Complete lattice parameters have been determined at a series of temperatures up to 950 °C using a furnace on a powder diffractometer for six synthetic supposed equilibrium albites, a nonequilibrium albite, two natural albites, and a heat-treated natural albite. The supposed equilibrium albites were prepared hydrothermally by MacKenzie (1957) for long periods of time and reached or nearly reached a state of no further change at their temperatures of synthesis between 500 and 1000 °C. A crystal of natural albite and three crystals of heat-treated natural albite were studied by high-temperature, single-crystal methods on a Buerger precession camera up to 1000 °C for comparison.

The lattice angles for each albite decrease smoothly with increasing temperature; the a-axis and the volume increase smoothly with increasing temperature. The rates of change increase with increasing temperature. The c and in some cases the b-axes appear to decrease initially in the range 25–300 °C and then subsequently increase. The thermal dilatation is highly anisotropic—the change in a is about twice that in b and about six times that in c over the whole range 25–850 °C. The effect of a temperature rise on the lattice parameters of the albites is very similar to the effect of the substitution of potassium on the room-temperature parameters.

Plots of α* against γ* for each albite at a sequence of temperatures up to 950 °C fall on nearly straight parallel lines, all albites becoming less oblique (triclinic) with increasing temperature. Of the equilibrium albites, only the one synthesized at 1000°C becomes monoclinic below the melting point. The ratio d₀₁₀/d₀₀₁ or c*/b* is a measure of order in the albites and does not vary with temperature. The similar behaviours of this ratio and of other lattice parameters as a function of the temperature of synthesis confirms the two-step ordering process in albite.