Coffee Science for CoffeePreneurs by CoffeeMind

Why Individual Organic Acids in Coffee are irrelevant

• Morten • Episode 18

Share episode

Since 2014 we have been skeptical about the inclusion of teaching, training, and testing coffee professional in individual organic acids. This episode is the story behind how we explored this question and clarified the issue in a scientific project proving this irrelevance in sensory education systems.

Organic acids in coffee

I wanted you - our cherished podcast listeners - to be the first to know, so this podcast script was written yesterday late evening. I wanted to tell you about the details of the upcoming scientific publication about organic acids in brewed arabica coffee as we got the final manuscript accepted a week ago, so I thought I had a bit of time to tell you before it was released. So I was quite surprised to find the official doi number (a specific identifier number given to scientific papers) in my inbox this morning (22nd of March 2023). But I’m sure nobody noticed yet, so you are still the first to know! If you want to go look for the article there is a link in the shownotes but you can also search for the title “Acids in brewed coffees: Chemical composition and sensory threshold” and add CoffeeMind to the search string and the article should show up. But I’m here to tell you about the background and content of the project so just stay tone for nowt. 

But first a word from our sponsor. This episode is brought to you by CoffeeMind. And what I’m about to tell you is not that I want to sell you some gadget that you don’t even need. I’m telling you about the biggest project CoffeMind has ever done, which is Ida Steen's industrial PhD project in CoffeeMind which we have been working on for 9 years now. Ida has extracted all the best practices from how professional sensory panelists are trained at universities worldwide and made her own training program based on it. For the last 6 years, we have been working hard on getting data measuring the effectiveness of the training program and these data were published last month in the prestigious “Journal of Sensory Studies” with the title “Evaluation of a sensory and cognitive online training tool for odor recognition in professional coffee tasters”. The paper covers data collected in a 12-week training period but now Ida has taken all her experience and created a 4-week sensory training program that is now available to purchase on CoffeeMind’s website so that you can get the latest sensory training program brought to your door, where you will get a box with different material for daily training for four weeks guided by Ida through our e-learning system. It is the first evidence-based training program the coffee world has ever seen and if you follow Ida’s methods you will never find yourself being dragged into pseudoscientific rabbit holes. And what you will find in the rest of this podcast is just one of the many misconceptions that have made it to the biggest education systems in the coffee world. This does not happen with CoffeeMind’s education programs. You can find this 4-week e-learning program on the front page of CoffeeMind’s webpage. The price of this 4-week training with a shipped sensory training kit is €250 + shipping. But I would like to extend the same offer to you that we released to our mailing list yesterday. In this offer, you can get Ida’s 4-week training program at the cost price of €250 + shipping but if you add the coupon code “podcast” during checkout you will get a free bonus with your purchase which is  Ida’s Sensory Basics e-learning courses (value €30) as well as a pdf containing all CoffeeMind’s scientific articles which is now a bundle of 11 scientific papers with a value of €100. So by using the coupon code “podcast” during checkout, you will get free bonuses for €130 on top of the base price of €250 of the 4-week sensory training kit. You need to add that 4-week training kit to the cart before applying the coupon code and if you apply the code before adding the 4-week sensory training kit you will get an error message. End of commercial. Now to the organic acids problem!

Acidity is an important sensory characteristic of coffee in general. But particularly important for the light roast wave in specialty coffee. And anybody who has spent just a bit of time in the specialty coffee business over the last two decades is familiar with Formic, Acetic, Lactic, Glycolic, Tartaric, Malic, Citric, Isovaleric, and Chlorogenic acids. As if it is relevant at all. Contrary to popular belief, there exists absolutely no public scientific evidence for relating the organic acid concentration to sensory perception of acid. The most interesting publication ever on this topic came from UC Davis In September 2021, supported by SCA’s own Coffee Science Foundation titled “Acids in coffee: A review of sensory measurements and meta-analysis of chemical composition”. Specialty coffee people who love to talk about individual organic acids in brewed Arabica coffee might be a bit disappointed that this article only contains insight into organic acids in green and roasted coffee and not brewed and also has a lot of focus on chlorogenic acid differences between arabica and robusta. Not exactly angles on this subject most interesting for us in the specialty coffee sector. But being a metaanalysis that gathers all research in the area, this is a necessary consequence of what has been published on the subject and not an angle chosen by the authors. So it seems interesting research is missing if you are interested in organic acids in brewed arabica coffee! In the literature study we did before our research into organic acids in coffee, we tried to get some insight into the research done by Joseph Rivera from Coffee Chemistry, who seems to be the most prominent teacher of organic acid in coffees, but he responded that his research was done in a commercial context so he could not share it publicly unfortunately. So we had to start from scratch to explore the relevance of individual organic acids in brewed arabica coffee since no published research exists on this topic

