Absolute Entropies from Calorimetry

Imagine cooling a crystal of sulfur dioxide to within one degree of absolute zero, then slowly warming it while tracking every joule that flows in. That single, painstaking experiment is how chemists build a number as ordinary-looking as $S^\circ_{298} = 248.2\ \mathrm{J\,K^{-1}\,mol^{-1}}$ . Behind every tabulated "standard molar entropy" sits a calorimetric integration that starts from the Third Law's clean reference point: zero entropy for a perfect crystal at $T = 0$ . This matters because absolute entropies feed directly into equilibrium constants, reaction spontaneity, and phase diagrams. Unlike enthalpies, which are always measured as differences, entropies can be pinned to an absolute scale. That is only possible because the Third Law hands us a true zero. This article shows exactly how calorimetric data become that number. You will see the integration of heat capacity through solid, liquid, and gas regions. You will see how phase-transition entropies are stitched in. Then you will work two full numerical examples, from raw $C_p$ data to a final $S^\circ_{298}$ value.

Key takeaways

The Third Law sets $S(0) = 0$ for a perfect crystal, giving entropy an absolute zero point.
Absolute entropy at temperature $T$ is built by integrating $C_p/T$ from near 0 K up to $T$ .
Each phase transition (fusion, vaporization, solid-solid) contributes a discrete jump $\Delta_{trs}H/T_{trs}$ .
Below the lowest measured temperature, the Debye $T^3$ law extrapolates $C_p$ to zero.
Real substances sometimes show a residual entropy at $T=0$ , signaling frozen-in disorder.
Key idea: Absolute entropy is a sum of calorimetric heating integrals plus transition-entropy jumps, anchored at zero by the Third Law.

The core insight

Because the Third Law guarantees $S(0) = 0$ for a perfect crystalline substance, the entropy at any temperature $T$ is not a difference but an absolute quantity, obtainable purely from experiment as the sum of $\int C_p\,d\ln T$ segments across each phase plus $\Delta_{trs}H/T_{trs}$ at every transition encountered on the way up.

The physical picture

Entropy measures the number of microstates consistent with a system's macroscopic state. Near absolute zero, a perfect crystal has essentially one accessible microstate: every atom locked into its ideal lattice position, no configurational choices left. The Third Law formalizes this as $S_m(0) = 0$ for any pure, perfectly ordered crystalline substance. As the crystal warms, thermal energy populates vibrational modes, and the number of accessible microstates grows. Heat capacity measurements track exactly how much heat enters per unit of temperature rise, and since $dS = dq_{rev}/T$ at constant pressure with $dq_{rev} = C_p\,dT$ , every calorimetric heating curve is really an entropy-generating curve in disguise. Integrating $C_p/T$ from 0 K to any target temperature accumulates the entropy the substance has gained. The complication is that real substances change phase. A solid melts, a liquid vaporizes, sometimes a solid rearranges into a different crystal form (a solid-solid transition, common in polymorphic materials like sulfur or ammonium nitrate). At each transition temperature $T_{trs}$ , heat flows into the system at constant temperature, and this latent heat contributes an extra chunk of entropy equal to $\Delta_{trs}H/T_{trs}$ . The total absolute entropy at some final temperature is therefore a staircase: smooth $C_p/T$ integrals within each phase, punctuated by discrete vertical jumps at each transition. Below the lowest temperature a calorimeter can reliably reach — typically 10–15 K — direct measurement fails, but Debye's theory rescues the calculation. Debye showed that at sufficiently low temperature the heat capacity of a nonmetallic solid follows

C_p \approx C_V = aT^3

so the missing sliver of entropy between 0 K and the lowest measured point, $T_1$ , can be computed analytically rather than measured.

Figure

The entropy staircase from 0 K to 298 K, showing smooth heat-capacity integrals and vertical transition jumps

Figure

Heat capacity versus temperature, showing the Debye T-cubed rise and the spikes at each phase transition

The core relations

S(0) = 0 \quad \text{(perfect crystal, Third Law)}

where: the reference for all absolute entropies is a flawless crystal at absolute zero.

