Thermodynamics
Heat Capacity Cv vs Cp
Cp − Cv = R for ideal gas (Mayer); γ = Cp/Cv ≈ 1.67 monatomic, 1.40 diatomic, 1.33 polyatomic — sets adiabatic exponent
Heat capacity is the amount of heat required to raise a substance's temperature by 1 K. Two flavours dominate every gas calculation: Cv = (∂U/∂T)V at constant volume, where all heat goes into internal energy; and Cp = (∂H/∂T)P at constant pressure, where the gas additionally does P·dV expansion work. For any ideal gas, Mayer's relation gives Cp − Cv = nR, with R = 8.314 J/(K·mol). The dimensionless ratio γ = Cp/Cv sets the adiabatic exponent in PVγ = const and reads 5/3 ≈ 1.667 for monatomic gases (He, Ar, Ne, Kr, Xe), 7/5 = 1.40 for diatomic gases at room temperature (H₂, N₂, O₂), and ~4/3 = 1.33 for polyatomic molecules with active rotations and partially-active vibrations. The descending sequence reflects increasing internal degrees of freedom soaking up heat that would otherwise raise temperature.
- MayerCp − Cv = nR (ideal gas)
- Monatomic γ5/3 ≈ 1.667 (He, Ar)
- Diatomic γ7/5 = 1.40 (N₂, O₂, air)
- Polyatomic γ~4/3 ≈ 1.33 (CO₂, NH₃)
- Sound speedc = √(γRT/M)
- Dulong-PetitCv ≈ 3R for most solids at 298 K
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 Cv vs Cp matters
- Sets sound speed in air. Newton calculated c = √(P/ρ) using isothermal compression and got 280 m/s — 18% too low. Laplace 1816 corrected by recognizing that sound compressions are adiabatic, giving c = √(γP/ρ) = √(γRT/M). For dry air at 293 K with γ = 1.401, M = 0.02897 kg/mol: c = 343.2 m/s, matching measurement. The first triumph of γ = Cp/Cv beyond calorimetry.
- Diesel engines exploit adiabatic heating. Compressing air 16:1 in a diesel piston with γ = 1.40 raises T from 300 K to 300·16^0.4 = 909 K — above the autoignition temperature of diesel fuel (~483 K). Spark plugs are unnecessary. Otto-cycle gasoline engines compress 8-12:1 to T ≈ 600-700 K, intentionally below autoignition so the spark plug controls combustion timing.
- Gas-turbine power plant cycles. Brayton-cycle compressors raise air from 300 K and 1 bar to ~600 K and 20 bar (compression ratio ~7), with the temperature rise determined entirely by γ and ratio. Combined-cycle plants achieve 60-62% net efficiency by stacking a Brayton (γ_air ≈ 1.4) topping cycle and Rankine (γ irrelevant — water phase change dominates) bottoming cycle.
- Joule-Thomson coefficient depends on Cp. μ_JT = (∂T/∂P)_H = (1/Cp)·[T·(∂V/∂T)_P − V]. The free expansion through a porous plug cools real gases (μ > 0) below their inversion temperature: 621 K for N₂, 23 K for He. The Linde air-liquefaction process (Carl von Linde 1895) uses J-T cooling to liquefy air at 78 K, a $30 billion/year industrial-gas market.
- Calorimetry of substances starts with Cv or Cp. Bomb calorimetry measures ΔU at constant V using the bomb's known C; coffee-cup calorimetry measures ΔH at constant P. The relation between them, ΔH = ΔU + Δn·RT, is just the integrated Mayer relation applied to the gaseous moles produced or consumed.
- Climate sensitivity hinges on atmospheric Cp. The dry adiabatic lapse rate is Γ_d = g/Cp = 9.81/1005 = 9.76 K/km (g = gravity, Cp of air = 1005 J/(K·kg)). Saturated lapse rate is ~5 K/km because latent heat of condensation reduces effective Cp. These two numbers underwrite tropospheric thermodynamics and weather-balloon analysis.
