#82 | Sensory and Chemical Explorations into Drip Brew Coffee Flavor Over Time | Expo Lectures 2019

This podcast was brought to you with the support of

The UC Davis Coffee Center is engaged in comprehensive sensory research using trained panel descriptive analysis to investigate how different parameters related to coffee brewing impact the flavor, and how these can be manipulated to an individual's desired effect. Of these factors, time can be an important variable—in multiple ways—and this talk will highlight two recent projects. The first investigated the time-evolution of coffee flavor extraction during the brewing process. By dividing a drip brew into eight parts and evaluating the flavor change every 30 seconds, it offers a more detailed picture of observable extraction changes. The second experiment investigated time elapsed post-brew, comparing how quickly perceptible changes occur when coffee is held in a carafe, in order to understand the stability of flavors depending on holding conditions. Both of these studies offer preliminary data that can be immediately useful to the industry in developing and preserving desired flavors in the coffee brewing process.

Today’s lecture is presented by Mackenzie Batali, a second-year graduate student in Food Science and Technology at the University of California, Davis, focusing her research in coffee sensory analysis. Previously she received a chemistry degree from Lewis & Clark College in Portland, Oregon, and worked as a flavor chemist before transitioning into the coffee world. She has spoken previously about her work on coffee fractionation at ASIC 2018.

If today’s lecture piques your interest, you can also read more about Mackenzie’s fractionation experiment in Issue 11 of 25 Magazine.

Special Thanks to Softengine Coffee One, Powered by SAP

This episode of the Expo 2019 Lectures podcast is supported by Softengine Coffee One, Powered by SAP. Built upon SAP's business-leading Enterprise Resource Planning solution, Softengine Coffee One is designed specifically to quickly and easily take your small-to-medium coffee company working at any point along the coffee chain to the next level of success. Learn more about Softengine Coffee One atsoftengine.com, with special pricing available for SCA Members. Softengine: the most intelligent way to grow your business.

Full Episode Transcript

0:00 Introduction

Heather Ward: Hello everyone! I’m Heather Ward, the SCA’s Senior Director of Content Strategy, and you’re listening to the SCA Podcast. Today’s episode is part of our Expo Lecture Series, dedicated to showcasing a curated selection of the extensive live lectures offered at our Specialty Coffee Expo. Check out the show notes for relevant links and a full transcript of today’s lecture.

This episode of the Expo 2019 Lectures podcast is supported by Softengine Coffee One, Powered by SAP. Built upon SAP’s business-leading Enterprise Resource Planning solution, Softengine Coffee One is designed to quickly and easily take your small-to-medium coffee company working at any point along the coffee chain to the next level of success. Learn more about Softengine Coffee One at softengine.com, with special pricing available for SCA Members. Softengine: the most intelligent way to grow your business.

The episode you’re about to hear was recorded live at the 2019 Specialty Coffee Expo in Boston. Don’t miss next year’s lecture series in Portland – find us on social media or sign up for our monthly newsletter to keep up-to-date with all our announcements, including ways to get involved in next year’s Expo and early-bird ticket release!

The UC Davis Coffee Center is engaged in comprehensive sensory research using trained panel descriptive analysis to investigate how different parameters related to coffee brewing impact the flavor, and how these can be manipulated to an individual’s desired effect. Of these factors, time can be an important variable—in multiple ways—and this talk will highlight two recent projects. The first investigates the time-evolution of coffee flavor extraction during the brewing process. By dividing a drip brew into eight parts and evaluating the flavor change every 30 seconds, we offer a more detailed picture of observable extraction changes. The second experiment investigated time elapsed post-brew, comparing how quickly perceptible changes occur when coffee is held in a carafe, in order to understand the stability of flavors depending on holding conditions. Both of these studies offer preliminary data that can be immediately useful to the industry in developing and preserving desired flavors in the coffee brewing process.

