Chapter 21 — Exercises

🍳 Kitchen Lab 21.1 — Cool the kettle, save the tea

Concept demonstrated: Tannin extraction is more sensitive to water temperature than caffeine extraction, especially for delicate green teas.

Time: 20 minutes Cost: ~$3 if you have to buy tea Allergen flags: None for unsweetened tea. If you add milk, dairy.

Materials: - Loose-leaf green tea (Japanese sencha or Chinese long jing/dragonwell), 4 g (about 2 teaspoons) - Two identical mugs or glass cups (clear glass is best for observation) - A kettle and water (filtered if your tap water is hard) - A thermometer (optional but illuminating) - A timer (your phone is fine) - Two teaspoons or a tea infuser

Protocol: 1. Bring water to a full rolling boil in your kettle. 2. Pour 240 mL (1 cup) of the just-boiled water (about 100°C / 212°F) into Mug A. Note the time. 3. Set the kettle aside and let it rest, uncovered, for about 90 seconds. The water will fall to roughly 80°C (175°F). If you have a thermometer, verify. 4. Pour 240 mL of the cooler water into Mug B. 5. Add 2 g (1 teaspoon) of tea leaves to each mug at the same moment. Start a 2-minute timer. 6. At 2 minutes, remove the leaves from both mugs at the same time. 7. Let the cups cool to a drinkable temperature (about 5 minutes). 8. Taste them side by side. Note color (Cup A often appears slightly more saturated/yellow), aroma (Cup B often smells fresher and grassier), and the dominant flavor on the back of your tongue (Cup A often more astringent, Cup B sweeter).

Expected result: Cup A (boiling water) tastes harshly astringent, with a bitter, tannic finish. Cup B (rested water) tastes sweeter and more floral, with much less back-of-tongue grip. Both contain similar amounts of caffeine; the difference you taste is almost entirely in the tannin extraction.

Troubleshooting: - Both cups taste the same. Did you let the water rest long enough? On a cold day, water cools faster; on a hot day, slower. Check with a thermometer if you have one. - Both cups taste bitter. The leaf may be over-old (oxidation degrades the green-tea profile, and what's left is mostly tannin) or you may have used too much leaf. Try 1.5 g per cup next time. - Both cups taste weak. Use more leaf, or steep a little longer.

Classroom adaptation (Pat's notes): Inexpensive grocery-store green tea works for the demonstration; the temperature effect is even more dramatic with cheap, broken-leaf tea than with high-quality whole leaf. Use 1 g sample servings (about ½ teaspoon) per pair of students. Keep the kettle in the front of the room and have one student team monitor the cooling water with a thermometer for the class. Total cost for a class of 28: under $20.

🍳 Kitchen Lab 21.2 — The extraction triangle (coffee)

Concept demonstrated: Grind size is the most powerful control on coffee extraction. Under-extraction tastes sour; over-extraction tastes bitter. You can find your home machine's sweet spot by triangulating.

Time: 30 minutes Cost: ~$0 beyond your normal coffee setup Allergen flags: None.

Materials: - Freshly roasted coffee (within 4 weeks of roast date), 60 g (about 2 oz) - A burr grinder, or access to one (a blade grinder will not give consistent results — you can still try, but expect noisy data) - Pour-over or drip brewer of any kind - Three identical cups - Filtered water - A scale that reads in grams - A timer

Protocol: 1. Set up your normal pour-over routine. Note your usual grind setting; we'll call it grind 0. 2. Grind 18 g (0.6 oz) of beans at grind 0. Brew with 300 g (about 10 oz) of water at 95°C (203°F), pouring in your normal pattern. Note how long it takes for the brew to finish. 3. Set the result aside in Cup 0. 4. Repeat with 18 g of beans at one setting coarser than your normal — grind +1. Brew identically. Note brew time. 5. Repeat with 18 g of beans at one setting finer than your normal — grind −1. Brew identically. Note brew time. 6. Let all three cups cool to roughly the same temperature (about 5 minutes — taste warm, not hot). 7. Taste them side by side, in this order: 0, +1, −1.

Expected result: Cup +1 (coarser) likely tastes thinner, more sour, less developed. Cup −1 (finer) likely tastes heavier, more bitter, possibly slightly muddy. Cup 0 sits in between. Whichever direction tastes better than your normal tells you which way to drift. If +1 tasted best, your home brew is over-extracted; coarsen up. If −1 tasted best, you're under-extracted; grind finer.

