Sub-system
A silicon carbide power device is mostly a substrate problem. The crystal grows slowly at extreme temperature, the material is nearly as hard as diamond, and slicing it wastes a large fraction of what was grown — which is why the wafer, not the device, has set the price of this technology.
In one sentence
A silicon carbide substrate is a single-crystal wafer of silicon carbide, grown by sublimation at very high temperature and sliced from a boule, used as the base for high-voltage power devices.
Growth is by sublimation rather than pulling from a melt: source powder is vaporised at over two thousand degrees and re-condenses on a seed crystal, at a rate of fractions of a millimetre an hour. A boule takes days to weeks and is far shorter than a silicon ingot, so each growth run yields relatively few wafers.
Then it has to be cut. Conventional wire sawing of a material this hard is slow and turns a substantial fraction of the boule into dust. Laser-assisted separation, which cleaves a wafer along a plane weakened by a focused laser, cuts that loss materially and is the main reason substrate cost has fallen.
Silicon carbide crystals contain dislocations and micropipes that propagate into devices built on them. A large device covers more substrate and is more likely to include a killer defect, so defect density limits how large a device can economically be — the same area-versus-yield arithmetic as logic, applied to the substrate rather than the process.
Moving from smaller to larger wafers spreads fixed per-wafer processing costs over more devices, and the industry's cost reduction has come substantially from that transition. Each step requires new growth, slicing and handling capability, and it is where most substrate capital investment goes.
Some device makers grow their own substrates to secure supply and capture margin; others buy them and focus on device design. Vertical integration protects against shortage and concentrates capital in a slow-moving process, and the industry contains both models arguing that theirs is correct.
Technology dependencies are solved by engineering; supply dependencies are solved by building something, which takes years.
Furnaces hold above two thousand degrees for days per boule, and a run interrupted part way through is lost rather than paused.
Grid infrastructureMicropipes and dislocations decide how large a device can be built on a wafer. A substrate maker that cannot map them ships defect density it cannot price.
Metrology and testOther pages in this map that name Silicon carbide substrates as something they cannot do without.
What each company supplies at this step, and — where a public figure exists — its share of this specific market — with what that share measures, the period it covers and who published it. Some rows also show the company’s own reported revenue for the segment covering this step, which is a different thing: it says how much this business matters to that company, not how much of the market it holds. Not a ranking and not a recommendation.
Grows silicon carbide substrates and manufactures devices on them.
Produces silicon carbide substrates alongside its wider compound semiconductor business.
Manufactures silicon carbide devices with integrated substrate capability.
Produces silicon carbide substrates for power and radio-frequency devices.
Supplies the epitaxial layer grown on the substrate, which sets device yield.
Chinese substrate maker whose capacity additions drove the recent price collapse.
Each page explains one technology in plain language, states what it depends on, and names companies by what they supply at that step. Company roles are described qualitatively and deliberately carry no market shares, revenue figures or rankings — those change faster than an explainer can, and a stale number is worse than none. Ticker links point at company pages on this site and are provided for reference only.
Nothing here is investment advice, a recommendation, or a forecast. A company named on a page about a technology is not thereby a good investment, and the chokepoints described are structural facts about supply chains rather than predictions about prices. Technology moves; where a page describes something as unresolved or in development, that was true when it was written.