Thermodynamics
The Clausius-Clapeyron Relation
The slope of a phase-boundary line — dP/dT = L / (T·ΔV)
The Clausius-Clapeyron relation is the equation that fixes the slope of a phase-coexistence line in a pressure–temperature diagram: dP/dT = L / (T·ΔV), where L is the molar latent heat of the transition and ΔV is the molar volume change between the two phases. Derived by Émile Clapeyron in 1834 and put on a thermodynamic footing by Rudolf Clausius in the 1850s, it tells you exactly how the melting, boiling, or sublimation temperature shifts when you change the pressure. For a liquid–vapor line — treating the vapor as an ideal gas and dropping the tiny liquid volume — it integrates to ln P = −L/(R·T) + const, a straight line in ln P versus 1/T, and it explains why water boils below 100 °C on a mountaintop.
- Exact (Clapeyron) formdP/dT = L / (T·ΔV)
- Differential formd(ln P)/dT = L / (R·T²)
- Integrated formln P = −L/(R·T) + C
- Latent heat of waterL_vap ≈ 40.7 kJ/mol at 100 °C
- Everest boiling point≈ 70 °C at 0.33 atm
- DiscoveredClapeyron 1834 · Clausius 1850s
Interactive visualization
Press play, or step through manually. The visualization is yours to drive — try it before reading on.
Watch the 60-second explainer
A condensed visual walkthrough — narrated, captioned, under a minute.
Why it matters
Every phase diagram you have ever seen — the P–T plane split into solid, liquid, and gas regions by three curves meeting at a triple point — is quantitatively controlled by a single equation. The Clausius-Clapeyron relation is the derivative of those boundary curves. It answers the practical questions that thermodynamics is built to answer: at what temperature does this liquid boil under this pressure? How much does the melting point move if I squeeze the sample? How steep is the vapor-pressure curve, and what does its steepness tell me about the energy locked up in the phase change?
The relation is the reason a pressure cooker works (raise the pressure, raise the boiling point, cook hotter), the reason the ice-water melting line leans backwards (local pressure nudges it toward lower temperature), and the backbone of meteorology, where the exponential rise of saturation vapor pressure with temperature — a direct consequence of the ln P versus 1/T law — governs cloud formation and how much moisture a warming atmosphere can hold (roughly 7% more water vapor per degree Celsius). It is also the standard laboratory method for measuring a latent heat: you never have to run a calorimeter, you just measure vapor pressure at a few temperatures and read the slope.
How it works — step by step
The starting point is phase equilibrium. Two phases coexist along a line in the P–T plane precisely when their molar Gibbs free energies (equivalently, their chemical potentials) are equal:
g_1(T, P) = g_2(T, P) along the coexistence line
Step 1 — Move along the line. Take a small step (dT, dP) that stays on the boundary. Both Gibbs energies must change by the same amount to keep them equal, so dg₁ = dg₂.
Step 2 — Use the Gibbs differential. For a single component, dg = −s·dT + v·dP, where s is molar entropy and v is molar volume. Applying it to each phase and setting the changes equal:
−s_1·dT + v_1·dP = −s_2·dT + v_2·dP
Step 3 — Solve for the slope. Collecting terms gives (s₂ − s₁)·dT = (v₂ − v₁)·dP, so
dP/dT = Δs / Δv
Step 4 — Replace Δs with latent heat. The transition is isothermal and reversible, so the entropy jump is the latent heat divided by temperature, Δs = L/T. Substituting gives the exact Clapeyron equation:
dP/dT = L / (T · ΔV)
where every symbol is defined in the table below. This form is exact for any first-order transition — melting, vaporization, sublimation, or a solid–solid structural change.
