Pure anhydrous caffeine melts at roughly 235 to 237 °C (about 455 to 459 °F). Multiple calorimetry studies have recorded the melting temperature in a tight band around 508 to 510 K, with slight variation depending on the sample’s crystal form and purity.1NIST Chemistry WebBook. Caffeine That number puts caffeine well above the temperatures encountered in brewing, baking, or roasting, which explains a lot about why the compound is so persistent in processed foods and beverages. But the melting point alone does not tell the full thermal story of caffeine, which undergoes a crystal-phase change, can sublime directly into vapor, and behaves differently depending on whether it is dry or hydrated.
The Number and Why It Bounces Around Slightly
If you search for caffeine’s melting point in different references, you will see values ranging from about 234 °C to 237 °C. That is not sloppy measurement. The NIST Chemistry WebBook compiles data from several independent research groups who all used differential scanning calorimetry, and their reported values span from 507.7 K to 510.2 K.1NIST Chemistry WebBook. Caffeine One well-cited study measured the solid-to-liquid transition at 509.00 K with an enthalpy of fusion of about 19.86 kJ per mole.2The Journal of Chemical Thermodynamics. The standard molar enthalpy of formation, molar heat capacities, and thermal stability of anhydrous caffeine Another group recorded 510.2 K, while yet another landed at 507.7 K. The spread comes down to minor differences in sample preparation, crystal purity, and instrument calibration, but the consensus sits comfortably near 509 K, which converts to about 236 °C or 457 °F.
That is hot. For context, 236 °C exceeds the smoke point of most cooking oils and is well above the boiling point of water. You are never going to melt caffeine by pouring hot water over coffee grounds or steeping a tea bag.
The Crystal-Phase Change That Happens First
Before caffeine actually melts, it goes through a structural rearrangement in its crystal lattice. At room temperature, dry caffeine sits in what chemists call its low-temperature crystal form (sometimes labeled form II or the α-polymorph, depending on the naming convention used). When heated to around 140 to 143 °C, the molecules rearrange into a high-temperature crystal form (form I, or the β-polymorph). One calorimetry study pinpointed this transition at about 143 °C as the onset temperature of an endothermic peak.3Thermochimica Acta. A kinetic study of polymorphic transition of anhydrous caffeine with microcalorimeter Another measured it at 413.39 K, or about 140 °C.2The Journal of Chemical Thermodynamics. The standard molar enthalpy of formation, molar heat capacities, and thermal stability of anhydrous caffeine
This crystal rearrangement absorbs a small amount of energy on its own, so if you are watching a heating curve on a calorimeter, you see a small dip around 140 °C before the much larger dip at 236 °C when the substance actually liquefies. The caffeine stays solid through both phases, just in a different internal arrangement. It then remains in the high-temperature crystal form all the way up until it melts near 509 K.4ScienceDirect (The Journal of Chemical Thermodynamics). Thermodynamic properties of caffeine: Reconciliation of available experimental data
This transition matters for pharmaceutical manufacturing, where caffeine is commonly added to pain relievers and stimulant tablets. The two crystal forms behave differently when compressed into pills and can dissolve at slightly different rates. If a batch of caffeine inadvertently converts between forms during storage or processing, it can change how the tablet performs.
Sublimation and Why Caffeine Can Skip the Liquid Phase
One of caffeine’s more interesting thermal quirks is that it can sublime: it goes from solid directly to vapor without passing through a liquid state. This happens at temperatures well below its melting point, starting around 178 °C under normal atmospheric pressure and becoming quite noticeable by 200 °C. Chemistry students who isolate caffeine from tea in laboratory exercises often encounter this firsthand when they see white, needle-like crystals forming on the cool side of a heated dish or on a cold finger condenser. Those crystals formed from caffeine vapor that re-solidified on the cooler surface.
Sublimation and melting are not in competition exactly, but which one you observe depends on the conditions. If you heat caffeine slowly in an open container with good airflow, a lot of the material sublimates before the temperature ever reaches 236 °C. If you heat it quickly in a sealed system where the vapor cannot escape, you will reach the melting point and see it liquefy. This is why the melting point is most reliably measured using sealed calorimeter pans, which prevent mass loss from sublimation during the heating run.
The practical consequence is that even though caffeine has a high melting point, it can escape from solid mixtures at temperatures you might encounter in a kitchen or industrial oven. Sublimation is the mechanism behind certain caffeine losses during high-heat food processing.
Does Caffeine Survive Cooking and Roasting?
