First a great thank you to Susan Coppin and the team of archivists at the University of Adelaide who made recovering these documents possible.
All these documents from Australia were contained in Series 600 of the Roseworthy Agricultural College archives. We should probably figure out how to digitize all of the papers.
Jamieson probably wrote the most thorough paper about brandy production weighing in at 51 pages. I re-typed large swaths of the paper so that it would be better indexed by search engines.
“Generally, pressings, leachings and wine from grapes unsuitable for brandy production are distilled at high proof (160-190% pf.) and is used for fortifying sweet wine and brandy is distilled at 135 – 140% pf. from the wine of selected grapes. The Australian Pure Foods Regulations specify that brandy must be the distillate from fresh fermented grapes.”
“… Some of these problems are bound up with the economics of production as the manufacture of cognac entails the use of equipment that is too slow and cumbersome for Australian methods of production. Many of these problems have been studied at Roseworthy Agricultural College from the angle of producing brandies more closely resembling cognac and it is the purpose of this project to include a re-survey of the work accomplished.” So if you can only read one Roseworthy paper, read this one!
“If we are to develop and expand our export markets we will have to modify our distillation procedure and methods of manufacture to comply with this existing demand. South Africa our sister dominion, and river on the overseas market with wines and brandies, is making strenuous efforts to capture this market.”
“The Govt. Brandy Board which has to pass all wines before distillation has lately declared itself in favour of a Cognac type brandy for South Africa. This will mean the exclusion of the two most prolific bearers of grapes, and from the varieties specified as being suitable for distillation for the production of beverage brandies.” It is really interesting how this stuff is regulated.
“In the eusuing [typo? ensuing?] controversy Professor C.J. Theron and Dr. C.J.G. Neehaus of the Shellenbosh-Elsenburg College of Agriculture in supporting the Govt. Brandy Board’s as to the necessity for conforming to this demand concisely summed up the situation thus.
(a) There is a great demand for certain types of South Africa brandies overseas and these resemble Cognacs. South Africa is in need of this demand and should foster it.
(b) Cognac type of brandy derives most benefit from maturation and the future of the South African wine industry lies in the production of well matured brandies. (4)”
Interesting names are introduced here and a new University to look into. No doubt these players carried out similar research to Roseworthy that we can probably track down.
“A French decree (May 16th 1936) limits the varieties which may be grown in the Charente for Cognac and these are Saint Emilion, Colombard, Folle blanche, Blanc romay, Montellis White Suavignon and Jurnac blanc. (13).. The first three varieties are the main ones and the Folle blanc which is considered the best for Cognac often attains 1.2% acid and only 7-8% alcohol by volume. Saint Emillion generally 709% alcohol by volume and Colombard 8-11%. (3)”
“In some years, during the harvesting period, many of the grapes are damaged by rain and mould and such damaged bunches are scrupulously rejected. The grapes are crushed on the day they are picked and the juice separated from the skins immediately. Fermentation of the free run juice is carried out in small, closed containers (up to 500 gals capacity).”
“No S.O.2 is used during fermentation as it is found that any wine fermented in the presence of considerable quantities of S.O.2 will contain a considerable quantity of aldehydes, even though the wine iteself may have only negligible S.O.2 content at the time of distillation. This is due to the formation of an aldehyde bi sulphite complex. The quantities of S.O.2 used to prevent distillation wines from becoming diseased will always result in too much aldehyde production. (4)”
“Fermentation is carried out at low temperature and the ‘silent’ fermentation takes about 3-4 weeks, the wine then containing about 8% alcohol by volume and up to 12 gms/litre of acid as tartaric. (13)”
“When the wine has fermented out dry it is left on lees but distilled as soon as possible to reduce risk of wine spoilage. Wines left too long on gross lees develop a typical stale lees flavour which may be imparted to the spirit. If the wine is racked from its lees the quality of the brandy suffers as distillation of the wines with large numbers of yeast cells is one of the essentials for imparting the Cognac character. (4) Just before charging the still wine and lees are agitated to give a uniform suspension.”
