Physical Chemistry
Van der Waals Equation
(P + a n²/V²)(V − nb) = nRT — finite molecule size + intermolecular attraction correct ideal-gas law (Nobel 1910)
The van der Waals equation (P + a n²/V²)(V − nb) = nRT corrects the ideal-gas law by adding a finite-volume term b for molecular size and an attraction term a for intermolecular forces. Published by Johannes Diderik van der Waals in his 1873 Leiden thesis Over de Continuïteit van den Gas- en Vloeistoftoestand and awarded the Nobel Prize in Physics in 1910, it is the simplest cubic equation of state that predicts liquid–vapor coexistence and the critical point. Tabulated constants span four orders of magnitude: a = 0.034 L²·atm/mol² for helium, 1.39 for nitrogen, 3.59 for CO₂, and 5.46 for water; b clusters between 0.024 and 0.045 L/mol.
- Equation(P + a n²/V²)(V − nb) = nRT
- Constantsa (attraction), b (volume)
- Critical ZZ_c = 3/8 = 0.375
- Publishedvan der Waals 1873
- Nobel PrizePhysics 1910
- OrderCubic in V
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 the van der Waals equation matters
- First equation to predict liquid–gas coexistence. Below the critical temperature the cubic in V has three real roots — the smallest is the molar volume of liquid, the largest is the molar volume of vapor, the middle is unphysical. No earlier equation of state showed two phases falling out of one continuous formula.
- Predicts a universal critical compressibility. Z_c = P_c V_c / (n R T_c) = 3/8 = 0.375 for every substance. Argon measures 0.291, nitrogen 0.290, water 0.229 — the prediction is wrong by 20 to 35 percent, but the existence of a universal value at all is a deep result that survives in modern equations of state.
- Foundation of corresponding-states theory. Rescaling P, V, T by their critical values collapses every gas onto one (P_r, V_r, T_r) surface. Pitzer extended this with an acentric factor ω in the 1950s and modern engineering charts still build on this idea.
- Mean-field benchmark for statistical mechanics. Van der Waals is the canonical mean-field theory of fluids — every renormalization-group calculation of critical exponents starts by quoting the van der Waals predictions (β = 1/2, γ = 1, δ = 3) and showing how they fail in three dimensions (β ≈ 0.326, γ ≈ 1.24, δ ≈ 4.79).
- Cheap and analytic. The equation is cubic, so its three roots are obtainable by Cardano's formula. Modern process simulators run millions of equation-of-state evaluations per second; a closed-form cubic is dramatically faster than any iterative scheme. This is why Peng-Robinson and SRK — both descendants of van der Waals — still dominate hydrocarbon process engineering.
- Inversion temperature for Joule-Thomson cooling. Setting (∂T/∂P)_H = 0 in the van der Waals form gives the inversion temperature T_inv = 2a/(Rb), which evaluates to 866 K for nitrogen and only 224 K for hydrogen. Below T_inv, JT expansion cools the gas; above, it warms it. This is the prediction Carl von Linde used in 1895 to liquefy air.
- Conceptual gateway to all real-gas physics. Every undergraduate physical chemistry course introduces non-ideality through the van der Waals equation. The two-parameter intuition — molecules have size, molecules attract — survives every later refinement.
Common misconceptions
- The constants a and b are universal for all substances. They are not. Each substance has its own a and b, tabulated by fitting to T_c and P_c. The dimensionless reduced equation (P_r + 3/V_r²)(3V_r − 1) = 8 T_r is universal; the dimensional form is not.
- b equals the molecular volume. b is roughly four times the molecular volume for hard spheres, because the excluded volume of a pair of spheres of radius r is the sphere of radius 2r around either center, which has 8 times the molecular volume but is shared between the pair, giving 4 V_molecule per molecule.
- Van der Waals gives accurate critical exponents. It does not. β_VdW = 1/2 versus β_exp ≈ 0.326. This is a feature of every mean-field theory; correctly capturing the singular behavior near T_c requires renormalization-group methods that Wilson, Fisher, and Kadanoff developed in the 1970s.
- The equation is valid below the critical point as written. The bare cubic predicts an unstable region with (∂P/∂V)_T > 0 — physically impossible. The Maxwell equal-area rule replaces this region with a horizontal tie line at the saturation pressure; the equation alone does not contain this stability condition.
- It works for water and ammonia. Polar and hydrogen-bonded fluids are poorly described by a single attraction parameter. Engineers reach for SAFT, PC-SAFT, or specialized cubic forms like CPA (Cubic Plus Association) for these substances.
- R has the same value in every term. R must use units consistent with the chosen P, V, T units: 0.08206 L·atm/(mol·K) when P is in atm and V in L, or 8.314 J/(mol·K) when P is in Pa and V in m³. Mixed unit choices are the most common student error.
