The science of great coffee

As a global leader in the specialty coffee scene, it is unlikely to surprise any knowledgeable observer that roasters in Melbourne are once again pushing the envelope to improve the quality of their coffee. While the specialty coffee industry right around the globe – and in activity centres such as Melbourne in particular – is always striving to find new ways to innovate and improve its offerings, it is somewhat surprising that the industry as a whole relies more on artistry than that traditional well of innovation, good old science, to break new ground. All that looks set to change, though, as more roasters begin to look to integrate more scientific processes into their business to achieve higher quality results on a more consistent basis. One person helping to lead the charge into the realms of caffeinated science is chemical engineer and Director of the Australian Coffee Science Lab, Dr Monika Fekete. Originally hailing from Hungary, Fekete moved to Melbourne to work for Australia’s leading scientific research body, the Commonwealth Scientific and Industrial Research Organisation (CSIRO).
However, it was not long before another of Fekete’s passions began to drive her professional development. “When I came to Melbourne I found myself in the middle of this amazing coffee scene that was so different to what was in Europe at the time,” she tells Global Coffee Report. As a way of indulging what was, until that point, a purely recreational interest, Fekete took a barista course with leading Australian specialty roaster Five Senses Coffee. “I realised there was so much passion, effort and art going into creating the perfect coffee, but little science just yet. This makes sense as coffee professionals often come from hospitality or business backgrounds, rarely science,” she says. Fekete knew from her work that there was a huge scope for applying scientific methods in order to improve coffee quality. “That contrasts with other food and beverage industries, such as wine, beer and dairy, which have been using modern chemical engineering knowledge to improve quality and consistency for years. I saw that as an opportunity to get involved.” Starting out, Fekete carved a niche for herself in the bustling Melbourne coffee scene as a traveling coffee expert, complete with mobile laboratory, which is complemented with sophisticated analytical instruments based at universities she can access through her academic network. Armed with a USB microscope with x400 magnification to examine coffee beans, a laptop to view the microscopic particles, pH meter to detect lime scale build-up in a coffee machine, and a hand-held coffee refractometer to measure total dissolved solids (TDS) in coffee to help calculate extraction yield, Fekete started out providing clients with basic measurement and diagnostic services. But she soon found that there were many other opportunities for somebody with her skill set in the booming specialty coffee industry. “Originally my idea was simply to provide clients with regular testing for things like caffeine levels, toxins and quality control methods in extraction and roast colour and so on,” she says. “However, I found that a lot of companies are interested in special research and development projects. They may be developing a new process, or a totally new coffee product, but they need help testing and validating it, turning an idea into reality. That’s where I come in.” One of her first collaborations was with Five Senses Coffee Roasters. She used statistical mining of roast data to rank the importance of roasting parameters, and see how they could better control the results. In particular, she was able to identify the most influential parameters, directly affecting measurable outcomes such as roast colour and TDS. Fekete says that she has found that the industry is right on the precipice of embracing a whole new area of scientific practices that will improve its output and enhance profitability. “Most large roasters tend to rely solely on sensory assessment as quality control, and of course taste is extremely important because that is the key experience for the customer,” she says. “But it really makes a huge difference for the roaster in terms of costs and quality if they have regular data collection. Measurements and well-designed experiments can help maximise efficiency in terms of output per unit of green bean, coffee extraction and coffee lifetime.” Fekete says she has come across a number of roasters who recognise the importance of gathering information in this way, and have even gone to the lengths of buying the necessary equipment to do so, but she finds that it is often under-utilised. “Unfortunately, I have found that some roasters see that it’s a good idea, so they buy the instruments. However, often they don’t have the time or the training to set up experiments that will yield meaningful results from that new instrument. In this case, the investment can just go to waste.” Fekete says the same thing applies to the information being collected by the sophisticated software that complements most industrial roasters nowadays. “Data collected through