GaN / GaAs CVD SiC Coated Epitaxy Susceptor Carrier
When you’re running GaN or GaAs epitaxy, the carrier takes a beating — high heat, aggressive gases, and repeated cycling. Semixlab’s CVD SiC coated graphite susceptors are purpose-built for these conditions. We designed them around the real-world demands of MOCVD and HVPE tools, focusing on three things: getting heat into the wafer evenly, keeping contamination low, and surviving long campaigns without falling apart.
Description
The basic construction is pretty straightforward. We use a high-density, isostatic graphite core and grow a thick β-SiC layer on top via CVD. That gives you the thermal conductivity you want from the graphite, while the SiC skin handles the chemistry. NH₃, H₂, and whatever else the process throws at it — the coating keeps all that away from the graphite underneath. The result is a part that holds up in GaN, GaAs, Micro-LED, RF, and power semiconductor fabs.
What makes them work in the real world?
On purity: we’re picky about the graphite raw material. The isostatic grades we use start out clean, and the CVD SiC coating ends up very dense. Metal contamination is low enough that you don’t have to worry about it fouling up a high-cleanliness epitaxy chamber.
Thermal behavior matters just as much. We put effort into matching the thermal profile so the wafer sees uniform temperature across its surface. Long runs at high temperature, rapid heating and cooling — the susceptor stays dimensionally stable through all of it. That consistency shows up in thickness uniformity and composition control run after run.
Chemical attack is probably the biggest concern in GaN MOCVD. If you’ve ever watched a bare graphite part erode away after a few cycles, you know what I mean. The SiC coating changes that dynamic. It resists the process gases much better, and because it doesn’t shed material as it ages, you get fewer particles floating around the chamber. Cleaner chamber, better yields — it’s that direct.
Mechanically, the coating sticks. We’ve put a lot of work into adhesion, so you don’t get blistering or peeling even when the part sees thermal shock. The dense CVD structure helps with that too. Users often see lifetimes several times longer than they’d get from uncoated graphite, which means less downtime for chamber cleans and part changes.
Where you’ll typically find them?

GaN MOCVD and HVPE — LEDs, power devices, RF chips, Micro-LEDs. Process temps sit around 1000 °C to 1100 °C, right in the sweet spot for this material system.
GaAs MOCVD — VCSELs, optical communication lasers, infrared sensors, RF parts. Similar thermal demands, slightly different chemistry, same need for a clean, stable carrier.
General high-temperature semiconductor tooling — beyond just epitaxy reactors, they turn up in wafer carrier assemblies and thermal zone components where cleanliness and thermal stability matter.
How we keep quality consistent
You can have the best graphite and the best coating chemistry, but if the process isn’t dialed in, the part won’t perform. We select graphite grades based on thermal conductivity, density, and long-term dimensional stability. Then the CVD step gets tuned to each design — especially important for large-diameter susceptors or anything with a tricky geometry. The goal is even coverage and strong adhesion, no thin spots or weak edges.
Every batch goes through several checkpoints: coating thickness, surface roughness, critical dimensions, purity, and overall coating integrity. It’s not glamorous, but it’s what keeps particle counts down and lets you reproduce your process without nasty surprises.
Why go with SiC coating instead of bare graphite?
Put simply: bare graphite gets eaten. SiC-coated graphite doesn’t. More specifically, you get:
● Way better resistance to corrosive gases
● Far less particle generation over time
● A longer useful life before replacement
● More stable process conditions
● Higher and more consistent wafer yields
It’s the combination that makes sense — graphite moves heat, SiC takes the chemical abuse. In compound semiconductor epitaxy, that’s become the go-to solution for a reason.
Customization
We almost never build the exact same part twice. Most of what we ship is made to a customer’s drawing or tuned to a specific reactor model. We can adjust:
● Wafer size and pocket layout
● Number and arrangement of pockets
● Hole patterns and through-holes
● Overall thickness
● Surface finish requirements
Prototypes and production volumes are both fine — we’ll work with you whether you need a couple of test parts or steady deliveries.
Frequently asked questions
What exactly does a SiC coated susceptor do?
It holds the wafers during epitaxial growth, spreads heat evenly across the wafer surface, and uses the SiC layer to protect the graphite from corrosive process gases.
Why not just use uncoated graphite?
Uncoated graphite corrodes too fast in gases like hot ammonia and hydrogen. That generates particles, shortens part life, and hurts yield. The SiC coating solves all three of those problems.
Can you build these for different reactors?
Absolutely. We tailor dimensions, pocket layouts, and hole patterns to match whatever MOCVD or HVPE tool you’re running.
What processes are they suitable for?
GaN LED, power, and RF epi; Micro-LED; VCSEL and GaAs epitaxy — anywhere you need a clean, thermally stable carrier at high temperature.
Our Services

Semixlab Product Shop


EN
EN
DA
NL
FI
FR
DE
IT
JA
KO
NO
PL
PT
RO
RU
ES
SV
TL
ID
SK
UK
VI
TH
TR
FA
BE
LA
UZ




