Part I: The Molecules in Your Cup
The coffee house is on a side street in Istanbul’s Kadıköy district, tucked between a spice shop and a store selling evil-eye talismans. I’m sitting on a low wooden stool, watching a man named Mehmet prepare Turkish coffee in a way that, as far as I can tell, hasn’t changed in centuries.
He spoons finely ground coffee — ground almost to powder, really, finer than flour — into a small brass cezve, adds cold water and a lump of sugar, and sets it on a bed of hot sand. He doesn’t stir. He watches. The liquid heats slowly, evenly, and just as it begins to foam — just as the surface starts to rise with a crown of dense, tawny froth — he lifts it off the heat. He lets it settle. He puts it back. He does this three times, with the patience of someone performing a ritual, which of course he is.
When he sets the small porcelain cup in front of me, I wrap my hands around it and look down. The coffee is opaque, thick, almost syrupy. There’s no filter of any kind between me and everything that was in those ground beans. Every oil, every suspended particle of cellulose, every molecule that hot water could coax out of that powder is right here, in this cup.
I take a sip. It’s extraordinary — rich, intense, slightly gritty against my tongue. And I’m thinking, as I always am in moments like these, about what I can’t taste. About the cafestol and kahweol dissolved in those oils floating on the surface. About the chlorogenic acids that have been breaking down in the heat. About how this cup, brewed without a filter, is a fundamentally different chemical object from the pour-over I made that morning at my hotel.
Same species of bean. Same basic process — hot water meets ground coffee. But the molecular outcome? Entirely different.
That difference is what this chapter is about.
Sometimes the most powerful experiments aren’t the ones we design in laboratories. They’re the ones that history runs for us, across entire populations, over decades, without anyone realizing what’s happening until the data starts to tell a story.
The Scandinavian countries — Finland, Norway, Sweden, Denmark — are among the highest per-capita coffee consumers on earth. Finland consistently tops the global rankings at over 12 kilograms per person per year. That’s roughly four cups a day for every adult in the country, every day of the year. These are serious coffee-drinking nations.
And for most of the twentieth century, the dominant brewing method across Scandinavia was boiling. Kokekaffe in Norwegian — literally “boiled coffee.” You put ground coffee in a pot of water, brought it to a boil, let it steep, and poured. No filter. No paper. Nothing between you and the full molecular payload of the bean.
Then something changed. Beginning in the 1960s and accelerating through the 1970s, paper-filtered drip coffee machines became widely available and hugely popular across Scandinavia. Within a generation, a significant portion of the population shifted from boiled to filtered coffee. The cultural transition was dramatic — Finland, for instance, went from a nation of boiled coffee drinkers to one where drip-filtered coffee became the overwhelming norm.
This wasn’t a controlled experiment, of course. Nobody designed it. But epidemiologists noticed something remarkable in the population health data. As paper-filtered coffee replaced boiled coffee across Scandinavian populations, average serum cholesterol levels shifted downward. The effect was measurable. It was consistent across multiple studies. And it was large enough to be clinically significant at the population level.
I want to be careful here about causation. Many things changed in Scandinavian diets and lifestyles during those decades — reduced butter consumption, increased awareness of dietary fat, changes in exercise habits. You can’t isolate one variable in a population-level natural experiment the way you can in a laboratory. But the timing was suggestive, and it prompted researchers to ask a very specific question: what, exactly, was the paper filter removing?
We now know the answer, thanks to the work I described in Chapter 2. The paper filter was catching the coffee oils — and with them, cafestol and kahweol, the diterpenes that research has identified as the most potent dietary cholesterol-elevating compounds known. A simple piece of paper, placed between the coffee grounds and your cup, was removing more than 95% of these molecules.
Finland’s transition from boiled to filtered coffee is, in retrospect, one of the most remarkable natural experiments in nutritional epidemiology. An entire nation changed its brewing method, and its cholesterol numbers moved. Not because of a public health campaign targeting diterpenes — nobody was thinking about diterpenes at the time. Not because of a pharmaceutical intervention. Because of a paper filter.
The numbers from Finland tell a striking story. In the 1970s, when boiled coffee was still the dominant preparation method, Finnish men had among the highest average serum cholesterol levels in Europe. Over the following decades, as filtered drip coffee replaced kokekaffe in Finnish kitchens, average cholesterol levels declined significantly across the population.