S_m(T) = S_m(0) + \int_0^{T} \frac{C_{p,m}}{T'}\,dT'

where: $C_{p,m}$ is the molar heat capacity at constant pressure, integrated over temperature within a single phase.

C_{p,m} = aT^3 \quad (T \lesssim 15\ \mathrm{K})

where: $a$ is a Debye constant fitted to low-temperature data, giving $S_m(T_1) = \tfrac{1}{3}C_{p,m}(T_1)$ .

\Delta_{trs}S_m = \frac{\Delta_{trs}H_m}{T_{trs}}

where: $\Delta_{trs}H_m$ is the molar enthalpy of transition (fusion, vaporization, or solid-solid) at the transition temperature $T_{trs}$ .

S^\circ_m(298\,\mathrm{K}) = \int_0^{T_1}\frac{C_p}{T}dT + \int_{T_1}^{T_{fus}}\frac{C_p(s)}{T}dT + \frac{\Delta_{fus}H}{T_{fus}} + \int_{T_{fus}}^{T_{vap}}\frac{C_p(l)}{T}dT + \frac{\Delta_{vap}H}{T_{vap}} + \int_{T_{vap}}^{298}\frac{C_p(g)}{T}dT

where: each integral is evaluated numerically from tabulated calorimetric $C_p$ data, plotted as $C_p/T$ against $T$ (or $C_p$ against $\ln T$ ).

Derivation AWhy the low-temperature Debye extrapolation gives

S = C_p/3

Goal. Find the entropy contained between $T=0$ and a small temperature $T_1$ , using only the Debye law $C_p = aT^3$ , without needing direct calorimetric data down to absolute zero.

Step 1. Start from the general heating relation at constant pressure,

S_m(T_1) - S_m(0) = \int_0^{T_1} \frac{C_{p,m}}{T}\,dT.

By the Third Law, $S_m(0) = 0$ for the perfect crystal, so $S_m(T_1)$ equals the integral itself.

Step 2. Insert the Debye low-temperature form $C_{p,m}(T) = aT^3$ , valid because at very low temperature only long-wavelength acoustic phonons are thermally excited, and their density of states forces this cubic dependence:

S_m(T_1) = \int_0^{T_1} \frac{aT^3}{T}\,dT = a\int_0^{T_1} T^2\,dT.

Step 3. Carry out the integration explicitly:

S_m(T_1) = a\left[\frac{T^3}{3}\right]_0^{T_1} = \frac{aT_1^3}{3}.

Step 4. Recognize that $aT_1^3$ is just $C_{p,m}(T_1)$ itself, the measured heat capacity at $T_1$ . Substituting,

S_m(T_1) = \frac{C_{p,m}(T_1)}{3}.

Step 5. This means a single heat-capacity measurement near, say, 10–15 K — combined with confidence that the $T^3$ law holds below it — gives the entire entropy content up to that point without ever measuring anything colder. This result is what lets tables report entropies with data that only start around 10–15 K rather than at absolute zero, an experimentally impossible target.

\boxed{S_m(T_1) = \tfrac{1}{3}C_{p,m}(T_1)}

Worked examples

Example 1 — Foundational: entropy of a solid warmed from 15 K to its melting point. Suppose a molecular crystal has been measured calorimetrically. At $T_1 = 15\,\mathrm{K}$ , $C_{p,m} = 3.9\ \mathrm{J\,K^{-1}\,mol^{-1}}$ , consistent with Debye behavior below this point. Tabulated $C_p/T$ values (from smoothed calorimetric data) integrate, by Simpson's rule, to

\int_{15}^{200} \frac{C_{p,m}}{T}\,dT = 45.0\ \mathrm{J\,K^{-1}\,mol^{-1}}

up to the melting point $T_{fus} = 200\,\mathrm{K}$ . The enthalpy of fusion is $\Delta_{fus}H_m = 8.20\ \mathrm{kJ\,mol^{-1}}$ . Step 1 — Debye contribution below 15 K.

S_m(15\,\mathrm{K}) = \frac{C_{p,m}(15\,\mathrm{K})}{3} = \frac{3.9\ \mathrm{J\,K^{-1}\,mol^{-1}}}{3} = 1.30\ \mathrm{J\,K^{-1}\,mol^{-1}}

Step 2 — Solid-phase heating integral (15 K to 200 K). Given directly: $45.0\ \mathrm{J\,K^{-1}\,mol^{-1}}$ . Step 3 — Fusion entropy.