Today’s lecture is presented by Mackenzie Batali, a second year graduate student in Food Science and Technology at the University of California, Davis, focusing her research in coffee sensory analysis. Previously she received a chemistry degree from Lewis & Clark College in Portland, Oregon, and worked as a flavor chemist before transitioning into the world of coffee. She has spoken previously about her work on coffee fractionation at ASIC 2018. If today’s lecture sparks your interest, you can also read more about Mackenzie’s fractionation experiment in Issue 11 of 25 Magazine.

Alright, let’s get started!

3:10 An overview of the coffee science research taking place at UC Davis

Mackenzie Batali: I’m Mackenzie Batali, Ph.D. student at UC Davis in the second year of my program, and I’m really excited to talk to you about my first project in my dissertation which will be Sensory and Chemical Explorations into Drip Brew Coffee Flavor Over Time and I wanted to give a little bit of background to how I ended up here because two years ago before I started at UC Davis I wasn’t in any way affiliated with the coffee industry. I have a background in chemistry. So, I’m coming at this from a little bit of a different perspective. I focused in organic chemistry in my undergraduate degree and then I worked for a few years at a chemical manufacturing company working research and development for artificial flavors and fragrances.

So, that’s where I got my introduction to sensory science. I spent two years optimizing the purification of artificial peach and this is some of our sensory panels for this peach fragrance I was working on and that got me really interested in sensory analysis, which led me to a graduate degree in food science to focus in sensory science, which brought me to UC Davis. When I got to UC Davis I was introduced to the UC Davis Coffee Center and the research opportunities there. S,o some of you may have seen Dr. Ristenpart’s or Dr. Frost’s talks yesterday. So, I’m sorry if this is redundant but basically the UC Davis Coffee Center started in 2013 with Professors of chemical engineering Bill Ristenpart and Tonya Kuhl developing an elective course called The Design of Coffee teaching basic chemical engineering concepts in a GE general education course. Using coffee as a medium and this caught the attention of the coffee industry. So, to make a long story fairly short over the years this has started the coffee initiative which brought professors from all sorts of disciplines at UC Davis from food science and chemical engineering, plant science, business law, sociology, dozens of professors who have expertise and interests that will relate to coffee together with shared research goals working with the coffee industry.

This has garnered a lot of excitement and funding and broadly we want to do what UC Davis has already done for wine and beer with our world-class teaching winery and pilot brewery for coffee by creating the UC Davis Coffee Center. In my particular lab studying coffee brewing research. It’s a marriage of engineering and food science. So, I work with a group of undergraduates and graduate students from mostly chemical engineering and food science and the engineering work is based on understanding the physical process of extraction and how that impacts basic properties like water absorption, percent extraction, TDS, pH, and all these sort of things. Then that moves forward into food science where we focus with sensory science and understanding the implications of extraction on the final consumable product with descriptive sensory analysis, consumer preference and chemical analysis in my particular work. I want to clarify that in this talk when I’m talking about censoring methodology it’s a little bit different than what the coffee industry might think of. So, we don’t work with the cupping system.

We use methodology more standardized within food science and there’s two different sorts of fields of sensory science within food science. We’ve got descriptive analysis which is more my focus which is purely quantitative. We’re selecting specific attributes and tracking how they change in intensity from one sample to another. It’s always comparative that we use trained panels. So, we go through a training system that’s relatively similar to the Q Graders but in this case, to contrast with the Q grading system there’s no quality judgment, there’s no liking scores. It’s all just quantitative. This is there in a certain intensity or it is not. That is separate from consumer science, which uses untrained coffee drinkers, people who the coffee industry would be marketing to and they rank on how they like certain products and how specific attributes may contribute to liking and you can correlate what kind of consumers like what kind of products and create different clusters for what type of products fit different people instead of a one-size-fits-all value judgment. So, that’s what I’m talking about when I’m talking about some of my work and sensory science and like I said, most specifically I’m doing descriptive analysis for this project.

So, some of the sensory work we’ve done so far at UC Davis if you didn’t see any of the other talks. We had the cone versus basket product project presented by Scott Frost last year and Bill Ristenpart yesterday. We have the project on updating and expanding the coffee brewing control chart that was presented Yesterday by Dr. Scott Frost and what I’m here to talk to you about today, which is fractionation of a coffee brew.

