Algorithmically Designed Optoelectronic Devices

算法设计的光电器件

• 批准号: RGPIN-2019-05130
• 负责人: Johlin, Eric
• 金额: $ 2.04万
• 依托单位: University of Western Ontario
• 依托单位国家: 加拿大
• 项目类别: Discovery Grants Program - Individual
• 财政年份: 2021
• 资助国家: 加拿大
• 起止时间: 2021-01-01 至 2022-12-31
• 项目状态: 已结题
• 来源: https://www.nserc-crsng.gc.ca/ase-oro/Details-Detailles_eng.asp?id=737307
• 关键词: Algorithmically; Designed; Optoelectronic; Devices

The program described herein explores the algorithmic design, experimental fabrication, and characterisation of nanophotonic structures for optoelectronic devices. The long term objective is to utilise these processes to create 3D structured devices, including photonic siphons, spectral sorters, and hybrid antennas, which would allow revolutionary advances for the fields of quantum computing, solar energy conversion, and machine vision. The major objectives of this program are threefold: 1-Algorithmic design At the nanoscale, light behaves both as a particle and a wave, complicating the design of structures with sub-wavelength features, to the point where even the function of a nanophotonic component cannot be determined intuitively. Algorithmic design addresses this, allowing full-wave optical simulations to dictate the geometry of the nanophotonic object. We previously demonstrated the use of an evolutionary algorithm to design ~2.5D dielectric structures. These design techniques will now be taken much further-incorporation of artificial neural networks, automatic differentiation-based simulations, and inverse design, will allow structures to be designed faster and with larger degrees of freedom. This allows us to expand into fully 3D designs, and even multi-material systems, thereby not only achieving further improvements in performance, but also introducing a wide range of unprecedented optical and electronic functionalities. 2-Fabrication in 3D Cognisant of the limited 3D structure in optoelectronic devices, our second objective is the creation of ways to fabricate the complex multi-material systems our algorithms design. We have previously demonstrated the ability to utilise 3D multi-photon lithography techniques to produce nanophotonic structures with sub-wavelength features. This will continue to be developed, along with exploration into new techniques as well; we are currently collaborating to introduce 3D printing using electron beam-induced deposition for feature resolutions of tens of nanometers. This is currently being explored for glassy materials, but will be expanded to metals as well. Furthermore, we are collaborating to utilise self-assembly of complex 3D geometries to allow parallel deposition of structures over a large substrate, and even deposition of semiconductor materials. 2-Nanophotonic characterisation Characterisation at high resolution is essential to the development of new devices, particularly those with novel properties. Previously we applied a super-resolution localisation microscopy technique to probe the photonic environment around simple nanowire structures. This will be further developed to understand the performance of the much more complex devices created in this program. Furthermore, an integrated in situ fabrication and measurement system will allow for development of novel structures in an automated system, providing a physical accompaniment to the algorithmic design process of the first objective.

本文所述的程序探索了用于光电器件的纳米光子结构的算法设计、实验制造和表征。长期目标是利用这些过程来创建3D结构设备，包括光子虹吸管，光谱分类器和混合天线，这将使量子计算，太阳能转换和机器视觉领域取得革命性进展。该计划的主要目标有三个方面：1-光子学设计在纳米尺度上，光的行为既像粒子又像波，使具有亚波长特征的结构设计复杂化，甚至纳米光子组件的功能也无法直观地确定。光学设计解决了这一问题，允许全波光学模拟来决定纳米光子物体的几何形状。我们先前展示了使用进化算法来设计~2.5D电介质结构。这些设计技术现在将被更进一步的结合人工神经网络，自动微分为基础的模拟，和逆向设计，将允许结构设计更快，更大的自由度。这使我们能够扩展到全3D设计，甚至多材料系统，从而不仅实现了性能的进一步提高，而且还引入了广泛的前所未有的光学和电子功能。2-认识到光电器件中有限的3D结构，我们的第二个目标是创造制造我们算法设计的复杂多材料系统的方法。我们先前已经证明了利用3D多光子光刻技术来产生具有亚波长特征的纳米光子结构的能力。这将继续发展，沿着探索新技术;我们目前正在合作引入使用电子束诱导沉积的3D打印，以实现数十纳米的特征分辨率。目前正在探索玻璃材料，但也将扩展到金属。此外，我们正在合作利用复杂3D几何形状的自组装，以允许在大型衬底上平行沉积结构，甚至沉积半导体材料。2-纳米光子表征高分辨率的表征对于新器件的开发至关重要，特别是那些具有新特性的器件。在此之前，我们应用了超分辨率定位显微镜技术来探测简单纳米线结构周围的光子环境。这将进一步发展，以了解在该计划中创建的更复杂的设备的性能。此外，集成的原位制造和测量系统将允许在自动化系统中开发新的结构，为第一个目标的算法设计过程提供物理伴奏。