Now, I need to emphasize: this wasn’t the only factor. Finland simultaneously launched aggressive public health campaigns promoting dietary changes — less saturated fat, more vegetables, the famous North Karelia Project that became a model for cardiovascular disease prevention worldwide. Disentangling the specific contribution of the brewing method shift from these other dietary changes is genuinely difficult.
But controlled studies have since confirmed the mechanism. Research consistently shows that switching from unfiltered to filtered coffee reduces serum LDL cholesterol by approximately 0.13 to 0.33 mmol/L, depending on baseline consumption. That’s a meaningful reduction — equivalent to or greater than what some dietary interventions can achieve.
What fascinates me about this story is the accident of it. Nobody told Finns to switch to paper filters for their health. They did it because drip machines were convenient, consistent, and modern. The health benefit was a side effect of a cultural shift in taste and technology. Sometimes the most important public health interventions are the ones nobody intended.
Let me take a moment to appreciate something that most of us take entirely for granted: the humble paper coffee filter.
It’s a cone or basket of porous cellulose fiber, usually white or natural brown. You can buy hundreds of them for a few dollars. You use one, throw it away, never think about it again. It seems like the least interesting part of your coffee setup.
But from a molecular perspective, that paper filter is doing something extraordinary. It’s performing a highly selective separation — allowing water-soluble compounds to pass through while trapping the lipid-soluble fraction. Coffee oils, which carry the diterpenes, are caught in the cellulose matrix. Water, which carries caffeine, chlorogenic acids, sugars, organic acids, and most of the volatile aromatics, flows right through.
The result is striking. Research consistently demonstrates that paper filtration removes greater than 95% of cafestol and kahweol from brewed coffee. A cup of French press coffee contains roughly 3 to 6 milligrams of cafestol. The same coffee, brewed through a paper filter? Studies indicate it contains near-zero — typically less than 0.1 milligrams per cup.
That’s not a subtle difference. That’s a near-total removal of a specific class of bioactive compounds, achieved by passing your coffee through a piece of paper. It’s one of the most efficient and accessible molecular separations you’ll encounter outside a chemistry laboratory.
And it changes your cup in ways that go far beyond diterpenes. Paper-filtered coffee has a cleaner, brighter flavor profile partly because those oils — which carry both diterpenes and many of the heavier flavor compounds — are absent. The body is lighter. The mouthfeel is thinner. Pour-over enthusiasts and drip coffee drinkers are, whether they know it or not, selecting for a specific molecular profile every time they reach for a paper filter.
This is the central insight of this chapter: your brewing method isn’t just about flavor. It’s a molecular decision. It determines which of the thousand-plus compounds in your coffee beans actually make it into your cup, in what concentrations, and in what ratios. And different brewing methods make that decision very differently.
I’ve spent more time than I’d like to admit standing in my kitchen with a thermometer, a scale, and a timer, brewing the same coffee six different ways and thinking about what’s happening at the molecular level. Here’s what I’ve learned — and what the research tells us.
The French press is the simplest illustration of what happens when you remove the paper filter from the equation. Coarsely ground coffee steeps in hot water for three to four minutes — a method called full immersion — and then a metal mesh plunger pushes the grounds to the bottom. You pour and drink.
That metal mesh stops the large coffee grounds. It does not stop the coffee oils. The mesh openings are far too large to catch the fine lipid droplets that carry cafestol and kahweol. Everything that dissolved or emulsified during those four minutes of full immersion — every diterpene, every oil, every fine suspended particle — pours right into your cup.
The result: 3 to 6 milligrams of cafestol per cup. A rich, full-bodied brew with a characteristic oily mouthfeel. Delicious, if you like that style. But molecularly, it’s about as far from filtered coffee as you can get while still using the same beans.
The French press also extracts more melanoidins — those large, brown Maillard reaction polymers that constitute roughly 23-25% of coffee’s dry weight — because the long steep time and lack of filtration allow these high-molecular-weight compounds to remain in the brew. The body you feel in a French press cup is partly those melanoidins.
Turkish coffee pushes the extraction even further. The grind is extraordinarily fine — almost powder. The water contact time, while variable, is extended by the slow heating process and multiple foaming cycles. And there is no filter of any kind. You drink everything that extracted, minus whatever settles to the bottom of the cup as sludge (which, traditionally, you leave behind — and which, incidentally, is where some of the heaviest molecular compounds end up).
