A Global Correction Process for Flat Optics With Patterned Polishing Pad

To improve the efficiency of flat optics fabrication, a global correction method with the patterned polishing pad is developed in this paper. Through creating grooves on a polishing pad, the contact pressure distribution on the optics surface can be adjusted to change the material removal rate (MRR) distribution during polishing. Using the patterned pad, the selective removal ability of the polishing process is greatly enhanced. The predictability and stability of the MRR distribution are the preconditions to efficiently implement the proposed global correction method. Relying on the MRR distribution prediction method proposed and validated in this paper, the pad pattern can be designed based on the original surface figure of the workpieces. The designed groove pattern is created on the polishing pad using the custom-developed equipment. Then, the optical glass is polished on the designed pad with the optimized polishing time. A flat optical glass sample (Φ 100 mm) is polished with the global correction method to show its feasibility and advantage. The correction instance shows that the peak-to-valley (PV) value of the surface profile (with 3 mm edge exclusion) dropped from 1.17 µm to 0.2 µm in 14 min using a polyurethane pad with two ring grooves. Comparing with the conventional polishing process, which usually takes hours or days, the global correction method proposed in this paper can improve the efficiency of the optics manufacturing significantly.

Numerical Analysis of Slurry Flow and Contact Pressure in CMP Considering the Effects of Retaining Ring

Retaining ring which keeps the wafer in place is an essential component in chemical mechanical polishing. Meanwhile, it helps to reduce the edge exclusion region where the material removal rate deviates significantly from that of the central region of the wafer. However, it may increase the slurry flow resistance and hence decrease the slurry flow rate. For properly designing a retaining ring of reasonable structure, the effects of retaining ring on slurry flow and contact pressure distribution in CMP process are analyzed by the mixed elastohydrodynamic lubrication model. It is found that the slurry flow is sensitive to the protrusion height of retaining ring used in the first generation carrier. The same as the first generation carrier, the slurry flow is obviously reduced with increasing pressure acting on the retaining ring in the second generation carrier. In addition, the floating retaining ring used in the second generation CMP carrier has better performance and is more controllable than the fixed retaining ring used in the first generation CMP carrier.

150 mm 4H-SiC Epitaxial Layer Growth in a Warm-Wall Planetary Reactor

Homo-epitaxial growth of 4H-SiC on 4o off-axis 150 mm diameter substrates has been performed in a commercial warm-wall multi-wafer planetary reactor. Based on our well developed 100 mm 4H-SiC epitaxial growth process, which can achieve excellent thickness and doping uniformities (δ/mean) of <1% and <5%, respectively, the growth process and hardware were further fine-tuned and improved for 150 mm 4H-SiC homoepitaxy. After the improvement, the 6 to7 μm thick epilayer uniformity has reached 1.1% with a 5mm edge exclusion while the doping uniformity has improved to 16.5% (<10%) with an edge exclusion of 5 mm (10mm), respectively. Surface roughness of the as-grown 150 mm 4H-SiC epitaxial layer has an RMS value of 0.12 nm scanned by AFM on 20×20 μm2 areas. Homo-epitaxial growth on C-face 150 mm 4H-SiC substrates has also been carried out. Other than the doping concentration and uniformity, the other results are very close to the epi-growth on Si-face.

50 μm-Thick 100 mm 4H-SiC Epitaxial Layer Growth by Warm-Wall Planetary Reactor

Homo-epitaxial growth of 50 μm-thick 4H-SiC on 4° off-axis 100 mm substrates have been demostrated by using a commercial warm-wall multi-wafer planetary reactor (Aixtron 2800 G4). With optimized process, epitaxial layer with an average thickness of 48.146 μm and doping level of 8.39×1014/cm3are obtained. The thickness uniformity with an edge exclusion of 5 mm are 1.30% (σ/mean) and 2.17% (max-min/max+min), and the doping level uniformity are 4.66% (σ/mean) and 6.95% (max-min/max+min), respectively. Surface roughness of the as-grown 50 μm-thick epitaxial layer has an RMS value of 0.606 nm with one step bunching on the 20×20 μm2areas. This initial effort on thick 4H-SiC homoepitaxial growth indicates that this comercial multi-wafer planetary reactor has the potential for mass production of SiC epiwafers for 5000 V and above power devices.

Wafer Edge Bead Cleaning with Laser Radiation and Reactive Gas

The purpose of this study is to investigate and optimize the process parameters for cleaning the top, bottom, and apex edges of silicon wafers using laser radiation and reactive gas. A secondary purpose is to conduct photoresist edge bead and post-etch polymer film removal (EBR) experiments to determine the minimum controllable edge exclusion in EBR processing to improve die yield. [ An overall purpose is to identify a robust and environmentally sound process for wafer edge cleaning and a hardware configuration (stand alone or track integrated) that can be cost effectively produced for device manufacturing.