The Wayback Machine -
Feature Science in the News Stereolab PubMeditation Culture Dish Sportsology Laberrations The Mad Gastronomist Mass Spectator LD50 Home Home

Science on a Grande Scale

By Pauline Fujita | 5.14.07

There are few places in America that a person of my stature (5’2”) could call herself tall. Enter the French-Italian stillborn language of Starbucks, Fre-talian. In an attempt to infuse the American coffee experience with a European mystique, Starbucks borrows token bon mots like grande from French and venti from Italian. The country where tall means small remains a mystery. The “venti triple nonfat mocha no whip latte,” however, is an American beast that would surely mortify any self-respecting European coffee connoisseur. So where has Fre-talian gotten us? Completely confused about how good coffee is really made.

After many years in grad school I think I’ve had at least as much coffee as science. Despite this devotion to caffeination, I realized recently that I know very little of the actual science behind the making of coffee. We’ll begin with the biology of the coffee plant, because I am first and foremost a biologist. Then we’ll delve into the chemistry of the compounds that make good coffee good. Finally we’ll look at the physics with the transfer of heat during roasting and we’ll learn how to use the force to make the tastiest sized coffee grinds.

Biology: The Plants

The coffee beans that we are familiar with are actually just the dried, roasted seeds of coffee fruit (Figure 1). Once the fruit is ripe it is dried and and removed, leaving the raw bean, green coffee.

There are two main species of coffee grown today. The first, Coffea arabica, also known as Arabica coffee accounts for 68-70% of the world’s coffee production (Anzueto et al. 2005). C. arabica is grown mostly in Central and South America and parts of Eastern Africa in cooler climates at medium to high altitudes (1000-2100m). The second, Coffea canephora, or Robusta coffee is grown mostly in Africa and South Asia in the warm humid climates of tropical lowlands and foothills (100-1000m).

To the average coffee consumer varietal names will seem more familiar. Although Hawaiian Kona may taste very different than Ethiopian Yerga Chaffe, genetic analysis has placed them in the same species, C. arabica, along with other familiar varieties like Typica and Bourbon, grown worldwide, Jamaican Blue Mountain, and Ethiopian Harar, among many others (Anzueto et al. 2005).

Another way in which C. arabica and C. canephora are different is in their ploidy, the number of copies of the genome present in each cell. C. canephora has the usual two copies (diploid) of its genome, common to all other species in the genus Coffea, but C. arabica is allotetraploid, it has two pairs of genome copies to a total of four copies of its genome. It is thought that C. arabica may have originated from a spontenous cross between two diploid species, followed by a duplication of the two genomes. And speaking of genomes, there is in fact a whole website dedicated to coffee genomics, A project to sequence the whole genome is currently underway.

Chemistry: Caffeine, Aroma, and Roasting

Drip Coffee vs. Espresso – Round One: Caffeine

Getting down to brass tacks, the coffee fact of most interest to many is one of caffeine content: Which has more caffeine – espresso or drip coffee? According to a study done by Bunker and McWilliams (1979) a 7 oz cup of coffee has more total caffeine but is more dilute than a shot of espresso:

Beverage Amount of Caffeine (mg)
Drip coffee (7 oz) 115-175
Espresso (1.5-2oz) 100
Instant coffee 65-100
Decaf, brewed 3-4
Decaf, instant 2-3
Tea, iced (12 oz) 70
Tea, brewed, imported 60
Tea, brewed, U.S. 40
Tea, instant 30
Mate 25-150

Table 1: Various caffeine contents. (from Bunker and McWilliams 1979).

So if you’re pulling that all-nighter and need a concentrated caffeine fix, espresso shots are the way to go.


I’ve always been suspicious of decaf coffee. Friends say I’m just being a coffee snob and those that know me even better say I’m just a caffeine addict. Now that I’ve looked into the decaffeination process, however, I stand firmly by my snobbery.

Most decaffeination processes involve soaking the green coffee beans in water to dissolve the caffeine, removing the caffeine from the water, and then putting the flavor molecules “back” into the beans by re-soaking them in the treated water ( Generally a solvent (like ethyl acetate or methylene chloride) is used to remove the caffeine. These solvents have low boiling points and are relatively inert so they can be boiled off from the water. One exception is the “Swiss Water Process” where caffeine is removed using activated carbon filters. Just as these processes are unable to remove all caffeine, they also fail to put all the flavor molecules back, not to mention the treated water contains traces of solvents that get added back to the beans. I’ll take my untreated beans any day, especially since they come with free caffeine.

