Silicon carbide single crystals used in power electronics begin with a material that is often overlooked: the SiC source powder loaded into the growth crucible.
In physical vapor transport, or PVT, the source powder is heated to an extremely high temperature. Si-containing vapor species and carbon-containing species are transported through the crucible and deposited on a SiC seed crystal. Although temperature distribution, pressure, crucible design and seed quality are all critical, the characteristics of the source powder directly affect vapor stability, impurity incorporation, boule growth rate and usable crystal yield.
For this reason, semiconductor-grade SiC powder cannot be evaluated by purity alone. Particle-size distribution, packing density, free silicon, free carbon, oxygen content, polytype composition and batch consistency must also be considered.
![]()
PVT growth typically uses a graphite crucible containing high-purity SiC powder. At high temperatures, the source gradually sublimates and generates vapor species that move toward the cooler seed region.
The source powder changes continuously during this process:
These changes affect the Si/C vapor ratio and the stability of mass transport. An unstable source can cause variations in boule diameter, growth rate, polytype stability and defect formation.
Research into PVT source optimization has shown that powder packing density and particle-size distribution can influence both source utilization and process stability. A high packing density increases the amount of usable source material in a fixed crucible volume, while a controlled particle-size distribution helps maintain a more stable sublimation behavior during growth.
SiC source powder is frequently described using purity grades such as 4N, 5N or 6N:
| Purity designation | Nominal purity |
|---|---|
| 4N | 99.99% |
| 5N | 99.999% |
| 6N | 99.9999% |
However, the number of nines does not provide enough information for selecting a crystal-growth source.
Two powders with the same total purity may behave differently if one contains boron and aluminum while the other primarily contains less electrically active contaminants. Buyers should therefore request an elemental impurity report instead of relying only on a nominal purity percentage.
Analytical results may also vary according to the testing method and detection limit. The test method, sampled batch and reporting threshold should be clearly stated on the certificate of analysis.
Boron is an acceptor impurity in SiC. Even at low concentrations, it can influence carrier compensation and electrical resistivity.
Boron control is particularly important when producing high-resistivity or semi-insulating SiC. The acceptable concentration depends on the target crystal specification and the relationship between boron, nitrogen and any intentionally introduced deep-level dopant.
Nitrogen acts as a donor in SiC and is commonly associated with n-type conductivity. Nitrogen may originate from the source powder, furnace atmosphere, graphite components or intentional doping.
Studies have shown that nitrogen concentration in synthesized SiC powder can be affected by powder-synthesis temperature and argon pressure. Nitrogen incorporated into the source must therefore be considered part of the overall doping-control strategy, rather than treated only as an external furnace-gas variable.
Aluminum is another acceptor impurity. It can enter the source through raw materials, powder-processing equipment or contaminated furnace components.
For n-type material, excessive aluminum may compensate donor carriers and make resistivity control more difficult. In high-resistivity material, uncontrolled aluminum can disturb the intended compensation balance.
Iron, titanium, vanadium, chromium, nickel and other metals may introduce electrically active energy levels or affect optical and electrical uniformity.
Some elements, such as vanadium, may be intentionally added when producing certain semi-insulating crystals. Intentional doping requires much tighter concentration and uniformity control than accidental contamination.
Oxygen may come from surface oxidation, raw materials, storage conditions or powder-handling environments. It may be present as surface oxygen, oxide compounds or contamination introduced during powder synthesis.
Because powder has a high total surface area, fine material can be particularly sensitive to oxidation and moisture exposure.
A powder can have a high nominal SiC purity while still containing undesirable free silicon or free carbon.
These phases affect the effective Si/C ratio of the source and may change its sublimation behavior. Therefore, semiconductor crystal-growth powder should be tested for phase composition—not only trace metallic impurities.
Research on SiC purification has found that powders containing high levels of free Si, free C, oxygen and metallic contaminants can present problems when used directly for semiconductor-grade single-crystal growth.
