Heat, Calorimetry & Phase Changes · Guide · glidepathscience.org
GuideHeat, Calorimetry & Phase Changes
Contents
This Guide explains the ideas and equations behind every Walkthrough step, Concept Challenge question and Predict round, in the order a textbook takes them. Part 1 prepares the Intro track and Part 2 the Advanced track; each takes about a quarter of an hour to read. Each section ends with a worked example that is not one of the sim’s own problems.
Part 1 · The essentials
Math you’ll need: algebra, ratios and percentages; no calculus.
Temperature and thermal equilibrium
Put a hot copper block into cool water and leave them alone in an insulated box. The copper cools and the water warms until nothing changes any more. At that point the two have reached thermal equilibrium, and the one thing they share is their temperature. They need not share anything else: a small hot coin dropped into a lake ends at the lake’s temperature while holding a tiny fraction of the lake’s energy.
This is why a thermometer works. If a thermometer is in equilibrium with a cup of tea, and the same reading appears in a cup of coffee, the tea and the coffee would not exchange energy if they touched: they are at the same temperature. That statement, that two bodies each in equilibrium with a third are in equilibrium with each other, is called the zeroth law of thermodynamics.
Temperatures are written in degrees Celsius in everyday life and in kelvins in physics. The two scales have the same size of degree; the kelvin scale starts at absolute zero, the lowest temperature possible:
$T_K = T_C + 273.15$
A difference of 20 °C is a difference of 20 K. But a ratio of temperatures only makes sense in kelvins, because only that scale has a true zero. The molecules’ average translational kinetic energy is proportional to the kelvin temperature, so “twice as hot” means twice the kelvins, not twice the Celsius number.
Temperature, heat and internal energy
Three words are easy to confuse, and the physics depends on keeping them apart.
- Temperature is what a thermometer reads: a measure of the molecules’ average kinetic energy. It is what evens out between bodies in contact.
- Internal energy is the total energy a body holds in its molecules: their kinetic energy and the potential energy of their attractions. It depends on how much material there is, not only on the temperature.
- Heat ($Q$) is energy moving from one body to another because of a temperature difference. A body never “contains heat”: once the energy has arrived, it is internal energy.
Heat is measured in joules. Chemistry and food labels also use the calorie, 1 cal = 4.184 J, and the food Calorie, which is a kilocalorie. In this sim heat is positive when it flows into a body and negative when it flows out.
Specific heat: Q = mcΔT
How much heat does it take to warm something? Experiments show that the heat is proportional to the mass $m$ and to the temperature change $ΔT$, with a constant $c$ that depends on the material, its specific heat:
$Q = mcΔT$ $C = mc$
$c$ is the heat needed to warm 1 kg of the material by 1 K; its unit is J/(kg·K). The product $C = mc$ is the heat capacity of one particular object, in J/K. Water’s specific heat, 4186 J/(kg·K), is unusually large: about 4.6 times aluminum’s (900) and 11 times copper’s (387). Ice (2090) and steam (2020) have about half of liquid water’s value. The formula holds only while the material stays in one phase.
The same heat put into equal masses of water and copper warms the copper eleven times as much. That is why the sea, with its huge heat capacity, keeps coastal climates mild.
Calorimetry: the heats add to zero
A calorimeter is an insulated container in which bodies exchange heat only with each other. Energy is conserved, so whatever heat leaves the hot bodies arrives in the cold ones. With heat positive into a body, the heats add to zero:
$ΣQ = 0$ $m_1c_1(T_f − T_1) + m_2c_2(T_f − T_2) = 0$
For two portions of water, $c$ cancels and the final temperature is the mass-weighted average, $T_f = (m_1T_1 + m_2T_2)/(m_1 + m_2)$: closer to the temperature of the larger mass. A light foam cup barely matters; a heavy mug matters a great deal.
The same equation turns around to measure a specific heat. Heat a metal sample in boiling water, drop it into water of known mass and temperature, and read the final temperature. Then the heat lost by the metal equals the heat gained by the water:
$c_m = m_wc_w(T_f − T_w)/(m_m(T_m − T_f))$
Latent heat and the heating curve
Put ice at −20 °C on a steady heater and watch a thermometer in it. The temperature climbs to 0 °C, then stops while the ice melts, climbs again, and stops a second time at 100 °C while the water boils. A graph of temperature against heat added, the heating curve, is a staircase of slopes and flat stretches.