5:45 How does brewing time impact perceptible sensory properties of coffee?

Mackenzie Batali: So, the question we’re looking at with this project is: “How does brewing time impact the perceptible sensory properties of coffee?”, and more specifically, “what is happening compositionally as time progresses in a coffee brew?” and there’s some chemical analysis in the current academic literature to this effect. There’s a study from 1992 by Lee, et al. looking at compositional changes in coffee brew as a function of brewing time. In this project, they did a one liter brew and found that 50% of all extractable materials extracted in the first fifth of the brewing time. A study looked at extraction of coffee antioxidants which can be sensory-relevant for bitter, astringent flavors and tastes and mouthfeel and found that the majority of antioxidants in espresso found in the first eight seconds of the pull and the majority and filter coffee or in the first 75-150 seconds of brew depending on the coffee origin and the antioxidant content of that particular coffee. A study by Mestdagh all looked at the kinetics of aroma extraction and look found that the polarity of flavor compounds was most highly correlated to extraction speed.

So, like I said, there’s some chemical analysis that informs some of what we expected to see but there haven’t really been sensory studies to this effect. So, my sensory study design looked at the fractionation of a batch brew changing the carafe under the Curtis Brewer every 30 seconds for the first four minutes of a brew, collecting eight samples total, which was also compared to a full brew under the same conditions. So, giving nine samples for sensory evaluation and along with the sensory evaluation, we did collect physical measurements, mass, total dissolved solids, pH, and titratable acidity. And we used a medium roast with an Agron score of 54 Colombian coffee donated by Java City in Sacramento, California.

So, first of all physical measurements. Predictably like I said in the literature review chemical studies showed the majority of the material being extracted very early on in the brew. so, the plot on the left here we see the total dissolved solids. Extremely high in the beginning though we do see a lot of variability in the first two fractions with the average of around 3% TDS which decreases towards the end of the brew to around, that’s about 0.4%t and then on the right here, we’re looking at the brew mass dispensed in each fraction. So, some of the previous studies looked at volumetric fractionation, but we are looking at time-based fractionation. So, first 30 seconds, as the grounds are saturating. The flow rate is much slower, and the flow rate increases as water is dispensed, and at the 6-minute mark after 3 minutes was where the water dispensation ended and the last minute is the drip out and you see the flow rate slow again. I can also look at acidity so we’ve got to titratable acidity here which is a slightly better measurement for acidity correlating to flavor and taste, the pH because it includes week organic acids that the pH measurements might not pick up. So, titratable acidity is the concentration of all acids based on the amount of base required to neutralize a solution and we see a continuous decrease in acidity; highly acidic early fractions and levels out sort of like TDS does.

So, like I said most of the previous studies did compositional analysis. So, from here we move into the sensory analysis and sensory analysis in food science, descriptive sensory analysis in food science does require pretty extensive training. We start with free term generation. We use the coffee taste or flavor wheel as a reference and the World Coffee Research Sensory Lexicon and instead of having a preset ballot of applicants were looking for all the panelists taste that all the coffees in this particular study and they get to come up with the terms that they think are relevant to these coffees. There’s no point in rating nuttiness if these coffees don’t have that character. So, they start with that they get to write down whatever terms they think are relevant and they discuss as a group to start to create a common consensus in the sensory attributes.

Then to help refine that consensus, we work with aroma standards and this is based on a mix of again the World Coffee Research Sensory Lexicon and food science literature for descriptive analysis work for other products. We have standards. You can see here that the panelists will smell alongside tasting the coffee. So, they learn to recognize these aromas and flavors in the coffee, and they can start to discuss maybe one panelist thinks something is more brown sugar. One panelist thinks it’s more maple and we have the standard right there. When they have that alongside it, they can start to agree more on what they’re recognizing and move forward with this to select the final attributes for evaluation.