What you've learned: You don't need a refractometer to dial in coffee. You need three cups, the discipline to change one variable at a time, and your tongue.

Variation for the curious: Repeat the experiment changing water temperature instead of grind. Brew at your normal grind, but with water at 95°C, 90°C, and 85°C. The hotter brew will extract more (bitterness rises); the cooler brew will extract less (sourness rises). The lever is different but the underlying physics is the same — diffusion rate scales with temperature.

🍳 Kitchen Lab 21.3 — Henry's law in your hand

Concept demonstrated: Carbon dioxide solubility depends strongly on temperature. A cold sparkling drink holds gas; a warm one releases it.

Time: 15 minutes Cost: ~$3 for a bottle of plain sparkling water Allergen flags: None.

Materials: - Two identical sealed bottles of plain sparkling water (12 oz / 350 mL is fine; brand doesn't matter) - A refrigerator - A warm spot (countertop, sunny windowsill — about room temperature, 20–22°C / 68–72°F) - A clear glass

Protocol: 1. The night before, place one bottle in the refrigerator. Leave the other on the counter at room temperature. 2. The next morning, observe both bottles. They should look identical — both clear, both sealed, both with no visible bubbles. 3. Open the cold bottle. Listen for the hiss; pour about 50 mL (1/4 cup) into the glass; observe the rate at which bubbles rise. Recap firmly. 4. Open the room-temperature bottle. Listen to the hiss — it should sound sharper, more aggressive. Pour 50 mL into a clean glass; observe; the rate of bubble release will be visibly faster, the foam more aggressive, and the drink will go flat noticeably sooner. 5. Taste both. The cold one is fizzy and sharply carbonic; the warm one is foamy at first sip but rapidly flat.

Expected result: The warm bottle releases CO₂ much faster than the cold one. Both started with the same amount of dissolved gas; the warm one couldn't hold it.

Why: Henry's law constant for CO₂ in water rises with temperature (gas solubility falls as temperature rises). At 4°C (refrigerator), water holds about 1.45 g of CO₂ per liter at 1 atm. At 20°C (room temperature), only about 0.84 g/L. The colder liquid simply has more capacity to keep gas dissolved.

Discussion questions

1. Tea processing fork. A green tea and a black tea start as the same leaf. What single chemical event is the dividing line between the two, and how is it controlled? Why is it inaccurate to say that black tea is "fermented"?

2. The bitter-vs-caffeine myth. The claim that "a one-minute steep gives you a low-caffeine cup of black tea" is widely repeated and largely false. Why? Which compounds extract on which timescale, and what does the short-steep cup actually have less of?

3. Coffee extraction directions. Your morning pour-over has tasted progressively sour for the last week. List three independent levers you could pull to push extraction higher (and therefore less sour). Rank them by how much each would change the cup, in your judgment, and defend your ranking.

4. Shaken vs. stirred. A martini is conventionally stirred, a margarita is conventionally shaken. Translate the convention into chemistry: what does each technique actually do to the drink in the glass?

5. Wine pairing as accumulated wisdom. A Tuscan Chianti is famously paired with a Tuscan ragù. Without invoking history, what about the chemistry of each makes the pairing work? Why does the same logic not lead a sommelier to pair a Chianti with a sushi platter?

6. Henry's law in beverages. A bartender claims that putting a champagne stopper on a half-finished bottle and refrigerating it overnight will keep the wine fully carbonated for the next day. Evaluate this claim. Which parts are right; which parts are oversold?

7. Distillation's upper limit. Why can't you distill ethanol-water beyond about 95.6% ethanol using a conventional still? What is special about that number, and what would you have to do chemically to get beyond it?

8. Roast level and origin character. A roaster sells the same Ethiopian Yirgacheffe in light, medium, and dark roast. Predict (with reasoning) which one will best preserve the bean's origin character, and which will taste mostly of "roast." What chemistry justifies the prediction?

9. The chai story. Indian masala chai uses milk, long brewing, and substantial sweetening. What chemistry does the milk contribute that water-only brewing of the same tea would not? Why does this approach favor strong, oxidized teas (Assam) over delicate green teas?

10. Yerba mate and the etiquette of language. Yerba mate is sometimes labeled "South American tea" in North American grocery stores. Why is that label both technically wrong and culturally clumsy? What would a more accurate labeling say?

