Healing Silicon Carbide (SiC) wafers to make them almost defect free and much higher quality.
Last updated: 2020-09-05
ACME Advanced Materials healed Silicon Carbide (SiC) wafers on parabolic flights of aircraft starting from 2014, but seems to be dormant and worthwhile advantages for healing wafers in orbit are lacking.
- Microchips that can operate at much higher temperature.
- Microchips that can withstand radiation much better.
- Automotive (used in Teslas).
Why & Solution
Silicon Carbide Wafers (SiC), a compound of silicon and carbon, can be used to produce wafers for the manufacture of computer chips that can operate at temperatures up to 1,000°C, can withstand 10 times the electric fields that standard semiconductors made of silicon can withstand, and offer high radiation resistance, high thermal conductivity, high maximum current density, and several other interesting properties that make them superior to standard semiconductors manufactured using silicon wafers (Department of Energy (DOE) 2015).1
SiC wafers can be produced in microgravity at much higher quality than those produced on Earth. On Earth, gravity prevents atoms from settling into their lowest energy states on a wafer, producing defects that interfere with the flow of electricity across the wafer. In microgravity, by cycling the pressure and temperature, these defects can be removed, resulting in “S-grade” (“space” grade) wafers with 99 percent of the original defects removed and little to no edge effects (Glover 2016; ACME 2016). With some redesign, SiC wafers can replace wafers made from silicon (DOE 2015). Because chips and other products manufactured from SiC wafers operate at very high temperatures, substituting SiC-based power electronics for traditional power electronics reduces required heat sinks (Hull 2013). Other uses for SiC include photovoltaic inverters, electric and hybrid vehicles, solar arrays, power grids, and wind turbines (Anagenesis 2015, expert interview).1
They claim to have multiple patents for development and application of wide band gap semiconductors and ZBLAN IR optical fiber, but ZBLAN fiber ones have not been found. They announced first production of superior Silicon Carbide wafers in microgravity in 2014, which is another promising product. They raised funding of €400,000 in 2015. Their social media platforms have been quiet for 2 years. The primary founder Rich Glover, who is still listed as the contact on website, has marked on LinkedIn that he stopped working there in 2018.
We assume the space station would be able to capture the profits generated by manufacturing wafers on orbit in the form of charges for leasing; it would also generate revenues for astronaut time.
For our low estimate, multiplying 75,000 wafers per year by $1,125 yields $84.4 million. Subtracting total costs of $73.7 million ($63.7 million in variable costs ($850 per wafer times 75,000) plus $9.9 million in astronaut time) yields $10.7 million available for lease payments. Adding in the $9.9 million in charges for astronaut time generates $20.6 million in revenues for the station.
For our high estimate, we assume 115,320 wafers are healed each year, yielding gross revenues of $129.7 million. Subtracting production costs of $113.3 million ($98.0 million of which is variable costs of $850 per wafer times 115,320) yields $16.4 million available for lease payments. The higher number of astronaut hours yields an additional $15.3 million in space station revenues, for potential total revenues of $31.7 million.1
After reprocessing in microgravity, ACME Advanced Materials sells S-grade SiC for $750 per wafer. Companies that produce better quality A-grade SiC wafers are able to charge $1,500 per wafer (Glover 2016). If substantially more A-quality SiC were to become available, this price would fall. In our analysis, we assume that a SiC wafer manufacturing operation in orbit would produce A-grade SiC wafers, but that it would be able to sell those wafers at the average of these two prices, $1,125 per wafer.1
Market Size Estimation
In 2015, over 75,000 SiC wafers were sold for total global sales revenues of $66 million (Anagenesis 2015 and references therein). For our low estimate, we assume this number of wafers (75,000) would be healed on the space station over the course of a year. For our low estimate, we look at the current market growth. The SiC wafer market rose 24 percent from $53.2 million to $66 million dollars between 2010 and 2015 (DOE 2015 and references therein). Assuming that the quantity of wafers grows at this same rate over the next 10 years, by the end of 2025, the number of wafers manufactured on orbit would be 115,320, if all production of wafers moved to space. We use this number for our high estimate.1
A future private station could generate revenues by leasing space for the production of high-grade SiC wafers. To determine how much revenue a private station could generate through such leases, we first estimate the costs of “healing” wafers in orbit. ACME Advanced Materials, based in Albuquerque, New Mexico, has been reprocessing or healing low-grade SiC wafers in microgravity to create high purity wafers with valuable material properties.
According to ACME, poorer quality wafers manufactured on Earth that can be healed in microgravity sell for $250 (Glover 2016).
We assume that a manufacturer in space purchases low-grade SiC wafers for $250 per wafer and then transports these wafers to the space station for processing. To calculate launch costs, we use the fact that SiC has a density of 3.21 grams per cubic centimeter (Patnaik 2009); hence, a 4-inch diameter wafer with a thickness of 1 millimeter weighs about 0.03 kg. At a cost of $20,000 per kg to transport wafers to the space station, each wafer would cost $600 to transport to the space station. Thus, the total variable costs for healing a low-quality wafer is $850, the sum of the purchase price of the wafer ($250) and transport costs ($600).1
Additionally, we assume some astronaut time would be needed for healing the wafers. For our low estimate, we assume that the manufacturing operation needs 5 hours per week of astronaut time for supervision and maintenance. At a cost of $38,000 per hour times 260 hours per year (52 weeks times 5 hours), the annual cost of astronaut time would be $9.9 million per year. We assume that the space station provides electric power and other services as part of the rental payment paid by the manufacturer to the station. For our high estimate, to produce a greater number of wafers, we assume a proportional increase in astronaut time to 7.7 hours per week to handle the increased production volume—in other words, 400 hours per year, which would cost $15.3 million.1
Earthly Solution Risk
Very high, see below for reasons.
- Parabolic flights are much cheaper, because SiC wafer healing is additive over multiple parabolas and a total of 6 minutes is enough. 1
- ACME Advanced Materials seems to be dormant based on social platforms and LinkedIn.