So, for this project, I selected 22 attributes at the five basic tastes, astringent mouthfeel and 17 different flavors. So, as you can see here with this project, we want to look both at flavors that might be more desirable. So, you see some things that you might see on a bag of coffee as a tasting note floral, fruity, honey, berry and some things that you might want to avoid rubber, meaty, brothy things like that because we want to quantify both like the positive and the negative. Because in understanding brewing you both want to know how to maximize the flavors you want and how to avoid the flavors you don’t. Then we work with panelists calibration. So, trying to calibrate everyone to the same level of intensity of these flavors and this is kind of the fun part for me not for my panelists. I spiked the coffee with different things that were on our attribute list. So, we have berry jam, brown sugar. I think that’s lemon juice then we move down to olive oil and can’t see it very well down here, but we’ve got salt, vinegar, MSG for umami flavor and we spike these at different intensities. So, the panellists not only know what they’re recognizing but there is some level of agreement of their recognizing it at this intensity on a scale of 1 to 10.

Then, for data collection moving to the isolated sensory piece. So, they’re no longer at a table discussing with other panelists. At this point, we can assume that they’re calibrated to each other and they’re served in these isolated booths under red lighting so that they are not biased by the color of the coffee and they receive samples in a random order, and they rate all sensory attributes on a sliding scale from low to high. They don’t see the exact numbers that come off of the scale and they rate everything in triplicates, so they come back to three sessions. This gives me the part that probably you’re actually more interested in the results. So, many attributes decrease over the course of the fractions. If you remember you saw a TDS decreasing. It’s pretty predictable that bitter, sour, smoky, burnt, astringent all these attributes, they were decreasing significantly over the course of the brew. This correlated to the TDS decrease. Alcohol and winey as well, my apologies. Some increased. We saw a significant increase in sweetness at the end of the brew and tea floral among some other flavors.

So. this was really surprising and another way we can look at this is looking at how attributes correlated to TDS. So, here we have the bitter correlations to total dissolved solid, Total dissolved solid upon the X-axis, bitter intensity on the Y-axis and you can see pretty clearly as TDS increases, bitterness increases. This would have a positive correlation. Then we can see the correlation between sweet and TDS. As TDS increases sweetness decreases or in the case of this project where TDS decreased over time. As TDS decreases sweetness increases. So, we can look at all of these attributes here and this has the correlation coefficient for each sensory attribute. So, correlation coefficient of close to 1 means a positive correlation.

So, these attributes decrease over the fractions as TDS decreases? So, we have these dried fruit, cocoa, smokey, burnt, astringent, sour, bitter, rubber, salty, umami, alcoholic, winey and meaty, brothy, and vinegar all decreasing as TDS decreases. The ones in the middle don’t have a significant correlation one way or another which I know these look like they’re pretty close but for these particular attributes, there wasn’t enough judge agreement to say that they were significantly correlated in one way or another. Then the last five attributes over here sweet, tea, floral, cereal, honey, and fruity flavors increase as TDS decreases. They have a correlation coefficient close to -1, meaning as TDS decreases they increase so that’s a negative correlation.

So, another way of looking at this is principal component analysis. So, if you’re not familiar with principal component analysis, this is a way of mapping all of the products. So, we see fraction 1, 2, 3, 4, 5, 6, 7, and 8 and the whole brew moving across the space with the attributes. So, these are the attribute loadings and the percentages here define how much of the variability between samples is encompassed by that dimension? So, we see pretty much the first principal component. So, the horizontal direction is where all of the variability lies, and you see the products moving neatly across that first principal component and the arrow length here corresponds to basically how important that attribute is to the differences between samples. So, on the first principal component, we see sourness, alcoholic, winey and bitter most characteristic of fractions one and two. In the middle fractions, three and four as well as the whole brew had a little bit more of the balance between the two directions on principal component 1 and then towards the end, 6, 7, and 8 all corresponding to the tea, floral, honey and sweet attributes. So, this is what these were all the attributes that were significant between the samples in either the early fractions or increasing in the later fractions.

Mackenzie Batali: We can also look at how attributes correlate to other attributes to get an idea of how different sensory properties might co-extract. So, again reminder the height of the correlation coefficient of close to 1 means that there are positively correlated at the correlation coefficients of close to -1 meaning that they’re negatively correlated. So, sweetness and tea floral increase together. Sweetness and sourness, bitterness are not going to be seen together as sweetness increases sourness is decreasing. Similar information to principal component analysis, but this looks specifically at attributes with other attributes.