A common assumption is that roasting or baking destroys caffeine because the process involves high heat. The melting point number alone suggests otherwise, and the research backs this up: caffeine is remarkably heat-stable compared to many other bioactive compounds in food. One study evaluating the thermal stability of compounds in coffee found that caffeine only began to degrade at 146 °C during re-treatment of coffee extract fractions, while chlorogenic acids (the other major bioactive compounds in coffee) started breaking down at lower temperatures, around 108 to 114 °C.5PubMed Central / Elsevier. Evaluation of the thermal stability of bioactive compounds in coffee beans and their fractions modified in the roasting process
But “degradation commences” does not mean “caffeine is destroyed.” That 146 °C figure marks the onset of measurable breakdown under specific laboratory conditions, not the temperature at which caffeine vanishes. In actual coffee roasting, beans reach internal temperatures of roughly 190 to 230 °C, yet finished roasted coffee retains most of its caffeine. The reasons include the short exposure time at peak temperature, the fact that caffeine molecules are embedded in a complex matrix of other compounds and cellular structures, and the relatively gradual nature of the thermal decomposition once it begins. Light roasts and dark roasts end up with broadly similar caffeine content per bean, contrary to the popular belief that dark roasting “burns off” the caffeine.
In baking, the situation is even less dramatic. Oven temperatures for cookies, brownies, or cakes usually sit between 160 and 190 °C, and the interior of the baked good stays much cooler than the oven air for most of the baking time. Caffeine added through chocolate, espresso powder, or tea is going to survive the trip through the oven with little loss.
Hydrated Versus Anhydrous Caffeine
Everything discussed so far applies to anhydrous caffeine, meaning the dry form with no water molecules trapped in the crystal. But caffeine readily picks up water from humid air, forming caffeine hydrate, a crystal that incorporates water molecules into its lattice at a fixed ratio (roughly four water molecules for every five caffeine molecules). This hydrated form behaves differently when heated.
Caffeine hydrate is less thermally stable than its dry counterpart. When exposed to low humidity, below about 30% relative humidity, the hydrate loses its water rapidly and converts into a metastable anhydrous state that resembles the α-polymorph on X-ray analysis but is not quite the same well-ordered crystal you get by carefully drying caffeine in a lab.6PubMed Central. Relative humidity-temperature transition boundaries for anhydrous β-caffeine and caffeine hydrate crystalline forms This transitional form eventually settles into one of the stable anhydrous polymorphs, but the path it takes depends on temperature and humidity conditions.
For the reader who just wants to know the melting point, the hydration issue introduces a wrinkle: if you try to measure the melting point of caffeine that has been sitting in a humid environment, you might first observe a dehydration event (water leaving the crystal) at a lower temperature, followed by the polymorphic transition, and only then the actual melt. Poorly stored caffeine samples can give confusing calorimetry curves with extra peaks that do not show up when the sample is properly dried first. This is one reason published melting points sometimes look inconsistent across older literature, where sample preparation standards were less rigorous.
Cocrystals and Engineered Melting Points
Pharmaceutical scientists have found that pairing caffeine with another molecule in a carefully designed crystal structure, called a cocrystal, can change its physical properties while keeping the caffeine molecule itself intact. By combining caffeine with carboxylic acids or plant-derived flavonoids, researchers have created cocrystals that are anhydrous and have very sharp, well-defined melting points that differ from pure caffeine’s 236 °C.7Crystal Growth & Design. Cocrystallization of Caffeine with Carboxylic Acids and Flavonoids: In Vitro Study to Control the “Caffeine Crash” The goal of that particular research was to control how quickly caffeine dissolves and enters the bloodstream, potentially smoothing out the spike-and-crash pattern that many coffee drinkers experience.
Cocrystals are a growing area in drug formulation because they let formulators tune dissolution rates without chemically modifying the active ingredient. For caffeine specifically, the idea is attractive because caffeine is already one of the most widely consumed pharmacologically active compounds on Earth. A cocrystal that slows absorption could, in principle, turn a quick jolt into a sustained release, all by changing the way molecules pack together in a crystal. The melting point of the cocrystal is a key quality indicator: a sharp, reproducible melting point tells formulators they have a single, well-defined crystal phase rather than a disordered mixture.
