Applications of Automated X-ray Topography and High-Resolution Diffraction in the Production of Cadmium Zinc Telluride Single Crystals

Zinc cadmium telluride (CZT) is a wide-bandgap II-VI group semiconductor material with excellent photoelectric properties and spectral resolution, which can be used to manufacture detectors for infrared rays, X-rays, γ-rays and other high-energy rays. CZT is currently the best substrate material for growing mercury cadmium telluride (MCT) thin films. By adjusting the zinc stoichiometry in CZT, the lattice constants of CZT and MCT can be perfectly matched (Figure 1). The growth of high-quality CZT substrates is technically challenging because crystal defects such as twins, subgrains and low-angle grain boundaries are prone to form during single crystal growth, and these defects will further extend to the epitaxial thin films, ultimately reducing the yield of devices on the substrates. Therefore, the large-scale manufacture of high-quality CZT substrate wafers requires fast and automated crystal defect monitoring equipment.
 
The JV-QCRT model X-ray topograph uses the latest numerical X-ray diffraction imaging (XRDI) technology and adopts a reflection mode to image the defects of single crystal substrates for identification (Figure 2a). Due to the principle of X-ray diffraction physics, unlike optical technology, single crystal wafers do not need to be polished and etched to see lattice defects. Different from traditional XRDI topographs, JV-QCRT uses a divergent X-ray beam to measure samples (Figure 3a). This means that areas with orientation changes on the substrate will appear as translated parts in the XRDI image, rather than areas with reduced intensity seen in the traditional parallel beam configuration (Figure 3b). The orientation direction can be identified by the edge contours in the XRDI image: there is a light-colored edge where the intensity is lost due to translation, and at the same time, there is a black edge where the intensity is gained due to the addition of the orientation change signal to the non-orientation change signal. The orientation direction points from the light edge direction to the black edge direction. Therefore, this technology can detect important lattice defects and orientation change areas in the sample without losing diffraction intensity due to the divergent beam configuration. The machine is also equipped with an automation program to achieve automatic calibration, measurement, data collection, processing of samples and generate final numerical topographic phase images. Samples can be quickly scanned in a 75µm resolution survey mode to quickly locate important defects (such as orientation change areas). The XRDI image can be interpreted by the grayscale change of the entire image, where areas with greater intensity (darker pixels) indicate local strain, and very dark/black pixels usually represent the presence of defects.
 
High-resolution X-ray diffraction measurement of rocking curves is a commonly used technique for characterizing the crystal quality of single crystal substrates or epitaxial materials. The width of the rocking curve at different relative intensities, such as the full width at half maximum (FWHM), reflects the crystallinity of the measured material. A wider FWHM indicates a decrease in material crystallinity and more defects; multiple peaks indicate the presence of regions with different lattice orientations in the measured area. The JV-QCVelox high-resolution X-ray diffractometer (Figure 2b) can perform fully automated measurement, collection, and data processing of rocking curves for substrates. In addition, the upgraded JV-QCVelox-E model (Figure 2c) of this machine can also be equipped with an automated robotic arm to achieve fully automated production line inspection.
 
In addition to identifying defects in single crystal substrates, the XRDI images measured by JV-QCRT can be used to determine which parts of the wafer are suitable for slicing to provide a basis for device manufacturing. The largest single rectangle without major defects can be determined from the XRDI image. This rectangle can be the area on the wafer suitable for device manufacturing (Figure 4a).
 
To verify the conclusions of the XRDI topographic phase, rocking curves of high-resolution diffraction were also measured in different crystallinity regions of the wafer. The representative scans of the uniform gray area and the orientation change area on the wafer are consistent with the XRDI topographic phase conclusions (Figure 4b). The rocking curve of the uniform gray area is typical of high crystallinity materials, a single narrow peak. The rocking curve of the area with orientation change is consistent with the result of the XRDI topographic phase, with multiple peaks. The rightmost peak is consistent with the peak of the uniform area, indicating that they are part of the same lattice, and the broad peak on the left indicates that the defective area is tilted by -0.1° from the main lattice, which corresponds to the left shift in the XRDI topographic phase.
 
Automated X-ray diffr