Derivation from corrections to the ideal gas
Start with the ideal-gas law PV = nRT and ask what physical effects it ignores. First, real molecules occupy non-zero volume, so the volume available for thermal motion is not V but V − nb, where nb is the total excluded volume of the molecules. Replacing V with V − nb gives the corrected equation P(V − nb) = nRT, sometimes called the Clausius equation. Second, real molecules attract each other through dispersion, dipole, and other intermolecular forces. A molecule near the wall feels a net inward pull from neighbors that reduces the impact pressure; this attractive correction is proportional to the density squared (one factor of n/V for the molecule of interest, another for its attracting neighbors), giving a pressure deficit a (n/V)². The measured pressure P is therefore less than the kinetic pressure by this amount, so P + a n²/V² is the right pressure to substitute back into the ideal form, yielding (P + a n²/V²)(V − nb) = nRT.
The critical point falls out of the requirement that the (P, V) isotherm have an inflection at horizontal tangent: ∂P/∂V|_T = ∂²P/∂V²|_T = 0. Algebra gives V_c = 3nb, T_c = 8a/(27Rb), P_c = a/(27b²). Inverting, a = 27 R² T_c² / (64 P_c) and b = R T_c / (8 P_c). For nitrogen with T_c = 126.2 K and P_c = 33.96 atm, this recovers a = 1.39 L²·atm/mol² and b = 0.0386 L/mol — close to the directly fitted values. The reduced equation (P_r + 3/V_r²)(3V_r − 1) = 8T_r emerges by dividing through, and reveals that all van der Waals fluids share one universal isotherm in reduced coordinates.
The equation is cubic in V: rearranging gives PV³ − (Pnb + nRT)V² + n²a V − n³ab = 0. Below T_c, three real roots exist; the smallest is the liquid molar volume, the largest is the gas molar volume, and the middle root is mechanically unstable. Maxwell's equal-area construction in 1875 supplied the missing thermodynamic input — the saturation pressure is fixed by requiring equal chemical potentials in coexisting phases, equivalent to making the two areas enclosed by the wiggle equal above and below the tie line.
Cubic equations of state — vdW vs Redlich-Kwong vs SRK vs Peng-Robinson vs ideal
| Equation | Year | Critical Z_c | Best for | Form (attraction term) |
|---|---|---|---|---|
| Ideal gas | 1834 (Clapeyron) | 1.000 | Low pressure, high T only | None — PV = nRT |
| Van der Waals | 1873 | 0.375 | Conceptual, undergraduate | −a/V² |
| Berthelot | 1899 | 0.375 | Older textbook usage | −a/(TV²) |
| Redlich-Kwong (RK) | 1949 | 0.333 | Non-polar gases above T_c | −a/(T^0.5 V(V+b)) |
| Soave-Redlich-Kwong (SRK) | 1972 | 0.333 | Hydrocarbons, vapor pressures | −a(T)/(V(V+b)) |
| Peng-Robinson (PR) | 1976 | 0.307 | Petroleum, natural gas | −a(T)/(V²+2bV−b²) |
| PC-SAFT | 2001 | varies | Polymers, associating fluids | perturbation theory |
Tabulated van der Waals constants for common substances
| Substance | a (L²·atm/mol²) | b (L/mol) | T_c (K) | P_c (atm) | Notes |
|---|---|---|---|---|---|
| Helium (He) | 0.034 | 0.0237 | 5.19 | 2.24 | Weakest attraction; quantum corrections matter |
| Hydrogen (H₂) | 0.244 | 0.0266 | 33.2 | 12.8 | Below 205 K JT inversion |
| Neon (Ne) | 0.211 | 0.0171 | 44.4 | 26.9 | Noble gas baseline |
| Nitrogen (N₂) | 1.39 | 0.0391 | 126.2 | 33.96 | Air's main component |
| Oxygen (O₂) | 1.36 | 0.0318 | 154.6 | 50.1 | Slightly stronger a than N₂ |
| Carbon dioxide (CO₂) | 3.59 | 0.0427 | 304.1 | 72.8 | Quadrupole moment boosts a |
| Ammonia (NH₃) | 4.17 | 0.0371 | 405.5 | 111.3 | H-bonding inflates a |
| Water (H₂O) | 5.46 | 0.0305 | 647.1 | 217.7 | Strongest H-bonded fluid |
| Methane (CH₄) | 2.25 | 0.0428 | 190.6 | 45.8 | Reference natural-gas component |
Applications
- Cryogenics and air liquefaction. Carl von Linde's 1895 industrial process for liquefying air relied on Joule-Thomson cooling — predicted by the van der Waals inversion temperature T_inv = 2a/(Rb). For nitrogen T_inv ≈ 866 K and air at 300 K cools on expansion. Hydrogen (T_inv ≈ 205 K) and helium (T_inv ≈ 40 K) require pre-cooling before JT expansion will liquefy them.
- Petroleum engineering and natural-gas pipelines. Phase equilibrium in oil and gas reservoirs uses Peng-Robinson or SRK — both direct descendants of van der Waals. Compressibility factors, vapor pressures, and dew points are routinely computed from these equations in process simulators like Aspen HYSYS and Pro/II.
