Ag–Ge films of thickness 1 to 3 microns were prepared by co-evaporation of Ag and Ge onto 0.005′′ thick vitreous silica plates held at liquid nitrogen temperature. The pressure during evaporation was 3 × 10⁻⁷ Torr or less. The X-ray scattering of the films and substrates was measured at room temperature with monochromatic Mo Kα and Cu Kα radiation. After correction for substrate scattering, the interference function I(K) was evaluated for each alloy as a function of K = 4π sin θ/λ. The Fourier transform of K[I(K)−1] yielded the reduced distribution function G(r) = 4πrV(r)[ρ(r)−ρ₀], where ρ(r) is the weighted atomic density at a distance r from a reference atom, ρ₀ is the average atomic density and V(r) is the size factor of the coherently diffracting domains in the sample. Both I(K) and G(r) indicate that the alloy films consist of the Ag solid solution and Ge. The Ag phase cannot be characterized as amorphous, but is micropolycrystalline. The sizes of the coherently diffracting domains or the correlation distances determined from G(r) increased from 12 Å in Ge to 30 Å in Ag-40 at. %Ge. Fourier analyses of the (111) peak profiles of Ge and Ag yielded particle sizes of 13 Å for Ge and 16 Å for the Ag phase in the Ag-83 at. %Ge alloy. The Ag phase particles increased with increasing Ag concentration, reaching a value of 45 Å in the Ag-28 at. %Ge alloy.

Amorphous films of AgCu and CuMg₂, approximately 3000 Å in thickness, were prepared by co-evaporation of Ag and Cu, and Cu and Mg, respectively, onto 25 μm thick Be sheets, held at liquid nitrogen temperature. Mo Kα X-rays were used as a radiation probe to determine the structure of the films, at room temperature, and of the liquid alloys of Cu with 50 at.% Ag and with 0 and 67 at.% Mg at 50°C above the liquidus temperature. With the transmission technique, the interference functions (or structure factors) I(K) were determined in the range of K = 4π sin θ/λ between 0.8 Å⁻¹ and 12.5 Å⁻¹, and then Fourier transformed to obtain the radial distribution functions (RDF). The I(K) and RDF of the amorphous AgCu and CuMg₂ films were compared with those of the liquid Ag–Cu and Cu–Mg alloys, respectively. It was found that the structures of the amorphous and liquid Ag–Cu alloys were similar with a more well defined short-range order occurring in the solid alloys, whose I(K) exhibited the well known shoulder on the second peak. The I(K) and RDF of the amorphous CuMg₂ and the liquid Cu–Mg alloys cannot be explained by a common structure, although I(K) showed a small premaximum below the first main peak in both the amorphous and liquid alloys, a feature observed in many liquid Mg alloys.

The long-range order parameter, S, of pure Cu₃Au and two Cu₃Au alloys containing 0.64 and 1.5 at. % Co was determined over the temperature range 325 to 387°C. In principle, the state of order in the ternary alloy (Cu₃Au)_1−XCo_X is described by two order parameters S₁ = Aα − Aβ, and S2 = Bβ − Bα where Aα is the fraction of α sites occupied by A atoms, etc. It is shown that S₂ − S₁ = Cα − Cβ = 0 as long as the Co atoms are randomly distributed on α and β sites. The values of S were measured at temperature using Warren's method [Warren, B. E. (1969). X-Ray Diffraction. Reading, Mass.: Addison-Wesley] applied to powder samples, employing the {110}, {210}, {211}, and {221, 300} superstructure and the {111}, {200}, {220}, {311}, and {222} fundamental reflections. The results on pure Cu₃Au are in relatively good agreement with earlier measurements, but in poor agreement with theoretical predictions, especially at temperatures above 360°C. Within the accuracy of the measurements, it is observed that cobalt, which is in solid solution in both alloys, has no effect on the experimentally determined values of S. The root-mean-square vibration amplitudes, 〈u²〉^1/2, of the Cu and Au atoms at 640°K were also estimated from the data, and it appeared that 〈u²〉^1/2_Cu decreased slightly with increasing Co content while 〈u²〉^1/2_Au increased.

A horizontal diffractometer has been set up for transmission studies of liquid samples. An asymmetric, singly bent LiF monochromator in the primary X-ray beam focuses the radiation on the receiving slit of the diffractometer. The sample rotates at half the angular speed of that of the X-ray detector to retain focusing at all angles of diffraction. A high temperature furnace assembly (camera) has been constructed which fits on the diffractometer. Its heating element consists of a thin-walled, pyrolitic graphite tube 38 mm in diameter. A special sample holder for the liquids is mounted on a boron nitride stool bolted to the bottom plate of the camera. The entire set-up allows the measurement of the diffracted intensity between 2θ = 2° and 2θ = 90° corresponding to K = 4πsinθ/λ = 0.3 Å⁻¹ at 12.4 Å⁻¹ when Mo Kα radiation is used. Correction procedures to obtain the scattered intensity per atom from the raw data are discussed, and a comparison is made between the transmission and reflection methods with liquid tin.