But let’s go back to when it all started for us in CoffeeMind as it started quite a long time ago. 9 years ago to be precise! Ida Steen joined CoffeeMind in 2014, right after graduating with a Master’s degree in Sensory and Consumer science from the University of Copenhagen and immediately got involved with the creator's group of SCA’s Sensory Skills curriculum. Immediately she raised the flag about the relevance of training and testing people's skills in differentiating and identifying organic acids in coffee. If this was relevant, why had she not heard anything whatsoever about this during her university studies? Furthermore, why does SCA include Citric, Malic, Tartaric, and Lactic acids in the exams. And why have they chosen different acids in the QGrading exam system where you work with Citric, Malic, Phosphoric, and Acetic acids but not Tartaric and Lactic acids? And on the SCA Flavour wheel, we have Citric, Malic, Acetic, Butyric, and Isovaleric acid. According to Ida, there is no evidence to support the choice of these specific acids instead of other acids in coffee or to justify why they are added to water and coffee and why they are all tested in a concentration of 0.4 g/L except the lactic acid which is tested in a concentration of 0.5 g/L

After many years of frustrating attempts to find evidence for this, we took the initiative to look into this ourselves In 2017. We got a student at the university involved and ran the first pilot study looking into this from both chemical and sensory perspectives of this subject. 

Let’s start with the findings amongst the sensory data in the pilot study. In triangle tests with watery solutions of different organic acids, we found that Acetic acid is the only acid cuppers where able to differentiate from the other acids. From a chemical perspective, this is not a surprise as Acetic and Formic acid differs from all the other acids by being detectable by nasal inhalation because they are volatile at room temperature. However, Formic acid could not be differentiated from other acids in the pilot study.

So Acetic acid was the only acid that could be differentiated from the other acids if it appeared individually in a solution up against other individually dissolved acids, but after mixing it with just one other acid, the assessors were not able to identify it anymore. This indicates that it will be challenging (I would say impossible) to recognize the acids in a cup of coffee that contains many other flavors and of course, a mix of all the different acids.

What we found in the chemical data did not make sense, so I dismissed it as a label error by the student. What do students know about rigorous procedures to not mess up labels in a science project? According to her chemical analysis there was much more citric acid in the Brazilian Pulped natural than the washed Kenya coffee! We had specifically chosen the washed Kenya because it is the king of perceived acidity in our mind, and the Brazilian Puplped natural was chosen because they typically have such a low acidity level that it is a safe choice in an espresso blend designed to have a low perceived acidity level. So coming out with a much higher chemical concentration in the Brazilian Pulped Natural coffee clearly pointed to a simple label error made by the student.

It was only in a conversation I had with Samo Smrke from the Coffee Competence Centre at Zürich University of Applied Sciences that I reconsidered the label error suspicion. He said something like: “It’s the funniest thing; I have seen data with similar trends, so I think it might be correct!”. I went home and talked to Ida, and we decided to give it another go. This time with another group from another university in Denmark, namely The University of Southern Denmark. We gathered a group and redesigned our approach to give the most interesting results based on the pilot study. We roasted the coffees, did the analysis, wrote the paper, and submitted our work to Food Chemistry in December 2021. In April 2022 (5 months after submitting) it was rejected with the argument that we needed more sensory data to make a full article. We redrafted the scope of the article and came up with relevant sensory test strategies and gathered more sensory data. We chose a fast processing time, open-access journal, where the publication process is paid for by the research group in order to make it available for everybody for free on the journal website. We submitted the new manuscript with added sensory data again in December 2022 and got minor revision requests already in mid-January this year and a final acceptance of the manuscript last week! So here it comes:

We determined the concentration of acids and sugars in brewed coffee of varying bean origin, namely two different Brazil Pulped Natural, one Bolivia, and two different Kenyas. The two low-perceived acidity Brazilians were taken from Kontra coffees (the roastery where we have CoffeeMind Academy). One of the Kenyan coffees where Coffee Collective’s Kieni, which is quite internationally recognized for being some of the most acidic (and absolutely delicious if you ask me) coffees you can get. The other Kenyan coffee comes from the Copenhagen-based coffee shop Original Coffee which also has a particularly high perceived acidity.

In the chemical data, the acid composition was affected more by roasting than by origin. Significant differences were found in a roast degree-dependent manner for all acids (except formic). The concentration of citric, malic, and chlorogenic acid decreased with darker roast, and acetic, lactic, phosphoric, quinic, and glycolic increased with an increasing roast degree. As for origin, Brazilian coffee was found to have a higher total concentration of organic acids than Kenyan coffee. Just like in the pilot study! There is no simple relationship between the total amount of acid in coffee and the perceived acidity! What do you think, then is the probability that you can pinpoint Malic acid or some of the other fancy acids in coffee? But we have much more detail about that in the following

Remember: For it to make sense to train and test people in differentiation and identifying individual organic acids, two criteria have to be met simultaneously (not just any of them):

  1. There has to be a big difference in concentration in different organic acids depending of origin. If there is more or less the same concentration for all arabica coffees, there is no fingerprint of organic acids for the cupper to identify!
  2. The different organic acids have to be above the sensory threshold! 