On the flat stretches the heater’s energy is not lost. It goes into pulling molecules out of the ice’s rigid structure, or apart into a gas, against their attractions. The molecules’ average kinetic energy, which is what the thermometer reads, stays the same; their potential energy rises. The heat needed per kilogram is the latent heat:
$Q = mL_f$ $Q = mL_v$
For water $L_f$ = 334 kJ/kg (melting, or fusion) and $L_v$ = 2256 kJ/kg (boiling, or vaporization), both at constant temperature. $L_v$ is enormous: boiling a kilogram of water away takes 5.4 times the energy of warming it from 0 to 100 °C. It is also why steam scalds: condensing on the skin it gives up $L_v$ before it even begins to cool.
Ice in a drink: decide the final state first
When a mixture can change phase, a single equation for $T_f$ is not enough, because you do not yet know which terms belong in it. Decide the final state first by comparing two energies:
- the heat the warm water (and the cup) can give in cooling to 0 °C;
- the heat the ice needs to warm to 0 °C, and then to melt completely.
If the water can give more than the ice needs, all the ice melts, and the energy left over warms the whole mixture above 0 °C: then write $ΣQ = 0$ with the ice warming, melting and the meltwater warming. If the water can give less, the cup stops at 0 °C with ice left over, and the amount that melted is the heat available divided by $L_f$. While ice and water are together, the temperature is pinned at 0 °C.
Evaporative cooling and humidity
Water evaporates at any temperature, not only at the boiling point, and each kilogram that evaporates carries away its latent heat, a little more than at 100 °C: about 2.42 MJ/kg at skin temperature. That is how sweating cools: the energy for evaporation comes from the skin. Sweat that drips off without evaporating carries almost nothing away.
$m_{ev}L_v = mcΔT$
How fast water evaporates depends on how much vapor the air already holds. The air’s relative humidity compares the vapor present with the most the air could hold at its temperature. In humid air, evaporation slows, sweat drips instead of evaporating, and the body cannot shed its heat. A thermometer with a wet wick in moving air, a wet-bulb thermometer, cools until evaporation can cool it no further: the wet-bulb temperature is the floor for evaporative cooling.
Part 2 · Going further
Math you’ll need: logarithms and exponentials, and some calculus (an integral in the Debye model and Clausius–Clapeyron).
Why solids have the specific heats they do
Each atom of a solid vibrates about its place in three directions. By equipartition, each direction holds on average $\tfrac12k_BT$ of kinetic energy and $\tfrac12k_BT$ of potential energy, so each atom holds $3k_BT$ and a mole $3RT$. The molar heat capacity is then $3R$ ≈ 24.9 J/(mol·K), whatever the element: the law of Dulong and Petit. Per kilogram, $c ≈ 3R/M$, so heavy atoms have small specific heats.
The law fails for light, stiffly bound atoms such as carbon in diamond. A vibration of frequency $f$ can take up energy only in quanta $hf$; when $hf$ is large compared with $k_BT$, few quanta are excited and the vibrations are “frozen out”. Debye’s model gathers this into one temperature scale, the Debye temperature $Θ_D$: well above it, $C → 3R$; well below it, $C ∝ T^3$. Copper’s $Θ_D$ is about 343 K, so it is near the classical limit at room temperature; diamond’s is about 2,230 K, so at 25 °C it holds only a fraction of $3R$. The single-Θ model has the right shape but not exact values: for diamond at 25 °C it gives about two-thirds of the measured value.
Real calorimeters: their own heat capacity, and the leak
A real calorimeter takes up heat itself. Its heat capacity $C_{cal}$ is found by calibration, either by mixing warm water into cold and solving
$C_{cal} = m_hc_w(T_h − T_f)/(T_f − T_c) − m_cc_w$
or electrically, by putting in a known energy $VIt$ and measuring the rise. Its value is often quoted as a water equivalent, the mass of water with the same heat capacity. Ignoring it makes every specific heat found by mixing come out low.
