Fruit Flavours and their Relevance to the Flavour of the Final Distilled Beverage

Follow along: IG @birectifier

Lets go back in time 40 years and take a look at the most helpful chapter of the ultra rare text Flavour of Distilled Beverages edited by J.R. Piggott. The book is unique, but I only found three chapters practical enough to spend the time with. I’m ignoring the chapter on gin because it only dealt with early experimental sensory evaluation techniques that look like the flavor wheel meets a crude version of semantic odor space. There were bits in other chapters that have caught me eye. Bound aroma and specifically rose ketones were on their radar in the early 1980’s; we still struggle to acknowledge them today.

The fruit flavors chapter is the most exciting and deserves to be out there and acknowledged by anyone distilling fruit today. One quote from near the end is worth considering in the very beginning:

Wine distillates have sensorically nothing in common with fresh grape flavours. The grapes can be regarded as a collection of precursors, which are transformed to the flavour of the distillate during its production and aging.

Should we frame more substrates as a collection of precursors? can that flip Molasses from salvage product to investment product? What about certain precursor rich parts of the sugar cane termed “trash“? (Higher in some varietals than others!) And then for my current interests in Portugul, arbutus fruit? The fruit is bland, but when fermented it becomes extremely characterful and can be thought of as adjacent to tequila. Basic low risk fast ferments do not seem to unlock all the aroma, but the best heritage slower ferments add dramatic persistence and a sophisticated cedary character not expected from any fruit that comes off a tree. If we understand the precursors, we can transform them better.

It is challenging to think of common fruit as a collection of precursors because of all there apparent aroma and how we are primed to make object comparisons. Wine grape varietals are very abstract to consumers, but an apple or pear? Do you engineer a fruit brandy to best represent a fruit as the consumer knows it already or do you just process it to unlock pleasure points related to persistence and the raw extraordinary?

Do more with quince if you can! A big thank you to the incredible friend that lent me this text!

Chapter 2

Fruit Flavours and their Relevance to the Flavour of the Final Distilled Beverage

Peter Dürr and Hans Tanner
Swiss Federal Research Station for Fruit-Growing,
Viticulture and Horticulture, CH-8820 Wädenswil

ABSTRACT

Volatiles of fruit origin in a distilled beverage can be classified as flavour impact compounds, flavour contributing, non important and secondary volatiles from precursors in the fruit. The relevance of volatiles of apples, pears, quinces, cherries, plums and raspberries to the flavour of the distillates is discussed in view of fruit ripeness and mashing parameters like crushing and treatments with acids and enzymes. Some quality control parameters are mentioned.

INTRODUCTION

A first class raw material is necessary to produce a top class fruit distillate. But quantitatively most of the European distillers raw material is far from what we desire, arising from agricultural surplus or from negative selection of table fruits. Nevertheless distillers are interested in producing a distilled beverage typical of the fruit. Therefore the question arises as to how far original fruit flavour volatiles contribute or even determine the quality of the final distilled beverage.

Since capillary gas chromatography, mass spectrometry and sensory methods have become common tools in flavour research, this question has found considerable interest among several research group. Fruit volatiles are rather well known, but subject to changes curing fruit ripening, storage and processing. Clearly, the odour of fruit juice can be very different from that of the ripe fruit. The fermentation of a fruit mash or juice changes the odour completely. Yeasts produce a whole range of volatiles (1) and metabolize original fruit compounds. Strong changes in the quality of volatiles occur during distillation and aging. The volatiles from the final distillate can be classified as follows:

Distillate volatiles:

—fruit origin
—produced by yeasts, bacteria and enzymes
—formed during aging
—added (from barrel oak, essences, “bonificateurs”).

Fruit volatiles in a distillate:

—original flavour impact volatiles e.g. esters of decanoic acids in pear brandy
—original volatiles contributing to the flavour e.g. linalool in plum brandy
—non-important original volatiles e.g. α-terpineol in apple brandy
—secondary volatiles from fruit precursors e.g. benzaldehyde in cherry brandy

The following fruits are discussed in some details: apple, pear, quince, cherry, plum, apricot and
raspberry.