\Delta_{fus}S_m = \frac{\Delta_{fus}H_m}{T_{fus}} = \frac{8200\ \mathrm{J\,mol^{-1}}}{200\ \mathrm{K}} = 41.0\ \mathrm{J\,K^{-1}\,mol^{-1}}

Step 4 — Total entropy at the melting point (liquid, just above 200 K).

S_m(200\,\mathrm{K}, l) = 1.30 + 45.0 + 41.0 = 87.3\ \mathrm{J\,K^{-1}\,mol^{-1}}

Sanity check: this magnitude sits comfortably in the typical 60–120 J K⁻¹ mol⁻¹ range for small organic molecules at their melting points, and the fusion term dominates as expected since latent heat entropy is usually the largest single contribution for molecular solids.

Example 2 — Applied: full standard molar entropy of a substance at 298 K, including vaporization. This mirrors the classic determination for a volatile liquid such as chloroform-type molecules. Data (illustrative, physically reasonable values):

Debye region: $C_{p,m}(13\,\mathrm{K}) = 2.70\ \mathrm{J\,K^{-1}\,mol^{-1}}$
Solid heating, 13 K to $T_{fus} = 210\,\mathrm{K}$ : $\int C_p/T\,dT = 62.4\ \mathrm{J\,K^{-1}\,mol^{-1}}$
Fusion: $\Delta_{fus}H_m = 6.40\ \mathrm{kJ\,mol^{-1}}$ at $210\,\mathrm{K}$
Liquid heating, 210 K to $T_{vap} = 335\,\mathrm{K}$ : $\int C_p/T\,dT = 38.9\ \mathrm{J\,K^{-1}\,mol^{-1}}$
Vaporization: $\Delta_{vap}H_m = 29.4\ \mathrm{kJ\,mol^{-1}}$ at $335\,\mathrm{K}$
Gas heating, 335 K to 298 K correction (the gas is superheated above its boiling point, then the final state is corrected down to 298 K using ideal-gas $C_{p,m}(g) \approx 33.6\ \mathrm{J\,K^{-1}\,mol^{-1}}$ ; since 298 K is below 335 K here the gas segment is actually a cooling correction)

Step 1 — Debye contribution.

S_m(13\,\mathrm{K}) = \frac{2.70}{3} = 0.900\ \mathrm{J\,K^{-1}\,mol^{-1}}

Step 2 — Sum solid, fusion, liquid, vaporization.

\Delta_{fus}S_m = \frac{6400}{210} = 30.5\ \mathrm{J\,K^{-1}\,mol^{-1}}

\Delta_{vap}S_m = \frac{29400}{335} = 87.8\ \mathrm{J\,K^{-1}\,mol^{-1}}

Running total at 335 K (gas, just above boiling point):

S_m(335\,\mathrm{K},g) = 0.900 + 62.4 + 30.5 + 38.9 + 87.8 = 220.5\ \mathrm{J\,K^{-1}\,mol^{-1}}

Step 3 — Correct from 335 K down to 298 K in the gas phase. Using $C_{p,m}(g) \approx 33.6\ \mathrm{J\,K^{-1}\,mol^{-1}}$ (roughly constant over this narrow range) and integrating $C_p \ln(T_2/T_1)$ :

\Delta S = C_{p,m}\ln\left(\frac{298}{335}\right) = 33.6 \times \ln(0.8896) = 33.6\times(-0.1170) = -3.93\ \mathrm{J\,K^{-1}\,mol^{-1}}

Step 4 — Final standard molar entropy.

S^\circ_m(298\,\mathrm{K}) = 220.5 - 3.93 = 216.6\ \mathrm{J\,K^{-1}\,mol^{-1}}

Sanity check: 216.6 J K⁻¹ mol⁻¹ is a very plausible standard entropy for a small polyatomic gas-phase molecule (compare CHCl₃(g), literature value near 295.7 J K⁻¹ mol⁻¹ — the difference is expected since the illustrative $\Delta_{vap}H$ and heat capacities here are simplified, not the exact literature data set). The vaporization step contributes by far the largest single jump, consistent with Trouton's rule intuition that $\Delta_{vap}S \approx 85$ –90 J K⁻¹ mol⁻¹ for normal liquids.

