Non-destructive neutron diffractometry for stress analysis will be a powerful experimental tool in material science research performed at the GKSS 5 MW reactor FRG-1. Arguments which show the advantages of the time-of-flight method are given and a suitable high-resolution neutron-efficient type of spectrometer is introduced. First results derived from this method are presented.

The macroscopic residual stress distribution in γ-quenched and stress levelled U-0.8wt% Ti alloy tubes was studied using neutron diffraction techniques. Residual strains were evaluated from the difference in d-spacings measured in the tubes and in small reference samples machined from each tube. Residual stresses were calculated with the isotropic bulk values of the elastic constants for polycrystalline α-U. Quenching from the γ field resulted in a nearly equi-biaxial stress state at every point across the wall thickness of the tube. The magnitude of the radial stress was very small compared with that of the axial and hoop stresses which were compressive at the surfaces and tensile in the interior. Stress levelling relieved almost completely the hoop residual stress without affecting the radial stress. The axial residual stress becomes tensile through the wall thickness and remains constant at about 20% of its magnitude in the as-quenched condition.

The high resolution neutron diffraction technique has been applied to determine, non-destructively, the residual stress distribution developed in the head of a railway rail after normal service.

Measurements were made, using the neutron strain scanner at the Institut Laue Langevin, Grenoble, on a transverse slice of rail 12mm thick taken from a section of straight track. The rail head was scanned in the three principal orientations in a series of parallel traverses sufficiently close to enable a two- dimensional matrix of data to be accumulated and vertical, transverse and longitudinal residual stress contours to be drawn.

Substantial residual compressive and tensile stresses and steep stress gradients were observed related to the depth below the top surface, distance from the running line and shape of the rail head. The highest stresses and stress gradients were observed in transverse and vertical directions but significant stresses were also retained in the longitudinal (12mm thick) direction of the rail.

The results demonstrate the effectiveness and unique characteristics of the neutron technique to determine non- destructively and continuously the residual stresses inside engineering components.

The triaxial stress field has been measured along the centre line of a compact tension specimen in the direction of cracking. The specimen had been subjected to ∼60,000 cycles at δK=31 and Kmax = 34 MPa mm½ and was bolted open at maximum stress. The field was remeasured after the stress had been fully relaxed. The results are discussed in terms of expectations from fracture mechanics calculations.

The thermal neutron reaction 10B(n,α)7Li is used to measure the boron concentration on the surface of silicon wafers. The technique, referred to as neutron depth profiling (NDP), requires no special sample preparation. Boron is determined on the as- received wafers at a level of 1012 to 1013 atoms/cm2. A boron level of about 2×1012 atoms/cm2 is found at the wafer surface after oxidation, epitaxial, or polycrystalline silicon deposition. Additional measurements are given from SIMS measurements of multilayer structures on silicon also showing the presence of boron. Ambient air appears to be a significant source of the boron.

The nondestructive neutron depth profile (NDP) technique has been used to measure boron (10B) distributions in Si and Hg0.7Cd0.3 Te after multiple energy ion implants. The NDP results are compared with simulations generated by TRIM and SUPREM computer codes. The influence of SiO2 films on the boron profiles was examined and the effects of thermal anneals are also described.

The single crystal diffractometer at LANSCE, SCD, provides an ideal ca- pability for the study of preferred orientation in geological samples by time-of flight(TOF) measurement of pulsed neutrons. The 2-d position sensitive neutron detector with the large wave length range allows one to measure the complete dis- tribution of intensities for several poles very quickly. Each histogram covers about πt2/16 radians of reciprocal space and contains information from all possible poles visible with the wave length range used, usually about 0.5 to 5.OÅ. With this method complete pole figures of many lattice planes can be constructed from only 12 to 20 sample orientations as compared to over 1000 sample settings per lattice plane using conventional diffractometers.