No calorimeter is perfectly insulated. After a hot sample goes in, the water’s temperature rises to a peak and then falls slowly as heat leaks to the room. The peak is not the temperature the mixture would have reached with no leak: heat was already leaving before it. The standard correction fits a straight line to the slow cooling after mixing and extrapolates it back to the moment of mixing. What the line cannot see is the heat lost during the mixing itself, which leaves a small bias.
Supercooling
Very pure, still water can stay liquid below 0 °C: it is supercooled, a metastable state that waits for a seed crystal to start freezing. Tap it, and part of it freezes at once. The latent heat released by the freezing warms the whole sample back to 0 °C, which stops the freezing. The share that freezes, $f$, follows from the books: $f\,mL_f = mc_wΔT$, so
$f = c_wΔT/L_f$
The mass cancels. Supercooled droplets are common in clouds, where they freeze onto aircraft wings and snowflakes.
Boiling, Clausius–Clapeyron and phase diagrams
Where latent heat goes. When a kilogram of water boils at 1 atm it becomes about 1.70 m³ of vapor and pushes the atmosphere back, doing work $pΔv$. Only about 7.6 % of $L_v$ goes into that work; the rest raises the internal energy by separating the molecules. At constant pressure the heat equals the change in enthalpy, $ΔH = ΔU + pΔV$. For many liquids the work share is similar:
$pΔv/L_v ≈ RT_b/(ML_v)$
Boiling is a pressure condition. Every liquid has a vapor pressure, the pressure of its vapor in equilibrium with it, which rises steeply with temperature. A vapor bubble can grow inside the liquid only when the vapor pressure equals the pressure above the liquid: that is boiling. Water boils at 100 °C only where the pressure is 101.3 kPa.
Clausius–Clapeyron. Along a coexistence line, $dp/dT = L/(TΔv)$. For a vapor treated as an ideal gas, with $L$ constant and $R_v = R/M$, it integrates to
$\ln(p_2/p_1) = −(L/R_v)(1/T_2 − 1/T_1)$
Phase diagrams show which phase is stable at each pressure and temperature, with lines where two coexist, a triple point where all three do, and a critical point where liquid and vapor become one. Below its triple-point pressure (about 5.1 atm for CO₂) a substance has no liquid: dry ice at 1 atm sublimes. Water’s melting line leans backward ($dp/dT$ ≈ −13.5 MPa/K) because ice is less dense than water, so pressure can lower its melting point, but only a little.
Vapor pressure, dew point and wet bulb
Air holding water vapor at partial pressure $e$ has relative humidity
$RH = e/e_s(T)$ $e_s(T_d) = e$
where $e_s(T)$ is the saturation vapor pressure. Cool the air without changing its vapor, and it saturates at its dew point $T_d$, where dew or fog forms (below 0 °C, frost forms at the frost point). A convenient formula near room temperature is Bolton’s,
$e_s = 611.2\,e^{17.67T/(T + 243.5)}$ Pa
with $T$ in °C, within 0.1 % of the reference (IAPWS) values from 0 to 35 °C but drifting above 60 °C. The wet bulb $T_w$ is where a wet wick in moving air settles: the heat the air brings balances the latent heat carried away. With mixing ratios $r$ (kilograms of vapor per kilogram of dry air),
$(c_{pd} + rc_{pv})(T − T_w) = L_v[r_s(T_w) − r]$
In unsaturated air $T_d ≤ T_w ≤ T$; all three are equal at 100 % humidity.
Evaporation, molecule by molecule
Molecules in a liquid have a spread of energies. To leave, one must overcome the pull of its neighbors, roughly the latent heat per molecule: for water about 0.42 eV, or 13 $k_BT$ at 100 °C. Only the most energetic molecules near the surface can do it, and each takes more than the average energy with it, so the liquid left behind cools. The share of molecules with that much energy goes roughly as the Boltzmann factor, $e^{−L/(Nk_BT)}$, which is why vapor pressure rises so steeply with temperature.