“The procedure is not in any way standardized and distillation practices are passed on from father to son. The distiller is called upon to show great judgement in the selection of wines for Cognac; and in deciding when to take off ‘heads’ and ‘tails’ from the wines made from the different varieties and grown in different areas. As the pot stills are direct fired great care must be taken in controlling the heat of the fire to avoid scorching of the contents.”
“The wine is fed into the cold still, the fire lit and distillation carried out at a very slow rate. It is essential that the pot be never any more than 75% filled with wine.”
“To minimize risk of spoilage the whole of the distillation wine for the season is sometimes put through as low wines before commencing the redistillings. The best brandy, however, is obtained by redistilling immediately and where there is only a single still in operation this is done when three charges of low wines have been collected, and the joint distillate is then redistilled. (4) The citation here is for the South African paper I haven’t yet come across and this information on not waiting to re-distill is something I’ve never seen before. I suspect the differences before waiting or not is due to the distillate approaching equilibrium. The fractions will be different due to the spirit evolving in terms of equilibriums like those of ethyl-acetate and acetaldehyde. Apparently the most extraordinary results are obtained when distilling as close to the previous distillation as possible. What I wonder is if a reference to this ideas can be found in whiskey or rum distillation.
“Sometimes the ‘tails’ are fractioned into a portion running from 50 – 20% alcohol by volume and this is mixed with the ‘broullis’ and the portion running from 20 – 0% alcohol by volume is mixed with the distillation wine.” This is interesting. I wonder if the very last fraction here contains any compounds related to “Cognac oil” that is supposedly produced by steam distilling the remnants. This fraction could also be higher in fatty acids which might benefit form the longer time under heat and the acids catalysts of the first distillation rather than the second. I haven’t seen this advice anywhere else. If this advice is also South African they must have produced some really interesting literature.
“Brandy distilled thus is totally unsuitable for immediate consumption while young, but has to mature for a number of years whereby its secondary constituents interact to remove the disagreeable, burning taste and to impart mellowness and palatability. This maturation can only take place where the wood of the container can contribute certain qualities to the brandy.” The wood ends up contributing a lot of acidity which influences the various equilibriums.
“The calcareous soils of the Charente are said to impart a delicate and characteristic bouquet to the Cognac. The exact explanation of why these soils give the grapes their desirable characteristics is not known but in the wine the delicate bouquet may be due to a pH effect.” I suspect the way the soil and climate stress the vines produces an abundance of diverse fatty acids while having a very high total acidity and low alcohol. Low alcohol creates a high ratio of aroma to alcohol as well as being correlated to high TA.
“Jocelyn and Amerine (1) suggest that red grapes contain larger amounts of the undesirable substances and if fermented on skins greater quantities of pigments and tannins are extracted. Possibly these substances are hydrolysed. The above workers also consider that the bloom of the grapes on hydrolysis yields a variety of fatty acids and other hydrocarbons and thus fermentation on skins of both red and white varieties will produce more of the hydrolytic products. More likely, as pointed out by Kuchel (12), contain a greater quantity of volatile constituents in the skin.” I don’t completely understand what he means by bloom here unless it simply is the skins. When you’d think you don’t really need to worry about compounds that aren’t volatile i.e. tannins or pectin the chemistry always gets a little deeper and what you thought wouldn’t be volatile gets subdivided into parts which may be volatile. Pectin gets hydrolized into methanol which needs to be looked out for. Amerine references this “bloom” on page 16 of Commercial Production of Brandies and has a better citation for Theron’s South African literature on the bottom of page 17. “Theron, D. de Wet. South African brandy. Wine and Spirit 11(97):4399, 4401. 1939.”
“Varietal aromas have a great influence on the character o brandies and this is generally considered undesirable in better class brandies.” This is a really interesting statement. In whiskeys we look for the character of specific grains but not in grapes? But this statement might just refer to varietals that have a foxy character like those of vitis lambrusco or some of the muscats.
