Fermentation: Recent Advances

Transcription

1 Fermentation: Recent Advances W. Mike Ingledew, Saskatoon (Canada) In the year 2000, I presented a talk at a local meeting on fuel alcohol about some of the major fermentation issues our industry had to understand. At that time in North America, 54 ethanol plants were operating and 6 billion L were made. There are now 111 plants in operation making 20.6 billion L, and 85 plants are in construction or expansion that will provide 23.5 billion more liters for a total of 44 billion/yr. In recent months the industry has seen ethanol profits as high as 86 cents (U.S.) per gallon when oil was $70/barrel and corn was only $2.50 per bushel (of 56 lbs or 25.4 kg). Today, with oil near $50 and corn priced above $4/bushel, it is claimed that every gallon of ethanol is made at a 21cent/gallon loss (Wall, 2007). Others have indicated that profit margins of 61 cents per gallon last year have slipped to 19 or 20 cents (Brasher, 2007). If true, this example points out the risk of this industry, the sensitivity of fermentation ethanol to both energy prices and substrate costs, and why every scientific advance possible in the manufacturing process must be exploited in a continual struggle to lower the cost price of production of this fuel. This also shows the even larger risk of cellulosic ethanol production that is claimed to be ready for production but at a significantly higher cost than grain-based alcohol. What have we as an industry learned so far about the fine tuning of the starch to ethanol process, and are we exploiting all these measures in current and future production? One of the first things we saw as the industry began to mature is that alcohol plants should strive to purchase high starch, low dockage, low moisture grain - free of mold infestation. Although the chemists have shown that ethanol production is not always proportional to starch, it is well known biochemically that you can t make more ethanol without more glucose, and glucose comes from starch molecules that are efficiently degraded by the exogenous enzymes provided by enzyme companies. In North America, we have also learned that if all feed grade corn and small grains were converted to ethanol we could only replace ~20% of the current consumption of gasoline with ethanol. This means that ethanol is not a complete answer to replacement of liquid fuel additional croplands must be planted and other alternate energy sources will have to step up to increase energy production for petroleum replacement, for carbon dioxide reduction and for energy security. We have also learned that this feed corn is removed from animal consumption but the feed industry is not ready to use distillers dried grains (DDG) or DDG with solubles (DDGS) in place of all this grain in spite of the fact that this byproduct of ethanol production contains a much higher protein content than the original grain (due to starch removal and an additional amount supplied by the addition of surplus dead yeast from the process) (Ingledew, 1999a). Demand for feed corn has led to higher prices and has complicated the plans and aspirations of this industry. My own laboratory has been attempting to lower ethanol cost price by allowing ethanol plants to make higher concentrations of ethanol in the same physical plant and with the same numbers of plant employees - using a technology that we called Very High Gravity (VHG) fermentation. In this process, higher dissolved solids mashes near 38 Brix must be made and handled in processing so that ethanol can be made at levels almost twice what was the norm in the mid 1980 s. It is now known that ethanol production at 23.8% is possible in lab situations, and that one company, Broin and Associates, now averages 20% v/v in the plants that have adapted with BPX technology and have been engineered for this process. A number of papers describing this VHG technology have been published. Two papers are particularly relevant (Thomas et al., 1993 and Thomas et al., 1996), and some of the steps needed to be taken to implement VHG fermentation are listed in Table 1. Calculations on VHG mashes are provided (Thomas et al., 1996).