- Quantum signature in gas thermodynamics. The temperature dependence of γ for H₂, N₂, O₂ — rising as vibrations freeze out at low T — was historically a major puzzle. Solving it required Planck's quanta and ultimately quantum partition functions: each vibrational mode contributes R only when k_B T ≫ ℏω, exactly the same Bose-Einstein cutoff that resolves blackbody radiation.
Common misconceptions
- "Cp − Cv = R holds for solids." No — Mayer's relation requires the ideal-gas equation of state. For solids and liquids the general relation Cp − Cv = α²·T·V/κ_T applies, where α is thermal expansion and κ_T is isothermal compressibility. Liquid water at 298 K: α = 2.57 × 10⁻⁴ K⁻¹, κ_T = 4.59 × 10⁻¹⁰ Pa⁻¹, V = 18.07 cm³/mol gives Cp − Cv = 0.466 J/(K·mol), versus Cp = 75.3 J/(K·mol) — only 0.6%, not the 8.314 J/(K·mol) of an ideal gas.
- "γ is a constant for each gas." γ depends on temperature because vibrational modes activate above their characteristic temperature θ_vib. CO₂ has γ = 1.295 at 300 K but 1.169 at 1500 K because three of its four vibrational modes (bending at 667 cm⁻¹ doubly-degenerate, asymmetric stretch at 2349 cm⁻¹, symmetric stretch at 1388 cm⁻¹) progressively activate.
- "Diatomic gases always have γ = 1.40." Only at temperatures where 5 modes are active (3 translational + 2 rotational, vibrations frozen). H₂ at 50 K has γ ≈ 1.67 because rotations also freeze (θ_rot(H₂) = 87.6 K is unusually high). At 5000 K, H₂ vibrations fully activate and γ → 9/7 ≈ 1.29. Air's γ varies from 1.401 at 300 K to 1.290 at 2000 K — significant for shock-tube and rocket-nozzle calculations.
- "Cv measures heat absorbed at constant volume." Cv is a state-function derivative (∂U/∂T)_V, defined regardless of process. The constant-V path is just the easiest to measure it via δq = dU. Adding δq at constant V is equivalent to dU because no expansion work is done.
- "Equipartition is exact." Equipartition is the high-T classical limit. Each quadratic degree of freedom contributes ½k_B T to U only when k_B T ≫ characteristic energy scale. For rotation θ_rot ~ ℏ²/(2Ik_B); for vibration θ_vib ~ ℏω/k_B. Below these temperatures the contribution shrinks toward zero. Equipartition's deviations were the first hints of quantum statistics for matter, predating Planck's blackbody result.
- "Cp/Cv tells you molecular structure unambiguously." γ measures only the count of active modes, not their identity. Methane (CH₄, 5 atoms) and ammonia (NH₃, 4 atoms) both report γ ≈ 1.31 at 298 K because both have 3 + 3 = 6 active translational + rotational degrees plus partially active vibrations — the spectroscopic and structural detail is invisible to a single thermal measurement.
Derivation
Start from the first law dU = δq − δw with δw = P·dV for quasi-static expansion. At constant volume, dV = 0 so δq_V = dU and Cv = (∂U/∂T)_V. At constant pressure define enthalpy H = U + PV. Then dH = dU + P·dV + V·dP, and at constant P this collapses to dH = dU + P·dV = δq_P. So Cp = (∂H/∂T)_P. Subtract: Cp − Cv = (∂H/∂T)_P − (∂U/∂T)_V. Using H = U + PV and the ideal-gas equation PV = nRT: (∂(PV)/∂T)_P = nR. The full general result is Cp − Cv = T·(∂P/∂T)_V·(∂V/∂T)_P, and for the ideal gas both partials simplify to give Cp − Cv = nR exactly.
