M.J.F. Healy

Modelling oxide thin films

Three simulation methodologies have been employed to investigate the growth, nucleation, and structure of oxides supported on oxide substrates, these are atom-by-atom deposition, layer-by-layer deposition and finally amorphisation of a structure followed by recrystallisation. The materials which have been investigated include the rocksalt-structured oxides; MgO, CaO, SrO, and BaO, the perovskite structured SrTiO 3 and also fluorite structured CeO 2 and ZrO 2. The work has shown that the substrate influences critically the structure of the supported thin film by determining the nature and interactions of defects, dislocations and grain-boundaries, as well as influencing the interfacial ion densities and various epitaxial relationships. In addition, graphical techniques have been employed to show the three-dimensional atomistic structure of each structural and epitaxial feature. Moreover, by considering large simulation cell sizes (approaching the mesoscale, 18 nm square), it has been possible to accommodate the synergistic interactions between neighbouring structural features, which can lead to changes in their basic structure. We also show that the particular surface termination of the substrate can influence the structure (and tentatively, the critical thickness) of the supported film through the example of SrO and TiO 2 terminated faces of a SrTiO 3 (001) substrate.

The crystal structure of CdS–CdTe thin film heterojunction solar cells

Abstract A detailed structural analysis of electrodeposited CdS–CdTe thin film heterojunction solar cells was undertaken. X-ray diffraction and Rutherford backscattering spectrometry were used to provide stoichiometric and microcrystalline data at increasing depths through the CdTe and CdS films. A model of the nature and extent of interdiffusion caused by a post deposition anneal is developed. A region in both pre-annealed and post-annealed samples which possesses a significantly different microstructure to that of the bulk CdTe, was identified. Within this region a stoichiometric gradient occurs and the grain size and preferred orientation decrease with increasing depth. Maximum CdTe film stress (post anneal) is estimated to be 140xa0MPa close to the interface and a shift in optical band gap of 6×10 −3 eV was also determined from structural measurements. We provide evidence that sulphur diffusion into CdTe is structurally rather than thermodynamically limited within these systems.

Minimising carbon contamination during ion beam analysis

Abstract The highly sensitivity 12C(d,p0)13C reaction makes Nuclear Reaction Analysis (NRA) an excellent tool with which to quantify trace amounts of carbon. However it is often the build up of carbon on the sample during analysis which limits the sensitivity of this technique in practice. Carbon deposition rates on silicon have been monitored for a range of experimental conditions in order to quantify the problem and minimise contamination during analysis. Under common experimental conditions used at the RMCS facility, 1 × 1014atoms cm−2 of carbon were found to be deposited on silicon for each microcoulomb of an incident 1.2 MeV deuteron beam. The total number of carbon atoms deposited over a fixed dose could be reduced by minimising the hydrocarbon content of the residual gas, heating the sample, minimising the analysis area and placing a range foil in the path of the analysing beam.

New data on the proton elastic scattering cross-section for silicon

Abstract A discrepancy between theoretical and experimental data for the proton elastic scattering cross-section from silicon was investigated. The proton elastic scattering cross-section for silicon was measured between 600 and 2500 keV for 110° and 170° backscattering by normalising yields from a thin natural silicon layer against corroborated cross-section data. The results obtained are in good agreement with the theoretical calculations. Thus the discrepancy is resolved and the power of the theoretical approach to cross-section data handling is confirmed by these measurements.

A simple accelerator calibration procedure

Abstract A calibration procedure for an accelerator is described which is based upon the principles of Rutherford backscattering spectrometry and uses existing experimental apparatus. The procedure enables calibration to be performed both rapidly and efficiently. Details of the calibration of a 2.5 MV Van de Graaff generator are given as an example, and the results are compared to the 19 F ( p ,αγ) 16 O resonant nuclear reactions at proton energies of 872 keV and 1373 keV.

Depth profiling sulphur in bulk CdTe and CdTeCdS thin film heterojunctions

Abstract Polycrystalline CdTeue5f8CdS heterojunction solar cells are a possible candidate for the low cost, high efficiency conversion of solar energy. The formation of an intermediate CdS x Te 1− x layer during a high temperature annealing stage is believed to increase optical absorption and decrease cell efficiency. S diffusion in single crystal CdTe has been investigated by NRA using the 32 S (d,p o ) 33 S nuclear reaction, at a deuteron energy of 2 MeV. Details of the NRA depth profiling procedure are given, which was found to be relatively straightforward and suitable for use on a small Van de Graaff accelerator. The resulting diffusion parameters are compared to those obtained by SIMS using a Cs + primary ion beam, examining negative secondary ions. The diffusion coefficients were found to be 1.1 × 10 −15 cm 2 s −1 at 450°C and ∼8 × 10 −15 cm −1 s at 550°C. S diffusion in thin films was also investigated by 2 MeV 4 He + RBS on annealed polycrystalline CdSue5f8CdTe multilayers.

A Model to Support CBRN Defence

There is no shortage of media alerting to the threat from chemical, biological, radiological and nuclear (CBRN) materials and their delivery modes, reflecting and fostering public perceptions of vulnerability. The threat is considered sufficiently real for many countries to have responded with considerable purchases of equipment as a demonstration of capability to satisfy public expectation and dissuade potential aggressors. Essential organisational structures and training programmes are being refined and reactive thinking is giving way to considered, informed planning. The wide range of components involved in CBRN defence (and the complexity within each component) is demonstrated by the wide range of stakeholder organisations, skills and equipment mustered to address the threat. Stakeholders are often challenged to appreciate the wider CBRN picture when busy developing their own capability.

Ion beam analysis of aluminium in thin layers

Abstract This work quantifies aluminium in thin surface and near surface layers. In one example, the layer overlies a thin gallium nitride layer on an aluminium oxide substrate and in a second example the aluminium exists just below the surface of an indium arsenide substrate. The technique of non-Rutherford elastic backscattering of protons was used for the samples where aluminum in the layer of interest needed to be resolved from aluminium in the sapphire substrate and the results were corroborated at the Technische Universitat Munchen using heavy ion elastic recoil detection analysis. In the second example, where it was unnecessary to isolate the signal of aluminium in the layer of interest (as the substrate contained no aluminium), then the 27 Al(d,p01)28 Al nuclear reaction was used. The elastic proton scattering cross section of aluminum was found to vary very rapidly over the energy range of interest.

Revised cross sections of the 32S(d,p)33S reactions

Abstract The differential cross sections of the 32S(d,p)33S nuclear reactions have been measured at a laboratory angle of 150° for incident ion energies between 1.0 and 2.7 MeV. Two different experimental methods were used, one using a thin CdS film and the other using a thick ZnS target. The resulting cross sections are compared to previously published data and the 32S(d,p0)33S and 32S(d,p1)33S cross sections are also compared directly to the Rutherford cross section. Some applications of the cross sections are discussed.

An electrostatic filter for high resolution nuclear reaction analysis

Abstract An electrostatic filter for the removal of the elastically scattered ions inherent in Nuclear Reaction Analysis is described, its transmission characteristics are shown and modelled. The consequences of obtaining improved energy resolution by use of the filter as opposed to conventional absorber foils are investigated, and the minimisation of data collection times for high resolution spectra using the filter is discussed.