Production, research and dummy grade SiC wafers may have the same diameter, polytype and nominal thickness, but they are not interchangeable.
A dummy grade wafer may be suitable for equipment calibration but unreliable for epitaxial growth. A research grade wafer may work well for early material experiments but produce misleading results in device-yield studies. A production grade wafer usually provides tighter defect, surface and geometry control, but it may be unnecessarily expensive for mechanical handling tests.
Choosing the correct SiC wafer grade therefore requires more than comparing prices.
Buyers must consider:
This guide explains the practical differences between production, research and dummy grade SiC wafers and provides an RFQ checklist for selecting the correct material.
One of the most important purchasing rules is that grade names are not fully standardized between suppliers.
The terms may include:
Two suppliers may both offer a “research grade” wafer but apply different limits for micropipes, surface scratches, bow, warp, resistivity uniformity or usable area.
Some suppliers use letter grades such as A, B, C and D, while others classify wafers by defect density or application. XKH, for example, offers SiC substrates identified as production, research and dummy grades across different wafer sizes and conductivity types.
The grade name should therefore be treated as a starting point. The purchase order must contain measurable acceptance criteria.
A wafer grade does not identify its complete material structure.
For example, all the following products may be supplied in different grades:
A buyer should specify grade together with:
A production grade 4H-N wafer for a power MOSFET is fundamentally different from a production grade semi-insulating 4H-SiC substrate intended for GaN RF epitaxy.
Production grade, sometimes called prime grade, is intended for processes where wafer defects and variation directly influence device yield, reliability and manufacturing consistency.
These wafers normally have the tightest available controls for:
Production grade SiC wafers are commonly selected for:
A production-grade RFQ may include limits for:
Production grade does not automatically mean “zero defects.” If zero micropipes or another specific defect limit is required, it must be stated separately.
Defects originating in the substrate can propagate into the epitaxial layer or create new surface defects during growth. Basal-plane dislocations, stacking faults, pits and other extended defects may affect leakage, forward-voltage stability, breakdown behavior and long-term reliability.
A 2025 review in the Journal of Applied Physics explains how multiple SiC epitaxial defects can contribute to device degradation and failure.
Production grade is therefore usually justified when the cost of epitaxy, device processing and failed dies is much higher than the additional substrate cost.
Research grade SiC wafers provide a balance between functional material quality and purchasing cost.
They may have the correct:
However, they generally allow more crystal defects, surface imperfections or geometry variation than production grade material.
Research grade wafers may be appropriate for:
Some research grade wafers are supplied with an epi-ready CMP surface. Others may contain polishing marks, scratches or subsurface damage that make them unsuitable for high-quality epitaxy.
Before ordering, confirm:
A low-cost research wafer with a poorly prepared surface may cause false conclusions during epitaxial development.
Research grade is often the most economical choice when:
For comparative experiments, all wafers should ideally come from a consistent grade and lot. Mixing different grades can make it difficult to determine whether performance changes come from the experimental process or the substrate.
Dummy grade SiC wafers are primarily intended for mechanical, equipment or non-device process testing.
They may have the correct nominal diameter and thickness but relaxed requirements for:
Dummy grade wafers are commonly used for:
Unless the supplier provides additional guarantees, dummy grade should not normally be used for:
A dummy wafer may survive the handling process but still contain defects that make meaningful electrical evaluation impossible.
| Item | Production Grade | Research Grade | Dummy Grade |
|---|---|---|---|
| Main purpose | Commercial devices and qualified epitaxy | R&D and prototype development | Equipment and mechanical testing |
| Crystal defect control | Tightest | Moderate | Relaxed or not fully specified |
| Micropipe/BPD/TSD limits | Normally required | May be relaxed | May not be guaranteed |
| Resistivity control | Tight range and uniformity | Functional but wider variation | May not be guaranteed |
| Surface finish | Normally epi-ready CMP | Epi-ready or standard polish | Lapped, polished or cosmetic |
| Surface scratches | Strictly limited | Some defects may be accepted | More defects may be allowed |
| TTV, bow and warp | Tight | Moderate | Relaxed |
| Usable area | High | Moderate | Not always specified |
| Inspection report | Detailed or available | Basic or optional | Limited |
| Lot traceability | Normally required | Often available | May be limited |
| Relative price | Highest | Medium | Lowest |
| Suitable for device production | Yes | Only after qualification | Normally no |
| Suitable for destructive testing | Possible but expensive | Yes | Yes |
| Suitable for equipment calibration | Unnecessarily expensive | Possible | Recommended |
The exact limits must be confirmed with the supplier because grade terminology can vary.
