The objective of this Phase I proposal is to develop multi-stacked wafer bonding techniques for wide-bandwidth anti-reflection (AR) treated silicon optics at terahertz (THz) frequencies. This process can enable high layer-count structures resulting in thick and large needed for the very wide-bandwidth AR treatment. At the end of the Phase I, the goal is to achieve <1% reflectance over a prototype of 4-layers AR structures by stacking with precision alignment and bonding techniques that Cactus Materials, Inc. has developed. Phase II of the project is to develop a complete wide-bandwidth AR treatment for silicon optics applicable for vacuum windows and it can be used in the future for powered optics by integration with a gradient-index lens architecture (GRIN) using wafer bonding, circumventing the challenge of AR-treating a curved surface. Transmission (T) and reflectance (R) on bonded wafers are expected to be 100% and <1% respectively. A precision alignment of <2 micron between wafers will be employed using automated lithographically defined alignment marks. To meet <1% reflectance, the bonding interface needs to be defect free, void free, chemicals and moisture free. In addition, bonding strength needs to be close to silicon bulk strength and withstand any vibration or stress as well as hold up as vacuum windows, so under deflection of 1.5-6 mm (depending on the diameters). For example, vibrational stress of a launch could damage the stacked Si lenses. A detailed testing and modeling will be incorporated to ensure the optics are robust enough for space platform. If successful, this technology development will be stepping stone towards making a high-performance, larger diameter, and thicker AR treated silicon optics.
A wide range of applications include studies of CMB polarization, of galaxy clusters using the Sunyaev-Zeldovich Effect, of galaxy evolution and the Epoch of Reionization using low-resolution spectroscopy and spectral line tomography. Specific spectral bands of interest for astronomy applications e.g. flat optical windows with 4-layer AR structures covering 4:1 bandwidth, specifically 100-400 GHz and 75-300 GHz, 7-layer AR structures covering ≥ 6:1 bandwidth, 80-420 GHz and 30-180 GHz, and a GRIN optic with 4-and 6-layer AR structure
The technology can be implemented in a cost-effective way for large optical elements in many applications in the range if FIR, MWIR, and THz. Silicon vacuum windows and gradient index silicon optics with integral AR treatment are two key products. Silicon is significantly cheaper, particularly as size increases compare to other materials e.g. Germanium vacuum windows are 2-3x higher in cost.