gal ethanol 18,000 tons DDGS This gives 7.19 lb of DDGS per gallon of ethanol a. SWECO-M.S.U. Stillgge (u) lbs, dry matter 995 §_x 3 0 995 > x > 438 1 1 438 > x T' 2.7 6.8 Adding the "Dry" and “Intermediate" fractions together and then dividing by the total gives (3.0 + 1.1)/6.8 = 0.603 or 60.3% $0.065 b. 0.603 7.19 lb DDGS 2.5 gal. ethanol = lB'DDGS gal ethanol x bu corn x $0.7045/bu corn 54 II. III. IV. FERTILIZER VALUE SAMPLE CALCULATIONS Basis: 5,000,000 gal ethanol 18,000 tons DDGS This gives 7.19 lb of DDGS per gallon of ethanol. SHECO-M.S.U. Stillage (p) lbs, dry matter 995 §_x 3.0 995 > x > 438 1.1 438 < x" 2;] 8 6. 8 In "Net" fraction or 438 > x, 2.7/6. = 0.397 or 39.7% Basis: 7.19 lb dry matter/gal ethanol SWECO-M.S.U. Stillage 7.10 lb dry matter/gal ethanol x 0.397 = 2‘85 1b dry matter gal ethanol 438 > x contains 2.85 lb dry matter/gal ethanol Basis: 2.5 gal ethanol/bu corn SWECO-M.S.U. Stillage: 2.85 lb dry matter gal ethanol 2.Sgal ethanol = 7.13 lb dry matter bu corn ‘bu corn X The "Net" Fraction material is 1.68% phosphorus, 2.2% potassium, and 4.78% nitrogen on a dry basis. SHECO-M.S.U. Stillage 7.13 lb dry matter = 0.12 lbrphosphorus bu corn x 0’0168 bu corn 7.13 lb dry matter = 0.157 lb potassium bu corn x 0'022 bu corn 7.13 lb dry matter _ 0.341 lb nitrogen bu corn x 0'0478 bu corn 55 _ ’_ Ann-.qu‘. I 56 Molecular Weight of Ca(0H)2 = 74 -5 88.41 l 4.995 x 10 rgmole x 74 g x 0.0022 lb bu corn 2 9 mole g = 7.189 x 10'4 lb bu corn 7.189 x 10'4 x $0.01575 = $1.204 x 10'5 bu corn lb bu corn ‘ ‘W's'. An". .‘A' £1.53ch P g j _n II. DISPOSAL COST OF ”NET" FRACTION SAMPLE CALCULATION Hauling cost range (depends on the number of miles) = 1.5¢/gal to 6.0¢/9al. Septic tank charges (may not be applicable) = 1.0¢/gal (high estimate) Assuming maximum costs, cost of disposal = 7.0¢/9al. Basis: 1 bushel of corn density of "Net" fraction = 8.34 lb/gal a. SWECO-M.S.U. Stillage (p) lbs, dry matter arg yield, lbs 995 < x 3.0 17.0 995 3'x > 438 1.1 9.3 438 > x —' “2:7 73.7, 6.8 In "Net" fraction, 2.7/6.8 = 0.397 or 39.7% From previous sample calculations, it is known that there are 7.19 lb of DDGS or dry matter per gallon of ethanol. b. 7.19 lb DDGS 73.7 lb al 0397 " gen ethanFl " 2.7 lb DDGS x _LT53.34 " $0 $7 = $1.64/bu corn 2.5rgal ethanol x . bu corn ga 57 APPENDIX C KJELDAHL ANALYSIS 58 MODIFIED KJELDAHL METHOD - M.S.U. DEPARTMENT OF ANIMAL SCIENCE Digestion of Samples for Auto Kjeldahl General. Weigh a sample of an amount sufficient enough so that 50-200 ppm may be obtained in a minimum of 25 ml. For each 1% crude protein (CP), figure 0.2 g in 250 ml final volume. Conc. For example: 10% CuSO4 K2504 H2504 10% CP use 1-2 9 (dry matter) in 250 ml 2 ml 2 g 15 ml 0.8 g (dry matter) in 100 ml 1 ml 1 g 12 ml 0.6 g (dry matter; in 50 ml 1 ml 1 g 10 ml 0.2 g dry matter in 25 ml 1 ml 1 g 4 ml To calculate the approximate ppm for your sample: a. 1 ppm = lg/1,000,000 g or 1 g/1,000,000 ml (using the approximation that 1 ml or 1 cc of H20 weights 1 g) Divide by 1,000,000. Then 1 ppm = 1 ug/g or 1 ug/ml b. CP% 9 N CP% x 1,000,000/(100 x 6.25 x (250 m1 f1ask)) = CP x 1,000,000 625 x flask size 100 = CP/g feed and CP/g e 6.25 = g N/g feed 1,000,000 = ug and ug/size of flask = ug/ml or ppm X For example: For a sample estimated to be 10% CP in a 250 ml flask: 10% x 1,000,000 100 x 5.25 x 250 = 54 g/m‘ 0? 54 PPm 1 g in a 250 ml flash would be between 50 - 200 ppm. 59 60 For a sample estimated to be 6% CP in a 250 ml flask: 6% CP x_1,000.000 100 x 6.25 x 250 38.4 ppm This amount is not within the 50-200 ppm range, so 2 9 must be used to obtain 76.8 ppm which is within the range. Weigh about 1.0 g of feed mix on a weighing paper. Record the exact weight. Do not touch the sample. Transfer this sample to a special 250 ml volumetric flask. Add 2 g of K2504 and then carefully add 2 ml of CuSO4. Add some concentrated H2504. To speed up digestion, carefully add 4-5 ml of 30% H202 to the above mixture and let it sit overnight if desired (see 4). Oxidation digestion. All nitrogen compounds (except N0; (nitrate)) are oxidized to NH3 and dissolved in the acid solution as (NH4)ZSO4. Allowing the sample to sit overnight reduces frothing, but this is unnecessary. Digestion may begin immediately. During the charring stage, if frothing occurs, shake well or swirl the sample. The frothing doesn't harm the analysis, it simply increases the odds the sample will puff itself right out of the flask. a. Set the heaters at #3-4 to begin. Gradually increase the heat. Turn to H1 when the sample starts fuming with white fumes of $02 which comes from the decomposition of the H2S04. These fumes are poisonous!! Shake down the charred mate- rial occasionally. b. Boil until most of the solid material is in solution. The mixture will be a dark reddish brown. c. Remove the two flasks and cool them. Add 30% or 50% H202 from a pasteur pipet. Try to wash down the bits of charred material. Return mixture to digester. 