Particle size affects source packing, available surface area, gas flow and the rate at which the source evolves during growth.
Commercial SiC powders used for bulk crystal growth can have particle sizes extending from tens to several hundred micrometres, but there is no single particle size that is optimal for every PVT reactor. The correct distribution depends on crucible geometry, thermal field, growth duration and target boule dimensions. A recent review describes commercial SiC growth powders with typical average particle diameters in an approximate range of 50–500 μm.
Excessively fine powder may:
Excessively coarse powder may:
A broad particle-size distribution can improve packing because fine particles fill the spaces between coarse particles. However, different particle fractions may also sublime at different rates.
A narrower distribution can make source behavior more predictable, provided that the selected size range produces sufficient packing density and vapor supply.
Particle-size selection should therefore balance four objectives:
Packing density determines how much powder can be loaded into a given crucible volume. A higher loading can support longer growth runs or larger boules without changing the external crucible dimensions.
However, maximum density is not automatically the best process condition. The powder bed must retain sufficient porosity for vapor transport. If packing is uneven, the source can develop local channels and regions with different sublimation rates.
Important loading parameters include:
For repeatable production, these parameters should be recorded as part of the growth recipe.
Commercial source powder may contain different SiC polytypes, including 3C, 4H and 6H phases. The source polytype does not simply determine the final boule polytype, because seed orientation and growth conditions play major roles.
Nevertheless, phase composition can influence source sublimation characteristics. Purified beta-SiC powder has been investigated as a source material because of its sublimation behavior and potential to support growth under modified temperature conditions.
X-ray diffraction is therefore useful for confirming the phase composition of each source-powder batch.
PVT yield should not be measured only by the total length or weight of the grown boule. A better assessment includes:
A powder that increases the initial growth rate but produces unstable vapor conditions later in the run may reduce total usable yield.
Similarly, a very pure powder may still perform poorly if its particle distribution causes low packing density or rapid source evolution. The best powder is the one that delivers repeatable mass transport throughout the complete growth cycle.
A complete incoming inspection plan may include:
| Inspection item | Typical method | Purpose |
|---|---|---|
| Trace elements | GDMS or ICP-MS | Measures metallic and dopant impurities |
| Particle-size distribution | Laser diffraction or sieve analysis | Evaluates powder consistency |
| Phase composition | XRD | Identifies SiC polytypes and secondary phases |
| Free silicon | Chemical or phase analysis | Checks Si-rich residual material |
| Free carbon | Carbon analysis | Checks excess carbon |
| Oxygen | Inert gas fusion or suitable elemental analysis | Evaluates oxidation and oxygen contamination |
| Bulk density | Gravimetric measurement | Supports crucible-loading control |
| Tapped density | Standard tapping test | Evaluates packing behavior |
| Morphology | SEM | Observes particle shape and agglomeration |
| Moisture | Controlled drying or moisture analysis | Checks storage and handling condition |
The specification should state the sampling procedure because one small sample may not represent a large powder batch, especially if particle segregation has occurred.
High-purity SiC powder can be contaminated after production. Contact with ordinary metal tools, open workshop air or unsuitable packaging can introduce particles and trace metals.
Recommended controls include:
Powder should also be handled carefully to avoid particle-size segregation. Excessive vibration during transportation can cause fine and coarse fractions to separate.
When requesting high-purity SiC powder for PVT growth, specify:
If possible, the powder should be qualified through a controlled growth trial. Chemical analysis is essential, but actual PVT performance remains the most direct way to evaluate source stability and usable crystal yield.
High-purity SiC powder is an engineered source material, not a simple consumable. Its purity, impurity profile, particle-size distribution, density, phase composition and handling history collectively determine how it behaves inside a PVT furnace.
For stable SiC boule production, manufacturers must look beyond the number of nines in the purity specification. The most suitable powder is one that provides controlled vapor generation, predictable source evolution, consistent electrical properties and high usable yield across repeated growth runs.