The microscopic content comes from equipartition. The internal energy of a gas in the high-T classical limit is U = (f/2)nRT where f is the number of quadratic degrees of freedom in the molecular Hamiltonian. So Cv = (f/2)R, Cp = ((f+2)/2)R, γ = (f+2)/f. Monatomic: f = 3 (translational), γ = 5/3. Diatomic with frozen vibrations: f = 5 (3 trans + 2 rot), γ = 7/5 = 1.40. Diatomic with active vibrations: f = 7 (adding 2 for vibration kinetic+potential), γ = 9/7 ≈ 1.29. Linear polyatomic CO₂: f = 7 frozen vib up to 9 active, γ between 1.29 and 1.18. Bent polyatomic H₂O: f = 6 (3 trans + 3 rot) frozen vib, γ = 4/3 = 1.33; with active vibrations γ → 13/12 ≈ 1.08 at very high T.
For an adiabatic quasi-static process (δq = 0): dU = −P·dV, with dU = Cv·dT and PV = nRT giving Cv·dT = −(nRT/V)·dV. Separating and integrating: ln T + (R/Cv)·ln V = const. Using R/Cv = γ − 1: TV^(γ−1) = const, equivalent to PV^γ = const and TP^((1−γ)/γ) = const. These three relations dominate gas-dynamics calculations from sound waves to rocket nozzles.
γ = Cp/Cv for representative gases at 298 K
| Gas | Type | Cv (J/(K·mol)) | Cp (J/(K·mol)) | γ | Note |
|---|---|---|---|---|---|
| He | Monatomic noble | 12.47 | 20.79 | 1.667 | Matches 5/3 to 3 decimals |
| Ar | Monatomic noble | 12.47 | 20.79 | 1.667 | 5/3, 78% of dry air's argon constituent |
| H₂ | Diatomic | 20.46 | 28.78 | 1.407 | θ_vib = 5980 K, vibrations frozen |
| O₂ | Diatomic | 20.95 | 29.27 | 1.397 | θ_vib = 2270 K — slight vib activation |
| N₂ | Diatomic | 20.79 | 29.10 | 1.400 | 78% of air; air γ ≈ 1.401 |
| CO₂ | Linear triatomic | 28.50 | 36.81 | 1.291 | Bending mode at 667 cm⁻¹ active |
| NH₃ | Pyramidal tetra-atomic | 27.30 | 35.61 | 1.305 | Inversion mode active above 200 K |
| CH₄ | Tetrahedral pentatomic | 27.40 | 35.71 | 1.304 | Multiple low-frequency vibs partially active |
| H₂O (steam, 373 K) | Bent triatomic | 25.95 | 34.30 | 1.322 | γ rises with T as bending mode freezes |
Equipartition predictions vs measurement
| Molecule type | Modes (trans + rot + vib) | f classical | Cv classical | γ classical | γ measured (298 K) |
|---|---|---|---|---|---|
| Monatomic | 3 + 0 + 0 | 3 | (3/2)R = 12.47 | 5/3 = 1.667 | 1.667 ✓ |
| Diatomic frozen vib | 3 + 2 + 0 (vib frozen) | 5 | (5/2)R = 20.79 | 7/5 = 1.400 | 1.40 (N₂, O₂, H₂) |
| Diatomic full classical | 3 + 2 + 2 | 7 | (7/2)R = 29.10 | 9/7 = 1.286 | Approaches at T ≫ θ_vib |
| Linear polyatomic (n atoms) | 3 + 2 + 2(3n−5) | 5 + 4(3n−5)/2 | varies | varies | CO₂ 1.29 partial vib |
| Bent polyatomic frozen vib | 3 + 3 + 0 | 6 | 3R = 24.94 | 4/3 ≈ 1.333 | H₂O steam 1.32 ✓ |
| Bent polyatomic full classical | 3 + 3 + 2(3n−6) | 6 + 2(3n−6) | varies | → 1 + 1/(3n) | CH₄ 1.31 partial vib |
Adiabatic compression results for various γ
| Compression ratio | Gas | γ | T_initial | T_final = T_i·r^(γ−1) | Application |
|---|---|---|---|---|---|
| 10:1 | Air | 1.40 | 300 K | 754 K | Otto-cycle gasoline engine |
| 16:1 | Air | 1.40 | 300 K | 909 K | Diesel engine — autoignition |
| 20:1 | Air | 1.40 | 300 K | 994 K | Marine slow-speed diesel |