So, to recap the sensory work. So, moving into the later fractions we had higher sweet, fruity, honey, tea, floral and cereal flavor as the TDS decreased. In the earlier fractions, we had the rest of the basic tastes were higher bitter, sour especially astringency and many of these other flavors including smokey, alcoholic, vinegar, meaty, rubber cocoa and dried fruit and these were the more these were the fractions of the more measurable titratable acidity.

So. after completing this first step of the experiment had to think about why this is happening. And the first concern is sensory carry-over as the concentration of the highly bitter soluble compounds decreases the weaker coffee samples taste sweeter by comparison. You can sort of see this in yourself if you drink lemon juice and then you drink a glass of water the water might start to taste kind of sweet. But the samples are served in a randomized block design. So, all panelists are getting samples in a different random order and every replicate to minimize this effect. So, dismissing this, for now, is a significant part of why this is happening.

Other explanations that we wanted to test was the difference in the mass transfer of different sensory attributes, the associated compounds, and some things might be extracting at different rates or there might be a masking effect. So, even if the compounds that contribute to sweetness, floral, honey flavors, even if they’re decreasing overall, if the sour and bitter compounds are decreasing faster. If the relative ratio changes that might make the apparent intensity increase.

So, to look more into extraction rates. I want to bring back the TDS box plots and bring up something I didn’t mention when I first showed this which was Sample 9. Sample 9 was not taken for sensory evaluation. So, after the 8 fraction, there’s still some coffee dripping out and I collected that. It wasn’t enough for panels to evaluate. It was probably about 20 milliliters of coffee total, but it’s enough for TDS measurements and for chemical analysis and what we consistently saw, and this was also seen in some of the fractionation work done more extensively by some of the engineering undergrads is you do see this significant increase in TDS in the last drip out which makes sense. A slower flow rate means more contact time with the grounds and more material extracting.

26:00 How does brewing time impact the chemical properties of coffee?

Mackenzie Batali: So, with this in mind we can move into the chemical analysis I did. So, we saw sweetness increasing. If sweetness is increasing obvious hypothesis may be the concentration of sugars is increasing. So, we were looking at monosaccharide content. So, monosaccharide are simple sugars that cannot be hydrolyzed into a smaller sugar. They’re building blocks of disaccharides so like sucrose, table sugar, lactose as well as the larger plant material the polysaccharides like cellulose and others that are found in coffee. So, coffee has a lot of monosaccharides. The question is how many of these are actually being extracted into the brew because a lot of these polysaccharides might not be soluble, might not break down.

So, we did two different analyses. We did the total monosaccharide composition which quantifies the monosaccharides that are found in these polysaccharides. We’ve got some polysaccharides here these individual little dots represent these single monosaccharides, disaccharides as well, such as sucrose. as well as the free monosaccharide composition which quantifies the unbound monosaccharides.

This gives us more of a picture into extraction patterns. This would give us a more accurate look at what’s actually contributing to sweetness. So, these are the 14 monosaccharides that our analysis was able to look at, so we’ve got fructose and glucose. These bind together to make sucrose as well as 12 others that are all known to be in coffee to some extent, some more than others and see they’ll have fairly similar chemical structures. They’re all five or six-membered rings.

Yes. So, for our methodology. This is a lot more extensive than you might think measuring sugar would be but yeah, so this is a full week of lab time and we’ve got this we’ve got a state-of-the-art $600,000 instrument and that doesn’t even take into account the cost of the upkeep and maintenance and whatnot.

So, we start with the sample standard preparation. This is a fancy way of saying brewing the coffee and diluting it. Then we have to use trifluoroacetic acid to hydrolyze and release the monosaccharides. So, this is how, for the total monosaccharide analysis we break down those polysaccharide structures. Then we have to derivative these hydrolyzed and free monosaccharides with PMP, which is a labeling compound because as you might know if you’re a roaster sugar breaks down when you heat it. So, to analyze it we need to label these compounds so that they don’t break down on the instrument and then we use this incredible equipment, these ultra-high performance liquid chromatography, triple quadrupole mass spectrometry for compound separation and resolution and extremely sensitive measurement.

