Equation of State of an Ideal Solid — The Grüneisen Parameter and the Mie–Grüneisen Equation of State

A solid expands when it is heated. That sounds far too obvious, but as long as the springs are ideal springs, a solid never expands at all. To expand, the force between atoms must be “asymmetric between stretching and compression” — that is, anharmonic. Here we squeeze that anharmonicity into a single number, the Grüneisen parameter, and derive the Mie–Grüneisen equation of state from a thermodynamic relation and the free energy of phonons. Then, once we impose the quasiharmonic approximation (QHA), the volumetric thermal expansion coefficient becomes αV = γCV/(BTV) — a surprisingly simple expression. Finally, we look at which factor in this expression actually governs thermal expansion, and why there are materials that shrink when heated (negative thermal expansion).

1. Putting “Expansion” into an Equation

The volumetric thermal expansion coefficient is how much the volume grows when the temperature is raised while the pressure is held constant.

αV≡ 1V (∂V ∂T)P (1)

Here V is the mean atomic volume, that is, the volume of the lattice divided by the number of atoms in it — this way, materials of different composition can be compared on an equal footing.

This definition is awkward to handle as it stands, so we rewrite it using the triple-product relation ((∂V/∂T)P (∂T/∂P)V (∂P/∂V)T =−1) as follows.

αV= − 1V (∂V ∂P)T (∂P ∂T)V = 1BT (∂P ∂T)V (2)

The first factor −1V (∂V/∂P)T is the compressibility κ, which is the reciprocal of the bulk modulus BT. In other words, to evaluate Eq. (1) we only need to calculate how much the internal pressure of the crystal rises when the temperature is raised at fixed volume — nothing more.

Why the internal pressure? When heated, atoms vibrate violently and try to push their surroundings outward. This is the thermal pressure. The amount by which the crystal actually expands is set by “the pushing force (how fast the thermal pressure grows)” ÷ “the pushing-back force (the bulk modulus).” The stiffer the material, the less it expands, because this denominator is large — the end of the story is already written here.

2. An Ideal Spring Does Not Expand

Take a pair of atoms and let x be the displacement from the equilibrium position. If the potential is a perfect parabola U(x)= 12kx2 then the Boltzmann distribution e−U/kBT is symmetric in x. No matter how high the temperature is raised,

⟨x⟩=0

— the atoms merely vibrate violently, and their mean position does not move by even a millimeter. However many harmonic oscillators you line up, thermal expansion never comes out.

So let us add a cubic term. The side where the atom stretches (x > 0) is easier to climb — conversely, the compressed side rises steeply. This is the asymmetry. However, if we add only the cubic term, then as x is made large the potential falls away and has no bottom (U → −∞). In that case the integral of the Boltzmann factor e−U /kBT (the partition function) diverges, and the average itself cannot even be defined. So we add a quartic term that makes a bottom.

U(x)= 12kx2 −gx3 +λx4 , ⟨x⟩≃ 3gkBT k2 (3)
The quartic term does not shift the mean λx⁴ is an even function of x, so in lowest-order perturbation it leaves ⟨x⟩ unchanged. What shifts the mean is only the cubic term, which is an odd function, and the expression on the right of Eq. (3) comes from it. The quartic term is included only to “make a bottom, so that the partition function converges.” In Fig. 1 we take λ = g²/k, and then the bracket in U′ = x(k − 3gx + 4λx²) is always positive (discriminant 9g² − 16λk = −7g² < 0), so there is only one valley, at x = 0.

The mean position shifts in proportion to the temperature. This is the most elementary picture of thermal expansion. Check it in Fig. 1 — set g = 0, and the mean position does not move however high you raise the temperature. Note, though, that Eq. (3) is a classical approximation that takes the average with the Boltzmann distribution, so it is valid only at high temperatures — quantum mechanically, the energy of an oscillator including its zero-point motion enters in place of kBT, and as T → 0 the shift of the mean position no longer depends on temperature (this corresponds to αV → 0 in Section 7).