POME FRUITS

Apples

Together with grapes, apples are the most widely used raw material to produce brandy and drinking
alcohol for different purposes. Apples loose their specific flavour rapidly during processing and relatively neutral distillates are produced. The most famous apple brandy, calvados, gets its specific character mainly from aging in oak barrels. Apples have no strict odour impact compounds, their odour is formed essentially by aliphatic esters (2). Once apples are crushed, their odour is rapidly changed to that of an unspecific juice odour, which is characterized by C6-unsaturated aldehydes (3). The enzymatic formation of these compounds from linolic and linolenic acids is well documented (4). During fermentation of apple juice the C6-aldehydes undergo a further enzymatic hydration to the C6-alcohols (5). The resulting 1-Hexanol occurs in rather high concentrations in the distillate and contributes strongly to its flavour (6). Schreier et al. (5) investigated the changes in the composition of neutral volatile components during the production of apple brandy. Quantitative determination of nearly 100 volatile components has shown that the flavour of fresh apple distillate is formed by the components resulting from mashing, fermentation and distillation. Genuine apple flavour volatiles are degraded and can be detected only in traces after distillation.

Methanol

Methanol occurs in all fruit brandies except that from grapes accounting for up to 2 vol% of pure alcohol, which is in the range of toxicological effects. It is formed by esterase cleavage of pectin and is part of the distillate character accounting for the sharp, burning off-flavour. Methanol formation can be reduced by a heat treatment of the fruit immediately after mashing, which deactivates the enzymes. The removal of methanol from the distillate is possible but results in a loss of flavour volatiles (7).

Technology of volatiles

As apple distillates smell rather neutral with little fruity notes, several attempts have been made to improve their quality. If full to overripe Golden-Delicious apples are properly processed, some of the typical esters of pentyl alcohols can be found in the distillate, which improves its fruity odour (8). Another approach was the addition of stripped sweet juice volatiles to the distillate (9). Usually such an addition changes unfavourably the distillate odour and so it has not become common practice.

Another question is how different is an apple distillate produced from concentrated juice, with out the juice volatiles which have been removed during concentration, from a distillate produced from fresh juice? Apple juice essence is a valuable by-product and could be used otherwise. Recently, four Swiss distillers ran an industrial scale experiment with apples and pears as outlined in figure 1.

Figure 1. —Apple and pear distillates from different pretreated juices.

Two companies processed apples and two processed pears. The fruits were of normal processing quality. Twenty coded distillates were submitted to three laboratories for chemical analysis and to a trained sensory panel of 12 assessors. While the chemists could not find significant differences in relation to the technological variables, the sensory analysis gave some interesting indications. The panelists were asked to identify each sample as apple or pear brandy and to rank the samples according to the intensity of the fruity odour. Firstly, the skilled panelists were not able to identify the samples as apple or pear. But in all rank orders, the sample from the untreated juice was ranked most fruity, except in the case of the samples from concentrated pear juice without volatiles, which were ranked equally. Back-adding the volatiles to juice from concentrate did not improve the fruityness of the distillate. The fruityness is more dependent on the heat treatment of the juice than on the removal of the juice volatiles. This removal can be an advantage in the case of off-flavoured juices from low quality fruits.

Pears

Apples and most of the pears give a varietal unspecific distillate, but with pears, there is a famous and well known exception: the Williams or Bartlett pear with its flavour impact esters from the decadienoic acid (10).

ethyl trans-2-cis-4-decadienoate

These esters have a relatively high boiling point, are retained during fermentation and are responsible for the typical Williams pear odour of the distillate. During fermentation, part of these esters are isomerized to the non-typical smelling trans-2-trans 4-iso-mers. In order to get sufficient flavour active decadienoate esters, the pears should be processed when they are nearly overripe, getting soft. This requires a storage time up to four weeks after picking. The pears are gently mashed, acid is added to prevent bacterial infections, properly fermented and distilled without delay.