So, looking at the results, I analyzed the starting material to figure out, so the grounds are used for the coffee to figure out what was there in the beginning. What we’re working with and predictably, so we saw arabinose, galactose, and mannose is the total monosaccharides and fairly high concentration which from literature makes sense. Coffee polysaccharides are mainly chains of these three monosaccharides: galactomannans, mannans, arabinogalactans, as well as fructose and glucose in fairly high concentrations. Everything else, these zeros mean everything was found in at least some concentration, but these zeros that they were less than 0.5% and that free monosaccharides, don’t see very much mannose anymore. Much more fructose and arabinose and galactose. Then we can look at the relative concentration across the fractions and we see some monosaccharides, some monosaccharide concentrations staying relatively constant, but we also see some that do significantly increase relative to other monosaccharides including galactose and we see a relative increase in fructose especially in fraction 7.

The relative concentration of free monosaccharides across all fractions stay mostly constant. So, we don’t see any particular monosaccharide changing much in terms of what percent of the overall material that is. We see something a little more interesting though when we looked at the absolute concentration. So, these are all monosaccharides in milligrams per milliliter and we see a decrease so the high TDS early fractions have a high monosaccharide composition and then it decreases but it’s the lowest in fraction seven and then we do see a statistically significant increase from fractions of seven, eight and nine.

This is seen even more starkly in the free monosaccharide absolute concentration with more increase with fraction 6 consistently the lowest and then an increase in the fractions seven, eight and the drip-out fraction nine as the flow rate of water slows.

So, we can break this down a little more too to look at specific monosaccharides. So, as I said, we had 14 monosaccharides that we analyzed. I want to look at the five most abundant monosaccharides in these samples to give an example of specific extraction pattern. So, we’ve got fructose and glucose again. This is sucrose in its disaccharide form. You’ve got arabinose, which is 60% of the sweet intensity of sucrose, lactose which is 30% of sweet intensity of sucrose. and mannose which can be perceived as sweet or bitter depending on the isomer. Like I said coffee polysaccharides are mainly these three monosaccharides.

Looking at the total concentration, the polysaccharides. So, you’re having galectins, galactomannans as indicated by the total concentrations of arabinose mass and galactose. They follow the same patterns more or less as the overall concentrations. But interestingly for dose and glucose extract pretty consistently across the brew. So, there’s not a decrease, we’re not seeing a high concentration of fructose and glucose in the early fractions which maybe goes back to the hypothesis of the relative ratio of compounds changing along a higher perceptible sweetness.

Free monosaccharides they’re all in very, very low concentration so these like I said are concentrations in milligrams per milliliter. You don’t see nearly any mannose, galactose and arabinose and now fructose follow the same extraction patterns as the overall concentration and then glucose stays consistent.

Mackenzie Batali: Another way of looking at this is the relative abundance of monosaccharides. So, this might not be a perfect comparison because the chemical analysis is going to analyze undissolved as well as dissolved solids, but this is presenting the monosaccharide concentration as a percentage of total dissolved solids. So, based on concentrations calculated on the mass spectrometer what percentage of what you’re extracting is sugars? We do see in the beginning it is a slightly lower percentage of the overall material than in some of these later fractions. though it is sort of an inconsistent pattern which does merit further exploration into relative extraction profiles over time. It Sparks some questions, but it’s definitely not the full picture and it would be interesting to quantify these relative ratios further.

So, this is an important caveat. So, I’ve talked about all of these sugars and the extraction pattern seen here and it does give us a lot of interesting information to jump off from and create future studies about coffee extraction overtime. But even the total monosaccharide composition, which I’ll remind you includes the polysaccharides that you’re not really tasting and a lot of these monosaccharides that don’t have as much sweet intensity. The entire concentration here is well below the documented human sensory perception threshold for the sweetest sugars, sucrose and fructose that are going to be extracted from coffee. So, basically that clearly means at least with these samples my panelists were not tasting sugar when they were perceiving sweetness.