Micropipes are hollow-core defects that can severely affect high-voltage devices. Modern production wafers may have very low or zero specified micropipe density.
Research and dummy grades may permit a higher density.
Ask for:
BPDs may propagate into the epitaxial layer or contribute to stacking-fault formation in bipolar devices.
Extended defects in SiC epitaxial layers have been linked to reduced device yield and reliability, making BPD limits important for high-value production wafers. Study of SiC defects affecting yield
TSDs can be associated with surface pits, leakage paths and local device failures. Production material normally requires a measurable maximum density.
A 4H-SiC wafer should maintain the required 4H polytype throughout the usable area. Local 3C or 6H inclusions may affect epitaxy and device performance.
Carbon-rich inclusions originating during crystal growth may produce surface defects after slicing and polishing or influence epitaxial defect formation.
A surface may look mirror-polished while still containing polishing damage below the surface. These defects can become visible after hydrogen etching or epitaxial growth.
For epitaxy, ask whether the wafer is:
TTV, bow and warp influence:
Even a dummy wafer requires adequate geometry if it will be used to qualify automated handling equipment.
A dummy grade wafer may be manufactured as new material but classified with relaxed defect limits.
A reclaimed wafer has already been used and then processed again through steps such as:
Reclaimed wafers may have reduced thickness, altered surface history or unknown contamination risks.
When purchasing low-cost wafers, ask whether the material is:
These categories should not be assumed to be equivalent.
Recommended starting point:
Production grade 4H-N SiC
Specify:
Recommended starting point:
Production grade high-purity semi-insulating 4H-SiC
Specify:
Recommended starting point:
Research grade with a guaranteed epi-ready surface
Use production grade reference wafers alongside research grade material when a reliable benchmark is required.
Recommended starting point:
Research grade wafer or diced SiC coupons
Confirm that defects within the test area will not invalidate the experiment.
Recommended starting point:
Dummy grade
Prioritize:
Electrical and crystal-defect specifications may be unnecessary.
Recommended starting point:
Dummy or research grade
Choose dummy grade for machine setup and research grade if the experiment must reflect realistic single-crystal behavior or surface quality.
A lower wafer price does not always reduce total project cost.
Potential hidden costs include:
If a dummy wafer causes the loss of a high-value epitaxial run, the substrate saving becomes insignificant.
Buyers should compare total experiment or production cost rather than only the price per wafer.