Pole figures from measurements of experimentally deformed standard samples of calcite and quartzite with a known history of deformation provide in- formation about deformation mechanisms and their temperature/strain history. This information can be applied to interpret preferred orientation of naturally de- formed rocks.

Neutron scattering has proved to be a powerful tool in the study of glasses and liquids: structures are investigated by neutron diffraction, and dynamics by inelastic neutron scattering. Evidence of current interest in this field is provided by the new instruments for diffraction from amorphous systems being constructed at IPNS, ISIS and the R2-reactor at Studsvik, in addition to existing instruments such as the D4B diffractometer at ILL.

A combination of small angle scattering (SAS) and neutron spin-echo (NSE) spectroscopy is used to characterize the structure and dynamics of polymeric silica aerogels during sintering. The SAS data indicate that densification at short length scales precedes the densification at longer scales (comparable to that of the pore structure). Interpreted within the fracton model, the NSE data are consistent with an initial decrease in connectivity during relatively early stages of densification.

Hydrogen vibration spectra have been measured by neutron scattering for several amorphous semiconductor materials, including a-Ge:H and a-SiC:H samples containing about 10 at. % H. The data for a-Ge:H are compared in detail with the results of realistic computer simulations.

Inelastic neutron scattering has been used to characterize the vibrational spectroscopy below 220 meV of para-nitroaniline adsorbed in the molecular sieve ALPO-5. Samples at loadings of both 3 and 13 weight %, which represent the onset of and the maximum in the nonlinear optical properties respectively, were studied. The torsional vibration of the amino (NH2) group has been identified at ca. 50 meV. The splitting and structure of this mode is sensitive to the loading level. This can be related to differences in the nature of the hydrogen bonding in these materials.

Coherent diffuse neutron scattering from single crystal UO2.13 has been investigated at temperatures up to 2100K. Above 850K the crystal as a single phase and the excess anions give rise to a characteristic pattern of elastic diffuse scattering. This can be interpreted in terms of uncorrelated simple clusters of defective anions first postulated by Willis. The intensity distribution changes with temperature increase but no quasielastic scattering has yet been resolved. On cooling to 293K a strong hysteresis is exhibited.

Neutron scattering experiments have been performed to study the structure and dynamics of three preparations of silica: a dense glass (fused quartz), a flame-hydrolyzed aggregate (Cab-O-Sil, grade M5) and a polymeric aerogel. The experiments were performed on the three materials at temperatures selected from 300, 77 and 4.2 K. Neutron inelastic scattering was measured at Q = 2.5, 2.9 and 4.0 Å−1.

The elastic scattering indicates a systematic decrease in positional correlations on progressing from fused quartz through Cab-O-Sil to aerogel. The inelastic scattering was analyzed with the Buchenau model [1] to obtain the sample, Q and temperature dependences of the density of states, g(v). In the aerogel g(v) increases with temperature at frequencies greater than 1.5 THz and is enhanced at Q = 2.5 Å−1. Thus a complete explanation of the inelastic scattering from aerogels must account for multiphonon processes, local antiphase motions and anharmonic effects.

The temperature dependent lattice parameter, c(T), was determined for samples of graphite, and HNO3 -graphite intercalates stage 3 and stage 4, using neutron diffraction. The temperature range studied is 12 to 300 K. The stage 3 and stage 4 intercalates exhibit a strong two-step discontinuity at the temperature of their structural phase transition (≈260 K). A stage-independent, temperature-dependent lattice parameter for the intercalant layer is extracted from the observed lattice parameters for graphite, stage 3, and stage 4 intercalates. The graphite and the intercalant lattice parameters are calculated using one-dimensional Debye theory in the framework of the Grüneisen relation. The Debye temperatures that yield best fits to the observed lattice parameters of the graphite and the intercalant are 500 K and 421 K, respectively. Using these values, thermal expansion coefficients for graphite, the intercalates and the intercalant are derived.