| 2:1 | He | 1.667 | 4 K | 6.35 K | Pulse-tube cryocooler stage |
| 5:1 | CO₂ | 1.291 | 300 K | 477 K | CO₂ heat pump cycles |
| 3:1 | Steam | 1.32 | 373 K | 522 K | Steam turbine first stage |
Famous experiments and applications
- Robert Mayer 1842 — first law from gas heat capacities. Julius Robert von Mayer published in Annalen der Chemie und Pharmacie that the difference between Cp and Cv of air, multiplied by the temperature change, must equal the work of expansion. This deduction predated Joule's paddle-wheel experiment by three years and established conservation of energy on theoretical grounds. Mayer's mechanical equivalent of heat (1 cal = 3.6 J) was crude; Joule refined it to 4.184 J/cal in 1845.
- Laplace 1816 — sound speed correction. Pierre-Simon Laplace recognized that Newton's isothermal calculation of sound speed (1687, c = √(P/ρ)) was 18% low because acoustic compressions in a sound wave happen too fast for heat exchange. Substituting γP for P recovered c = √(γRT/M) and the measured 343 m/s in air. The first physical use of the adiabatic exponent.
- Dulong-Petit 1819 — solid-element heat capacities. Pierre Louis Dulong and Alexis Petit measured Cv of 13 solid elements and found a near-universal molar value of ~3R = 24.94 J/(K·mol). Failures (diamond, beryllium, boron) were unexplained until Einstein's 1907 quantum oscillator model and Debye's 1912 phonon spectrum, which introduced the Debye temperature θ_D and predicted Cv/3R as a universal function of T/θ_D.
- Eucken 1912-1913 — first low-T heat capacity of H₂. Arnold Eucken at Berlin measured the Cv of H₂ from 30 K to 290 K and showed it fell from 5R/2 to 3R/2 as rotational modes froze, consistent with quantization of rotation. This was a key experimental input for Otto Stern's 1933 ortho/para hydrogen demonstration and supported Bohr's old quantum theory before Heisenberg/Schrödinger.
- Linde 1895 air liquefaction at Munich. Carl von Linde commercialized Joule-Thomson cooling using counter-current heat exchange and free expansion through a throttle valve. Air at 200 bar and 290 K cooled to 78 K and partially liquefied. The whole cycle's design hinges on Cp(T,P) and the J-T coefficient (1/Cp)·[T(∂V/∂T)_P − V], making accurate Cp data for air essential. Linde's company became the world's largest industrial-gas supplier — currently $30+ billion in annual O₂, N₂, Ar, He.
Frequently asked questions
Why is Cp greater than Cv?
At constant volume, all heat input δq = dU goes into raising internal energy and hence temperature. At constant pressure, the gas expands as it heats, so input δq = dU + P·dV — the system spends some heat on doing expansion work against the surroundings. For 1 mole of ideal gas, PV = nRT means at constant P the expansion work per K is P·dV/dT = nR. Mayer's relation Cp − Cv = nR follows immediately, with R = 8.314 J/(K·mol). For solids and liquids, where thermal expansion is small, Cp − Cv is typically ~3-10% of Cv (e.g. liquid water at 25 °C: Cp − Cv ≈ 0.5 J/(K·mol) versus Cp = 75.3 J/(K·mol)) — 6 ‰ — but for ideal gases the gap is large enough to give γ in the 1.3 to 1.7 range.
