Things I don’t understand #3214

I am not particularly ashamed of the phrase “I don’t know” but there comes a point in the day when you’ve said it five or six times and you feel you really ought to do something about it.

The cause of my embarrassed ignorance:  the change in flavour when coffee cools.

The change in a coffee as it cools is familiar to anyone, especially those who’ve cupped a lot.  I think I might have made the rookie error of associating the change with the cupping process too much – the continued extraction, the constant slight agitation of spoons.  I hadn’t really thought much more about it until customers at Penny University started asking and I realised the change was independent of brew method, filtration type (metal, cloth or paper) and common to all coffees.  Clearly something else is going on here.

I understand a few things about how temperature affects taste – the classic example being Coca-Cola.  Cold Coke is (shamefully) delicious.  Warm Coke is too sweet.  The amount of sugar hasn’t changed, merely our tongues capacity to detect it.  Though with coffee it clearly isn’t simple sugars, and one also experiences changes in taste, flavour and mouthfeel.

Perhaps there is something in the books that I own and I may have missed it, or maybe there are some good archived discussions online I haven’t seen.  It just seems like this is something important in coffee that we all talk about, enjoy and appreciate but don’t really understand.

Links, insight, indepth technical explanations, invitations to seminars in exotic locations and casual abuse for my ignorance all welcome!

35 Comments

  1. I am pretty sure there is something about this in Brillat-Savarin – and maybe McGee as well.

    If my memory is not flawed, it’s a combination of factors. These would include:

    Nerves associated with taste have been shown to have thermal sensitivity. There was a research project I read years ago that showed that applying heat or cold to specific areas of the tongue could cause a perception of taste. For example, where warming a specific area caused a sensation to be registered as “Sweet” or cooling another caused a perception of “Sour”.

    There is also taste receptor interaction / interference. While some molecules always interact well with the associated receptors, others have a “weaker” interaction and bond. Some of these molecules interact better at higher temperatures from what I understand. And I think it might be possible that others interact better at lower ones (but it’s been a while sense I read this stuff).

    Changes in convection change density. As a liquid cools, convective currents diminish and then cease. This causes the liquid to separate and layer by density. As a result, the composition of each sip / spoon starts changing over time.

    And finally, of course much of what we perceive as taste is actually aromatics. And aromatics are volatile and will change at different temps (the mix of volatiles will change) resulting in changes in smell as the coffee cools – changing the perceived flavour.

    I don’t think it’s any one of these but rather the combination that creates the (often radical) changes in perceived flavour as temperature of coffee changes. The fact that it’s a combination like this makes understanding it incredibly complicated. Daunting, for me at least.

  2. I don’t have a full answer, or even a sure answer, but I would assume that different chemicals are released gasseously at different temperatures ergo affecting the taste. The effect on body and mouthfeel are different, maybe oils joining together to form longer strings in the less active liquid?

    The aroma may explain why coke becomes sickly at room temperature and why beer becomes unpleasantly bitter likewise.

    Maybe an experiment would be to try tasting at different temperatures with a blocked nose (or comedic clothes peg)?

    And as I’m pulling this out of the air (no pun intended) I could be horrifically wrong

  3. First off, love the honesty and humility coming from one whom many of us consider somewhat a genius when it comes to coffee… That is refreshing!
    Secondly, my completely uneducated guess of a theory – in that perhaps there is a temperature, neither too hot nor too cold, at which human flavour receptors are at their peak? So when a drink is above that temperature, the brain registers the heat before the flavours, so the heat takes priority? The same would apply to extreme cold – as in cold coke, the brain registers the cold temperature in priority over the flavours (which in the example of coke is a good thing, as when the temperature settles out, the extreme sweetness which has always been present is just more distinguishable)
    As I said, a thory with no scientific backup, rather just my whimsical ideas… But a theory nonetheless!