Fig. 1. In an anharmonic potential, the mean position shifts. The light shading is the Boltzmann distribution e−U/kBT, and the vertical line is its mean ⟨x⟩. We fix k = 15 eV/Ų and take the stabilizing quartic term to be λ = g²/k (for a reduced mass of 30 u, this spring constant corresponds to about 11 THz — the frequency is not determined by the spring constant alone; the mass is needed). Note also that the shift is very small — thermal expansion is, by nature, a phenomenon of this size.

In a crystal, anharmonicity shows up as the volume dependence of the vibration frequencies For a pair of atoms a cubic term is enough, but in a crystal N atoms are connected to one another. So we usually change our viewpoint and measure the anharmonicity by how the phonon frequencies change when the volume of the crystal changes. For an ideal spring, changing the volume does not change the frequency — because the spring constant is the spring’s own business. In a real crystal, expanding it loosens the bonds and lowers the frequencies.

3. The Grüneisen Parameter

For a phonon with wave vector q and band ν, we define the mode Grüneisen parameter as follows.

γqν≡ −V ωqν (∂ωqν ∂V)T = − ∂ln⁡ωqν ∂ln⁡V (4)

The form on the right is easier to read — it is by what percentage the frequency drops when the volume is increased by 1 %. The minus sign is there so that “it drops when expanded” comes out positive.

4. The Helmholtz Free Energy of a Crystal

The Helmholtz free energy of a crystal with volume V and temperature T can be written as the sum of the static energy at 0 K, E(V), and the phonon part.

Using the same normalization for every quantity Below, E, F, U, CV, and the mode sum ∑q,ν are all written in the same normalization — everything for the whole crystal, or everything per atom (in that case the sum carries a factor 1/Natom in front of it). The αV that appears at the end is an intensive quantity, so as long as CV and V are taken in the same normalization, Eq. (11) holds as it stands, whichever way you write it. Since we defined V as the mean atomic volume in Section 1, please read everything on this page per atom.
F(V,T)= E(V)+ ∑q,ν [ ℏωqν2 +kBT ln⁡ (1− e− ℏωqν /kBT ) ] (5)

The first term is the zero-point motion, and the second term is the part excited by heat. This expression itself follows from the partition function of the harmonic oscillator — I wrote it up in the derivation of the harmonic oscillator and the derivation of the specific heat. The internal energy of one mode is

Uqν= ( 1 eℏωqν /kBT −1 +12 ) ℏωqν (6)

The quantity in parentheses is “the number of phonons in that mode + 1/2.” Even at absolute zero, the 1/2 remains — the zero-point motion cannot be removed.

5. The Mie–Grüneisen Equation of State

The pressure is obtained by differentiating the free energy with respect to the volume: P=− (∂F/∂V)T. The second term of Eq. (5) does not contain V explicitly and depends on the volume only through ωqν, so by the chain rule we have

∂Fph ∂V = ∑q,ν ∂Fph ∂ωqν ∂ωqν ∂V , ∂Fph ∂ωqν = ℏ2+ ℏeℏω /kBT −1 = Uqν ωqν

Here is the satisfying part — when we differentiate, exactly Uqν/ωqν from Eq. (6) appears. All that remains is to substitute ∂ωqν /∂V= − γqν ωqν/V from Eq. (4).

P+ dEdV = 1V ∑q,ν γqν Uqν (7)

This is the Mie–Grüneisen equation of state (G. Mie, 1903 / E. Grüneisen, 1912). This is to a solid what PV = NkBT is to an ideal gas.

How to read Eq. (7) The term dE/dV on the left-hand side comes only from the elasticity of the bonds, that is, it is the pressure at 0 K (E(V) is a static energy, so it contains no temperature). The right-hand side is the pressure with which the phonons push the crystal outward. Uqν contains the zero-point energy ħω/2, so this does not vanish even at 0 K — which is why it is more accurate to call it the phonon pressure rather than the “thermal pressure” (if you want to refer only to the part that rises with temperature, take Pvib(T) − Pvib(0)). Each individual mode contributes just “its own energy Uqν × its own γqν.” Modes with positive γqν push outward, and modes with negative ones pull inward. What tab ① of the simulator shows is the result of this tug of war.

If we leave the crystal without external pressure (P = 0), it settles at the volume where the right-hand and left-hand sides of Eq. (7) balance. If we approximate the elastic energy at 0 K by the parabola E(V)= 12BT (V−V0)2 /V0 then the amount of expansion itself can be written down.