2 It is also possible to move towards increased fermentation rates while eliminating the grain residues which after dextrinization are not required in the fermentor other than for buffering (Abbott and Ingledew, 2004). Grain removal results in an approximate 11% volume gain in each fermentor and this volume is better used to make more alcohol. In addition, mash viscosity is reduced, and no cooling of the insoluble grain particles in the fermentor, or heating of insoluble grain particles in the still are needed, saving energy. In new plants, the still design would not require the consideration of large amounts of solid particles, leading to liquid-liquid extraction rather than liquid-liquid-solid extraction. No dry grind fuel ethanol plant is yet known to carry out spent grain removal prior to fermentation. Technology in wet milling, of course, does lead to solids-free mashes, and in this process, spent yeast is easily removed at the end of fermentation leading to a liquid-liquid distillation. The brewing industry, and the wine industry (white wine at least) remove insolubles prior to fermentation, and use a variety of specific equipment for the process. In this industry, use after mash preparation of a decanter centrifuge (with rinse) or a rotary drum vacuum filter would appear to be the best choice. These can be used while the mash is still hot, minimizing bacterial contamination in this microbiologically difficult process step. In this way, the grain obtained could at least be dried and used for animal food, and would likely be lighter in color and perhaps nutritionally advantageous. It may also have other uses as its microbiological content will be minimal. These very reasons provide a solid basis for the fractionation of grain prior to processing using either dry grind, dry milling or wet milling techniques or using a simple one pass debranning to remove husk (a low starch, high fiber component). Techniques like this are described by Singh (2006) and colleagues in a large number of publications and presentations, and somewhat independently by Broin and Associates in their proven B-frac processing and commercialized ethanol production technology. Fermentation plants are known to be rather polluting. This has led to governmental scrutiny of the whole ethanol industry and it is now very much less likely that surplus spent grain, excess thin stillage and other chemicals from ethanol plants will be able to be spread on land or passed into water bodies. Gaseous emissions are now treated by thermal oxidizers after being scrubbed to remove organics including ethanol. Some plants provide their own sewage disposal systems, but most find that the liquid wastes can be processed by methanators (removing organics in water by anaerobic digestion providing methane for internal plant use). Thin stillage from the plants can also pass into the methanators, but often is handled by triple effect evaporators that create syrups that can either be burned in boilers, dried with DDG to produce DDGS, or sold outright for animal feed. As VHG procedures are brought on board, it will be necessary not just to increase cooling capacity in plants (probably with external heat exchangers), but also to plan plants with excess evaporative capacity so that, at times, all or more thin stillage can be removed from the site and NOT recycled into new mash. This process, termed backsetting, can be very detrimental to fermentations due to the organic acids, ions and other inhibitors that build up in concentration as backset is continually recycled. The benefits of thin stillage in fermentation in this author s opinion are grossly exaggerated. Yeasts have already had one to a multitude of chances to utilize any nutrients that might have been present in each fermentation as the backset was offered to them over and over again. Most fermentation problems disappear when backset is eliminated (or even reduced) for one or more cycles. Other plants at least remove the 6-10% insolubles from backset that confound yeast counts in newly pitched fermentors (the dead yeast appears live by the poor methylene blue microscope technique used). However, soluble ions and acids and other inhibitors that may be there in the substrate mash do not need to be recycled at increasing concentration each round. The best reason for excess evaporative capacity, however, will be to decrease fresh water useage, and to move towards zero effluent technology.