Pear distillates are also characterized by high amounts of 1-butanol and 1-hexanol. Both occur in small quantities in the fresh fruits. They are formed mainly during fermentation by hydrolization of the corresponding acetates (11). The addition of sweet juice essence improves the fruityness of a Williams pear distillate, but increases the danger of rancidity, as the unsaturated esters are known precursors for rancid smelling compounds.

Quinces

Quinces are well known for their characteristic and intense odour. Typical distillates can be produced. Ethyl-2-methyl-2-butenoate (12) and the so-called marmelo-lactones (13) have been described as strong contributors to quince odour.

Quinces have also to be fully ripe or slightly overripe at processing. They should be washed and brushed to remove the small hairs from the skin. The hairs contain an oil which is responsible for off-flavours of the final beverages (14). Quinces, as hard fruits, are somewhat difficult to mash, and breakage of the pips should be avoided. Quince pips contain a considerable amount of amygdalin, which is split into benzaldehyde, prussic acid and two molecules of glucose. Prussic acid is rather toxic and benzaldehyde affects the quince odour of the distillate.  The cleavage of amygdalin is a major parameter in the production of distillates from stone fruits.

STONE FRUITS

Cherries

Both types of cherries, sweet cherries (Prunus avium) and morello cherries (Prunus cerasus) are used for brandy production. The distillate from black sweet cherries, known as Kirsch, is famous from the international region around Basel: Black Forest, Alsace, central and northwestern Switzerland. Kirsch does not show the typical fruit odour. It has got its own character. The fermented fruit mash is left in contact with air for several weeks to achieve a limited amount of acetic acid, which subsequently forms large amounts of acetates. Therefore the main odourous components of Kirsch are low molecular aliphatic esters, mainly ethyl acetate, acids, benzaldehyde and benzylalcohol. TUTTAS and BEYE (15) studied the volatiles of sweet cherries and their occurrence in the distillate. They identified 83 volatiles in ripe cherries, none of them has a typical cherry odour. They refound 20 of them in corresponding concentrations in the distillate. Among these benzylacetate, ethylbenzoate, benzaldehyde and benzylalcohol are typical of stone fruits. Typical to cherries are cinnamyl alcohol, β-phenylethyl formate, benzyl formate and dodecyl acetate. The concentration of another 23 genuine components increases during fermentation.

Major odour components of Kirsch and other stone fruit distillates are benzaldehyde and benzyl alcohol. The sensory recognition threshold of benzaldeyde in stone fruit distillate is around 4 ppm (16). Table 1 presents a selection of data taken from several reports (16-20).

Table 1. — Concentration of amygdalin fragments in stone fruit beverages

The legal upper limit of prussic acid in ready to drink cherry distillates is 40 ppm (Codex Alimentarius). The dosis letalis of prussic acid is 70 mg, whereas for ethanol it is 200 ml. Therefore the toxic factor of stone fruit distillates is ethanol, not prussic acid (8).

When processing cherries, benzaldehyde and prussic acid are already released from the intact stones during mashing and fermentation, whereas with plums and apricots it only comes from crushed stones. As benzaldehyde is less stable than prussic acid, its content in the final distillate depends on technological parameters, as indicated in Table 2 (15). Benzaldehyde is metabolized during fermentation to benzyl alcohol and benzyl acetate (20,21).

Table 2 – Benzyl components in fermented cherry juice and mash (15)

The content of prussic acid and benzvl components increases also by an excessive acidification with mineral acids, enzyme treatment or storage time of the fruit mash. At mashing, not more than 5% of the stones should be broken.