V−V0= V0 BTV ∑q,ν γqν Uqν (8)

Since Uqν is always positive, whether the crystal expands or shrinks is determined only by the sign of γqν — but what decides it is the sum. A single mode with negative γqν is not enough; ∑γqν Uqν<0 must hold.

6. What Is the Quasiharmonic Approximation (QHA)?

Eq. (7) is beautiful, but it cannot be used as it stands. This is because both ωqν and γqν really depend on both temperature and volume. The assumption we make is the quasiharmonic approximation (QHA).

The quasiharmonic approximation comes down to two lines The reason it is called “quasiharmonic” rather than “harmonic” is that, although the crystal is harmonic at each volume, the frequencies change when the volume changes (that is, some anharmonicity remains).

An actual first-principles calculation proceeds in the following steps under this assumption.

  1. Calculate the harmonic phonons at several volumes obtained by changing the lattice constant a little at a time (this gives ωqν(V)).
  2. For each volume and each temperature, build F(V, T) from Eq. (5).
  3. For each temperature, search for the volume that minimizes F(V, T). That is the equilibrium volume V(T).
Imaginary phonons must not be present At a volume where a frequency becomes imaginary, the free energy of the harmonic phonons cannot be defined as a real number. If only one of the trial volumes, far out in compression or expansion, is unstable, it can be dropped and the free energy built from the rest, but if imaginary modes appear within the range of volumes needed to find the equilibrium volume, the standard QHA cannot be used as it is. This is the reason why four polymorphs of SiO2 are excluded from the analysis in the paper.

Look at how “the valley moves” in Fig. 2. Relative to the valley at 0 K (V0), the valley bottom shifts as the temperature is raised. If you make γ negative, the valley bottom moves to the left — it shrinks even though it was heated; this is negative thermal expansion.

Fig. 2. In the quasiharmonic approximation, the bottom of the free-energy valley moves with temperature. This is Eq. (5) itself for an Einstein model (3 modes per atom, ΘE = 500 K), with ω(V)= ω0(V/ V0)−γ. V0 = 12 ų/atom. The two curves are drawn so that each has its own valley bottom at 0 — because what we want to see is not the depth of the valley but the sideways shift of the position of its bottom. We also show that the αV obtained by searching numerically for the valley bottom agrees with the αV obtained from Eq. (11). Note that the “0 K valley” is the valley bottom of the free energy including the zero-point motion, F(V, 0) — it differs from the valley of the static E(V), V0 = 12 ų, by an amount due to the zero-point pressure whenever γ ≠ 0.

In Fig. 2, BT increases a little when heated Looking at the readout, we set 150 GPa, yet at 600 K it reads 157 GPa. First, the 150 GPa of the slider is the curvature parameter of the static parabola E(V), whereas the readout BT = V∂²F/∂V² is a quantity that also includes the curvature of the phonon free energy — so, to begin with, they are different quantities. On top of that, the main reason it increases is that, even if E″ is constant for a parabola, B = V E″, so it grows with volume. A parabola lacks the property that “it gets stiffer when squeezed” (∂BT /∂P≈4) so it does not get softer even when it expands. A real solid gets softer when heated. The simulator, on the other hand, displays the BT(T) given by the QHA calculation as it is up to 595 K, and only above that extrapolates with the Anderson–Grüneisen relation BT= B0(V/ V0)− δT (δT = 5). In the simulator, BT decreases with temperature — for MgO, the 150 GPa at 300 K drops to 126 GPa at 1000 K (BT(1000)/BT(300) = 0.840; the experimental value is 0.84).

7. The Volumetric Thermal Expansion Coefficient Takes a Simple Form

Recall that what Eq. (2) requires is (∂P/∂T)V — so we differentiate Eq. (7) with respect to temperature at constant volume. Since E(V) does not contain the temperature, it drops out, and we get

(∂P ∂T)V =1V ∑q,ν [ (∂Uqν ∂T)V γqν + Uqν (∂γqν ∂T)V ] ≃ ∑q,ν CV,qν γqνV = γCVV (9)

In the second line we used the quasiharmonic approximation — at constant volume the frequencies do not depend on temperature, so γqν does not depend on temperature either, and the second term drops out. The remaining first term (∂Uqν /∂T)V is nothing other than the specific heat at constant volume of that mode, CV,qν. Finally, we define the total specific heat and the specific-heat-weighted average of the Grüneisen parameter as

CV= ∑q,ν CV,qν , γ= ∑q,ν CV,qν γqν ∑q,ν CV,qν (10)

Then Eqs. (2) and (9) give the result in a single line.