Step 5 — Specialize to a vapor. For a liquid–vapor or solid–vapor boundary, the gas is enormously more voluminous than the condensed phase, so ΔV ≈ V_gas. Treat the vapor as ideal, V_gas ≈ RT/P. Then
dP/dT = L·P / (R·T²) → d(ln P)/dT = L / (R·T²)
Step 6 — Integrate. Assuming L is roughly constant over the temperature range, integrate to obtain the celebrated Clausius-Clapeyron equation:
ln(P_2 / P_1) = −(L/R) · (1/T_2 − 1/T_1)
equivalently ln P = −L/(R·T) + C
A plot of ln P against 1/T is therefore a straight line of slope −L/R. This is where the "just read the slope" latent-heat measurement comes from.
Symbols and units
| Symbol | Meaning | SI unit |
|---|---|---|
| P | Pressure along the coexistence line (vapor pressure for a liquid–gas line) | Pa |
| T | Absolute temperature (always Kelvin) | K |
| dP/dT | Slope of the phase-boundary curve | Pa/K |
| L | Molar latent heat (enthalpy) of the transition, L = T·ΔS | J/mol |
| ΔV | Molar volume of the higher-T phase minus the lower-T phase | m³/mol |
| ΔS | Molar entropy change of the transition | J/(mol·K) |
| R | Universal gas constant = 8.314 | J/(mol·K) |
Sign convention: L and T are always positive. ΔV is positive for a normal transition where the higher-temperature phase occupies more volume (liquid→gas, most solid→liquid), which makes dP/dT positive. When the higher-temperature phase is denser — famously ice→water — ΔV is negative and the boundary slopes backwards.
The three coexistence lines
Near the triple point, three curves radiate outward, one for each pairwise transition. Their relative steepness is set entirely by their latent heats and volume changes, per the Clapeyron equation.
| Boundary | Transition | Latent heat (water) | Slope character |
|---|---|---|---|
| Sublimation line | solid ⇌ vapor | L_sub ≈ 51.1 kJ/mol | Steepest of the three (largest L, huge ΔV) |
| Vaporization line | liquid ⇌ vapor | L_vap ≈ 40.7 kJ/mol | Steep, ends at the critical point |
| Fusion (melting) line | solid ⇌ liquid | L_fus ≈ 6.01 kJ/mol | Nearly vertical; tiny ΔV → enormous dP/dT |
Because ΔV for melting is thousands of times smaller than for vaporization, the fusion line is almost vertical: it takes a huge pressure change to move the melting point even a little. For water the fusion line is not just steep but leans slightly left, because ice is less dense than liquid water (ΔV_melt ≈ −1.6 cm³/mol). Note that L_sub ≈ L_fus + L_vap when all three are evaluated at the same temperature (energy is a state function, so subliming equals melting then boiling); near the triple point this gives L_sub ≈ 6.0 + 45 ≈ 51 kJ/mol.
Worked example — boiling point on Everest
How hot must water get to boil at the summit of Mount Everest, where atmospheric pressure is about P₂ = 0.333 atm? Water boils where its vapor pressure equals the ambient pressure. At sea level, P₁ = 1 atm at T₁ = 373.15 K (100 °C). Use the integrated Clausius-Clapeyron equation with L_vap = 40.7 × 10³ J/mol and R = 8.314 J/(mol·K):
ln(P_2/P_1) = −(L/R)·(1/T_2 − 1/T_1)
ln(0.333) = −(40700/8.314)·(1/T_2 − 1/373.15)
−1.0996 = −4895.4·(1/T_2 − 0.0026799)
1/T_2 − 0.0026799 = 2.246 × 10⁻⁴
1/T_2 = 0.0029045 → T_2 ≈ 344.3 K ≈ 71 °C
So water boils at roughly 71 °C on Everest — cool enough that a boiled egg never fully sets and pressure cookers become essential. The measured value is about 68–71 °C depending on the exact summit pressure, so a one-line calculation with a constant latent heat nails it to a couple of degrees. The small discrepancy comes from L_vap actually rising a little at lower temperatures, which the constant-L assumption ignores.