If BOTH these criteria are not met, it does not make sense to teach, train and test people on individual organic acids! At all.

When you get your fingers on the article, you will see on Figure 2 that Criteria one - regarding the necessary difference between coffees - that none of the acids really differ between samples except Citric acid. But remember that the Brazilian coffees have the highest concentration - not the Kenyan. It does not make sense. To quote this passage from our resarch paper

“the differences in acid concentration between the five coffee samples were generally not strongly dependent on the geographical origin except for citric and chlorogenic acid. Citric acid had larger and directly statistical significant variations between the sample origins. Brazil 1 had a much higher concentration of chlorogenic acid than the other samples. The two Kenyan coffees, which are from different regions in Kenya, differed slightly in chlorogenic acid concentration, but showed no significant differences between any other acids. In contrast, the two Brazilian coffees from the same region were significantly different in the concentration of chlorogenic, citric, malic, phosphoric, and acetic acids. Brazil 1 had the same concentration of malic acid as both Kenyan coffees, and Brazil 2 did not differ in the concentration of acetic, phosphoric, formic, and glycolic acid to both Kenyan coffee samples”

Criteria two - regarding the necessary exceeding of the sensory threshold of the individual organic acids - are not described in any research project yet, so we had to make it part of the project. We found that most of the identified threshold values for the different acids (Acetic, Phosphoric, Lactic, and Malic) were above the actual concentrations of these compounds in coffee identified based on the chromatographic data. Only Citric acid was found in concentration clearly above the sensory threshold. If threshold values are ABOVE concentration, you can’t taste the individual acids in brewed coffee, and training, teaching, and testing students is not relevant and is a waste of everybody's time. You will see these data in Figure 6 in the final article.

But we did not stop there. Even though the thresholding data was convincing enough, we also wanted to test coffee professionals’ ability to identify individual acids in water and brewed coffee to get the final dataset to convince the coffee world that this is irrelevant when teaching, training, and testing students in sensory education systems.

The identification of five acids in water was found by adding 0.50 g/L of lactic acid and 0.40 g/L of citric, malic, acetic, and phosphoric acids in water (according to the training from SCA). The identification of five acids in coffee was set up by adding the average measured concentrations in brewed coffee. The argument behind spiking the coffee with the average concentration our study found - and thereby doubling the concentration of the particular acid under investigation - was that this would make the investigated acid concentration way above the average concentration and thereby it should really stand out in that particular cup of coffee. This also factors in the possible counterargument that there might be some coffees not included in the study that contains an even higher concentration than the coffees included in this study. If you double the concentration of an acid it should be perceivable and count as a concentration that is safe on the high side of what you can find out there.

We were lucky enough that The Coffee Collective in Copenhagen agreed to participate in this study as a follow-up on a training session they received from Ida Steen a few weeks before, where she worked with all the bar managers from their different coffee shops. This made us a panel of 13 coffee experts (mean age 29, 46% women), who all have coffee as their profession. They are all regularly trained and calibrated on sensory evaluations of coffee, including trains in individual organic acids identification in their standard bar manager training program. The experts were a perfect representation of the target group for testing the identification of organic acids in coffee. If it made no sense for them with that background and that career position, we can’t see it make sense for anybody. The peer reviewers of our scientific publication objected that none of these experts were neither SCA nor Qgrader certified to which we responded that here in Scandinavia, these organizations are not very dominating. The Coffee Collective have created its own internal training program and participated in 24 competitions of which they have won the Danish championship 10 times and the world championship twice (in 2006 Klaus Thomsen won the World Baristas Championship and in 2008 Casper Engel Rasmussen won World Cup Tasters Championship). In my humble opinion, this beats passing any exam.

The Coffee Collective bar manager group evaluated each of the five acids in water and in brewed coffee. Immediately before evaluating the samples, they underwent a 30-minute intense training session in detecting and memorizing the five acids in water and was encouraged to take notes to help them identify the acids later in the test. The coffee experts tasted the water solutions and coffee samples based on the identification of tastes test from ISO standard 3972:2011, a similar test setup used in the SCA and CQI training on organic acids. They tasted the samples one at a time in a complete randomized block design and were asked to name the specific acid added out of the five possible acids while being allowed to keep their personal descriptive notes for each acid from the training session.

From the sensory detection threshold values found, only citric acid can be detected in concentrations above average measured concentration. In the identification test, none of the five acids was correctly identified. Not even close, actually! And we know that they are skilled cuppers. If it is not relevant for them with their education and career path, it is not relevant generally in the specialty coffee business worldwide. How can this have been standard practice in education systems worldwide for more than two decades?

I can’t tell you how happy Ida and I are to share this with you after 9 years of hard work on so many levels. With our publication released today, the cat is out of the bag!