αV= γCV BTV (11)

The thermal expansion of a solid can be written with just four quantities. Moreover, CV, BT, and V are always positive, so it is only γ that determines the sign of αV.

No expansion at absolute zero Since CV → 0 as T → 0 (the derivation of the specific heat), Eq. (11) gives αV → 0. This agrees exactly with what the third law of thermodynamics requires — see Section 5 of the derivation of the harmonic oscillator for details. Conversely, at high temperatures CV levels off at 3kB (Dulong–Petit). That does not mean, however, that αV becomes constant — Eq. (11) also contains γ, BT, and V, and at high temperatures BT decreases and V increases, and γ also changes because of the intrinsic anharmonicity. Only in the range where these temperature changes can be neglected, does αV settle at a nearly constant value.

8. What Governs Thermal Expansion?

Four quantities appear on the right-hand side of Eq. (11). Which of them matters most? Lining up how much each of them varies among real materials makes the answer clear.

FactorRange at room temperatureFold differenceRole
CV0.7–3 kB/atom about 4×Levels off at high temperatures (Dulong–Petit). A material with a high ΘD, such as diamond, has not yet leveled off at room temperature
γ−0.06 to 3.3 Changes sign
37 positive materials:
0.38–3.34, about 9×
Determines the sign. Set by the crystal structure and the coordination number
V5.7–71 ų 12×Anticorrelated with BT, so it has little effect on its own
BT2.9–438 GPa 150×This one dominates

The ranges were taken from the 38 cubic materials in Tables S2/S3 of the paper (you can see the same table in tab ② of the simulator). Only BT differs by more than two orders of magnitude. This is why, when αV is plotted against BT as a scatter plot, an inverse-proportion curve αV ∝ 1/BT appears cleanly. To be precise, among these 38 cubic crystals, the factor that most strongly affects the spread in the “magnitude” of positive thermal expansion is BT — the sign is still determined by γ, and, as in ferroelectrics, magnetic transitions, and electronic-state transitions, there are also materials in which something other than lattice vibrations governs thermal expansion.

The αV of diamond (BT = 438 GPa) is 3.3 ppm/K, while that of potassium (BT = 2.9 GPa) is 245 ppm/K — 75 times larger. Hard things do not expand. Soft things expand a lot. This agrees with everyday intuition, and the denominator of Eq. (11) says the same.

The mean atomic volume V matters less than it appears Looking only at Eq. (11), it seems that the larger V is, the smaller αV should be. In real materials, however, the larger V is, the smaller BT is — there is a strong anticorrelation (the farther apart the atoms are, the looser the bonds). Since BT changes more steeply than V in the denominator BTV, αV consequently shows a weak positive correlation with V. You can check this by switching the horizontal axis in tab ② of the simulator to “Mean atomic volume V.”

9. Materials That Shrink When Heated — Negative Thermal Expansion

In Eq. (11), CV, BT, and V were all positive. Therefore, the only way to get αV < 0 is γ < 0. This is the necessary condition for negative thermal expansion (NTE).

By Eq. (4), γqν < 0 means a mode whose frequency rises when the volume is expanded. This is counterintuitive, but it does happen. And having a few negative modes is not enough — it is only when the sum weighted by the specific heat satisfies ∑CV,qν γqν<0 that the whole crystal shrinks. The weights change with temperature, and this is why the temperature range of negative thermal expansion is limited even for the same material (Si, which shrinks only at low temperatures, is an example).

The tension effect Consider an oxygen atom sandwiched between two atoms, as in A–O–A, swinging in a direction across the bond. When the oxygen escapes sideways, the A–O–A distance shrinks (just as pinching a string pulls its two ends together). If the crystal is now stretched, the string is pulled taut and transverse vibration becomes harder — the frequency rises. This is exactly γqν < 0. And the higher the temperature, the more violent the transverse vibration becomes, and the crystal shrinks.