Silicon carbide single crystals used in power electronics begin with a material that is often overlooked: the SiC source powder loaded into the growth crucible.
In physical vapor transport, or PVT, the source powder is heated to an extremely high temperature. Si-containing vapor species and carbon-containing species are transported through the crucible and deposited on a SiC seed crystal. Although temperature distribution, pressure, crucible design and seed quality are all critical, the characteristics of the source powder directly affect vapor stability, impurity incorporation, boule growth rate and usable crystal yield.
For this reason, semiconductor-grade SiC powder cannot be evaluated by purity alone. Particle-size distribution, packing density, free silicon, free carbon, oxygen content, polytype composition and batch consistency must also be considered.
![]()
PVT growth typically uses a graphite crucible containing high-purity SiC powder. At high temperatures, the source gradually sublimates and generates vapor species that move toward the cooler seed region.
The source powder changes continuously during this process:
These changes affect the Si/C vapor ratio and the stability of mass transport. An unstable source can cause variations in boule diameter, growth rate, polytype stability and defect formation.
Research into PVT source optimization has shown that powder packing density and particle-size distribution can influence both source utilization and process stability. A high packing density increases the amount of usable source material in a fixed crucible volume, while a controlled particle-size distribution helps maintain a more stable sublimation behavior during growth.
SiC source powder is frequently described using purity grades such as 4N, 5N or 6N:
| Purity designation | Nominal purity |
|---|---|
| 4N | 99.99% |
| 5N | 99.999% |
| 6N | 99.9999% |
However, the number of nines does not provide enough information for selecting a crystal-growth source.
Two powders with the same total purity may behave differently if one contains boron and aluminum while the other primarily contains less electrically active contaminants. Buyers should therefore request an elemental impurity report instead of relying only on a nominal purity percentage.
Analytical results may also vary according to the testing method and detection limit. The test method, sampled batch and reporting threshold should be clearly stated on the certificate of analysis.
Boron is an acceptor impurity in SiC. Even at low concentrations, it can influence carrier compensation and electrical resistivity.
Boron control is particularly important when producing high-resistivity or semi-insulating SiC. The acceptable concentration depends on the target crystal specification and the relationship between boron, nitrogen and any intentionally introduced deep-level dopant.
Nitrogen acts as a donor in SiC and is commonly associated with n-type conductivity. Nitrogen may originate from the source powder, furnace atmosphere, graphite components or intentional doping.
Studies have shown that nitrogen concentration in synthesized SiC powder can be affected by powder-synthesis temperature and argon pressure. Nitrogen incorporated into the source must therefore be considered part of the overall doping-control strategy, rather than treated only as an external furnace-gas variable.
Aluminum is another acceptor impurity. It can enter the source through raw materials, powder-processing equipment or contaminated furnace components.
For n-type material, excessive aluminum may compensate donor carriers and make resistivity control more difficult. In high-resistivity material, uncontrolled aluminum can disturb the intended compensation balance.
Iron, titanium, vanadium, chromium, nickel and other metals may introduce electrically active energy levels or affect optical and electrical uniformity.
Some elements, such as vanadium, may be intentionally added when producing certain semi-insulating crystals. Intentional doping requires much tighter concentration and uniformity control than accidental contamination.
Oxygen may come from surface oxidation, raw materials, storage conditions or powder-handling environments. It may be present as surface oxygen, oxide compounds or contamination introduced during powder synthesis.
Because powder has a high total surface area, fine material can be particularly sensitive to oxidation and moisture exposure.
A powder can have a high nominal SiC purity while still containing undesirable free silicon or free carbon.
These phases affect the effective Si/C ratio of the source and may change its sublimation behavior. Therefore, semiconductor crystal-growth powder should be tested for phase composition—not only trace metallic impurities.
Research on SiC purification has found that powders containing high levels of free Si, free C, oxygen and metallic contaminants can present problems when used directly for semiconductor-grade single-crystal growth.