  4. I don’t know, but:

    “After a century of investigation, the role of temperature on taste perception is still unclear. The general view is that perception is optimal at normal mouth temperature. For example, cooling reduces the sweetness of sugars and the bitterness of alkaloids (Green and Frankman, 1987). However, the bitterness (and astringency) of tannins is well known to be more evident at cool temperatures. This apparent anomaly may relate to the different receptors involved in tannin and alkaloid bitterness.
    Another important factor affecting taste perception is PH. It both directly influences the ionization of salts and acids, and indirectly affects the shape and biological activity of proteins. Structural modification of receptor proteins on gustatory neurons could significantly affect taste responsiveness.”

    Jackson, Ronald S. Wine Tasting: A Professional Handbook. San Diego, CA: Academic Press, 2009. 140-141. Print.

  5. Whoops. Good point Ben – I forgot about the pH issue (and don’t remember the science on that at ALL anymore).

  6. Lots of interesting stuff here! Thank you!

    I’m not sure I agree on the mouthfeel issue – I am pretty sure that you could swirl the vessel/stir before each sip and still see a big change in mouthfeel through cooling. A simple experiment to run, if I can coerce someone into helping me run it blind.

    The idea of the mix of volatiles is interesting. I guess what bugs me about it is that when coffee is hot there should be more volatile compounds available, compared to cooler temperatures. The inverse (when it comes to perception) appears to be true, with complexity increasing as temperature decreases. Very odd!

  7. I think the oils are unlikely to play a role as this change is noticeable in paper filtered coffee that doesn’t really let any suspended oils through into the brew. In fact it has been more noticeable for me with some paper filtered coffees than with other filtration methods (which is a bit confusing, and perhaps a result of my own psychological flaws!

  8. I was talking to the Singapore barista champion, Keith Loh, at the show last week and he said something that stuck with me: “If you act smart then no one will ever teach you anything.” A good motto to live by!

    I think your guess is probably right, and is inline with what Ben posted below. It is interesting that improvements to the coffee continue even as the coffee cools below body temp down to room temperature. But that might simply be “interesting” rather than important!

  9. pH is really interesting and makes me want to buy a decent meter and see if anything correlates in terms of cup quality, temperature and pH.

  10. I don’t think eliminating one of these (swirling the cooled coffee for example) is going to result in the flavour not changing. As I noted – I think it’s the combination that is the big issue.

    It’s not a matter of more or less aromatics – but rather a different mix of volatiles.

  11. Solubility, at least in the case of carbonated beverages, and this possibly affects crema as well. (CO2, or any gas, is more soluble in cold water, so you end up with more carbonic acid and more “bite” to balance the sweetness of the Coke. Warm Coke also goes flatter faster.)

  12. It is very annoying when you catch yourself looking for a simple answer to a question, when you know the truth is far more nuanced and complex!

  13. precipitation. as a solution cools, some of what might have dissolved just fine at a higher temperature precipitates out. that is part of why there is almost invariably some sludge at the bottom of a slow-sipped coffee mug regardless of brew method.

    I’ve long suspected (but am just theorizing) that some of that solute acts as an inhibitor to the perception of sweetness and as it starts to leave solution during cooling it changes the mouthfeel and taste – and often in a positive way.

  14. Here’s an interesting addition to the discussion: try brewing a pot of chemex, pouring most into your cup but leaving some in the bottom, which will inevitably cool faster. Drink enough of the coffee in the cup that you’re right on track with it’s cooling stages and can predict what the next sip will be like. Then pour the cooled coffee into the cup.

    Surprisingly, you get an addition of flavors that you’d normally find at the end of cooling. I’ve only done this two or three times, and never by design but it definitely works. I, too, am completely stumped.

  15. So if one triple filtered, or filtered better, or filtered at different temperatures – then one would probably be less confused than he was when he started typing this sentence.

    I’ve had some brews with absolutely no precipitate – I now want to go and play with papers and filtration to see what happens.

  16. bear in mind that you can also filter things _out of_ solution. there is quite a bit of gradient between soluble and insoluble.