The diterpene concentrations in Turkish coffee are among the highest of any preparation method. The fine grind creates enormous surface area, which accelerates extraction. The repeated heating cycles provide additional energy for dissolving lipid-soluble compounds. It’s essentially the maximum-extraction scenario.
Scandinavian kokekaffe and similar boiled preparations — cowboy coffee, Greek coffee prepared without straining — follow the same logic. Long contact times, no paper filtration, maximum molecular transfer from bean to cup.
Espresso fascinates me, because it doesn’t fit neatly into either the “filtered” or “unfiltered” category.
The extraction process is unique: water at approximately 90-96°C is forced through a densely packed puck of finely ground coffee at 8-10 bars of pressure. The contact time is short — typically 25 to 30 seconds. The volume is small — roughly 25 to 30 milliliters for a single shot.
So what about diterpenes? Studies indicate that a single espresso shot contains approximately 1 to 2 milligrams of cafestol. That’s less than French press (3-6 mg per cup) but more than paper-filtered coffee (near-zero). Why the middle position?
Two factors. First, the short contact time limits extraction — there simply isn’t enough time for the water to dissolve as many lipid-soluble compounds as it would in a four-minute French press steep. Second, the compressed coffee puck and the crema that forms on top act as partial filters. They’re not paper — they don’t catch everything — but they do retain some of the coffee oils and the diterpenes dissolved in them.
The result is a brew that is genuinely its own category. Not filtered. Not unfiltered. Something in between — a high-pressure, short-contact extraction that produces a unique molecular profile found in no other brewing method.
I’ve had more arguments about espresso than about any other topic in coffee science. Is it filtered or unfiltered? Healthy or unhealthy? The answer, frustratingly, is: it depends on what you’re measuring.
From a diterpene perspective, espresso occupies a genuine middle ground. At 1 to 2 milligrams of cafestol per shot, it delivers more than paper-filtered drip but substantially less than French press. If you drink a single shot, the absolute amount is modest. If you drink six shots a day — and I know people who do, myself occasionally among them — the cumulative load starts to add up.
But espresso’s uniqueness goes beyond diterpenes. The high-pressure extraction creates compounds and concentrations that no other method produces. The crema — that golden-brown foam on top — is an emulsion of CO₂ gas, water, and coffee oils, stabilized by melanoidins and proteins. It’s essentially a micro-laboratory of surface chemistry sitting on top of your drink. The concentration of chlorogenic acids per milliliter is higher in espresso than in any other common preparation, simply because the same amount of compound is dissolved in a much smaller volume of water.
So when someone asks me, “Is espresso good or bad?” I tell them it’s the wrong question. Espresso is a distinct chemical system. Comparing it to drip coffee is like comparing a concentrated stock to a clear broth — they started from the same ingredients, but the process created fundamentally different molecular outcomes.
At the other end of the spectrum sits the paper-filtered pour-over — Hario V60, Chemex, Kalita Wave, or a standard automatic drip machine. Hot water passes through a bed of medium-ground coffee and through a paper filter before reaching your cup.
The paper does its work. Greater than 95% of diterpenes are removed. The cup is clean, bright, and light-bodied compared to French press or espresso. If the beans are good and the technique is right, the flavor clarity can be stunning — you taste the origin characteristics of the bean without the oily heaviness of an unfiltered brew.
But here’s something I find important to note: the paper filter doesn’t remove everything. Caffeine passes straight through — it’s highly water-soluble and has no particular affinity for cellulose. Chlorogenic acids pass through. Most of the Maillard reaction products, including many melanoidins (at least the smaller ones), pass through. The volatile aromatics that create coffee’s aroma pass through.
What you lose, besides the diterpenes, is body. Mouthfeel. That viscous, oily quality that French press drinkers love. The paper catches the oils that create that sensation. Whether that’s a loss or a benefit depends entirely on your preference — and, perhaps, on whether you’re thinking about it from a molecular perspective.
Cold brew adds another variable to the equation: temperature. Instead of hot water extracting compounds quickly, cold or room-temperature water works slowly — typically 12 to 24 hours of steeping.
The kinetics of extraction change dramatically at lower temperatures. Chemical reactions — and dissolution is a chemical process — generally slow down as temperature decreases. A rough rule of chemistry is that reaction rates roughly halve for every 10°C drop in temperature. Cold brew compensates for this by extending the contact time enormously.