Advanced sidebars expanded

A1. Diffusion in a coffee bed: more on Fick's law

We introduced Fick's first law in the chapter. The full kinetic story of brewing is governed by Fick's second law, which describes how concentration changes over time:

∂C/∂t = D · ∂²C/∂x²

For a coffee particle of characteristic size L, the time scale for soluble compounds to diffuse out is approximately τ ≈ L²/D. Notice the L² — halving the particle size reduces the diffusion time by a factor of four. This is why moving from a coarse French-press grind to a fine espresso grind dramatically accelerates extraction. It also explains why "perfectly even" grinds (low fines fraction) extract more cleanly: a population of particles with widely varying sizes will produce a brew where the smallest particles are over-extracted while the largest are under-extracted, even though the average is "right."

The diffusion coefficient D itself depends on temperature according to an Arrhenius-like relation, D = D₀ · exp(−E_a / RT). For aqueous diffusion of small organic molecules, D roughly doubles for every 25°C increase, which is why brewing temperature has such large effects on extraction kinetics.

A2. The stoichiometry of fermentation

In Chapter 21 we wrote alcoholic fermentation as a single equation: glucose → 2 ethanol + 2 CO₂. The full pathway (glycolysis followed by ethanol fermentation) involves about a dozen enzymatic steps, regenerates the cofactor NAD⁺ that glycolysis consumed, and yields 2 ATP molecules per glucose. The yeast keeps the energy; the cook gets the byproducts.

The mass-balance is approximately: 1 g of glucose → 0.51 g ethanol + 0.49 g CO₂. For a wine starting with 220 g/L sugar fermenting to dryness, that's roughly 112 g/L (~14% w/v) ethanol and 108 g/L CO₂. The CO₂ in a still wine bubbles off; the ethanol stays.

For comparison, malolactic fermentation: malic acid (C₄H₆O₅) → lactic acid (C₃H₆O₃) + CO₂. One CO₂ per acid molecule, accompanied by a softening of perceived acidity (because lactic is roughly half as strong an acid as malic, with a higher pKₐ, so for the same molar concentration the resulting pH is higher).

A3. Foam stability and crema

A foam is gas dispersed in liquid, stabilized by surfactants at the gas-liquid interfaces. For espresso crema specifically:

• CO₂ from roasting provides the gas. Beans degas continuously after roast; a bean roasted six months ago has lost most of its CO₂ and produces little crema.
• Lipids extracted under pressure (which a paper-filter pour-over largely retains in the filter) provide one class of surfactant.
• Proteins and melanoidins from the roasted bean provide additional surface activity.

The crema's stability — the time before it collapses — depends on bubble size (smaller bubbles, more stable), liquid viscosity (more viscous, more stable), and the surfactant concentration at the interface. A typical good espresso has crema that holds for several minutes; a poor extraction or a stale bean can give a thin crema that collapses in 30 seconds.

🥖 Mastery food checkpoint

Bread track: Bread baking and beer brewing are sister fermentations — both rely on Saccharomyces cerevisiae converting sugar to ethanol and CO₂. Many bread recipes (especially European traditions) use beer or its byproducts as a flavor and leavening source. The yeast biology you'll learn for bread in Chapter 31 is the same biology that produces wine.

Cheese track: Wine and cheese are old companions. The malolactic fermentation in many wines is performed by lactic-acid bacteria — the same family of microbes that turn milk into yogurt and cheese. Cheese-and-wine pairings work because the wine's acid and tannin cut through the fat and protein of the cheese; the chemistry of food-and-wine matching applies directly to your cheese-tasting board.

Chocolate track: Coffee and chocolate share a fermentation step (Chapter 34) and a roast step (Chapter 8). The flavor compounds developed during roasting are recognizably similar — both rely on Maillard reactions on amino acids and reducing sugars, producing pyrazines, furans, and melanoidins. Tasting a single-origin coffee and a single-origin dark chocolate side by side reveals the family resemblance.

Fermented vegetables track: The malolactic fermentation in wine — Oenococcus oeni converting malic to lactic acid — is performed by relatives of the same lactobacilli that ferment kimchi, sauerkraut, and kosher dill pickles. Same family of microbes, very different food matrix; we'll revisit this in Chapter 33.

Coffee track: This chapter is the centerpiece for the coffee track. You now understand grind, water, temperature, time, pressure, and the kinetic principle that ties them all together. Chapter 34 will go deep on coffee fermentation (the wet-vs-dry processing decision); Chapter 24 will give more on the Maillard chemistry of roasting; Appendix F gives a structured progression for taking your home coffee from "drinkable" to "intentional."