“Grapes with a high sugar content are to be avoided for the production of quality brandy. Of two wines with the same percentage composition of volatile constituents, the wine with the lower alcohol content will yield a brandy having a greater percentage of volatile constituents if both are distilled at the same proof. (1). The ratios of aroma to alcohol. The mosto-verde technique used in Peru is a way to skew it further.
“This may be due to the fact that the weaker the alcoholic solution the higher the temperature at which is boils and the more of the less volatile constituents pass over in the vapour. If the wine is of comparatively high alcoholic strength the higher boiling point substances remain behind in the pot.” Here Jamieson gets a little confusing with his speculation. The strength of the alcoholic solution in the pot is also effected by run-back which is the recycling of previous fractions. Higher proof distilling material produces higher proof distillates and as proof increases congeners get more bunched up and less evenly distributed. This make the cuts more of an all or none proposition. What people didn’t understand back then was that aroma precursors would change significantly as ripening happened. Grapes in cooler climates keep more noble fatty acids as they ripen relative to warm climates.
“The grapes that are high in total acids are superior to those containing lesser amounts. They should be picked when mature – fully ripe or over ripe grapes do not give the same quality brandy. Wines low in total acids and high in pH will give less aroma and have a more undesirable aroma, if distilled at the same proof, than those brandies made from wines high in total acid and low in pH. Here “mature” needs to be elaborated. It could probably be stated as the point when favorable aroma precursors are at their maximum. This will be at a point when total acidity is still very high and potential alcohol is not maximized.
“The subsequent wine made form grapes with the desirable features will be thin and very acid with no special qualities.” It takes a lot of foresight to wrap hour head around this.
“The grapes are fermented “off skins” either by medium of a foredrainer or by fitting a false bottom for immediate drain off. It is imperative that no S.O.2 be added and should only be used as a last resource and then the wine should be distilled at high proof and not used in the production of beverage brandy. Mouldy grapes will impart their taint to the wine and thus to the spirit. Where S.O.2 is used the risk of mercaptan formation is veyr great and apart from the danger of these evil smelling compounds being produced, the aldehydes formed during fermentation are fixed in the wine by the sulphur and released during distillation increasing the aldehyde content of the spirit.”
“Wines fermented at high temperatures show less retention of the more desirable volatile products and in addition may give a taint to the spirit. (1) Under Australian conditions where the grapes arrive at the crusher hot and are fermented in large vats it is impossible to keep the temperature down to a reasonable degree without the use of coolers, particularly if no S.O.2 is used.”
“In order to bring the acid content of the wine to a level favouring maximum ester production it is advantageous to make acid additions to the wine prior to distillation.” I think this advice comes only from the other Roseworthy projects.
“It was also found that by putting through the distillation wine as a low wine and incorporating the ‘heads’ and ‘tails’ from the brandy run into the next low wine charge caused an increase in esters. This was particularly noticeable at pH 3. Acid catalyzed esterification and the recycling of aroma precursors.
Jamieson goes on to summarize more of the Roseworthy projects and explains in some circumstances the conflicting results. He spends a nice amount of time summarizing the sodium flouride as a sulphur alternative project done by Shipster which was a paper I didn’t request from the archives.
“Comparisons of brandies from the different stills are practically meaningless as methods of distillation and wine type used are not known when the finished brandy is presented from analysis at the laboratory.”
“Graham pointed out that slow distillation gives a more selective separation of the secondary constituents their distribution being governed by (a) their respective volatilities, (b) their respective miscibilities in water/alcohol mixtures of different strengths. Also that slow distillation accounts for the nature or composition of these secondary constituents as well as their amounts.” Graham ends up demonstrating how a slower distillation and thus a longer time under heat produces more extraordinary results.
