The role of point defects in multinary chalcogenide compound semiconductors

The electronic properties of multinary chalcogenide compound semiconductors, like chalcopyrite type ternary Cu(In,Ga)Se2 and quaternary kesterite type Cu2ZnSnSe4, depend strongly on their intrinsic point defects. For instance it is generally believed that in CuInSe2 the copper vacancies (VCu) cause p-type conductivity, whereas copper on interstitial positions (Cui) or InCu anti-sites act as donors and promote a n-type character. These defects are resulting from deviations from the stoichiometric composition. In order to keep the charge balance in the non-stoichiometric compounds, only a number of cation substitution reactions are possible: for example the transition to Cu-poor CuInSe2 goes via the defect pair 2VCu+InCu, whereas the transition to Cu-poor and Zn-rich Cu2ZnSnSe4 goes via the substitution 2Cu+->ZnCu + VCu. The presentation will give a comparison of the role of cationic point defects in chalcopyrite and kesterite type compound semiconductors concerning the following features: (i) Phase stability: the chalcopyrite type structure is very flexible to hold defects and can adapt itself to substitutions. Beyond a given copper vacancy rate, a vacancy compound (for instance CuIn3Se5) is formed, thus avoiding the occurrence of binary secondary phases (like copper selenides). For kesterite type Cu2ZnSnSe4 the situation is different: due to the lower flexibility of the kesterite type structure and the absence of vacancy compounds, secondary phases, like ZnSe, occur when the compound becomes Cu-poor. (ii) Atomic disorder: The cationic point defects cause an atomic disorder on the short range level which also influences the electronic properties (for instance the bandgap energy). For instance in Cu(In,Ga)Se2 defects such as antisites or interstitials lead to variations in the local atomic arrangements and thus broaden the bond distance distribution due to static disorder. The discussion will be underlined by the experimental results of neutron diffraction [1], anomalous scattering of synchrotron X-rays [2] as well as X-ray absorption spectroscopy [3].

Investigating the annealing dependency of Al/Si distribution in Eifel sanidine

Sanidine is the monoclinic high-T modification of K-rich alkalifeldspars. Annealing at T > 900⁰C usually causes disordering of Al/Si distribution at the two non-equivalent tetrahedral sites, but it is supposedly possible to disorder samples of sanidine from Volkesfeld/Eifel at notably lower temperatures and shorter times [1]. To investigate this behavior and compare various approaches to obtain the Al/Si distribution, samples from different Eifel locations and Madagascar have been studied. Al/Si order was determined by direct and indirect methods, including X-ray and neutron diffraction of powder and single crystal samples. Neutron powder diffraction experiments were executed at the Fine Resolution Neutron Powder Diffractometer E9, single crystal neutron diffraction at the 4-circle Diffractometer E5 and diffuse neutron scattering experiments at the Flat-Cone Diffractometer E2, all located at the Berlin Research Reactor BERII. The Al/Si distribution was determined directly, refining site occupancies by applying Rietveld analysis to powder diffraction data and XTAL for single crystal data. This approach is inapplicable when using X-ray data, due to similar atomic form factors of Al3+ and Si4+, thus indirect methods [2,3,4] were applied. X-ray powder diffraction was performed at Helmholtz Centre Berlin, single crystal X-ray diffraction was done at Ruhr-University Bochum. The obtained data was processed using Rietveld refinement and ShelXL software, respectively. It was possible to verify a dependency of decreasing Al/Si order on increasing annealing times and temperatures. Interestingly we observed different results from direct and indirect methods, regardless whether samples were untreated or annealed. Applying the direct determination method, a stronger change of Al/Si distribution during annealing was revealed. Moreover, diffuse scattering of untreated and annealed samples was detected, which may arise from hydron incorporated in the crystal lattice.

Commissioning the Neutron Laue Diffractometer in Berlin

The Fast Acquisition Laue Camera for Neutrons (FALCON) is a thermal neutron Laue diffractometer situated in the experimental hall of the BER-II reactor at HZB in Berlin. The thermal beamtube, D1S, delivers a stream of neutrons direct to FALCON just 8m from the reactor core but with a low gamma radiation count. FALCON benefits from a beam that does not pass through any objects upstream whilst a beam definer delivers a highly focused neutron beam to the instrument with <1⁰ divergence. The instrument comprises two scintillator plate detectors coupled to four iCCD cameras each. The neutron beam passes through the detector units enabling one detector to be placed in the backscattering position and the second detector in the transmission position. The image-intensified CCDs are capable of obtaining 20-bit digitization Laue images in under ten seconds and variable sample table and detector positions allow a full range of sample environments to be utilised. Scientifically, FALCON offers the opportunity to study samples from a wide range of fields for example; low-temperature magnetic studies, high-temperature structural phase transitions, in-situ kinetics studies and point-defect analysis in compound semiconductors. Data from FALCON can be used to solve crystallographic structures and as a neutron instrument it has all the advantages of neutrons as a probe for condensed matter, for instance, identification of the location of hydrogen atoms within structures and differentiation between electronically similar elements. FALCON will now enter the commissioning phase using in-house samples to test both ambient and sample environment conditions on the instrument. We present here details of the upcoming commissioning tests and invite users to submit proposals for Laue diffraction experiments.

Cation disorder in CZTS materials from anomalous diffraction at KMC-2 (BESSY)

The compound semiconductor Cu2ZnSnS4 (CZTS) is a promising alternative for absorber layers in thin film solar cells, as it has a nearly ideal band gap of about 1.5 eV, a high absorption coefficient for visible light, and contains only earth abundant and non-toxic elements. Besides chemical composition and phase purity, the efficiency of CZTS thin film solar cells depends strongly on the concentration of Cu- and Zn-antisites and copper vacancies in the kesterite-type structure. However, Cu(I) and Zn(II) are isoelectric and thus cannot be distinguished by conventional X-ray diffraction. In prior work we determined Cu-Zn-distribution successfully from neutron scattering [1]. Here we present experiments utilizing anomalous X-ray diffraction on the K-edges of Cu and Zn. Anomalous scattering coefficients are heavily wavelength-dependent close to the absorption edges of the respective element. This is utilized for contrast enhancement. Usage of multiple wavelengths above, below and between the absorption edges of Cu and Zn ensures significant overdetermination, so that the Cu-, Zn-, and vacancy concentrations can be refined reliably for the independent crystallographic sites. Experiments were conducted at the diffraction end station of the KMC-2 beamline [2] at BESSY (Berlin, Germany). KMC-2 provides X-ray radiation with both very stable energies and intensities. The accessible energy range of 4 - 14 keV is ideally suited for the K-edges of Cu (8979 eV) and Zn (9659 eV). A 6-circle goniometer in psi-geometry allows both powder and grazing incidence diffraction, so that bulk samples and thin films can be measured. The instrument can be equipped with either a scintillation point detector (Cyberstar) or an area detector (Bruker Vantec), allowing to optimize resolution and intensity to the needs of the experiment.