- Refrigeration cycle design. The cooling capacity of a vapor-compression cycle depends on the latent heat at the evaporator pressure and the JT throttling behavior at the expansion valve. Cubic equations of state derived from van der Waals give engineering-grade predictions of both.
- Supercritical fluid extraction. Above the critical point liquid and vapor merge into a supercritical phase with tunable density. Decaffeinating coffee with supercritical CO₂ (above 304 K and 73 atm) and dry-cleaning with supercritical water are designed using van der Waals-class equations.
- Astrophysics of gas giants. Jupiter and Saturn interiors are described to first order by extended van der Waals forms; metallic hydrogen at megabar pressures requires fully ab-initio molecular dynamics, but the conceptual handoff begins with the same a, b parameters.
Frequently asked questions
What do the constants a and b represent physically?
The constant a quantifies the strength of intermolecular attraction. Each pair of molecules in a volume V contributes an attractive correction; integrating over the gas gives the term a n²/V², which adds to the measured pressure to recover the pressure the gas would exert if attractions were absent. Tabulated values track polarizability and hydrogen bonding: He at 0.034 L²·atm/mol², N₂ at 1.39, CO₂ at 3.59, H₂O at 5.46. The constant b is the excluded volume per mole — roughly four times the actual molecular volume because two hard spheres of radius r cannot approach closer than 2r. b is around 0.024 L/mol for He, 0.039 for N₂, and 0.043 for CO₂. Both a and b are determined by fitting to critical-point data.
How does van der Waals predict the critical point?
At the critical point the first and second derivatives of P with respect to V at constant T both vanish — the isotherm has an inflection point with horizontal tangent. Solving (∂P/∂V)_T = 0 and (∂²P/∂V²)_T = 0 simultaneously with the equation of state gives V_c = 3nb, T_c = 8a/(27Rb), and P_c = a/(27b²). The compressibility factor Z_c = P_c V_c / (n R T_c) is therefore exactly 3/8 = 0.375 for every substance — a strong prediction. Real gases give Z_c around 0.27 to 0.30 (water 0.229, argon 0.291), so the equation is qualitatively right but quantitatively off by 20 to 30 percent at the critical point. The constants a and b are usually back-fitted from measured T_c and P_c.
What is the law of corresponding states?
If you rescale by critical values — reduced pressure P_r = P/P_c, reduced volume V_r = V/V_c, reduced temperature T_r = T/T_c — the van der Waals equation becomes (P_r + 3/V_r²)(3V_r − 1) = 8 T_r, with no a or b in sight. Every gas obeying van der Waals plots on the same surface in (P_r, V_r, T_r) coordinates. Real gases of similar shape and polarity follow this universal curve to within a few percent — the basis for engineering charts that estimate compressibility for any substance once T_c and P_c are known. Pitzer's acentric factor ω corrects for non-spherical shapes and brings the agreement to under 1% for most non-polar fluids.
When does van der Waals fail badly?
Two regimes. First, near the critical point: the equation predicts the wrong scaling exponents — classical mean-field exponents like β = 1/2 for the coexistence curve, where experiment gives β ≈ 0.326. This is a fundamental issue with all mean-field cubic equations of state and is corrected by renormalization-group treatments. Second, for strongly polar or hydrogen-bonded fluids: water, methanol, ammonia. Here a single attraction parameter a cannot capture both dispersion forces and directional hydrogen bonds. Modern engineering uses Peng-Robinson (1976) or SRK (Soave 1972) for hydrocarbons, and SAFT or PC-SAFT for associating fluids. Van der Waals is rarely used in production plants today but remains the conceptual starting point for every textbook.
What is the Maxwell equal-area rule?
Below T_c the van der Waals isotherm in the (P, V) plane has a wiggle — pressure increases with volume in part of the range, which is mechanically unstable and unphysical. James Clerk Maxwell argued in 1875 that the real isotherm is a horizontal line (constant pressure during phase change) drawn so the two areas enclosed between it and the wiggle are equal. This construction picks out the actual saturation pressure at temperature T and the volumes of coexisting liquid and vapor at the endpoints of the line. The areas-equal condition is equivalent to setting the chemical potentials of the two phases equal — a thermodynamic consistency requirement that the bare equation does not enforce.
Why did van der Waals win the 1910 Nobel?
The Nobel committee cited his work on the equation of state for gases and liquids — specifically, the unification of the gaseous and liquid states under a single equation. Before 1873 these were thought of as distinct phenomena requiring different theories. Van der Waals showed in his Leiden doctoral thesis that introducing two molecular constants reproduced the entire phase diagram qualitatively, including the existence and continuity of the supercritical region — the regime above T_c where liquid and vapor are indistinguishable. This conceptual unification underpins all of modern fluid thermodynamics and inspired Heike Kamerlingh Onnes (Nobel 1913) to liquefy helium in 1908 by following van der Waals' predictions for the inversion temperature.