3 Another concept that requires clarity is the oxygenation of the mash. Brewers understand the need for oxygenation from 8-20 ppm in fresh wort because they determined early that growing yeasts produce ethanol faster than non-growing yeasts. In this semi-anaerobic fermentation, a small amount of oxygen is needed for synthesis of the sterol and unsaturated fatty acids used to make cell membranes, and IF not supplied, the yeasts are less viable and less vital during and at the end of fermentation. Brewers recycle yeasts for long periods of time (many cycles - termed by them to be generations) because yeasts are almost completely viable in their ~ 5% ethanol beer, and due to stringent sanitation and sterility of wort they normally have little contamination in the recycled yeast cream. In fuel alcohol production, we only need to supply oxygen when the yeasts are growing, but we do not recycle yeasts from one fermentation to another as they are virtually all dead in our 12-20% v/v ethanol beers. Good practice should be the provision of 8-20 ppm oxygen to mash (perhaps at 5-20 ppm per hour due to 3-4 hour generation times) but only during the time of continued yeast growth. It is therefore suggested that any aeration over this rate when yeast growth is still happening is likely a waste of money. Moreover, the best place to aerate may be on the cold side of an external heat exchanger, taking advantage of the temperature and the turbulence to dissolve the oxygen most effectively and mix it into the fermentor mash. Note also that oxygen residuals become a problem in those plants that have installed carbon dioxide recovery systems. Temperature staging is another interesting advance. Some years ago, we published a series of four graphs ( Jones and Ingledew, 1994) in which it was shown that temperatures of over 27 C did not permit the production of high concentrations (near 20%) ethanol under any conditions even when usable nitrogen in mash is provided. The clever application of this research has been the procedure of temperature staging. In the many plants in North America that use this procedure, the temperature used in fermentation is controlled at C initially, and as the ethanol content of the fermentor increases, the temperature is ramped downwards (presumably to less than 27 C) as the alcohol rises to the point (10-13%) where inhibition of yeast growth occurs, so that continued metabolism not affected at lower ethanol values proceeds to near 20 or 21% v/v. This kind of thought process leads one to revisit the computer modeling of the ethanol process to increase plant efficiency. Only now are some plants turning attention to such processes so that the utmost lowering of costs can occur as overall plant efficiencies are increased. Another important aspect of ethanol production is the yeast used in the process. The producers of yeast should be able to provide yeasts with low levels of contamination such that ethanol processing is not compromised by bacteria and wild yeasts co-inoculated with purchased yeast. Yeasts from the major suppliers will be well researched and chosen for ethanol manufacture, and representatives from these companies will provide appropriate recommendations and advise on use. Nevertheless, as an incoming commodity, in-house ethanol plant microbiologists, if present, should monitor incoming batches of active dry yeasts, compressed yeasts or yeast creams (stabilized or not) for viable yeasts per gram, viable bacteria per gram, wild (non-saccharomyces) yeasts per gram, and perhaps other important parameters such as generation time, maximum cell number in fermentation of typical mash, fermentation rate, maximum temperature tolerated, and tolerances to ethanol and sugar. In this light, it is important that propagation or conditioning tanks be operated in such a way that bacteria and wild yeasts are not favored by conditions imposed. It is also important not to contemplate continuous propagation in an ethanol plant where the less sanitary conditions present will lead to the growth of any contaminant that has a faster generation time than the culture yeast used. Some yeasts sold may be better than the yeasts used in your plant. Keep in mind that the new and smaller ADY yeasts (Ingledew and Bellissimi, 2003) have higher cell numbers per gram