Plums

Plums, Prunus domestica, subsp. domestica, form a large group of cultivars, differing in colour, shape and taste, but with a rather unique, typical odour. Plum brandy is well known from many regions, but mainly as Sliwowitz from Eastern Europe. When plum brandies are not too much alterated by aging, they show an intense flavour reminiscent of the fresh fruit. The relation between fresh plum flavour, juice flavour and distillate flavour has been comprehensively investigated by ISMAIL et al. (17,18,22). They identified GC-peaks from 2 effluent regions of the chromatogram with a characteristic plum-like odour. In all three flavours they found benzaldehyde and linalool in the first region and γ-decalactone and methyl cinnamate in the second region of the chromatogramm. They concluded that the plum-like comments associated with the two regions of the gas- chromatogramm are the result of a fortuitous co-elution of compounds, benzaldehyde with an almondy odour and linalool with a fragrant, spicy odour, which together create the plum-like character (18). The removal of linalool was associated with the disappearance of the plum-like odour (17). They also found, that fermented plum juice has a stronger plum-like odour than fresh juice (18). The neutral volatile components of Czechoslovak plum brandy have been analyzed by VELISEK et al. (23). In addition to the fruit volatiles mentioned by ISMAIL (22), they describe some formates and propionates possessing plum-like odours and many contributing factors formed during fermentation. Some conclusions relating to mashing technology and the fermentation of plums can be drawn. Late ripening plum varieties are preferred. The fruits are best when soft, nearly overripe, with maximum sugar and flavour content. The fruits should not be milled, the stones left intact or removed. Otherwise too many components from the cleavage of amygdalin and unsaturated fatty acids affect the fruityness of the distillate. Soft fruits are simply mashed by their own weight or with little agitation. Pectinase helps to disintegrate the fruit tissue. Fermentation and distillation should be performed with out delay.

Apricots

As only fully ripe table fruits are suitable for the production of a fruity distillate, this fruit is not widely used for these distillates in Western Europe.

In eastern Europe some famous distillates are produced, namely the Barazk of Hungary. The volatile flavour components of apricots have been investigated by TANG and JENNINGS (24) and CHAIROTE et al. (25). Their work shows that apricot odour is dependent the correct balance of several components.

Table 3 shows the odour quality and the main components of several silica gel fractions of an apricot extract (25). Mixing these fractions gave a product possessing a typical apricot odour.

Table 3 – Fractions collected after silica gel chromatography, pooled according to their odour (25)

Practical experience shows that the type of raw material, mashing, fermentation and distillation has to follow the same rules as with plums. Mashes from overripe fruits, having a high pH, are subject to bacterial infections and formation of taints from butyric acid or acrolein. For prevention an acid treatment or a rapid processing of the mash is
required.

Raspberries

Raspberry is one of the few fruits containing a real flavour impact compound: 1-(4-hydroxy phenyl)-3- butanone, called raspberry ketone.

This raspberry ketone, together with the two ionone-isomers, essentially contribute to the raspberry (1) odour. The concentration of raspberry ketone has been found to be three times higher in wild berries than in cultivated berries (26). Distillates from fermented raspberries show the typical odour, but due to the low sugar content of the berries, most food regulations allow the production of distillate from alcoholic extracts of fresh berries (27). If the berries are left too long in contact with ethanol too many of the higher fatty acids, namely palmitic, linolic and linolenic acid are extracted from the pips and esterified to acetates, which affect the raspberry odour. A limited amount of these esters seems to improve the odour intensity of distillates from unfermented raspberries (28). Fruit and distillate volatiles have been listed by POSTEL & ADAM (35).

Grapes

Grapes for wine production together with apples are the most widely used fruits in the production of drinking alcohol. With a few exceptions, by proper processing of extremely flavourful grapes, already the wine shows an outstanding character. Wine distillates have sensorically nothing in common with fresh grape flavours. The grapes can be regarded as a collection of precursors, which are transformed to the flavour of the distillate during its production and aging. If the husks of grapes are further processed to a distillate, known as Marc, they should not be left in contact with air and be used without delay to prevent the formation of methanol and acetaldehyde.