Limiting cases & common traps

Situation	What to watch for
Substance has multiple solid polymorphs	Each solid-solid transition adds its own $\Delta_{trs}H/T_{trs}$ term; omitting one under-counts entropy substantially
Very low $T_1$ (below 10 K) unreachable	Debye $T^3$ extrapolation is essential; a poor fit constant $a$ propagates error into every entry above it
Glass-forming liquids that vitrify instead of crystallizing	The substance may never reach a true perfect-crystal reference state; the Third Law's zero may not apply cleanly
Molecules with orientational disorder (e.g. CO, N₂O in the solid)	A nonzero residual entropy can appear at $T=0$ , meaning the extrapolated integral undershoots the true spectroscopic entropy
Confusing $C_p$ and $C_V$	Calorimetry measures $C_p$ ; the difference from $C_V$ is negligible for solids at low $T$ but matters for gases

The chemist behind it

William Giauque spent much of his career pushing calorimetry to temperatures nobody else could reach, developing adiabatic demagnetization to cool samples into the millikelvin range and using painstaking low-temperature heat-capacity measurements to test the Third Law itself. His comparisons of calorimetric ("Third-Law") entropies against spectroscopic entropies for gases like CO and N₂O revealed the residual entropy discrepancies that later made sense through Pauling's statistical treatment of orientational disorder in molecular crystals. Giauque's meticulous experimental work turned the Third Law from a theoretical postulate into a quantitatively verified, load-bearing pillar of chemical thermodynamics.

Origins · 1 minWilliam F. Giauque and how this was discovered

William F. Giauque

1895-1982 · Pioneered low-temperature calorimetry and adiabatic demagnetization, testing the Third Law experimentally and discovering residual entropy in molecular crystals

Ask the tutor: How does residual entropy in molecular crystals connect to Pauling's statistical treatment of ice and carbon monoxide?

Why is the Third Law needed at all if we could just measure entropy differences like we do for enthalpy?

Entropy differences alone would work for many purposes, but chemists need absolute entropies to compute equilibrium constants and reaction entropies from tabulated single-substance values, using $\Delta_rS^\circ = \sum \nu S^\circ_{products} - \sum \nu S^\circ_{reactants}$ . This only works if each $S^\circ$ is on a common absolute scale. The Third Law supplies that scale by declaring $S(0)=0$ for a perfect crystal, turning every subsequent calorimetric integral into an absolute rather than relative quantity.

Why does the calorimetric integration use Cp and not Cv?

Nearly all calorimetry is done at constant (typically atmospheric) pressure, so $C_p$ is the directly measured quantity, and $dS = C_p\,dT/T$ follows directly from $dH = TdS$ at constant pressure. For solids at low temperature, $C_p$ and $C_V$ differ negligibly because thermal expansion is tiny. For gases the difference becomes significant, but by the time integration reaches the gas phase, ideal-gas relations and known $C_{p,m}$ values handle the correction directly.

What happens if a substance does not crystallize on cooling but instead forms a glass?

A glass is a frozen, disordered liquid structure rather than a true crystal, so it does not reach the ordered, single-microstate condition the Third Law assumes. Its calorimetric entropy at very low temperature approaches a nonzero, path-dependent residual value rather than zero. Reported Third-Law entropies for glass-forming substances therefore carry an extra uncertainty, and careful work often reports the entropy of the corresponding crystalline form instead, when it exists.

How accurate are Third-Law entropies compared to values from statistical mechanics and spectroscopy?

For many well-behaved substances, agreement is excellent, typically within 1 to 2 J K⁻¹ mol⁻¹. Discrepancies of several J K⁻¹ mol⁻¹ signal genuine residual entropy from molecular disorder, as found for carbon monoxide, nitrous oxide, and ice. These comparisons were historically crucial. They confirmed the Third Law's validity for ordered solids while also revealing its limits, motivating the statistical treatment of configurational disorder in imperfect crystals.