4 than traditional strains, but if you buy by weight, and viabilities are equally high, the same amounts of viable protoplasm will be inoculated with either new or traditional yeast strains. It is strange that little research data is available to show that any of the current strains really are different regarding tolerances to ethanol, temperature, organic acids, sugars or ions. In fact, the existing but small amounts of data seem to indicate similarities between the new yeasts leading one to believe that some, at least, have common lineage. At this time, it would seem safest to purchase yeast from those companies that stand behind their products and have research capabilities to ensure quality. Another point should be made about yeast propagation. Most yeasts of the Saccharomyces genus used in alcohol production have generation times in the range of 2.6 to 3.6 hours in a good mash. The lag time prior to growth is usually 5-7 hours. Both numbers are dependant on the nutrition of the mash. In any case, in fuel alcohol plants where conditioning is done, less than 12 hours are provided for this activity. One reason for conditioning is to grow more yeasts, thus saving some money in the processing. The above figures, if true in most mashes, show that at best, two to three generations of yeasts take place. This means that if the amount in the condition tank inoculum was enough to provide 1 million cells per ml when inoculated directly into mash, a suitable conditioning of the same yeasts would provide only 2 to 4 million yeasts/ml (Bellissimi and Ingledew, 2005a,b). At risk would be the contaminants that co-culture with the active dry yeasts due to lack of cleanliness of tanks and pipes used to condition. Labor and time and cost of vessels should also be factored into this procedure. Little information seems to be available tracking the revival of ADY under plant conditions. Recent literature (Soubeyrand et al., 2006) shows that the only rehydration indicators showing early recovery of the yeast are carbon dioxide release and proton (ph) release. Rehydration temperatures were also recommended to be above 30 C and 43 C with 38 C being safe (not causing death). The effects of nutrition are also mandatory to really understand yeasts, their growth rates and their production of ethanol under semi-anaerobic conditions. In our lab, we have investigated mashes made from almost all grains, from molasses and from syrups. All mashes were found to be deficient in usable nitrogen compounds - which for yeasts includes only amino acids, and low molecular weight inorganic ammonia and urea - but not proteins and not peptides (Thomas and Ingledew, 1990). Addition of yeast foods containing usable nitrogen increases fermentation rate and makes fermentation rate less dependent on the temperature. A calculation of the amount of yeast growth occurring in a fermentor demonstrates the very large amount of cell mass made, and therefore the need for specific nutrients that are part of the composition of yeast cells. If we assume a 250,000 US gallon (950,000 L) fermentor and yeast growth to 250 million cells/ml under conditions of ethanol production, the fermentor at end fermentation would contain ~2.3 X yeasts. Thomas and Ingledew (Ingledew, 1999a) have shown that one gram of yeast contains ~ 4.87 x yeasts. This fermentor would contain 4,846,000 grams or 4864 Kg (10,660 lbs) of yeasts. The inoculum used was approximately 250 Kg (550 lbs), so the net growth of new cells could easily be calculated. Knowing the composition of a yeast cell (Ingledew, 1999b), one could estimate how much usable nitrogen would have to be present in the mash to grow these cells. Similarly, levels of other essential nutrients could also be calculated. Our work has shown that with little available nitrogen species in mash, almost 0.5 grams of urea per L would have to be added to fully provide enough of the essential nitrogenous nutrient. Yeast cells contain approximately 6% nitrogen when grown semi-anaerobically. When one has considered yeast composition and yeast nutrients in mash and supplied as yeast food, a fuller understanding of yeast, the magnitude of nutrients required and fermentation rate.

5 So very important is the fact that a growing yeast makes alcohol faster than non-growing yeast (Kirsop, 1982). Put another way, the glycolytic pathway functions to take more than 90% of the sugar in fermentation through the pathway with the production of a net two ATP (energy currency of the yeast used for growth and metabolism) and two molecules of NADH + H + per molecule of glucose used. Pyruvic acid is the end product of glycolysis, and it is then converted through acetaldehyde to ethanol, reoxidizing NADH + H + back to NAD that participates again in glycolysis of another glucose molecule. The pyruvic acid acts as a sink for the reduced enzyme cofactor made in glycolysis which otherwise would accumulate and temporarily inhibit further metabolism. This is why growing cells (growth limited by