roasting sotftware is really important, but the output can be confusing. I can provide a statistical assessment and look at the key controllable variables that are really important in order to achieve a consistent outcome,” she says. One of Fekete’s favourite projects in 2016 was working with Australian Specialty Coffee Association’s (ASCA) Australian Brewers Cup Champion Devin Loong to fine-tune Loong’s mineral composition in preparation for his World Brewers Cup (WBrC) routine in Dublin. To begin, Fekete recreated the water participants are required to use in the compulsory round of the WBrC for Loong to practice with ahead of time. She then adjusted the mineral composition of the water used for his presentation, adjusting calcium, pH and carbonate levels. “It was very important to get the water recipe just right for Devin, given that water accounts for more than 98 per cent of a brew,” Fekete says. After a lengthy process of elimination, Fekete found that a 1:1 molar ratio of calcium to magnesium worked best to highlight the tasting notes of Noria, Loong’s La Esmeralda Geisha coffee. The general hardness was 150 parts per million, and the carbonate hardness was relatively low at 30 parts per million. The end result was a brew water with a TDS reading of 190, pH of 8, general hardness of 150 and equal parts calcium and magnesium. Fekete also worked with Loong to find the “sweet spot” on the Bunn grinder, the official grinder machine of the WBrC. Using 425-micrometre and 800-micrometre sieves, Loong determined that a grind setting of nine gave him the largest amount of grinds in this range (the smallest amount of grinds below 425 micrometre and boulders above 800 micrometre) and the best taste. Loong practiced with these parameters, but when he got to Dublin he found the same grinder reacted differently. His back-up plan was to adjust the grinder settings by one to two places until he found his “sweet spot”. But unfortunately the outcome wasn’t as expected. Fekete says the results from Loong’s competition brew were “confusing” and didn’t follow the same pattern they had determined. She held a late-night emergency session back in Melbourne analysing different weight ratios Loong sent her via email. She cross-examined and plotted the different results and figured he actually had to adjust the grinder by six settings finer to get the best profile. He tried this at his last practice session, and got the “sweet spot” straight away – just in time for his WBrC routine. Fekete will continue to work with Loong and other baristas to assist their competition routines. She says while it’s good for coffee professionals to embrace sensory skills in their workplace and in competition training, it’s important to appreciate that science is a complementing component. Having worked with such high-profile clients, Fekete says she is now aware of the huge range of work that is required by the industry and is expanding her offering accordingly. “I find that a lot of roasters prefer to do their own quality testing, which is really exciting.Then what I can do is to provide them with the best value instruments for the job, as well as a complete package of training, service, maintenance, calibration and consultation about their results,” she says. Another new service she plans to offer in 2017 is to run workshops to help smaller roasters and baristas get an insight into how they can use science to improve their products. “Often I find that people will have some great ideas, but they need to improve the systematic rigour it often takes to realise those ideas,” she says. Since getting involved in the coffee industry, Fekete says the scope of her work has been continually expanding. As well as consulting with roasters about how to improve their quality and consistency, she has worked on improving grinder performance, looking at ways to extend shelf life, and even matching water to coffee to improve results based on the water types available in different parts of Australia. Another area Fekete says scientific input can have a huge benefit is efficiency. “When a roaster is looking to find ways of reducing their costs, we can find ways to maximise the output of green beans to, say, produce 2 per cent more coffee and that’s already a huge saving,” she says. “Another approach is to improve roasting and grinding to maximise extraction output, which again can lead to significant savings.” For those roasters looking for a simple way to apply the benefits of scientific testing and experimentation to their business, Fekete has a simple, three-step process. “As a first step, everybody should assess what quality controls they have in place and evaluate their strengths and weaknesses and think about their goals. As a next step, it’s important to set up and apply some well-designed experiments to help them achieve those outcomes. Then I think the final step is really about commitment – it’s not enough to just play around every now and again, it can be fun, but it won’t achieve consistent results that way,” she says. “Think about where you are now and where you want to be, design your experiments accordingly, and just keep at it.” GCR