The absence of ½ 111-type superlattice spots in the [−110] SAED patterns of both samples (not shown here) indicates a lack of ordering on the 111A planes. We associate the absence of extra spots in S25 sample to the

smaller size of the layer, which could lead to a reduction of its intensity beyond detectable limits. Figure #OICR-9429 chemical structure randurls[1|1|,|CHEM1|]# 3 [110] SAED patterns of samples (a) S25 and (b) S100. (a) The conventional pattern for the ZB structure, (b) the additional ½ 111 superlattice spots associated of a CuPtB-type ordering. The inset corresponds with the ½ 111 superlattice spots, magnified and filtered to improve the visualizations. Due to the difficulty in obtaining representative SAED patterns from the different regions of the GaAsBi layers, HRTEM images were acquired in the [110] zone axis in both samples to detect CuPtB-type ordering in the layers. Figure 4a displays an HRTEM image taken at the lower GaAs/GaAsBi interface of sample S100, and Figure 4b,c depicts the corresponding FFTs of the GaAsBi and GaAs regions of the image, respectively. The ½ 111-type spots in Figure 4b confirm the presence of CuPtB ordering. This was also observed in sample S25, confirming the formation AZD2281 mw of CuPtB-type ordering that was too weak to be detected in

the SAED pattern and highlighting the danger of relying on SAED analysis alone. Figure 4 Degree of ordering in sample S100. (a) Cross-sectional selleck chemical HRTEM image taken along [110] at the lower interface of sample S100. The dashed line marks the interface between GaAs (below) and GaAsBi (above). (b,c) depict the FFT of (a) corresponding to GaAsBi area and GaAs, respectively. (d) The Bragg-Williams long-range order parameter (S) estimated along the layer of sample S100. The dashed circle mark the corresponding Bragg mask used to obtain the numerical moiré fringe maps of Figure 5. In order to obtain an estimate of how the ordering is distributed along the layer, we have analysed the intensity of ½ 111-type and 111-type spots in FFTs and calculated the order parameter from

the bottom, middle and top of the layer in sample S100 (Figure 4d). The analysis revealed the absence of ordering within experimental error in the GaAs region (as expected) with an average LRO of 0.1, while the LRO was S ≅ 1 for both 111B families in the region closer to the bottom GaAs/GaAsBi interface (region I) in all HTREM images. Conversely, in the middle and top parts of the GaAsBi layers, regions both with and without ½ 111-type spots could be found and when present the LRO parameter varied between 0.3 and 1. It can therefore be concluded that there is a higher degree of ordering near the bottom interface. Ordering map Figure 5 shows the ordering distribution map of the different regions of the GaAsBi layer obtained from HRTEM images.