DISCUSSION

The question arises as to how far the present knowledge on fruit-borne flavour components of distillates can be used for quality control, for instance to check the identity and possible alteration or adulteration of a distilled beverage. The analyst is faced with a product, where many processing parameters namely fermentation, distillation and aging, specify the composition of the beverage. Therefore, the use of fermentation alcohols and esters as quality factors, as proposed by several workers (19, 29-34), has not been very successful. Cherry distillates contain a considerable amount of 1-propanol (8,19). 1-Butanol is a typical component of pome fruit distillates and occurs clearly in lower concentrations in stone fruit distillates (11,31,32). The natural content of benzaldehyde cannot be higher than the corresponding concentration of prussic acid (16). But with modern flavour analysis, capillary gas chromatography and mass spectrometry, it is possible to rely more on fruit-typical volatiles such as decadienoates from pears, marmelo lactone from quinces, cinnamyl alcohol from cherries, methyl cinnamate from plums and 4-hydro-xypheny 1-2-butan one from raspberries.

References

1. Suomalainen, H.; Lehtonen, M.; The production of aroma compounds by yeast, J.Inst.Brew. 1979, 85, 149-156

2. Paillard, N.; Comparaison de l’arôme de differentes variétés de pomme: relation entre les differences d’impression olfactives et les aroma grammes, Lebensm.Wiss. Technol. 1975, 8, 34-37

3. Dürr, P.; Schobinger, U.; The contribution of some volatiles to the sensory quality of apple and orange juice odour, in Schreier, P., ed., Flavour ’81, Walter de Gruyter, Berlin 1981

4. Drawert, F.; Heimann, W.; Emberger, R., Tress1‚R.; Enzymatische Bildung von Hexen-2-a1-1, Hexanal und deren Vorstufen, Justus Liebigs Ann. Chemie 1966, 994, 200

5. Schreier, P.; Drawert, F. Schmid, M.; Changes in the composition of neutral volatile components during the production of apple brandy, J.Sci. Food Agric. 1978, 29, 728-736

6. Pfannhauser, W.; Eberhardt, R.; Aromastoffe in Williamsbirnen- und Apfelbranntweinen, Alkohol-Industrie 1979, 104-107

7. Tanner, H.; Brunner, H.U.; Obstbrennerei heute, Verlag Heller Chemie- und Verwaltungsgesellschaft, Schwäbisch Hall 1982

8. Woidich, H.; Pfannhauser, W.; Eberhardt, R.; Ergebnisse von gaschromatographischen Untersuchungen der flüchtigen Inhaltsstoffe von Apfelbränden, Mitt. Klosterneuburg 1978, 28, 56-63

9. Neuartiges Verfahren zur Herstellung von Alkohol mit rohstofftypischem Aroma, Branntweinwirtschaft 1970, 110, 502

10. Jennings, W.G.; Creveling, R.K.; Heinz, D.E.; Volatile esters of Bartlett pear. IV. Esters of trans-2-cis-4-decadienoic acid. J.Food Sci. 1964, 29, 730

11. Bricout, J.; Sur les constituents aromatiques de l’eau de vie de poires, Ind.alim.agric. 1977, 94, 277-281

12. Schreyen, L.; Dirinck, P.; Sandra, P.; Schamp, N.; Flavor analysis of quince, J. Agric. Food Chem. 1979, 27, 872-876

13. Tsuneya, T.; Ishihara, M.; Shiota, H.; Shiga, M.; Isolation and identification of novel terpene lactones from quince fruit, Agric.Biol. Chem 1980, 44, 957-958

14. Schobinger, U.; Dürr, P.; Aeppli, A.; Quittensaft – eine interessante Möglichkeit zur Diversifikation des Getränkesortiments, Flüss. Obst 1982, 49, 10-14

15. Tuttas, R.; Beye, F.; Die genuinen Aromastoffe von Cerasus avium und ihr Schicksal von der Frucht bis zum Destillat, Branntweinwirtschaft 1977, 117, 349-355

16. Bandion, F.; Valenta, M.; Kain, W.; Zur Beurteilung des Benzaldehydgehaltes in Steinobst-Edelbranntweinen und Steinobst-Fruchtlikören, Mitt. Klosterneuburg 1976, 26, 131-138

17. Ismail, H.M.; Williams, A.A.; Tucknott, O.G.; The flavour of plums (Prunus domestica L.). An examination of the aroma components of plum juice from the cultivar Victoria, J.Sci.Food Agric. 1981, 32, 613-619

