Interview

The Science of Flavour

Harry Riffkin is the head of Tatlock & Thomson, a Scottish consultancy which has offered scientific services to the wine and spirits industries since 1891
By Gavin D. Smith
The advent of lab testing transformed the whisky industry into what it is today.
The advent of lab testing transformed the whisky industry into what it is today.
When opening a new bottle of whisky, one focuses on the aroma and flavour of the spirit without having to worry that it might contain anything untoward. However, this was not always the case. During the 19th century, adulteration of food and drink was rife, as evidenced by the Glasgow Whisky Scandal of 1872. The editor of the North British Daily Mail obtained 30 samples of whisky from public houses in Glasgow and had them subjected to analysis. Only two were found to be genuine ‘whisky,’ while the others were either greatly diluted with water or, more worryingly, methylated spirits, turpentine, furniture polish and even sulphuric acid.

These practices occurred despite the existence of the Adulteration Act (1860) and, ultimately, such exposés led to the appointment of public analysts across Britain. In Glasgow, leading analytical chemist Robert Rattray Tatlock and his nephew, Robert Thomson, established a business as food and drink analysts in Bath Street during 1891, going on to be appointed public analysts for the city. The partners were key witnesses in the 1908 Royal Commission which determined the first legal definition of Scotch whisky. The firm remains in business, safeguarding the integrity of spirits and assisting distillers in optimising the quality of their products. Now based on a farm site at The Teuchats near Leven in Fife, Tatlock & Thomson is headed up by Dr Harry Riffkin, who purchased the business with the late Dr Jim Swan in 1993.

Dr Harry Riffkin


Riffkin studied at Edinburgh’s Heriot-Watt University, before undertaking a PhD at the University of Edinburgh and subsequently being recruited by Pentlands Scotch Whisky Research Ltd (now the Scotch Whisky Research Institute) as head of distillation studies in 1985.

Looking back on his time at Pentlands, Riffkin notes that one of the key issues of the day was spirit quality. There was a perceived loss of desirable ester-y notes in new-make spirit and a problem of potential toxicology due to ethyl carbonate, a nitrogen compound.

He explains, “The Canadians had noticed high levels [of ethyl carbonate] in US Bourbon. This was not a major problem for Scotch itself, but because we were using ex-Bourbon barrels it became an issue – there is up to 18 litres of ‘in-drink’ left in a cask so there was potential for contamination.”

However, the loss of ester-derived aromas was something that couldn’t be ascribed to American barrels. By the early 1990s, research had begun to indicate that the new mashing systems installed to deliver faster turnaround were allowing malt to be ground finer, which increased turnaround time and yield. It soon become clear that ester formation was prevented by cloudy wort, which was caused by rapid turnaround. The clearer the wort, the greater the presence of esters.

“We also found that difficulty controlling wash distillation could deliver high levels of ethyl carbonate in the low wines. This was noticed in what we might term ‘class 1’ distilleries. Clear wort is more difficult to control than cloudy wort during distillation,” explains Riffkin. “Modern distilleries with cloudy wort were producing spirit that was not as good quality as that of the 1960s and 70s. Distilleries with old mash tuns that were operated very slowly, such as Bruichladdich, Deanston and Glenfiddich, produced clear wort, which was then distilled very carefully.”

By the late 1990s, it was understood that if the industry went back to creating clear wort then spirit quality could improve. This was contrary to the belief expressed by some at the time that the generational change in quality had come about as a result of moving away from using brewers’ yeast. Armed with this new understanding, distilleries went back to slowing down mashing and using a coarser grind. “Once we had addressed controlling wash distillation by ensuring properly balanced distillations, careful control of running rates and maximisation of copper contact, we got rid of the ethyl carbonate issue,” he adds.

Temperature-controlled wash backs at Dalmunach Distillery in Speyside, Scotland


Riffkin believes that the application of scientific principles to distilling can enable both quantity and quality to be achieved, noting that most distilleries were quite unclean before 1970 and this contributed to lower yields. Combined with the widespread use of poor-quality brewer’s yeast, this poor hygiene made for very inefficient fermentations. Essentially, there was a lack of understanding of bio-chemical and micro-biological processes in the whisky industry. After scientists started promoting modern sterilisation techniques, fermenters began to be properly cleaned for each batch. What’s more, the optimum temperatures for mashing were examined and yeast was improved, meaning that when this optimised wash was introduced to clean wash backs there was little standing in the way of efficient fermentation. These practices improved yields enormously and modern malting varieties of barley further increased the volume of pure alcohol that could be extracted per tonne of malt.