    The coffee community needs a full time food chemist who can chime in on this sort of thing. I feel like most of my hamfisted research in this stuff just begets more questions, and many potentially useful experiments fall by the wayside.

  17. I’m having a hard time wrapping my head around this one. Wish I were smarter ’cause it’s really complex now.

    The whole “taste changes are different depending on speed of cooling” really throws a monkey wrench into everything.

    I wish I had a real food scientist handy (or was a real food scientist).

  18. Unless I’m mistaken, I think my coffee cuppers handbooks says that chlorogenic acid, being fairly unstable, starts to break down as temperatures cool and becomes quinic and cafeic acid. That’s the only big thing that jumps in my mind, but I’m sure there are more that are somewhat related in that sense.

  19. Soluble solids are part of the liquid at a molecular level though. If you clarify through paper a dozen times, you should have the same resulting liquid (measured solids) as filtering through paper twice. The only difference is that you will either: A. have a liquid that tastes of paper filter or B. have a liquid that has been diluted from the water already present in the filter from rinsing.

    Am I wrong or am i wrong?

  20. The SCAA Brewing Handbook (Lingle) provides much of the answers to this, and other related questions. It’s sort of my secret weapon when discussing coffee brewing science… a lot of info packed into that thin little booklet!

    The compounds in coffee each have their own characteristics, including a certain stability (or lack thereof) within certain temperature ranges. A great example is the one James mentions above: at the 75-85*C range (roughly, from memory), chlorogenic acid readily breaks down to its component caffeic and quinic acids, both of which are generally received as “less pleasant” tasting organic acids. Below that temperature range, the de-esterization rate of the chlorogenic acid is much lower, and is therefore preserved.

    The smaller the quantity of coffee (beverage), the generally faster it will cool. If you were to brew a couple liters of coffee, you’d find a noticeably different array of flavors as it cooled to room temperature, by simple fact that the coffee will have spent more time in certain temperature ranges because of the thermal mass and slower cooling.

    This, and the before-mentioned relationship between taste perception and temperature, essentially answers your question. I’d again vigorously recommend the Brewers’ Handbook, though I’d also recommend trying this simple experiment: taste some coffee hot, let the coffee cool, taste it, and then throw it back in the bottom flask of your siphon and heat it up. See how it tastes both hot and cool again. I’d suspect the four samples will each yield a distinct taste experience. Similarly, you could brew a large-ish (liter?) quantity of coffee, separate a small cup of it, and taste both the small cup and the main brew as the two cool. Anyway, that’s my two cents on the subject.

    Yay for coffee nerds!

  21. I”m talking purely as informed guess work, but the comments so far seem to capture a lot of the issues – but particularly Nick’s comment. Part of the complexity of investigating taste is the inability to separate variables. It will potentially matter not just what temperature it’s at, but also how long it took to reach that temperature, how long it has spent at different temperatures, how it was cooled, whether that cooling can be considered to be equal across the entire liquid, whether the cooling allowed greater access to air (or other substance for chemical interaction) and whether the cooling and the pouring off (for tasting) procedures accidentally impacted in another way (e.g. Extra decanting of the liquid) – to name just a few things. We’re also dependent upon a measuring system which is subjective – so psychology, mood, expectations etc. which change with time might impact it.

    That said, I think Nick, James, Ben and the other commenters have probably identified the main factors. A big thankyou for raising this and to everyone who’s commented – I know a bit more about this now, and have many more questions :)

  22. One more thing that has a dramatic effect on your ability to taste coffee over time: sensory plasticity. This concept was introduced to me by food scientist Dr. Terry Acree from Cornell, and it blew my mind as a coffee taster. Two concepts are important here: adaptation and habituation.

