从生产过程实现碲镉汞红外探测器的性能提高

Performance Improvement of Mercury Cadmium Telluride Infrared Detector from Production Process

高性能碲镉汞红外探测器需要生长高质量的外延层。由于与碲镉汞具有完美的晶格匹配,体碲锌镉被认为是理想衬底。晶体材料在亚晶界缺失、位错密度、均匀锌分布和低微缺陷密度等方面实现非常高的性能,对于获得优异的图像质量至关重要。法国Sofradir公司利用自己生长的碲锌镉晶体作为衬底,以控制碲镉汞外延层的质量,从而实现高性能成像。实际上,通过掌握从原材料到焦平面阵列的整个制造链以及所有前端和后端步骤,可以改进整个制作流程。介绍了如何将最新的工艺改进转化为探测器图像质量和可靠性的提高,其重点是前端工艺(衬底和外延层)。首次展示了衬底微缺陷与焦平面阵列(Focal Plane Array, FPA)图像质量之间的相关性。这得益于Sofradir公司和法国CEA--LETI研究中心之间的通力合作。对每个工艺步骤进行了大量表征(例如用于衬底检查的红外显微镜观察、位错的化学显示以及外延层的X射线双晶摇摆曲线衍射),由此完成了这种工艺的整体优化。在有效像元率和过噪声方面对图像质量进行了检测。最后,除了改进流程之外,了解每个关键步骤如何影响后续步骤并转化为最终图像质量,实现在正确的流程步骤中对单元进行划分,从而保证产量及产品质量。在中波红外和短波红外技术上,Sofradir垂直整合模型的这些优点得到了体现。

The growth of high-quality epilayers is necessary for high-performance HgCdTe infrared detectors. Due to perfect lattice match with HgCdTe, bulk CdZnTe is considered as the ideal substrate. Crystal materials should achieve very high performance in terms of subgrain boundary absence, low dislocation density, homogeneous zinc distribution, and low micro-defect density, which is essential for acquiring excellent image quality. In order to control the quality of HgCdTe epitaxial layer, Sofradir uses CdZnTe crystals grown by itself as the substrate, thereby achieving high performance imaging. In fact, the whole production process can be improved by mastering the entire manufacturing chain from raw materials to focal plane array(FPA) including all front-end and back-end steps. The method for translating the latest process improvements into detector image quality and reliability improvements with a focus on front-end processes(substrates and epitaxial layers) is introduced. Under the collaboration between Sofradir and CEA-LETI, the correlation between substrate microscopic defects and FPA image quality is firstly demonstrated. Though a large number of characterizations for each process step, including IR-microscopy for the substrate inspection, chemical revelation of dislocations and X-ray double-crystal rocking curve mappings for the epitaxial layer, the overall optimization of this process is achieved. Image quality detection in terms of effective pixel rate and excess noise is completed. Finally, in addition to improving the process, understanding how each key step affects the subsequent steps and translates them into the final image quality is helpful for sorting units in the correct process step, which serves yield and product quality. These advantages of the Sofradir′s vertical integration model are demonstrated on the mid-wave infrared and short-wave infrared technologies.