37:00 Summary and Conclusions

Mackenzie Batali: So, on that note what is happening? So, to recap what we’ve got so far, we have the higher sweet taste fruity, honey, cereal, tea, floral flavors in later fractions, and the early fractions were higher in most of the rest of these attributes including bitter and sour most significantly as well as winey, vinegar and some other attributes. They were more acidic than the higher monosaccharide composition. We did see the lowest carbohydrate concentration in the 6th fraction, then a decrease later but not as high as the early fractions. So, with all of this in mind some hypotheses for what’s causing the increase in perceived sweetness and the increase in these other attributes.

So, we saw this slight increase in extraction of polysaccharides in the later fractions in the seventh and eighth fraction and other higher molecular weight molecules, especially things that are carbohydrate derived like Maillard reaction products might be increasing as well as we saw the relative extraction profile potentially changing increasing the perception of honey and cereal flavors that might be attributed to these products. Then, on that note perception of sweetness could be related to aroma compounds that are generally more associated with sweetness. So, when you taste something that has a fruity flavor, a honey flavor your brain might associate these with things like fruit and honey that do have sugar and you think you’re tasting sweetness when your sweet taste receptors are not actually being stimulated by sugar. It is a perceived sweetness and that as well as the contrast with the more bitter, more sour taste as that decreases. That might cause the increase in perception of sweetness. So, these are some theories based on this work so far.

I want to sort of wrap up with if I could continue this project further. This project is mostly complete for now. It’s being worked for publication and this will be released from the soon. But some jumping off points that I or if you have if you want to do some experiments yourself would love to investigate further is how much is it related to TDS versus extraction changes. Some of our other Sensory Studies have shown a negative correlation as well between sweetness and total dissolved solids. So, if I could do another sensor study on this I would want to dilute that high TDS early fractions and see it to the same total dissolved solids concentration as the late fractions that we received as sweet to compare the perceived sweetness intensity Just based on TDS versus what’s actually being extracted.

Of course, if I had the funding and time I would love to analyze the entire chemical composition. But as I showed you that can be both time and cost prohibitive, unfortunately. We did not hit a super high percent extraction in this in this experiment. Our overall percent extraction was I think about 17%, so on the low side and as we saw in that 9th fraction, there was an increase in extraction. So, there’s more material to come out and I would it would be interesting to see what happens just if you keep going. Just keep pouring water on the ground to keep taking fractions every 30 seconds, and that what’s happening at 9, 10, 15 fraction. Another thing that I want to know is we saw that increase in monosaccharide extraction at the point where the flow rate decreased and we’ve done some work on different pulsing cycles on these batch bars and it could be very interesting to explore how changing these pulsing cycles water on, water off to give decreases in flow rate and an earlier point in the brew if you slow it down slow it way down to fraction three or four to see if that changes the extracting profile as well. So, those are some jumping off points, some questions I would love to answer in the future if I had the opportunity.

With that I’d like to acknowledge my colleagues at the UC Davis Coffee Center, the students with Dr. Carlito LeBrilla who helped me with chemical analysis. We had donations of the brewers from Wilbur Curtis Co. and Java City Coffee Roasters donated the coffee and most importantly I have to thank the Specialty Coffee Association and underwritten by Breville for the funding for this project would not be possible without that financial support.

42:30 Outro

Heather Ward: That was Mackenzie Batali at Specialty Coffee Expo in April 2019. Remember to check our show notes for a full episode transcript of this lecture and a link to coffeeexpo.org for more information about this year’s event.

This has been an episode of the SCA Podcast’s Expo Lecture Series, brought to you by the members of the Specialty Coffee Association, and supported by SAP’s Softengine Coffee One. Thanks for listening!

ListenSCA StaffDecember 9, 2019Research, Expo 2019, Brewing Fundamentals

#82 | Sensory and Chemical Explorations into Drip Brew Coffee Flavor Over Time | Expo Lectures 2019

Special Thanks to Softengine Coffee One, Powered by SAP

Related Links

Table of Contents

Full Episode Transcript

#MakeCoffeeBetter