Silica-clay composites were prepared by reacting a 40 Å silica sol with aqueous clay suspensions. The clays which were studied included Na+-montmorillonite, fluorohectorite, and the synthetic clay Laponite. The aggregation mechanism of the sol particles and the degree of dispersion of the clay layers were monitored by small-angle neutron scattering. The data for powdered silica-montmorillonite products showed the presence of highly dispersed clay platelets and spherical sol particles. The results suggest that the interaction between these two components inhibits the sol aggregation process. Studies on the products prepared from other types of clays, however, showed that this interaction is highly dependent on the morphology and charge density of the clay platelets.

Incoherent, inelastic neutron scattering has been used to study the vibrational spectra of tetramethylammonium montmorillonite and trimethylammonium vermiculite in the energy range 20–140 meV. For both systems peaks are observed due to the internal modes of the intercalate and to the excitations of the hydroxyl groups within the host layers. For the montmorillonite sample, it is found that the steric constraints imposed on the tetramethylammonium ion by the bounding clay layers contribute an additional 28 meV to the rotational barrier of the methyl groups. This additional barrier is shown to be strongly related to the volume that the tetramethylammonium ion occupies. For the trimethylammonium vermiculite sample normal mode analysis of the internal modes of the intercalated ion shows that the N-H bond is parallel to the c-axis of the host.

The small angle neutron scattering intensity distribution from the empty pores of the silica glass Vycor can be satisfactorily interpreted using a simplified late stage spinodal decomposition model, due to Cahn', if one allows for a distribution of wave numbers in the material's density function. The result confirms that the porosity of the glass is 28% and gives the mean wave number of the density wave to be k0 = 0.0285 Å-−1 with a dispersion, δk / k0, of about 0.29. Small angle neutron scattering measurements of the anionic surfactant SDS in the Vycor pores indicate that when the surfactant is introduced into the pores at 80 × CMC in bulk solution, the surfactant forms micelles with the same average micelle separation distance as in the bulk solution, but the distribution of micelle separations is much greater in the glass pores.

Small angle neutron scattering measurements have been performed on several samples of silica controlled pore glasses with pore sizes ranging from roughly 7 to 30 nm. The scattering intensity is strongly peaked at small Q and shows approximate Porod law behavior at large Q. Contrast variation measurements have shown that the pore space in these samples is entirely interconnected and thus forms a bicontinuous microstructure. The scattering data have been analyzed using the leveled wave method based on an early scheme for representing two-phase microstructures resulting from spinodal decomposition. In this approach interfaces are modeled by the contours of a stochastic standing wave composed of plane wave components propagating in random directions with random phases and having wave numbers distributed according to a given probability density, P(k). We have determined model P(k) functions by fitting the SANS data with the leveled wave scattering function and then used these to construct leveled wave images of the corresponding porous structures. The average pore sizes obtained by measuring chord lengths in the computer models turn out to agree with the values determined for these glasses by mercury porosimetry.

Small angle neutron scattering has been used to characterize the structure of linear hydrocarbon chains chemically grafted to the internal pore surfaces of microporous silica particles. The aim of this work has been to relate the structure of the bonded adsorbate layers in these particles to their performance in, for example, reverse-phase liquid chromatography. By filling the pore space in the modified silica with a solution that matches the scattering density of the silica framework, the scattering from the adsorbate layers is enhanced and provides a sensitive probe of the effective thickness, uniformity and degree of solvent penetration in the layers. Results are presented for both monomeric and polymeric phases of alkyl chains ranging from C8 to C30 bonded to silica particles with a mean pore size of 100 nm.

The microstructure of low-density porous silica precursor (unsintered) bodies was studied as a function of starting chemistry. The ratio of colloidal silica sol to potassium silicate is known to have a marked effect on the size distribution of pores in this material, which in turn have a major impact on the resultant physical properties of the sintered product. In the present research, the sizes and the size distributions of particles underlying the pore microstructure was investigated, and it was found that the lower the amount of colloidal silica, the greater the size distribution of particle aggregates.