History
Benoît Paul Émile Clapeyron, a French engineer, published the exact form dP/dT = L/(T·ΔV) in 1834 in his memoir reworking Sadi Carnot's ideas — it predates the formal concept of entropy. Rudolf Clausius, having introduced entropy in the 1850s, recast the equation on a rigorous thermodynamic basis and derived the integrated vapor-pressure form that now carries both names. The relation is one of the earliest quantitative triumphs of thermodynamics, connecting a purely mechanical measurement (how pressure shifts a transition temperature) to a caloric quantity (latent heat) without any microscopic model.
Common misconceptions
- "It's an approximation." The Clapeyron equation dP/dT = L/(T·ΔV) is exact for any first-order transition. Only the integrated ln P versus 1/T Clausius-Clapeyron form is approximate, because it assumes an ideal vapor, a negligible liquid volume, and a constant L.
- "You can use Celsius." Never. T appears as an absolute temperature (and squared in the differential form), so it must be in Kelvin. A Celsius substitution gives nonsense, especially near 0 °C.
- "Vapor pressure depends on how much air is above the liquid." Vapor pressure is a property of the liquid–vapor equilibrium at a given temperature, independent of any inert gas present. Boiling occurs when this vapor pressure equals the total ambient pressure.
- "The straight line works all the way to the critical point." It breaks down there: ΔV → 0 and L → 0, the ideal-gas approximation fails badly, and the line simply terminates. The exact Clapeyron slope stays finite (0/0 resolved), but the log-linear plot curves.
- "Latent heat is constant." L_vap of water falls from 45.0 kJ/mol at 0 °C to 40.7 kJ/mol at 100 °C to zero at the 374 °C critical point. Over a narrow range the constant-L integral is fine; over a wide one you must keep L(T) inside the integral.
- "ΔV is the volume change of the liquid." ΔV is the difference between the two phases' molar volumes. For vaporization it is dominated by the gas (≈ 30,000 cm³/mol for steam versus ≈ 19 cm³/mol for water), which is exactly why the V_gas ≈ RT/P shortcut is so good.
JavaScript — Clausius-Clapeyron calculations
const R = 8.314; // J/(mol·K)
// Exact Clapeyron slope of a coexistence line, in Pa/K
// L in J/mol, T in K, dV (V_high - V_low) in m^3/mol
function clapeyronSlope(L, T, dV) {
return L / (T * dV);
}
// Vapor pressure at T2 given a reference point (P1, T1), integrated form
// Assumes constant latent heat L (J/mol)
function vaporPressure(P1, T1, T2, L) {
return P1 * Math.exp(-(L / R) * (1 / T2 - 1 / T1));
}
// Boiling temperature at a given ambient pressure (invert the integrated form)
function boilingTemp(Pambient, P1, T1, L) {
// ln(P/P1) = -(L/R)(1/T - 1/T1) -> solve for T
const invT = 1 / T1 - (R / L) * Math.log(Pambient / P1);
return 1 / invT;
}
// Latent heat from the slope of a ln P vs 1/T fit
function latentHeatFromSlope(slope) {
return -slope * R; // slope is d(ln P)/d(1/T)
}
const L_vap = 40700; // J/mol, water at 100 C
const P0 = 101325, T0 = 373.15; // 1 atm, 100 C
console.log(vaporPressure(P0, T0, 343.15, L_vap).toFixed(0), 'Pa at 70 C'); // ~32,000 Pa
console.log((boilingTemp(0.333 * P0, P0, T0, L_vap) - 273.15).toFixed(1), 'C on Everest'); // ~71 C
console.log(latentHeatFromSlope(-4895).toFixed(0), 'J/mol from slope -4895 K'); // ~40,700
Where it shows up
- Meteorology. Saturation vapor pressure of water follows Clausius-Clapeyron; a warmer atmosphere holds ~7%/°C more moisture, intensifying rainfall.
- Cooking and altitude. Pressure cookers raise P to raise the boiling point (≈121 °C at 2 atm); high-altitude kitchens fight the opposite effect.
- Vacuum and freeze-drying. Lyophilization uses the sublimation line — low pressure lets ice go straight to vapor without melting.
- Meteorology of ice / glaciology. The backward-sloping fusion line means high local pressure can lower the melting point of ice.