Jamieson’s table 5 reproduction of Grahams work is a must see. I first heard of Graham’s work through Amerine and then finally inter library loaned the summary of his thesis from the Australian Brewing Journal. I was really motivated to track down the original thesis to see if there was any great data to look at. It is really a great work to help out beginning distillers.
“If early heads are undesirable in brandy and later heads desirable then the effect of lees is to give a clearer and cleaner separation between the two heads thus more of the heads could be included in the brandy.
The effect of lees summed up by Angove is to cause a marked increase in the ester and acid content of the low wines, and, in the brandy run, a stronger tendency for these constituents to pass into the heads.”
“In his literature review Angove (7) outlines the following procedures that may take place.
(a) The heads and tails may be run back to the next wine charge.
(b) The heads returned to the next brouillis charge and the tails to the next wine charge, or,
(c) High strength tails may be divided into two, the highest strength portion going to the next brouillis charge and the lower strength going to the wine charge; finally,
(d) The heads and tails may be bulked and redistilled and the heads and tails from this going back to the wine charge.” Apparently there are a lot of options and we should probably evaluate each option in relation to recycling aroma precursors, differences in time under heat, and availability of an acid catalyst. I suspect some options might maximize ester content while others not. You probably do not always want to maximize ester content. Probably only from the greatest source material where the highest amount of noble precursors are found.
“Aldehydes. Heath (22) found that there was a slight cleansing of all fraction down to brandy due to the effect of run back and that the extra aldehyde tended to collect in the tails.
This is contrary to the Angove’s results where it was found that aldehydes accumulated in the heads and this was accentuated by run back.
Tails were almost free of aldehyde and they were not effected by run back.” So there is some disagreement here. I suspect we can look to Robert Léauté’s James Guymon lecture for the definitive explanation.
“The chemical changes are due to oxidation alcoholysis and esterification. Bellet (26) found that the total ester content remained essentially constant but the more volatile esters rose very slowly at the expense of the less volatile esters of the higher alcohols. The consequent liberation of higher alcohols is termed ‘alcoholysis’. Some of the higher alcohols released are converted to aldehydes. The secondary importance of oxidation, and the relatively greater importance of alcoholysis gives reasons why the rapid oxidation of new brandies by the introduction of oxygen or ozone under pressure has not been successful.” This is all in regards to maturation.
Here Jamieson quotes a very large passage of Amerine:
“…The proportions of the various higher alcohols present in brandy is not constant due to differences in the composition of the raw material, to the use of various types of yeasts and of degrees of purity of the strain, to the extremely variable fermentation conditions, to the methods of distillation and to various other factors. As far as brandy is concerned, comparative analysis of spirits produced from raw materials of known composition, with a uniform method of fermentation and distillation would do much to elucidate the problem of the relative proportions of the higher alcohols present, and the factors which influence their formation.
Yeasts will preferentially assimilate inorganic nitrogen compounds such as ammonium phosphate and ammonium sulphate and the production of higher alcohols is reduced” [end Amerine]
“The addition of an organic source of nitrogen should give a heavier bodied brandy.”
“The tails or aftershots contain the major portion of the higher alcohols and consequently they will come over towards the end of distillation.
Iso propyl alcohol and tertiary butyl alcohol are readily soluble in ethyl alcohol and will come over during the brandy run. In a very weak alcoholic solution some of the higher boiling point higher alcohols will also tend to come over early because they are relatively insoluble in water compared to ethyl alcohol and do not suffer a reduction of vapour tension due to water admixture as does alcohol.” Amerine explains some of this well as does Robert Léauté
“The secondary constituents in brandy are never objectionable when in a pure state but their reactions with acids constitutes esters of flavouring substances which add considerably to the fullness and aroma of brandy.” I don’t really understand the logic of this statement. So higher alcohols aren’t objectionable when alone nor are they objectionable when they form esters because they contribute to “fullness” so don’t fear higher alcohols? I do remember reading that column stills have a tendency to produce more esters of higher alcohols of which some are considered big flaws. One higher alcohol ester associated with column still brandy production has a highly attention, plebeian, and ordinary aroma of bananas.