61 H202 addition should not cause the mixture to sizzle and steam. It should not spit out of the flask. Cool longer if spitting occurs. Continue adding H202 until mixture is blue-green. d. If the mixture is still somewhat yellow—green, repeat the peroxide procedure twice. e. Boil 15 minutes past the last peroxide addition. f. Cool to room temperture. Standards. Make 5000 ppm. Weigh on an analytical balance 23.6 g (NH4)ZSO4 into a one-liter volumetric flask. Add some dionized water and mix. Make the solution up to the mark on the flask. Mix well and transfer to a glass stoppered bottle. Label and store in a cold room with parafilm or plastic wrap around the stOpper. a. For each set of 10 samples, digest one 10 ml aliquot of the above standard in a 250 ml volumetric flash and one 5 ml aliquot in another 250 ml flask. b. 10 ml of 1000 5 ml of 1000 1/100 or 0.236 g/250 ml flask. 200 N ppm 1/100 or 0.118 g/250 ml flask. 100 N ppm Dilution: a. Dilute samples and standard to the 250 ml mark while swirling with deionized distilled water. b. Mix well. c. Pour some of the solution into two screw cap bottles-- 50 cc is plenty. Label correctly and cap tightly. d. Run samples on an Auto Kjeldahl unit. Calculations and traces. Two aliquots from each bottle will go into the instrument. Sometimes there is a slight drift. The double injection will average the drift. Count and average the two peaks for each aliquot. 62 a. Average count x 2 = ppm of N or ug/ml b. pg/ml x 250 ml = total ug in the flask c. Total ug/ml : sample weight = ug N/g sample d. pg x 1,000,000 = g e. g N/g sample x 100 = g% N f. 9% N x 6.25 = % CP 7 CP = avg count x 2 x 250x 100 x 62 ° sample weight x 1,000 ,000 x (Hry matter)= avg count x 0.3125 s.w. x dLm. g. (1) Run a linear regression on the average transmittance percentage versus the concentration such as: ____3£____ __JL__. 0 0 50 ppm 24.6 100 ppm 49.5 150 ppm 73.0 200 ppm 99.5 (2) Ratio of actual slope to the theoretical value (3) Divide the factor in f above by 1.01 and complete the calculations. h. Average the % CP and report if the match is 90% or better. 1000 ml of dilution 8. mg protein/g sample = chart reading (ppm) x 2 e 6.25 sample weight in g BIBLIOGRAPHY 63 BIBLIOGRAPHY Chemical Marketing Reporter. April 1983. Schnell Pub. Co., Inc. Corn Industries Research Foundation. Corn Gluten Food and Corn Gluten Meal. Washington, D.C., 1957. Davis, Dr. Mackenzie L. Personal conversation, Department of Civil and Sanitary Engineering, Michigan State University, East Lansing, MI, July 1983. Distillers Feed Research Council. Distillers Feeds. Cincinnati, Ohio. Drovers Journal. Shawnee Mission, Kansas, April 1983. Farlin, Stanley D. "Wet Distillers' Grains: An Excellent Substitute for Corn in Cattle Finishing Rations.“ Animal Nutrition and Health (April 1981): 35. Geiger, Joseph W. "Design, Energy, and Economic Analysis of Small Scale Ethanol Fermentation Facilities." Master's Thesis, Michigan State University, 1981. National Research Council. Feeding Value of Ethanol Production By- Products. Washington, D.C.: National Academy Press, 1981, p. 15. Perry, R. H., and Chilton, C.li. Chemical Engineers' Handbook, 5th edition. New York: McGraw Hill, 1973. SWECO, Inc. SWECO Vibro-Energy Separators. Los Angeles, Ca. Waller, Dr. J. C., et al. "Use of Fuel Ethanol By-Products in Livestock and Poultry Diets." Paper prepared for Michigan State University. "Separation and Storage of High Moisture Distillers Feeds." Paper prepared for DOE, Michigan State University, 1983. , and Weber, Gary M. "Development of a 'Controllable' Farm Scale Research Still and Assoc. Research Package." Paper prepared for Michigan State University, 1981. 64 65 Webber, Gary M. Personal Conversation, Michigan State University, April 1983 and August 1983. Whatman Paper Division. 1979 Laboratory Catalog Paper Products Publication 800. Clifton, New Jersey. "11111111111144“