The fastest a liquid can evaporate is into a vacuum, where no molecule comes back. Kinetic theory gives the flux at which molecules from saturated vapor would strike the surface, the Hertz–Knudsen rate, with an evaporation coefficient α ≤ 1:
$j = αp_s\sqrt{M/(2πRT)}$
In air, the vapor must diffuse away, and molecules return at a rate set by the vapor present, so the net rate goes as $p_s − p_v$: humidity slows evaporation.
The same sharing of energy molecule by molecule explains mixing through a wall: two gases separated by a light wall that passes energy reach one temperature, $T_f = Σf_iN_iT_i/Σf_iN_i$, where $f$ counts the ways each molecule holds energy (3 for helium, 5 for nitrogen, which also rotates).
Reference
Common mistakes
Part 1
- Taking temperature for energy: a hot cup does not hold more energy than a warm bath.
- Speaking of a body’s “heat content”: heat is energy on the move; a body holds internal energy.
- Taking ratios of Celsius numbers: ratios need kelvins.
- Expecting equal heats to give equal temperature changes: $ΔT = Q/(mc)$.
- Averaging temperatures without weighting them by $mc$.
- Forgetting the latent heat, or assuming all the ice melts.
- Thinking sweat cools by being wet: only evaporation cools.
Part 2
- Taking the peak reading of a leaky calorimeter for the final temperature.
- Taking 100 °C as water’s boiling point everywhere.
- Using Bolton’s formula far outside the range it was fitted for.
Equations
| Part 1 | |
|---|---|
| Kelvins | $T_K = T_C + 273.15$ |
| Specific heat | $Q = mcΔT$; $C = mc$ |
| Calorimetry | $ΣQ = 0$ in an insulated cup |
| Latent heat | $Q = mL_f$; $Q = mL_v$ |
| Evaporative cooling | $m_{ev}L_v = mcΔT$ |
| Part 2 | |
|---|---|
| Dulong–Petit | $c ≈ 3R/M$ |
| Calibration | $C_{cal} = m_hc_w(T_h − T_f)/(T_f − T_c) − m_cc_w$ |
| Supercooling | $f = c_wΔT/L_f$ |
| Enthalpy | $ΔH = ΔU + pΔV$; $pΔv/L_v ≈ RT_b/(ML_v)$ |
| Clausius–Clapeyron | $\ln(p_2/p_1) = −(L/R_v)(1/T_2 − 1/T_1)$ |
| Humidity | $RH = e/e_s(T)$; $e_s(T_d) = e$; $e_s = 611.2\,e^{17.67T/(T + 243.5)}$ Pa |
| Wet bulb | $(c_{pd} + rc_{pv})(T − T_w) = L_v[r_s(T_w) − r]$ |
| Equipartition | $T_f = Σf_iN_iT_i/Σf_iN_i$ |
| Hertz–Knudsen | $j = αp_s\sqrt{M/(2πRT)}$ |
$Q$ heat into a body; $m$ mass; $c$ specific heat; $C$ heat capacity; $L_f$, $L_v$ latent heats; $R$ = 8.314 J/(mol·K); $M$ molar mass; $T_b$ boiling point (K); $e$ vapor pressure; $e_s$ saturation vapor pressure; $T_d$ dew point; $T_w$ wet bulb; $r$ mixing ratio; $c_{pd}$ = 1005.7 and $c_{pv}$ = 1870 J/(kg·K); $f$ ways of holding energy; $α$ evaporation coefficient.
Sources and further reading
- OpenStax, University Physics Volume 2, Chapter 1, “Temperature and Heat,” CC BY-NC-SA 4.0: the order of Part 1 and its notation.
- OpenStax, Chemistry 2e, Chapter 5, “Thermochemistry,” and College Physics 2e, “Humidity, Evaporation, and Boiling,” CC BY-NC-SA 4.0.
- The IAPWS releases on the saturation pressure of water and of ice; D. Bolton (1980); D. M. Murphy and T. Koop (2005).
- The Field Manual (H): the model, its constants and what it leaves out.
Heat, Calorimetry & Phase Changes v1.1 · © 2026 Dr. David Meier · CC BY-NC-SA 4.0 · glidepathscience.org