Drip Coffee vs. Espresso – Round Two: Aroma

Espresso based drinks aren’t just for yuppies and Europhiles anymore. Looking at the profile for aroma producing compounds in figure we see that they are present in greater concentrations in espresso. Is espresso tastier? Perhaps the proof is in the pudding, or in this case, the chromatogram.

Roasting: Is it hot in here or … ?

Most of the aroma we associate with coffee is created during the roasting process. Longer roasting times mean coffee that is more bitter and less acidic and darker in color (Fortin 1999). Green, or un-roasted coffee contains about 300 volatile organic compounds (Bonnländer et al. 2005 pp. 198) whereas over 1000 such compounds have been found in roasted coffee. The green bell pepper-like “aroma” of green coffee can be attributed primarily to the compound isobutylmethoxypyrazine. In contrast, the aroma of roasted coffee is thought to result from a combination of about 25 volatile organic compounds, the “aroma compounds”, found at a total concentration of only 1g/kg of coffee and ranging in individual concentration from the lower part per million range down to as little as parts per trillion.

So where do all these extra compounds come from? During the roasting process many different chemical reactions occur, the most important of which can be classified as one of two types of reactions. The first, Maillard or “browning” reactions, produce aroma compounds as well as colored compounds (melanoidins), and the second, caramelization reactions, involve the chemical reduction of sugar compounds, the same tasty process that, you guessed it, makes caramel.

Physics: Heat Transfer and Grinding

Would you like some physics with your coffee?

Sounds easy enough. We heat green coffee beans and get flavorful coffee. But like so many seemingly simple tasks, the science of roasting coffee is actually quite tricky. The roasting process we’ve been talking about takes anywhere from 90 seconds to 20 minutes depending on the roasting method used and involves temperatures ranging from 20 °C to 220 °C. The aroma developing reactions typically occur at temperatures upward of 160 °C. What’s so tricky? We have to heat the green beans to over 160 °C in a short window of time and in such a way that they heat evenly. Suddenly we find ourselves in the deep-running still waters of heat transfer physics.

Like any good physics problem we begin by assuming something is a sphere, in this case a coffee bean. Not such a bad approximation considering I’ve heard that card-carrying physicists will assume any shape is a sphere. The difficulty is that a bean does not have a uniform heat capacity. As it begins to dry and roast, the outer dry parts heat faster than the inner green parts, thus exacerbating the problem of heating the inside of the bean enough while being careful not to over roast the outer parts. We can even write an equation for heat energy, Q, transferred to the bean:

Q = α ∫ SA * Tdiffdt

Where α is the heat transfer coefficient – a measure of what proportion of heat is transferred and which is specific to the material, in this case a coffee bean. SA is the surface area of the bean and Tdiff is the temperature difference between the air in the roaster and the bean.

A partial solution to the problem is to have the beans constantly mixing while being heated. There are complex machines built to heat the beans and mix them with precision. Sadly these are beyond the financial means of a grad student. But this doesn’t mean we can’t try to roast our own coffee anyway. In fact many people use popcorn poppers to roast their own small quantities of coffee.

Bumping and Grinding: Why size really does matter

Especially in the case of espresso, the distribution of sizes of coffee grinds can mean the difference between a spiritual coffee experience and trauma for your taste buds. The size of the grinds determines the surface area of contact between the coffee and the hot water or steam being used to extract the tasty compounds from the coffee. Figure 3 shows the microstructure of coffee grinds, up close and personal.

What we really care about is extracting the coffee the right amount. If the water is too hot, the grinds too fine, or the brewing goes on for too long, the coffee will be over-extracted. This means that compounds that are more difficult to extract start making an appearance in our over-extracted coffee. Unfortunately these are often the bitter, astringent and unpalatable compounds.

Grinding is usually done by one of two methods: impact grinding or gap grinding (Petracco 2005). Impact grinding is the breaking of coffee beans by collisions and is the principle behind cheap grinders with a spinning blade. Gap grinding involves passing the coffee between two moving parts usually wheels with teeth. A limitation of collision grinding is that it cannot produce fine grinds and the size distribution of grinds is hard to regulate. This is particularly important for espresso where both fine and coarse particles are needed for a good brew. The small particles are essential to get sufficient extraction from the brief blast of steam used in espresso making. The larger particles are necessary to form a sort of sieve structure within the espresso cake that allows the steam to pass efficiently through the grinds. If the grinds are too tightly compacted the espresso will be under extracted.