The paper systematically investigated 46 elemental solids and 45 binary oxides and showed the following five points.

  1. A crystal with a large V tends to have a small BT.
  2. The main role of γ is to determine the sign of αV, and its value is determined more by the crystal structure and coordination number than by the element itself.
  3. Transverse acoustic (TA) phonons and low-frequency transverse phonons contribute strongly to negative thermal expansion.
  4. Negative-thermal-expansion materials have relatively large V and voids in their crystal structure.
  5. What negative-thermal-expansion materials have in common is the presence of phonons that vibrate toward the voids.

Y. Mochizuki, H. Koiso, K. Nagamatsu, S. Bae, T. Isobe, A. Nakajima, J. Phys. Chem. C 128, 525–535 (2024) — summary of the conclusions.

In other words, the design guideline for negative thermal expansion is “voids, a small coordination number, and softness.” Voids give the atoms room to swing into, a small coordination number makes the bond angles easy to bend, and softness (a small BT) makes the denominator of Eq. (11) small and amplifies the effect.

MaterialStructureγ BT (GPa)αV (ppm/K)Remark
Cu2OCuprite type Pn3̄m−0.061 117−1.40Cu has a coordination number of 2. O–Cu–O swings sideways
ReO3ReO3-type Pm3̄m0.384 2443.47A perovskite with the A site vacant. The octahedra rotate
SiDiamond type0.39 90.07.22At low temperatures αV becomes negative
MgORock-salt type1.63 15035.9Densely packed, with no voids
KBody-centered cubic1.30 2.89245Just plain soft

It is especially important that the same composition can have a different γ when the structure changes. The paper compares the rock-salt and zinc-blende types of CaO: the rock-salt type has only positive γqν, whereas the zinc-blende type has negative γqν at the X and L points. γ is not a property of the element but a property of the arrangement.

10. Where Does the Quasiharmonic Approximation Break Down?

Let us keep, rather than drop, the term (∂γqν /∂T)V that was dropped in the second line of Eq. (9). Define the thermal Grüneisen parameter δqν and the anharmonic parameter Δqν by

δqν≡ −T ωqν (∂ωqν ∂T)V , Δqν≡ −VT ωqν ∂2ωqν ∂T∂V (12)

This δqν is a different quantity from the Anderson–Grüneisen parameter δT, which was used in the simulator to extrapolate BT above 595 K — the letter is the same, but the definition and the meaning are different.

With these two, the temperature derivative of γqν can be summarized neatly.

(∂γqν ∂T)V = 1T ( γqν δqν + Δqν ) (13)
Where Eq. (13) comes from γqν= − (V/ω) ωV: just differentiate it with respect to temperature, which gives T(∂γ /∂T) = −VT ωVT/ω +VTωVωT /ω2, where the first term is exactly Δqν and the second term is [− (V/ω) ωV] [− (T/ω) ωT]= γqν δqν. The key point of this expression is that γ is attached in front of δ, and this carries over unchanged into the definitions of the averages below.

Putting this back into the first line of Eq. (9), the volumetric thermal expansion coefficient is extended as follows. Here we write the averages of the mode sums, weighted by Uqν, as

δ≡ ∑q,ν Uqν γqν δqν ∑q,ν Uqν , Δ≡ ∑q,ν Uqν Δqν ∑q,ν Uqν (14)

— δ contains γqν inside it. Setting U=∑Uqν, we obtain

αV = 1BTV ∑q,ν [ CV,qν γqν + UqνT ( γqν δqν+ Δqν ) ] = γCV BTV + 1BTV UT (δ+Δ) (15)

The first term is the QHA, and the second term is the deviation from it.

Be careful how you read the high-temperature limit For a single phonon mode, at high temperatures (ħω ≪ kBT), we have Note that 3kB is the sum over the 3 modes per atom, not the specific heat of a single mode. Therefore, at high temperatures the weight of the second term of Eq. (15) is, for every mode, Uqν /T≃ kB the same, and what matters is the size of γqν δqν+ Δqν itself.