What is Mayer's relation and where did it come from?
Cp − Cv = nR for any ideal gas. Julius Robert von Mayer (1842) was a German physician who, while working as a ship's surgeon in Java, noticed venous blood was unexpectedly red — interpreted as less oxygen consumed at high ambient temperatures. He used the difference between the empirical Cp and Cv of air to back out the mechanical equivalent of heat: knowing Cp/Cv from sound-velocity measurements (Laplace had derived c = √(γP/ρ) in 1816) and the absolute Cv from calorimetry, the difference Cp − Cv equals the work done by 1 mol of gas expanding by ΔV at constant P, which is RΔT — providing one of the earliest derivations of the first law of thermodynamics, three years before Joule's paddle-wheel result.
Why does γ equal 5/3 for monatomic gases?
Equipartition assigns ½k_B T to each independent quadratic degree of freedom in the Hamiltonian. A monatomic gas has only 3 translational degrees of freedom (vx, vy, vz) — no rotations because a point-like atom has no moment of inertia, and no vibrations because there are no internal bonds. So U = (3/2)nRT, Cv = (3/2)R = 12.47 J/(K·mol), Cp = Cv + R = (5/2)R = 20.79 J/(K·mol), and γ = Cp/Cv = 5/3 ≈ 1.667. Helium, neon, argon, krypton, and xenon all measure within 0.1% of this value at 298 K. Mercury vapour at 600 K, the only common monatomic non-noble gas, also falls in line.
Why does γ for diatomic H₂ depend on temperature?
A diatomic molecule has 3 translational + 2 rotational + 2 vibrational (kinetic + potential) = 7 quadratic degrees of freedom in the classical limit, predicting Cv = (7/2)R and γ = 9/7 ≈ 1.29. Real diatomics measure γ ≈ 1.40 at room temperature (5 active modes — vibrations are 'frozen') and approach 1.29 only well above the characteristic vibrational temperature θ_vib (5980 K for H₂, 3354 K for N₂, 2270 K for O₂). Below θ_rot (87.6 K for H₂) rotations also freeze, leaving only translations and γ → 5/3. This stepwise quantization of accessible modes was the first quantitative evidence for quantum mechanics in thermodynamics, debated by Boltzmann and Planck in the 1890s.
What is the adiabatic exponent and why does it appear everywhere?
For a quasi-static adiabatic process (δq = 0) in an ideal gas, PV^γ = const, TV^(γ−1) = const, and TP^((1−γ)/γ) = const. The exponent γ = Cp/Cv determines: speed of sound c = √(γRT/M) (so air at 293 K: c = √(1.40·8.314·293/0.029) = 343 m/s); compressor and turbine work in Brayton/Diesel/Otto cycles; temperature rise in adiabatic compression (in a diesel engine compressing air 16:1, T rises from 300 K to T·16^0.4 = 909 K — high enough to ignite injected fuel); ratio of bulk-modulus stiffness to atmospheric pressure for sound waves; and supersonic flow relations (Mach-angle, normal shocks).
What is the Dulong-Petit law for solids?
Pierre Louis Dulong and Alexis Petit (1819) measured the heat capacities of 13 solid elements and found that molar Cv ≈ 3R = 24.94 J/(K·mol) for almost all of them at room temperature. The classical interpretation: each atom in a 3D lattice has 3 vibrational modes, each contributing R per equipartition (½k_BT kinetic + ½k_BT potential = k_BT), giving Cv = 3R independent of element. The law breaks down for light elements (diamond's Cv at 298 K is 6.1 J/(K·mol), one quarter of 3R) and at low temperatures. Einstein's 1907 quantum solid model and Debye's 1912 phonon-spectrum refinement explained the deviations: Cv/3R = D(T/θ_D) where θ_D is the Debye temperature (88 K for lead, 215 K for silver, 343 K for copper, 1860 K for diamond).