Biochemistry is rarely quantitative for given reactants and products unless a specific process is excised from a known pathway and performed in vitro.
For instance, enzyme isolation and purification, together with a clean substrate, can yield a small number of products. One example is the use of β-amylase to generate maltose from starch.1 Now this can be an aspect of starch hydrolysis using an “amylase mixture,” but in the case of the latter, the resulting product is likely to be a mixture of glucose, maltose, and a minor contribution of smaller oligosaccharides and limit dextrins. The point is that while the product spectrum is broader, the mono-, di-, and tri-saccharides are amenable to fermentation to produce alcohol and therefore satisfy conventional fermentation in terms of ethanol yields. A mixture of these fermentation inputs offers little resistance to their processing by fermentative organisms.
The process is ostensibly simple. Mix a fermentable substrate at an appropriate temperature with a microorganism species or two (or even more, especially if there is a performance-significant population of “contaminating” organisms), sit back and wait. Usually the onset of fermentation becomes evident with the evolution of bubbles,2 mainly carbon dioxide, and when bubbling ceases we can assume that fermentation has finished or is close to finishing.
But we all know that, to quote an ancient proverb, “There’s many a slip twixt cup and lip”. The use of microorganisms, which for the purposes of this article is restricted to Saccharomyces cerevisiae (for the sake of clarity and to limit the generality required to encompass the plethora of fermentative microorganisms), results, in a sense, in a contaminated fermentation broth. It’s contaminated in the sense that the primary goal is the production of ethanol, with a co-product of carbon dioxide, the latter sequestering a third of the carbon from the fermentable carbohydrates available that remain after activities such as cell growth.
However, other pathways that generate compounds known as secondary metabolites (or congeners) are produced alongside ethanol and carbon dioxide. Congener production is a consequence of the necessity for the completion of many biosynthetic pathways, such as the production of proteins — structural and enzymatic — that require a specific amino acid complement. A malted barley extract has a respectable complement of essential amino acids but still requires fine-tuning of the amino acid profile during fermentation for adequate yeast growth and function. This is generally dealt with via the classic amino acid biosynthetic pathways, although there is a subset of amino acids — termed essential amino acids — that yeast requires in the broth as it cannot synthesize these de novo.
The quantity of given chemical components by mass is less important than how it relates to the flavor threshold. The encyclopedic work compiled by Nykanen and Suomalainen (1983) lists the major products from various alcoholic drink categories. Some of these congeners will ultimately contribute flavor and mouthfeel to new-make spirits and can be expressed as flavors in the final distilled products. No matter the source of the raw materials or the mode of distillation, fermentation contributes to the breakdown products and metabolites of yeast and bacteria. Their contribution may be more subtle in more highly rectified spirits, such as those from continuous distillation, but nonetheless fermentation can, with volatiles from the raw materials and extractives from wood, be considered to be the major flavor contributors to most spirits. Indeed, this drives the stipulation for spirits such as cognac and scotch whisky that the ethanol content of new-make spirits cannot exceed 94.8 percent ABV so that at least some of the flavor from the raw materials and fermentation are retained.
While distillers demand a consistent and high-quality spirit, they also want as high an ethanol yield as possible from their fermentations, which implies avoiding excessive yeast growth. Distillers yeast sources its nitrogen requirement from the various amino acids and small peptides released during cereal malting and mashing. Alternatively, for potentially nitrogen-deficient fermentations, yeast extracts or the simple salt diammonium phosphate (DAP) can be helpful supplements. Yeasts also require other elements, such as phosphorus and sulfur (available from wort in the form of phosphates and sulfate), as well as a plethora of metal ions (most importantly, potassium, zinc, manganese, and iron). Yeasts are able to synthesize all of their required vitamins with the exception of biotin, although malt and grain worts contain sufficient levels of biotin to ensure reliable fermentations. In addition, ethanol is only produced during the anaerobic respiration.
The metabolic pathways for the production of ethanol are well described in many biochemistry texts and by Boulton and Quain (2006). Briefly, for the Embden-Meyerhof-Parnas (EMP) pathway, glucose and fructose are phosphorylated, whilst maltose is diphosphorylated during uptake before hydrolysis. In any case, glucose-6-phosphate is isomerised to fructose-6-phosphate. This is essential, as this latter can be split to yield two identical molecules of glyceraldehyde 3-phosphate, which is converted to pyruvate, recovering additional molecules of ATP. The pyruvate undergoes decarboxylation to yield CO2 and acetaldehyde, which in turn is reduced to ethanol by NADH. An alternative pathway, exemplified by homofermentative lactic acid bacteria, sees pyruvate reduction to lactate without the loss of carbon as CO2. A minor alternative pathway is that terminating in the formation of the non-volatile glycerol. In terms of flavor congeners, the EMP pathway yields mainly ethanol and acetaldehyde, but other compounds, such as higher alcohols and esters, are produced as by-products from other yeast biosynthetic pathways. While this discussion has focused on the EMP pathway, it is important to appreciate that other pathways can exist, especially under aerobic conditions at the beginning of fermentation.