How the Melting Point Is Measured
Almost every modern measurement of caffeine’s melting point uses differential scanning calorimetry, often abbreviated DSC. The instrument heats a tiny sample, usually a few milligrams in a sealed aluminum pan, at a controlled rate and compares the heat flow into the sample against an empty reference pan. Whenever the sample absorbs energy to undergo a phase change, like the crystal rearrangement at 140 °C or the actual melting at 236 °C, the instrument registers it as a dip in the heat-flow curve.8Russian Journal of Physical Chemistry A. Determining the main thermodynamic parameters of caffeine melting by means of DSC
Sealed pans are important for caffeine because of the sublimation issue mentioned earlier. In an open pan, some caffeine vaporizes as temperature climbs, reducing the sample mass and distorting the measurement. Hermetically sealed pans trap the vapor and force the system to reach the actual melting temperature. Heating rate matters too: very fast heating can shift the apparent onset of a transition to a higher temperature, while very slow heating gives the most thermodynamically accurate reading but is more susceptible to noise. Most studies use heating rates in the range of 2 to 10 °C per minute, with 5 or 10 being common choices.
Older references sometimes report caffeine’s melting point using a simple capillary tube method, where a small amount of powder is placed in a thin glass tube and heated in an oil bath while an observer watches for the powder to collapse into a liquid. This method is less precise and more subjective, which contributes to the wider range of values you find in older handbooks.
Caffeine Compared to Its Chemical Relatives
Caffeine belongs to a family of naturally occurring compounds called methylxanthines. Its two closest relatives are theobromine, the stimulant in chocolate, and theophylline, a compound historically used to treat asthma. All three share a similar two-ring molecular backbone, differing only in where methyl groups are attached. These small structural differences translate into noticeably different physical properties.
Theobromine has a melting point around 350 °C, substantially higher than caffeine’s 236 °C. That might seem counterintuitive since the two molecules are so similar, but the position of the methyl groups affects how tightly the molecules can pack together in a crystal lattice and how strongly they interact with their neighbors. Theophylline melts at roughly 270 to 274 °C, sitting between its two relatives. These differences are relevant for anyone working with these compounds in food science or pharmaceutical applications, because processing temperatures that leave caffeine unaffected might behave differently with theophylline or theobromine.
The solubility patterns also diverge. Caffeine dissolves far more easily in hot water than theobromine does, which is part of why a cup of coffee delivers a more concentrated stimulant hit than an equivalent weight of dark chocolate. But solubility and melting point are distinct properties governed by different molecular interactions, so one does not directly predict the other.
What Happens Above the Melting Point
Once caffeine melts at around 236 °C, the resulting liquid is not particularly stable. Continued heating pushes the molecule toward decomposition. At temperatures above roughly 285 °C, the molecule begins to break apart, with the purine ring system fragmenting into smaller nitrogen-containing compounds and carbon dioxide. By around 350 to 400 °C, depending on the atmosphere and heating rate, the decomposition is rapid and essentially irreversible.
This decomposition window is one more reason caffeine survives ordinary food processing. Coffee roasting peaks at around 230 °C for the darkest roasts, which is just below caffeine’s melting point and well below its decomposition range. The molecule might lose a few percent to sublimation during an aggressive roast, but it is not being chemically torn apart. Espresso machines operate at water temperatures of about 90 to 96 °C, which is not even in the same thermal neighborhood. And home baking, as discussed, rarely pushes the interior of food above 100 °C until the very end of the bake.
Industrial decaffeination processes, by contrast, do not rely on heat to remove caffeine. They use solvents or supercritical carbon dioxide to selectively dissolve caffeine out of the bean at relatively mild temperatures, typically below 100 °C. If you had to thermally destroy caffeine to decaffeinate coffee, you would also destroy everything else that makes it taste like coffee long before the caffeine broke down. The molecule’s stubbornness at high temperatures is precisely why chemical extraction is necessary.
Why This Number Keeps Coming Up in Unusual Places
Caffeine’s well-characterized melting point and its interesting polymorphic behavior have turned it into something of a reference standard in thermal analysis. Instrument manufacturers sometimes include caffeine as a calibration compound for calorimeters because its transitions are clean, reproducible, and happen at convenient temperatures. The crystal-phase change near 140 °C and the melt near 236 °C are both sharp enough to serve as benchmarks for checking whether an instrument is reading temperature accurately.
In university teaching labs, caffeine shows up for a different reason: it is easy to extract from tea using basic chemistry equipment, it sublimes beautifully for demonstration purposes, and its melting point is high enough to be interesting but low enough that standard lab equipment can reach it. Students who isolate caffeine from tea bags are often asked to confirm their product’s identity by checking the melting point. A reading near 236 °C, combined with the white needle-like crystal habit that caffeine forms upon sublimation, is a satisfying confirmation that the extraction worked.
The compound’s dual role as a universally familiar stimulant and a well-behaved laboratory specimen gives it an outsized presence in chemistry education relative to its complexity. It is one of the few organic molecules that the average person has both consumed and could, in principle, isolate, purify, and characterize with equipment found in a first-year chemistry lab.