Protocol for a Good Cup of Coffee: Judge a book by its cover

“Black as the devil, hot as hell, pure as an angel, sweet as love.” – Talleyrand

This quote from an eighteenth century French diplomat is certainly not quantitative enough for the analytical mind, although it is catchy. For all that science we are still wondering how to improve our cup of coffee. Like any good scientist I don’t know the right answer, but I do know lots about the wrong answer. Figure 4 shows three espressos, 4b being perfectly extracted. Any error in grinding, percolation, temperature or extraction level will be reflected in the appearance of the espresso foam. A perfectly extracted espresso will have the characteristic “tiger skin” foam seen in figure 4b, formed by tiny gas bubbles and cell wall fragments from the extracted beans.

An under-extracted espresso will lack foam while an over-extracted espresso may have dark foam or white foam from bubbles that are too large.

As for improving your filter or French press coffee, I can provide some storage tips. For long-term storage (greater than two weeks) put it in the freezer in an airtight container as freezing can slow the “staling“ reactions that destroy compounds important for that fresh coffee aroma. Air and water are the enemy. When potent sulphur compounds in the coffee that contribute to aroma come into contact with air they begin oxidizing. Oxidation of lipid compounds also leads to rancid tastes and smells. If you are consuming the coffee sooner than that room temperature, air tight storage is recommended, as the freezer also contains increased levels of water vapor and renegade smell compounds from other things in your freezer (fish anyone?). Finally, storing the coffee as whole beans helps as the structure of the bean partially protects it from reacting so badly with air.

One of my all time favorite coffee quotes is from the famous mathematician Paul Erdös: “A mathematician is a machine for turning coffee into theorems.” He is so famous in fact, that those in the math world play a variant on six degrees of Kevin Bacon: they calculate their Erdös number. An Erdös number of 1 indicates coauthorship with the number theory guru and a number of 2 indicates that you published with someone who published with Erdös. Small Erdös numbers are big status symbols. Sadly I think my number is close to infinity, so at best I might be a machine for turning coffee into email. Now that I’ve discovered a little of the science behind good coffee though, small doesn’t seem so bad. At the very least it’s tastier than tall where coffee sizes are concerned.


Anzueto F, Baumann TW, Graziosi G, Piccin CR, Söndahl MR, van der Vossen HAM. 2005. “The plant.” In: Illy A, Viani R. eds. Espresso Coffee: the science of quality. 2nd ed. London: Elsevier Academic Press.

Bee S, Brando CHJ, Brumen G, et al. 2005. “The raw bean.” In: Illy A, Viani R. eds. Espresso Coffee: the science of quality. 2nd ed. London: Elsevier Academic Press.

Bonnländer B, Eggers R, Engelhardt UH, Maier HG. 2005. “Roasting.” In: Illy A, Viani R. eds. Espresso Coffee: the science of quality. 2nd ed. London: Elsevier Academic Press.

Bunker ML, McWilliams M. 1979. Caffeine content of common beverages. J Am Dietetic Assn. 74(1): 28-32.

Fortin F (ed). 1999. The Complete Food Guide. Montreal: QA International.

Hoffman P. 1998. The Man Who Loved Only Numbers: the story of Paul Erdös. New York: Hyperion.

Maughan RJ, Griffin J. 2003. Caffeine ingestion and fluid balance: a review. J Hum Nutr Dietetics. 16(6): 411-20.

Nicoli MC, Savonitti O. 2005. "Storage and packaging." In: Illy A, Viani R. eds. Espresso Coffee: the science of quality. 2nd ed. London: Elsevier Academic Press.

Petracco M. 2005. “Grinding.” In: Illy A, Viani R. eds. Espresso Coffee: the science of quality. 2nd ed. London: Elsevier Academic Press.

Petracco M, Viani R. 2005. “Coffee consumption and health.” In: Illy A, Viani R. eds. Espresso Coffee: the science of quality. 2nd ed. London: Elsevier Academic Press.

Philip P, Taillard J, Moore N, Delord S, Valtat C, Sagaspe P, Bioulac B. 2007. The effects of coffee and napping on nighttime highway driving. Ann Int Med. 144(11): 785-91.