Jamieson finally starts his experiment on page 41.
“By the addition of inorganic and organic sources of nitrogen to the must before fermentation it was hoped to ascertain the effect on the composition of the resultant distillate.” Interesting!
“Amounts of ammonium phosphate, ammonium sulphate and peptone containing the same percentage of nitrogen were added to the grape juice and this was seeded with 2% yeast culture and allowed to ferment.”
Jamieson has a simple table that displays his results.
“There appears to be no significant difference between the fermentation rates but this may be due to the fact that they were all fermented at low temperatures. Perhaps ammonium sulphate gave a slightly more efficient alcohol production pH.”
“Ammonium sulphate gave a big reduction in pH, while peptone raised the pH slightly above the control. Ammonium sulphate is the salt of a weak base and a strong acid and on hydrolysis there would be a slight reversion to the strong acid thus lowering pH. Peptone on being decomposed by the yeast would yield amino acids and finally higher alcohols and ammonia. Ammonia on being released would raise the pH slightly.”
“Peptone has given a decided increase in ester content in the wine over the controls. Throughout the whole series where there has been a rise in the volatile acid content, there has also been in a rise in the ester content.” This regards the wines but who knows if these were noble esters or not. Maybe Jamieson will provide organoleptic analysis.
Jamieson goes on to detail some of the limitations of his distillation apparatus that might skew the results.
Alcohol – There are considerable differences in the strength of alcohol which is not entirely due to the original strength of the wine. Probably rate of distillation played some part in the variance of the strength of the distillate.
Titratable Acidity – This represents volatile acidity for the most part as there would be little or no entrainment of fixed acid when using a rectifying column. There appears to be a decrease in volatile acids in all samples when compared with the controls – the wines fermented with ammonium phosphate giving the least volatile acidity.
Esters – There is no significant difference in the ester content of any of the distillates. This is rather surprising in view of the increased ester content of some of the samples of wine, particularly those fermented with ammonium phosphate.
Aldehydes – Again no significant differences except perhaps those fermented on less (see later)
Higher Alcohols – Not determined as method unsatisfactory.”
Now Jamieson goes on to analyze everything in regards to lees distillation:
“Distillation On Lees – Represented by C – the third jar in the triplicate of each experiment.
Titratable Acidity – No significant differences. With the exception of the control those distilled on lees may have given slightly less volatile acidity than those distilled with wine only.
Esters – No significant differences.
Aldehydes – With the exception of the control all wines distilled on lees gave a considerably lower aldehyde content than those distilled with wine only.
DIFFERENCE DUE TO ESTER CONTENT IN THE WINE AND IN THE DISTILLATE.
It is difficult to give a reasonable explanation for this difference. A possibility may be that the esters were high boiling point esters which would come over in the ordinary distillation apparatus fitting with a splash head used in the analysis of ester content of the wines, but with a still fitted with a reflux column would remain behind in the DISTILLATE.” I capitalized this last term because it is a typo and crossed out in the text but not corrected. I think he means left in the pot. Jamieson is getting into the deeper end of chemistry and analysis here and I don’t really know how to clarify anything. Not all esters are created equal and some are noble and some are not. Back in the day their analysis was a lot more simplified and they didn’t have the ability to sub divide esters into the specific types.
Jamieson mentions an appendix to the project where he will discuss more results but it isn’t included in the PDF. Jamieson had problems with his higher alcohol analysis which really limited his work. It would be really fun to perform some of these very basic experiments again with modern analysis to help beginning distillers.
Unique references from the bibliography:
“4. C.J. Theron & C.J.G. Niehaus “Wine-making” Farmers’ Bulletin, Series No.130.”
“6. F.C. Lloyd “The art & technique of wine””
“13. R.A.C. Distillation Notes”
“17. Francois De Castella. “The Grapes of South Australia Reports to the Phylloxera Board.”