18. Ismail, H.M.; Williams, A.A.; Tucknott, O.G.; The flavour components of plum, an examination of the aroma components present in a distillate obtained from fermented plum juice, Z.Lebensm. Unters.Forsch. 1980, 171, 24-27

19. Beaud, P.; Ramuz, A.; Composition générale des eaux-de-vie de griottes: Comparaison avec les kirsch, Mitt.Gebiete Lebensm. Chem.Hyg. 1978, 69, 536-543

20. Albagnac, G.; Vangheesdaele, G.; Etudes preliminaires sur l’origine de l’alcool benzylique present dans les kirschs, Ind.alim.agric. 1975 92, 1271-1274

21. Tuttas, R.; Nahrstedt, A.; Bedeutung der Cyanglykoside fur den Benzylalkoholgehalt in Kirschbranden, Z. Lebensm.Unters.Forsch. 1977, 193, 257-259

22. Ismail, H.M.; Williams, A.A.; Tucknott, O.G.; The flavour components of plums: an examination of the aroma components present in the headspace above four cultivars of intact plums, Marjorie’s Seedling, Merton Gem, NA 10 and Victoria, J.Sci.Food Agric. 1981, 32, 498-502

23. Velisek, J.; Pudil, F.; Davidek, J.; Kubelka, V.; The neutral volatile components of czechoslovak plum brandy, Z.Lebensm.Unters.Forsch. 1982, 174, 463-466

24. Tang, C.S.; Jennings, W.G.; Volatile components of apricot, J.Agr .Food Chem. 1967, 15, 24-28

25. Chairote, G.; Rodriguez, F.; Crouzet, J.; Characterization of additional volatile flavor components of apricot, J.Food Sci. 1981, 46, 1898-1906

26. Honkanen, E.; Pyysalo, T.; Hirvi, T.; The aroma of finnish wild raspberries, Rubus idaeus L., Z.Lebensm.Unters .Forsch. 1980, 171, 180-182

27. Pieper, H.J.; Bruchmann, E.; Kolb, E.; Technologie der Obstbrennerei, Verlag Eugen Ulmer, Stuttgart 1977

28. Schöne, H.J.; Sparrer, D.; Zur Untersuchung von Himbeergeist, Alkohol-Industrie 1975, 16, 372-375

29. Nosko, S.; Zur Beurteilung von Williams-Christ-Branntweinen, I. Analysendaten authentischer Branntweine von Williams-Christ- und anderen Birnensorten sowie deren Verschnitten, Deutsche Lebensm.Rundschau 1974, 70, 397-400

30. Nosko, S.; Zur Beurteilung von Williams-Christ-Branntweinen, II. Analysendaten und Beurteilung von Handels proben, Deutsche Lebensm.Rundschau 1974, 70, 442-447

31. Reinhard, C.; Beitrag zur Untersuchung und Beurteilung von Obstbranntweinen, Deutsche Lebensm. Rundschau 1978, 74, 299-301

32. Hildenbrand, K.; Beitrag zur Beurteilung von Obstbranntweinen, Zwetschgenwässer aus dem Handel, Branntweinwirtschaft 1982, 122, 2-8

33. Wencker, D.; Louis, M.; Laugel, P.; Hasselmann, M.; Versuche zur Identifizierung von Obstbranntweinen auf Grund ihres Gehalts an höheren Alkoholen und an Methanol, Deutsche Lebensm.Rundschau 1981, 77, 237-238

34. Pieper, H.J.; Röhrig, G.; Zur Zusammensetzung flüchtiger Komponenten sowie sensorischen Charakterisierung von Brombeergeisten unter besonderer Berücksichtigung der Extraktionsbedingungen, Deutsche Lebensm. Rundschau 1982, 78, 356-364

35. Postel, W., Adam, L., Chromatographische Charakterisierung von Himbeergeist, Deutsche Lebensm. Rundschau 1983, 79, 117-122