Particle size affects source packing, available surface area, gas flow and the rate at which the source evolves during growth.
Commercial SiC powders used for bulk crystal growth can have particle sizes extending from tens to several hundred micrometres, but there is no single particle size that is optimal for every PVT reactor. The correct distribution depends on crucible geometry, thermal field, growth duration and target boule dimensions. A recent review describes commercial SiC growth powders with typical average particle diameters in an approximate range of 50–500 μm.
Excessively fine powder may:
Excessively coarse powder may:
A broad particle-size distribution can improve packing because fine particles fill the spaces between coarse particles. However, different particle fractions may also sublime at different rates.
A narrower distribution can make source behavior more predictable, provided that the selected size range produces sufficient packing density and vapor supply.
Particle-size selection should therefore balance four objectives:
Packing density determines how much powder can be loaded into a given crucible volume. A higher loading can support longer growth runs or larger boules without changing the external crucible dimensions.
However, maximum density is not automatically the best process condition. The powder bed must retain sufficient porosity for vapor transport. If packing is uneven, the source can develop local channels and regions with different sublimation rates.
Important loading parameters include:
For repeatable production, these parameters should be recorded as part of the growth recipe.
Commercial source powder may contain different SiC polytypes, including 3C, 4H and 6H phases. The source polytype does not simply determine the final boule polytype, because seed orientation and growth conditions play major roles.
Nevertheless, phase composition can influence source sublimation characteristics. Purified beta-SiC powder has been investigated as a source material because of its sublimation behavior and potential to support growth under modified temperature conditions.
X-ray diffraction is therefore useful for confirming the phase composition of each source-powder batch.
PVT yield should not be measured only by the total length or weight of the grown boule. A better assessment includes:
A powder that increases the initial growth rate but produces unstable vapor conditions later in the run may reduce total usable yield.
Similarly, a very pure powder may still perform poorly if its particle distribution causes low packing density or rapid source evolution. The best powder is the one that delivers repeatable mass transport throughout the complete growth cycle.
A complete incoming inspection plan may include:
| Inspection item | Typical method | Purpose |
|---|---|---|
| Trace elements | GDMS or ICP-MS | Measures metallic and dopant impurities |
| Particle-size distribution | Laser diffraction or sieve analysis | Evaluates powder consistency |
| Phase composition | XRD | Identifies SiC polytypes and secondary phases |
| Free silicon | Chemical or phase analysis | Checks Si-rich residual material |
| Free carbon | Carbon analysis | Checks excess carbon |
| Oxygen | Inert gas fusion or suitable elemental analysis | Evaluates oxidation and oxygen contamination |
| Bulk density | Gravimetric measurement | Supports crucible-loading control |
| Tapped density | Standard tapping test | Evaluates packing behavior |
| Morphology | SEM | Observes particle shape and agglomeration |
| Moisture | Controlled drying or moisture analysis | Checks storage and handling condition |
The specification should state the sampling procedure because one small sample may not represent a large powder batch, especially if particle segregation has occurred.
High-purity SiC powder can be contaminated after production. Contact with ordinary metal tools, open workshop air or unsuitable packaging can introduce particles and trace metals.
Recommended controls include:
Powder should also be handled carefully to avoid particle-size segregation. Excessive vibration during transportation can cause fine and coarse fractions to separate.
When requesting high-purity SiC powder for PVT growth, specify:
If possible, the powder should be qualified through a controlled growth trial. Chemical analysis is essential, but actual PVT performance remains the most direct way to evaluate source stability and usable crystal yield.
High-purity SiC powder is an engineered source material, not a simple consumable. Its purity, impurity profile, particle-size distribution, density, phase composition and handling history collectively determine how it behaves inside a PVT furnace.
For stable SiC boule production, manufacturers must look beyond the number of nines in the purity specification. The most suitable powder is one that provides controlled vapor generation, predictable source evolution, consistent electrical properties and high usable yield across repeated growth runs.