But not all compounds respond to temperature the same way. Some molecules extract efficiently even in cold water. Others need heat energy to dissolve. Caffeine, for instance, is reasonably soluble in cold water, so cold brew still delivers plenty of it — often more than hot-brewed coffee per volume, because of the extended steep time and the typical concentrate-then-dilute preparation method.
Chlorogenic acids, however, are less efficiently extracted at lower temperatures. Some studies suggest that cold brew contains lower concentrations of certain chlorogenic acid subtypes compared to hot-brewed coffee. The Maillard reaction products — melanoidins and their precursors — also extract differently. And the volatile aromatics? Many of the delicate, heat-dependent volatiles that create coffee’s complex aroma are simply not released at cold temperatures. This is why cold brew often tastes smoother and less acidic but also less aromatically complex than hot-brewed coffee.
As for diterpenes: cold brew is typically filtered before serving, often through paper or cloth. If paper-filtered, the diterpene content is low, similar to hot-brewed drip. If filtered only through a metal mesh — as some cold brew devices use — the diterpene content will be higher, though the lower extraction temperature may partially offset this. The research on cold brew’s specific diterpene content is still emerging, and I expect we’ll have better data in the coming years.
I want to step back from the mechanics of individual brewing methods and consider the global picture, because the scale matters.
An estimated 500 million people worldwide drink unfiltered coffee as their primary preparation method. That’s not a niche. That’s not a curiosity. That’s a significant fraction of the world’s coffee drinkers.
Consider the geography. Turkish coffee is the dominant preparation across Turkey, much of the Middle East, North Africa, the Balkans, and parts of Central Asia. We’re talking about hundreds of millions of people for whom coffee means a small cup of thick, unfiltered brew prepared in a cezve — exactly like the one Mehmet made for me in Kadıköy. This isn’t a quaint tradition practiced by a few enthusiasts. It’s the daily reality for entire regions.
In Scandinavia, despite the shift toward filtered coffee, kokekaffe hasn’t disappeared entirely. Many Norwegians, Swedes, and Finns still prepare boiled coffee, particularly in rural areas, at cabins, and during outdoor activities. The tradition persists alongside modern drip machines.
French press coffee has surged in global popularity over the past few decades, particularly in North America, Europe, and Australia. It’s marketed as a premium, artisanal brewing method — and it is, from a flavor perspective. But it’s also unfiltered, and every one of those full-bodied cups delivers 3 to 6 milligrams of cafestol.
Then there are the regional variations: Greek coffee (essentially identical to Turkish in preparation), cowboy coffee in the American West, various traditional preparations across Africa and Asia where coffee grounds are boiled or steeped without paper filtration.
When I think about this number — 500 million people — I think about what it means for public health research. The epidemiological studies that link coffee consumption to various health outcomes typically ask participants whether they drink coffee and how much. They rarely ask how the coffee is brewed. But as we’ve seen in this chapter, the brewing method is arguably as important as the quantity. Three cups of pour-over per day and three cups of French press per day are, from a diterpene perspective, entirely different exposures. Lumping them together under the category of “three cups of coffee” obscures a molecular difference that research suggests has measurable physiological consequences.
This is one of the frustrations I encounter when reading coffee epidemiology. The studies are large, well-designed, and statistically rigorous. But the exposure variable — “coffee consumption” — is often too coarsely defined to capture the molecular reality. I’ll return to this problem later in the book when we discuss how computational modeling might help us design better epidemiological studies by predicting which molecular differences between brewing methods are most likely to matter for specific health outcomes.
I’ve focused heavily on diterpenes in this chapter because they’re the most dramatic example of how brewing method changes the molecular content of your cup. But I want to be clear: diterpenes are just one chapter of the story. Brewing method changes everything.
Caffeine extraction varies significantly between methods. Espresso, despite its reputation as a caffeine bomb, contains roughly 60 to 80 milligrams per shot — less than a standard 8-ounce cup of drip coffee, which typically delivers 80 to 120 milligrams. The difference is contact time and volume: drip coffee extracts caffeine over a longer period and into a larger volume of water. Cold brew concentrate can contain even more, depending on the steep time and bean-to-water ratio.
Chlorogenic acid profiles shift with brewing method and temperature. These compounds — which we’ll explore in much greater detail in later chapters — are sensitive to heat, pH, and extraction time. A light-roasted pour-over and a dark-roasted espresso deliver very different chlorogenic acid spectra, even before you account for the roasting differences. The extraction method adds another layer of variation.