    Adaptation has to do with how we perceive aromas and tastes. Here’s a simplification about how aromas work: there are areas within your aromatic receptors that are designed to detect specific odor chemicals. Think of them as a hole of a specific shape, ready to accept a peg of the same shape. Once the peg “fits” into the hole, your body knows that particular chemical has been detected, and sends the signal to your brain. As more of that hole-shape get filled with more pegs, your brain knows there is more of that aromatic chemical in the environment, and the smell appears “stronger”. Problem is, it takes a while for these receptors to clear themselves out, and once all the “holes” are filled with “pegs”, you can no longer perceive that particular aroma. In practice, once you hypersaturate your nose with a particular aroma, you actually physically lose the ability to perceive it for a while. Dr. Acree demonstrated it for me in this way. He showed me a Sauvignon Blanc from New Zealand which had a strong green bell pepper note in the nose. After a sip, he then handed me a small piece of actual green bell pepper, which I chewed and swallowed. I then returned to the wine, which tasted DRAMATICALLY different. The green pepper note was gone, and I could taste nuances in the wine that were previously obscured by the green pepper note. Amazing. We then repeated the experiment with a Gverstraminer and a lychee. Similar result. Dr. Acree explained that this is a major factor when tasting anything over time- you gradually become “adapted” to the tastes and aromatics, and your body loses the ability to perceive some of the flavors, making other less-intense flavors seem to “appear” out of nowhere.

    Habituation has a similar effect, but happens in your brain. There are all sorts of stimuli around us all the time- sounds, smells, sights, etc. Our brains learn to screen certain stimuli out, once we have registered them. This happens dramatically with smells. For example, when you walk into your house, you may smell the odor of floor cleaner. Within a few minutes, you can no longer smell the cleaner. The aroma hasn’t disappeared- it’s just that your brain is screening it out. If you walk outside, and “de-habituate” yourself, you can return and smell the cleaner again.

    With both habituation and adaptation, our “instrument” (our sense of taste and smell) changes over the course of our interaction with the flavorant (in this case, coffee). In other words, our ability to perceive the coffee is changing, even as the coffee itself is changing (as has been outlined above). Our mouth and throat become coated with coffee, and the vapors rise retronasally through our nasal cavity, and continue to stimulate our taste buds and aromatic receptors. As we become habituated and adapted, previously hidden flavors seem to “emerge” over time.

    In other words, as the coffee itself is changing over time, and so are we. The first time you taste a coffee, it will taste different than the second or fourth our eighth. Step away, some of the habituation and adaptation clears up, and returning to the coffee it seems radically different again. It’s really difficult to separate what is intrinsically changing about the coffee (due to changes in aromatic volatility, chlorogenic acid breakdown, etc.) and what is changing about our perception of the coffee.

    Peter G

    p.s. I couldn’t agree more that we need more food scientists “at the table” in the coffee industry. Dr. Acree himself is enthusiastic about coffee, and he attended the Symposium last year. That’s not enough, however, which is why we are working on getting the Global Coffee Quality Research Initiative- which could actually fund academic research on coffee quality- started. Apologies for the shameless plug, but we’re going to need a lot of support throughout the community in order to really get this thing going. We’ll get food scientists involved in quality coffee as soon as we can fund and direct their research- plain and simple.

  23. Great comment thanks Peter G very interesting

    I hadn’t thought to link the adaptation and habituation process to the transformation of flavour experienced during cooling.

  24. all of this begs the question….are we going to start serving coffees, or at least recommending them at a certain temperature? Could it be argued that the flavor change from 190′ishF to say 98.6F brewed temp. is greater than the flavor difference of brewing at 195F instead of brewing at 200F ? And if so, isn’t cooling effect on flavor more affecting, or at least noticeable than the temp we are brewing at? Although, it wouldn’t be since the former affects the latter.

    Art and Science seem to be colliding at this point…

  25. I hate to admit I never had any idea that the flavor changed with the temperature. I suppose I never really thought about it but reading all of your posts is really enlightening. I’ll have to do more research on this. Let me know if any of you would like to write a post on my blog on this topic!

Submit a comment