ATP concentration) make ethanol about 33 times faster than cells that have ceased growth and no longer need ATP at the same levels. This is also a large part of the reason why glycerol is made when intermediates taken for growth from the pathway subsequent to the production of NADH + H + lead to a lower level of sink and alternate mechanisms to reoxidize excess NADH + H + lead to conversion of dihydroxy acetone phosphate to glycerol. The understanding of these concepts is why we have made the statement that, in ethanol fermentations, the objective is to grow as many cells as possible in as short a time possible such that by the time growth ends (10-13% ethanol v/v), there are 20 to 30 times more cells in the fermentor than inoculated, and they continue metabolizing glucose (albeit at a rate approximately 30 times less than cells in exponential phase) until glucose is gone. These cells have high levels of glycolytic enzymes, and Very High Gravity levels of ethanol result. Another area of great concern in the ethanol industry is one imposed partly due to the presence of contaminants, and partly due to engineering procedures designed to reduce pollution and overall reuse of water in the process. The procedure leads to recycle of water in the form of backset and, in some plants, recycled methanator water. These water sources used in very high amounts in some plants contain unused substrates and mash constituents, boiler additives,, organic acids such as lactic, acetic and succinic acids, and ions (especially sodium from sodium hyroxide used as a cleaning compound). Mycotoxins and process condensate compounds can also be present. Any compound that increases in concentration above that inhibitory to yeast can cause stress, and therefore serious stuck or sluggish fermentations. Antibiotic use to combat serious bacterial infections that can lead to ethanol yield losses of more than 17% are not discussed in this article, as in Europe, the use of such antibiotics are banned. It is foreseen that as the alcohol industry expands in such circumstances, other antimicrobial compounds will have to be employed to reduce such losses. For now, few such compounds have been identified. Hop extracts are in use but appear to be less antibacterial and somewhat more expensive than antibiotics. Chemicals such as hydrogen peroxide, chlorine dioxide and even urea hydrogen peroxide have been tested and some of these hold some promise for the industry. In the meantime, the ethanol industry in Europe will likely experience some heavy yield losses and the only fix might eventually be the establishment of serious and expensive clean-in-place (CIP) cleaning and sanitation procedures as used by the brewing industry for the manufacture of beer. The alcohol industry is a multidisciplinary business encompassing a broad range of expertise including plant science, food science, chemistry, biochemistry, microbiology, engineering, animal nutrition, business administration, finance, economics, and politics. Due to the growth of the industry world wide, there is a severe deficiency of labor to staff new plants and to re-staff raided manpower. A program of education is therefore an absolute necessity for the industry. So far, there is only one textbook covering fuel production (The Alcohol Textbook, 4 th Edition, Ethanol Technology Inc.), two schools (The Alcohol School Europe and The Alcohol School North America, Ethanol Technology Institute being most established) and one Workshop (Fuel Alcohol Workshop by BBI International). Many other meetings take place around the world, but

6 most concentrate on marketing and political matters - not on the science of alcohol production. No research journals or Associations exist that cover production and quality assurance issues. For this reason, a focus on education of this industry is most important. Consultants and speakers can be used to advantage by companies wishing to continue the learning process. My concern as well rests with the engineering design teams around the world. There have been so many new developments in processing and engineering that it is just as difficult for engineering firms to keep up with advancements as it is for plant operators and managers. If this statement is even partially correct, then how will the plants in construction and expansion (as well as any future plants) ever be sure that the technology engineered into their new plant will be compatible and state-of-the-art? How will the wishes of the owners of the plants be interpreted into a turnkey workable facility free of problems on startup? The hundreds of questions located in the BBI Plant Development Handbook