How a computational chemist and an understanding of water helped a coffee shop owner to become the 2014 UK Barista Champion, set to take on the world. Guest post by Chris Hendon.

Christopher Hendon and Maxwell Colonna-Dashwood at Colonna and Smalls, Bath
© tomsmith photography

Brewing coffee might be the most practiced chemical extraction in the world. But within this process there are many variables, all of which dictate the flavour of the resulting coffee. I’ve summarised just a few of them here:

Variable Effect
Bean origin Not all beans have the same chemical composition.
Bean roast The chemical composition of the coffee bean changes throughout the roasting process.
Size of coffee grindings A consistent particle size is important as the higher the surface area, the faster the extraction.
Dry mass of coffee grindings A different extraction composition.
Temperature of extraction The temperature dictates both the rate and composition of the extraction.
Pressure of extraction Has a similar effect as temperature.
Time of extraction Increasing extraction time allows for a greater extraction.
The water This variable is less obvious, but it is clear that the chemical composition of water (i.e. dissolved ions) play a very important role.


Analysis of extracted coffee by gas chromatography–mass spectrometry (GC-MS) suggests that the average coffee bean contains upwards of 500 chemical compounds, excluding all of the heavy cellular material. Of this complex array, the bean is primarily a mixture of weakly acidic molecules, and the weaker acids are more desirable. A ‘bad coffee’ can be the result of any combination of the aforementioned variables going awry.

Maxwell Colonna-Dashwood, 2014 UK Barista Champion and co-owner of the specialty coffee shop Colonna and Small’s, in Bath, UK, carefully controls most of these variables. He grinds beans to a consistent particle size, weighs the dry mass, uses a constant temperature and pressure for extraction (which  he dials in for each bean on the day), and defines the extraction by the mass of the extracted coffee. However, it’s almost impossible for him to control the chemical composition of the incident water. Water’s ionic content fluctuates dramatically depending on region and quantity of rain – and it rains a lot in England. The coffee industry has designed some ways to deal with this problem, with filtration units and vague guidelines on what chemical composition to aim for.

The coffee industry has concluded that an ionic concentration of 150–300 parts per million (ppm) is ideal for coffee extraction. But that is inherently flawed; water could contain 150ppm of HCl and would certainly not taste very nice. Conversely, water could contain 150ppm of NaHCO3, which might make you burp as the acid/base reaction in your stomach rapidly releases gaseous CO2. So in search of the perfect coffee, it would be wrong to accept these guidelines without some thought as to the impact of different chemical compositions. Unfortunately, not every local barista has access to an atomic absorption spectrometer (we do). Instead this measurement is often collected (if at all) using an ionic conductivity probe, which makes two absolutely fatal assumptions:

  1. The ratio between dissolved ions (Ca2+, Mg2+, Na+, H+, HCO3, CO32- etc) is approximately constant
  2. The ionic conductivity of all ions is approximately the same

Both of these statements are incorrect. For instance, in Bath there is much more Ca2+ present than Mg2+ (approximately 300ppm:5ppm, respectively); but in Melbourne the Ca2+:Mg2+ ratio is approximately 20ppm:20ppm. Secondly, the conductivity of an ion is dependent on, among other things, the size and charge of the ion.

Colonna-Dashwood is certainly right in challenging this accepted guideline, but what interaction do these molecules have with coffee particulates and more importantly, does it even matter?

Yes. It matters, a lot.

Christopher Hendon and Maxwell Colonna-Dashwood at Colonna and Smalls, Bath
© tomsmith photography

As is often the case in science, it was an element of serendipidy that brought me into the picture. I’m a computational chemist at the University of Bath, and overheard this discussion while waiting for my coffee. I thought I might be able to contribute to the problem, at very least help to rationalize this in terms of quantum mechanics. I had learned from fundamental chemistry that electron-rich motifs interact with electron deficient motifs. Essentially all molecules in coffee feature a heteroatom (an electron-rich motif), which should interact strongly with dissolved cations in water. Dissolved anions may act as bases, but are not expected to interact with coffee particulates. Thus, water with a high cationic concentration should facilitate a greater extraction of flavorsome notes in coffee. Along with Maxwell and his partner Lesley, we’ve recently published our results.

The world of coffee is an unusual place. As mentioned earlier, Maxwell is the 2014 UK Barista Champion. To be crowned this, you must submit an espresso, a cappuccino and a signature drink to a panel of four judges. You have 15 minutes to do so, and are judged on knowledge of the coffee, your overall presentation and cleanliness, flavour, technical ability and so on. Armed with the knowledge gleaned from our research into the interactions of ions in solutions, I designed different waters for different extractions, to bring out different flavours. This was particularly intriguing for the signature drink which featured an espresso shot mixed with two grape extracts brewed in different water – one with high cation content, one with high base content – to extract different flavours from the grapes. With this victory (for science, I like to think), we are now headed to the world barista championships in Rimini, Italy, June 8–12. I hope that our artillery of scientific knowledge will see Maxwell becoming the 2014 World Barista Champion.

If you find yourself in Bath and fancy a coffee, I would highly recommend you head down to Colonna and Small’s to see and taste the result in person, you’ll be pleasantly surprised. At very least, you can use this story to prove that work really does get done in the coffee shop.

VN:F [1.9.10_1130]
Rating: 8.5/10 (35 votes cast)
VN:F [1.9.10_1130]
Rating: +18 (from 20 votes)
Schrödinger’s water for the perfect cup of coffee, 8.5 out of 10 based on 35 ratings
Digg This
Reddit This
Stumble Now!
Share on Facebook
Bookmark this on Delicious
Share on LinkedIn
Bookmark this on Technorati
Post on Twitter
Google Buzz (aka. Google Reader)