Melanoidin concentration depends on both the roast level (which creates them) and the brewing method (which extracts them). French press, with its long immersion and lack of filtration, extracts and retains more melanoidins than paper-filtered methods. Espresso concentrates them into a small volume. Cold brew, with its lower temperatures, may extract some melanoidin subtypes less efficiently.
Volatile retention — the aromatic compounds that create coffee’s smell and much of its perceived flavor — is highly sensitive to temperature, pressure, and exposure to air. Espresso, brewed quickly under pressure and consumed immediately, retains many volatiles that dissipate from a cup of drip coffee left on a warming plate. Cold brew, produced without heat, develops a fundamentally different volatile profile.
Each brewing method, in other words, is a different chemical experiment performed on the same starting material. You’re controlling the variables — temperature, time, pressure, particle size, filtration — and each combination produces a different molecular outcome. If coffee were a drug and brewing methods were formulations, we’d be talking about this the way pharmaceutical scientists talk about different drug delivery systems: same active ingredient, entirely different bioavailability.
In Chapter 13, I’ll introduce the ADMET framework — Absorption, Distribution, Metabolism, Excretion, and Toxicity — as a way of thinking systematically about these differences. It’s a tool borrowed from pharmacology, and it turns out to be remarkably useful for understanding what happens to coffee’s bioactive compounds after you swallow them. But the ADMET story starts here, with the brewing method, because ADMET begins with what’s actually in your cup. And what’s in your cup depends, more than most people realize, on how you made it.
I once spent an afternoon at a specialty coffee conference, talking to baristas about diterpenes. These were talented, passionate people who could discourse for twenty minutes about extraction yield, total dissolved solids, and refractometer readings. They knew more about coffee flavor chemistry than most food scientists.
Not one of them had heard of cafestol.
This isn’t a criticism — it’s an observation about how differently the coffee industry and the biomedical research community think about the same beverage. Baristas optimize for flavor. Researchers optimize for understanding biological effects. The two worlds rarely overlap.
When I explained that their beautiful French press brews delivered 3 to 6 milligrams of a cholesterol-raising compound per cup, and that a simple paper filter would remove more than 95% of it, I watched a room full of coffee professionals recalibrate their understanding of what they were serving. One barista said to me, “So when I recommend a French press to a customer, I’m making a health decision for them without knowing it?” I told her: essentially, yes.
The science of coffee brewing and the science of coffee bioactivity exist in parallel, rarely intersecting. One of the goals of this book is to build a bridge between them — to give both coffee professionals and coffee drinkers the molecular vocabulary they need to understand what their brewing choices actually mean.
Let me bring this back to where we started: Mehmet’s cezve in Kadıköy, and the simple question of what’s in your cup.
Your brewing method is your first and most powerful molecular decision. Before you choose your beans, before you decide on a roast level, before you measure the dose or set the water temperature — the moment you reach for a French press instead of a paper filter, or order a Turkish coffee instead of an Americano, you’ve already determined the broad molecular profile of what you’re about to drink.
A paper filter isn’t just about taste. It’s a molecular gatekeeper that removes specific compounds with measurable biological effects. Research consistently demonstrates that paper filtration removes greater than 95% of diterpenes — compounds that studies have linked to elevated serum cholesterol at the population level. That’s not a subtle effect, and it’s not a theoretical concern. It’s a measurable molecular difference with documented physiological consequences.
But — and this is the paradox I keep returning to — those same diterpenes also show potentially beneficial properties in laboratory studies. And the paper filter, while removing diterpenes, may also alter the profile of other bioactive compounds in ways we’re only beginning to understand. The full picture is more complex than “filtered good, unfiltered bad.”
What I can tell you with confidence is this: the difference matters. The choice isn’t neutral. And understanding what your brewing method does at the molecular level is the first step toward making that choice with full information rather than by default.
I’m not here to tell you to throw away your French press. I still drink espresso every day — a brewing method that, as we’ve seen, occupies its own strange middle ground in the diterpene spectrum. What I am here to tell you is that every cup of coffee is a chemical experiment, and your brewing method is the experimental protocol. Understanding that protocol — understanding what it extracts, what it filters, and what it delivers to your body — is what transforms a daily habit into an informed choice.
In the next chapter, we’ll move from the cup to the body, following these molecules as they’re absorbed, distributed, metabolized, and excreted. Because getting the molecules into your cup is only half the story. What happens after you swallow is where the real complexity begins.
Next: Chapter 4 — “When Molecules Meet Receptors”