are only a start. Many basic process alterations might lead to lower cost alcohol IF they could be easily inserted in plant design in modules. There appears to be no answer to such questions. One can only hope that each advance will, if successful, be used as it is proven in design and operation. References 1. J.K. Wall. Making Ethanol Not Risk-free After All. Indianapolis Star February 18, P. Brasher. Report: Squeezing Profit from Ethanol to Get Harder. Register Washington Bureau March 7, W.M. Ingledew, 1999a. Yeast - Could You Build a Business on this Bug? In: Under the Microscope. Focal Points for the New Millennium. Biotechnology in the Feed Industry. Edited by: T.P. Lyons and K.A. Jacques. Nottingham University Press. Thrumpton, UK K.C. Thomas, S.H. Hynes, A.M. Jones and W.M. Ingledew, Production of Fuel Alcohol From Wheat by VHG Technology. Effect of Sugar Concentration and Fermentation Temperature. Applied Biochemistry and Biotechnology, 43: K.C. Thomas, S.J. Hynes and W.M. Ingledew, Practical and Theoretical Considerations in the Production of High Concentrations of Alcohol by Fermentation. Process Biochemistry, 31: D. A. Abbott and W.M. Ingledew Buffering Capacity of Whole Corn Mash Alters Concentrations of Organic Acids Required to Inhibit Growth of Saccharomyces cerevisiae and Ethanol Production. Biotechnology Letters, 26: V. Singh Past, Present and Future of Dry Grind Process Bioenergy Symposium. Purdue University, West Lafayette IN June 6, A.M. Jones and W.M. Ingledew, Fuel Alcohol Production: Optimization of Temperature for Efficient Very-High-Gravity Fermentation. Applied and Environmental Microbiology, 60: W. M. Ingledew and E. Bellissimi, Active Dry, High Alcohol Yeast : The New Yeasts on the Block. Ethanol Producer Magazine, 9(6): V. Soubeyrand, A. Julien and J-M. Sablayrolles Rehydration Protocols for Active Dry Wine Yeasts and the Search for Early Indicators of Yeast Activity. American Journal of Enology and Viticulture, 57(4): K.C. Thomas and W. M. Ingledew Fuel Alcohol Production: Effects of Free Amino Nitrogen on Fermentation of Very-High-Gravity Wheat Mashes. Applied and Environmental Microbiology, 56: I.S. Pretorius Tailoring Wine Yeast for the New Millennium: Novel Approaches to the Ancient Art of Winemaking. Yeast, 16: E. Bellissimi and W.M. Ingledew. 2005a. Metabolic Acclimatization: Preparing active dry yeast for fuel ethanol production. Process Biochemistry, 40:

7 14. E. Bellissimi and W.M. Ingledew. 2005b. Screening of commercially available active dry yeast used for industrial fuel ethanol production. American Society of Brewing Chemists Journal, 63: W.M. Ingledew, 1999b. Alcohol Production by Saccharomyces cerevisiae: a Yeast Primer. In: The Alcohol Textbook. A Reference for the Beverage, Fuel and Industrial Alcohol Industries. 3 rd Edition. Edited by: K.A. Jacques, T.P. Lyons and D.L. Kelsall. Nottingham University Press. Thrumpton Nottingham UK. 16. B.H. Kirsop, Developments in Beer Fermentation. Topics in Enzyme and Fermentation Technology 6: Address of author: W. M. Ingledew Professor of Industrial Microbiology Department of Applied Microbiology and Food Science University of Saskatchewan Saskatoon, SK S7N 5A8 Canada Table 1: Steps to be taken to implement VHG Fermentation. Prepare mashes with increasingly high solids (less water) - you can t make higher ethanol without more potential sugar Work up slowly in specific gravity to ensure that problems don t occur Remove solids (with rinse) prior to fermentation - if it can be done Supply sterile oxygen to the fermentation - cold side of heat exchanger - 20 ppm or ~ 5 ppm/hour - needed for strong yeast cell membranes Supply enough nitrogen so that mashes are not usable N deficient - LOTS! More stimulates fermentation rate and increases yeast cell numbers Supply other nutrients as needed. Gelatinize and liquefy but don t saccharify starch in mash before fermentation Ensure that pumps can handle mashes of more than 32 Brix Use a yeast which tolerates alcohol well and thrives in high sugar media Carry out SSF in fermentors (Simultaneous Saccharification and Fermentation) Condition or prepare yeast in lower gravity mashes to inoculate VHG mash Do not reuse yeast they will be dead at the end of fermentation Keep fermentor temperature down (without much loss in rate of ethanol made due to adequate nitrogen supplied) and/or use temperature staging Keep mash free from contaminating bacteria and their end products (STRESS)

8 From: Pretorius, 2000 Substrates End products ATP Thousands of other enzymes are also located in the cell. Figure 1: Enzymic steps of glycolysis for ethanol production of yeast (Pretorius, 2000).