When it comes to the work now regularly carried out by Tatlock & Thomson, Riffkin notes that most tasks are relatively standardised. A common task is using gas chromatography and mass spectroscopy to identify specific compounds in samples submitted by distillers: “Every day we get samples which we monitor to check that they meet with requirements for export markets where there may be stringent regulations,” says Riffkin. This is a big change from when he and the late Dr Jim Swan bought the company; at that time they could only process four samples and one control each day. Today that number is closer to 40 and a much more detailed analysis is possible, largely because analytical equipment has improved significantly over the past two decades.

“We work with most [wine and spirit] companies in the UK, plus producers of Caribbean rum, vodka from the USA and whisky from Canada. We have a core analytical team of 10 people, which includes two very senior analysts,” says Riffkin, adding that Tatlock & Thomson now do production analysis for 15 or 16 distilleries – including assessment of malt quality, fermentation efficiency, and distillation – on a weekly, monthly or quarterly basis.

“It’s nice to be in at the beginning of something,” he says, noting that Tatlock & Thomson has been involved with many of the recently established Irish distilleries. “We worked on the Waterford project from the very start, getting a former Guinness brewery to make high-quality spirit.”

Much of this work is sensory and the business now has a dedicated ‘sensory panel’ which helps distilleries (principally new ones) achieve the highest levels of quality. The panel undertakes analysis of new-make spirit and sometimes mature whisky with the aim of detecting any issues that have arisen or may arise in the future.

Riffkin admits to being a convert to mash filters, as opposed to mash tuns, the likes of which can be found at Inchdairnie Distillery in Fife, Scotland, and Waterford Distillery in Ireland. Teaninich was the first distillery in Scotland to have one and, according to Riffkin, these pieces of kit can be integral to producing very high-quality spirit because they produce very clear wort. However, he notes that there are other ways to achieve the same aims: “The new distillery at Burnbrae, East Kilbride, that we’re working with has a properly designed, shallow bed, wide-body lauter tun in conjunction with a mash conversion vessel. This will produce clear wort with rapid turnaround times.’’

The mash filter at Waterford Distillery is used instead of a more traditional mash tun


Another issue he is keen to flag up is fermentation temperature control. “Usually there is very little control over fermentation temperatures,” he says. “You used to have ‘summer whisky’ and ‘winter whisky,’ which could be quite different.” In order to combat this cause of variation, temperature-controlled fermenters are coming into use which allow distillers to optimise conditions for the yeast during fermentation, before increasing the temperature after 48 hours to promote secondary lactic acid fermentation. “It’s very important for flavour,” he adds. “I think it should be a criminal offence to build a new distillery without fermentation temperature control! Proper control of condenser water temperatures throughout the year also helps.”

In addition to the more routine day-to-day tasks, Tatlock & Thomson is currently undertaking two wood-related PhD studies. Dr Jim Swan left Tatlock & Thomson to establish his own international spirits consultancy in 2002, and one aspect of whisky production in which he specialised was maturation. “He had done a great deal of work on that,” says Riffkin. “A lot of wood expertise was lost when Jim Swan died.”

The late Dr Jim Swan, who consulted on distilleries such as Kavalan, in Taiwan and Kingsbarns, in Scotland


Today, Tatlock & Thomson are now looking at the role of ex-wine casks in relation to organo-sulphur compounds. This is because ex-wine casks often have had sulphur candles in them for fumigation purposes to inhibit growth of bacteria which might infect any wine filled into the cask. This generates the potential for organo-sulphur compounds to appear in spirit that’s matured in these casks at a later date.

“We now think that there are three possible sources for these compounds, which include sulphur candling, the previously filled wine, and the new-make spirit itself. All or some of these may be the source of the vegetable-type organo-sulphur aromas. We are also looking at differences between worm tubs and shell-and-tube condensers in relation to organo-sulphur compounds and feints character.”

The second PhD project relates to filling strength and whether this should be varied depending on the type of cask being filled. “If someone is using a second, third, or even fourth-fill cask, is 63.5% necessarily the optimum filling strength?” asks Riffkin, before going on to explain that this is just one area of wood research that is benefiting from new techniques and technologies. “The introduction of ICPMS, inductively coupled plasma mass spectrometry, allows us to look very specifically at elements of the periodic table and the correlations between calcium and magnesium in the wood and how that affects maturation. Where has the wood been grown and what has it got from the soil there? We’re exploring the concept of terroir from an analytical point of view.”

With so many regular duties to perform and new areas to research, Tatlock & Thomson has never been busier and this means expansion has been necessary: “We’re constructing a second building that will house a molecular biology lab which can look at genetically modified compounds. We’re also installing a trial distillery to offer training facilities for staff and to conduct test distillations,” says Riffkin.

A great deal of whisky is marketed by emphasising its heritage and apparent ‘craft’ provenance, even if the distillery that makes it is operated by one man and an iPad. However, it is really the application of science to varying degrees that enables consistently high-quality whisky to be produced. For that, we have to thank scientists of rigorous intellect and perpetual curiosity like Harry Riffkin and his colleagues.