For modes of lower frequency, when the temperature is raised, ∂ω/∂T becomes relatively larger (δqν is a quantity divided by ω). Therefore soft materials with many low-frequency modes deviate more from the QHA.

Limits of this extended formula itself Eq. (15) was obtained by letting ω depend on temperature and then differentiating formally. To do this rigorously, one must decide whether CV,qν is to be regarded as “the harmonic specific heat at fixed frequency” or as “the total derivative including the temperature-dependent frequency,” and moreover, merely substituting ω(V,T) into the harmonic free energy does not guarantee thermodynamic consistency with the entropy and the free energy. A proper treatment requires a framework such as self-consistent phonon theory, in which temperature-dependent phonons and thermodynamic potentials are consistent with each other. Please read it here as a signpost saying “the QHA has a definite pattern of deviation.”
Materials for which the QHA cannot be trusted For very soft materials such as rock-salt-type NaBr, the QHA alone cannot reproduce the measured αV. Compounds containing heavy, soft elements such as Pb and Bi, and bromides, iodides, selenides, and tellurides become doubtful for the same reason. On the other hand, most of the elements and binary oxides treated in this derivation are not as soft as NaBr, so the QHA agrees well with them. However, beginning with K (BT = 2.9 GPa), the alkali and alkaline-earth metals such as Li, Na, Ba, Sr, and Ca are as soft as NaBr or even softer, and, like the Pb mentioned just above, they call for the same caution.

11. Summary

  1. An ideal spring (a harmonic solid) does not expand, however much it is heated. Thermal expansion is anharmonicity itself.
  2. In a crystal, the anharmonicity is summarized by the Grüneisen parameter γqν= − ∂ln⁡ω/ ∂ln⁡V.
  3. Differentiating the free energy with respect to the volume gives the Mie–Grüneisen equation of state, Eq. (7). The right-hand side is the pressure with which the phonons push the crystal outward — it includes the zero-point motion, so it does not vanish even at 0 K (Section 5).
  4. Imposing the quasiharmonic approximation (if the volume is held fixed, the frequencies do not depend on temperature) gives αV = γCV/(BTV), Eq. (11), in a single line.
  5. Of the four factors, the only one that varies by orders of magnitude from material to material is BT. What governs thermal expansion is stiffness.
  6. Since CV, BT, and V are positive, the only way to shrink when heated is γ < 0. Voids, a small coordination number, and transverse phonons create that condition.
The size of the vibration in ① is a different quantity from thermal expansion The amplitude with which the atoms shake in tab ① of the simulator is the square root of the atomic displacement parameter (ADP) estimated from the Debye model, Uiso (the mean-square displacement per direction). Uiso retains the zero-point part, so the shaking does not stop even at 0 K — it is a quantity different from the “expansion” we have been following on this page (the movement of the equilibrium volume V(T)). The formula itself is given in Section 8 of the derivation of the specific heat. Note also that the size of the box (the lattice constant) in ① is a calculated value built from V(T) of the QHA calculation, and only the way the atoms shake is a schematic display based on the Debye model. However, the stretching on the screen is exaggerated 25 times by default (the factor is shown on the screen as “Expansion ×25” and can be changed with the slider) — the actual thermal expansion is a far smaller quantity, as we saw in Section 2.

You can work with the actual numbers in the simulator for the equation of state of an ideal solid — in ① (“Heat it and shake the crystal”) raise the temperature while rotating the crystal, in ② (“Stiffness controls expansion”) look at the αV–BT correlation of the 38 materials, and in ③ (“What changes with temperature”) follow the temperature dependence. The V(T), αV(T), BT(T), and CV(T) that appear in ①–③ are not just the 300 K cross-section but the output of the QHA calculation itself, over 0–595 K. The data come from Tables S2/S3 of Y. Mochizuki, H. Koiso, K. Nagamatsu, S. Bae, T. Isobe, A. Nakajima, “Thermal Properties of the Element and Binary Oxides toward Negative Thermal Expansion: A First-Principles Lattice-Dynamics Study,” J. Phys. Chem. C 128, 525–535 (2024) (open access, CC BY 4.0) and from the phonopy-qha output used for that calculation. Related pages: specific heat derivation, harmonic oscillator and second quantization, entropy and free energy.