The presence of free amino nitrogen (FAN) is essential for both yeast growth and for flavor development. The amount of FAN present is dependent both on mashing-in conditions and the degree of modification of the malt. Excessive FAN levels stimulate yeast growth to an extent that there are appreciable losses of carbon into cell growth rather than ethanol production. Yeast cells do not have active uptake mechanisms for all amino acids, and the venerable Jones and Pierce (1964) classification persists today as a useful guide to the uptake rates of amino acids by yeast. Aspartate/asparagine, glutamate/glutamine, and lysine do have dedicated transport enzymes. Arginine, serine and threonine share a single permease but all of these aforementioned amino acids, Group A in the Jones and Pierce classification are taken up at a sufficiently rapid rate to satisfy the requirements of the growing cell. The B group share a common permease and are slowly absorbed. Group C, relying on a general permease for transport, are only taken up late in fermentation, too late for significant cell growth. The final Group D are absorbed slowly, but at rates sufficient to satisfy the biosynthetic needs of the yeast cell. As there is a general deficiency of Groups B and C, the yeast cell must supplement these amino acids via additional biosynthetic pathways. In fact, lysine cannot be utilized for this purpose, so the yeast cell relies on glutamate/glutamine and aspartate/asparagine to fulfill this role. However, if the supply of amino nitrogen is depleted, the intermediate α-keto acids are readily decarboxylated to their corresponding aldehyde and then in turn reduced to the corresponding higher alcohol.
One of the most important flavor congeners formed during fermentation is diacetyl. This is formed by the chemical oxidation of secreted acetoin and 2,3-butanediol, which in turn are derived from acetolactate, an intermediate in the synthesis of the Group B amino acid valine. While the conversion rate to diacetyl is low yielding, it is of disproportionate importance both because of its low flavor threshold in spirit and the fact that its volatility is so similar to ethanol that it is not possible to distill diacetyl from ethanol completely in conventional distillery setups.
Esters are also important flavor congeners in spirit. They form a chemical equilibrium with their component carboxylic acids and alcohols, but during fermentation they are a by-product of coenzyme A (CoA-SH) recycling when the acyl CoA is produced but not required for lipid and protein synthesis. Ethanol and acetate are the most abundant alcohol and acid present during fermentation, so naturally ethyl acetate is the most abundant ester. However, it is not the most important in terms of flavor, as higher esters are more flavor-active and have a greater impact on the flavor attributes of the new-make spirit and the final whiskey.
Some of the most flavor-active organic compounds contain one or more sulfur atoms. In the context of whiskey production, the main contributors to the production of sulfur-derived flavor compounds are the biosynthesis of the sulfur amino acids methionine and cysteine, and the reduction of sulfate and sulfite in the wort. Unlike beer production, the whiskey industry has kept faith with copper in their stills, which can partially, but not completely, remove sulfur volatiles in the distilled spirit.
Finally, it is worth considering oxygen’s pivotal role during fermentation. Yeast cells require oxygen for the synthesis of unsaturated fatty acids and sterols that are essential for the development of cell membranes. This limits growth under anaerobic conditions unless the wort is supplemented with these components directly. However, because of a phenomenon known as the Crabtree effect, when there are appreciable sugar levels present, fermenting yeasts convert sugars anaerobically, while the yeast can use dissolved oxygen to grow even when aerobic conditions no longer prevail. Some work has suggested that oxygen helps to restore mitochondrial function and, in principle, such yeasts could perform satisfactorily even without wort oxygenation. Nevertheless, this is likely to influence the flavor profile of the final whiskey.
All of the above is predicated on the contribution of yeast. Distillery yeast is purchased from any number of suppliers. It is typically grown on molasses supplemented with ammonium salts to provide a nitrogen source. It is important to achieve growth aerobically, so the sugar content of the medium should be kept low to avoid the Crabtree effect. Propagation is continued until it is not possible to maintain aerobic conditions. The yeast is recovered by rotary vacuum filter and supplied in one of three forms: cream (18 percent dry weight), compressed yeast (24-30 percent dry weight) and, more recently, dried yeast (92-95 percent dry weight). The former two must be kept cool and used within three weeks, while dried yeast can be kept at ambient temperatures for up to two years. However, dried yeast requires careful rehydration to ensure that its viability is not unduly lost.
The distiller is often in an enviable position of being able to manipulate spirit composition during distillation. Higher alcohols and esters can, with opulent distillation resources, be stripped from spirit to an arbitrary degree. Nevertheless, this “cure rather than prevent” strategy can backfire, as in the case of diacetyl. In this regard, it is perhaps prudent to take a leaf out of the brewers’ playbook, where fermentation and post-fermentation “maturation” can not only manage higher alcohols and esters downwards, but also control diacetyl. Admittedly, this does extend the duration of brewery fermentations, typically four to seven days, compared with distillery fermentations that can ferment to dryness in as little as 44 hours.
Perhaps we need a sign over the distillery door: “Prevent or cure — our choice!”
____________________
Paul Hughes, Ph.D. is assistant professor of food science and technology at Oregon State University in Corvallis, Oregon. For more information visit www.oregonstate.edu or call (541) 737-4595.
____________________
FOOTNOTES
1 This is a hypothetical example for various reasons, including that the residual limit dextrins would potentially reduce the efficiency of the process.
2 This should always be the case for conventional yeast fermentations for alcoholic beverages
____________________
REFERENCES
Boulton, C. and Quain, D.E., Brewing Yeasts and Fermentation, Blackwell, Oxford, 2006.
Jones, M. and Pierce, J., Absorption of amino acids from wort by yeasts, J. Inst. Brew., 1964, 70, 307-315.
Nykanen, L. and Suomalainen, H., Aroma of Beer, Wine and Distilled Alcoholic Beverages, Riedel, Dordrecht, The Netherlands, 1983.