- Materials science. Solid–solid transition lines (e.g. graphite–diamond) obey the exact Clapeyron form, guiding high-pressure synthesis.
- Refrigeration. The vapor-pressure curve of a refrigerant sets the pressures at which it evaporates and condenses in the cycle.
Frequently asked questions
What is the Clausius-Clapeyron equation?
It is the exact relation for the slope of a coexistence line between two phases in a pressure–temperature diagram: dP/dT = L / (T·ΔV). Here L is the molar latent heat of the transition, T is the absolute temperature, and ΔV is the molar volume difference between the two phases. It follows directly from equating the Gibbs free energies (equal chemical potentials) of the two phases along the boundary, so it holds for melting, vaporization, and sublimation lines alike.
What is the difference between the Clapeyron and Clausius-Clapeyron equations?
The Clapeyron equation dP/dT = L/(T·ΔV) is exact for any first-order phase boundary. The Clausius-Clapeyron equation is the approximate integrated form for a liquid–vapor (or solid–vapor) line: you neglect the condensed-phase volume, treat the vapor as an ideal gas so ΔV ≈ RT/P, and assume L is roughly constant. That yields d(ln P)/dT = L/(RT²), which integrates to ln P = −L/(RT) + C — a straight line in ln P versus 1/T.
Why does water boil at a lower temperature at high altitude?
A liquid boils when its vapor pressure equals the surrounding atmospheric pressure. Atmospheric pressure drops with altitude — about 0.7 atm at 3,000 m — so the vapor pressure needed to boil is lower, and Clausius-Clapeyron predicts that lower vapor pressure occurs at a lower temperature. On top of Mount Everest (≈0.33 atm) water boils near 70 °C, which is why food takes far longer to cook and pressure cookers are used at altitude.
How do you get a straight line from vapor-pressure data?
Plot the natural logarithm of vapor pressure (ln P) against the reciprocal of absolute temperature (1/T). The integrated Clausius-Clapeyron equation ln P = −L/(R·T) + C predicts a straight line with slope −L/R. Multiplying the measured slope by −R (8.314 J·mol⁻¹·K⁻¹) directly yields the molar latent heat of vaporization. This is a standard undergraduate lab method for measuring enthalpies of vaporization.
Why does the melting line of water slope backwards?
For most substances ice-to-liquid ΔV is positive, so dP/dT is positive and the melting temperature rises with pressure. Water is anomalous: ice is less dense than liquid water, so ΔV_melt is negative. Since L_fusion and T are positive, dP/dT = L/(T·ΔV) is negative — the melting line leans left. Increasing pressure therefore lowers the melting point of ice by about 0.0074 K per atmosphere, an effect that helps explain why ice under high local pressure can melt.
What is the latent heat in the Clausius-Clapeyron relation?
L is the latent heat (enthalpy) absorbed to convert one mole (or one kilogram) from one phase to the other at constant temperature and pressure: L = T·ΔS. For water, L_vap ≈ 40.7 kJ/mol (2,257 kJ/kg) at 100 °C, L_fusion ≈ 6.01 kJ/mol (334 kJ/kg) at 0 °C, and L_sublimation ≈ 51.1 kJ/mol near the triple point. Because energy is a state function, L_sub ≈ L_fusion + L_vap when both are evaluated at the same temperature (near 0 °C, L_vap ≈ 45 kJ/mol, so 6.0 + 45 ≈ 51). Being the largest, L_sub makes the sublimation line the steepest of the three coexistence curves near the triple point.
Does the Clausius-Clapeyron relation work near the critical point?
No. The integrated form assumes the vapor is ideal, the liquid volume is negligible, and L is constant — all of which fail as you approach the critical point, where liquid and vapor densities converge, ΔV → 0, and L → 0. There the coexistence line simply ends. The exact Clapeyron equation dP/dT = L/(T·ΔV) remains valid right up to the critical point (both L and ΔV vanish together), but the convenient ln P versus 1/T straight line does not.