| RFQ Item | Information to Specify |
| Application | Device production, epitaxy, R&D or equipment test |
| Grade | Production, research or dummy |
| Material condition | New, reclaimed or recycled |
| Polytype | 4H, 6H or other |
| Conductivity | N-type, P-type or semi-insulating |
| Dopant | Nitrogen, vanadium, high-purity undoped or other |
| Diameter | 2, 3, 4, 6, 8 inch or custom |
| Thickness | Nominal value and tolerance |
| Orientation | On-axis or off-axis |
| Offcut | Angle, direction and tolerance |
| Wafer face | Si-face or C-face |
| Resistivity | Range or minimum value |
| Surface | SSP, DSP, CMP, lapped or epi-ready |
| Roughness | Maximum Ra |
| MPD | Maximum density |
| BPD | Maximum density |
| TSD/TED | Maximum density |
| Inclusions | Carbon and polytype limits |
| TTV | Maximum value |
| Bow and warp | Maximum values |
| Edge quality | Bevel, chips and cracks |
| Usable area | Minimum percentage |
| Inspection | Defect map, XRD, AFM and geometry report |
| Packaging | Single-wafer box or cassette |
| Quantity | Qualification and production volume |
Application: SiC MOSFET production
Grade: Production grade
Material: 4H-N SiC
Diameter: 150 mm
Orientation: 4° off-axis toward ⟨11-20⟩
Surface: Si-face epi-ready CMP
Resistivity: Defined production range
Defects: MPD, BPD, TSD and TED limits required
Geometry: TTV, bow and warp limits required
Inspection: Full wafer defect and geometry report
Quantity: Qualification lot followed by monthly production
Application: CVD epitaxial process development
Grade: Research grade
Material: 4H-N SiC
Diameter: 100 mm
Surface: Si-face epi-ready CMP
Accepted Defects: Relaxed crystal-defect limits
Critical Requirement: No deep scratches in the central test area
Inspection: Basic surface and geometry report
Quantity: 10 pieces
Application: Wafer-handling robot calibration
Grade: New dummy grade
Diameter: 150 mm
Thickness: Matched to production wafers
Critical Requirements: Diameter, thickness, bow, warp and edge profile
Surface: Standard polish sufficient
Electrical Properties: Not required
Quantity: 25 pieces
Yes, provided the wafer has an epi-ready surface and its defect levels are acceptable for the experiment. Production-level yield should not be assumed.
Dummy wafers are normally intended for handling, equipment or destructive process tests. Device fabrication is not recommended unless the supplier provides suitable electrical, crystal and surface guarantees.
Not always. Zero-MPD may be a higher or separately defined grade. The purchase order should state the actual micropipe-density limit.
Yes. It is often the most cost-effective choice for material studies, process development and small-area prototypes.
Yes, when they are used to test robot handling, cassettes, load ports, vacuum chucks or equipment clearance.
No. Dummy material may be a new wafer with relaxed specifications. Always confirm whether the wafers are new, reclaimed or recycled.
Production grade high-purity semi-insulating 4H-SiC is normally the safest starting point for qualified RF epitaxy. Research grade may be suitable for early development.
Production, research and dummy grade SiC wafers serve different purposes.
Production grade provides the tightest control of defects, surface condition, geometry and electrical uniformity. It is the appropriate choice for commercial epitaxy, device manufacturing and high-reliability applications.
Research grade offers a practical balance between quality and cost for process development, laboratory testing and prototype fabrication.
Dummy grade is the most economical choice for equipment qualification, robot calibration, dicing trials and other mechanical or sacrificial processes.
The correct purchasing decision should be based on total process risk—not simply the lowest wafer price.
When requesting a quotation, buyers should provide:
A complete specification allows the supplier to recommend the lowest-cost grade that can still perform reliably in the intended process.
Production, research and dummy grade SiC wafers may have the same diameter, polytype and nominal thickness, but they are not interchangeable.
A dummy grade wafer may be suitable for equipment calibration but unreliable for epitaxial growth. A research grade wafer may work well for early material experiments but produce misleading results in device-yield studies. A production grade wafer usually provides tighter defect, surface and geometry control, but it may be unnecessarily expensive for mechanical handling tests.
Choosing the correct SiC wafer grade therefore requires more than comparing prices.
Buyers must consider:
This guide explains the practical differences between production, research and dummy grade SiC wafers and provides an RFQ checklist for selecting the correct material.
One of the most important purchasing rules is that grade names are not fully standardized between suppliers.
The terms may include:
Two suppliers may both offer a “research grade” wafer but apply different limits for micropipes, surface scratches, bow, warp, resistivity uniformity or usable area.
Some suppliers use letter grades such as A, B, C and D, while others classify wafers by defect density or application. XKH, for example, offers SiC substrates identified as production, research and dummy grades across different wafer sizes and conductivity types.
The grade name should therefore be treated as a starting point. The purchase order must contain measurable acceptance criteria.
A wafer grade does not identify its complete material structure.
For example, all the following products may be supplied in different grades:
A buyer should specify grade together with:
A production grade 4H-N wafer for a power MOSFET is fundamentally different from a production grade semi-insulating 4H-SiC substrate intended for GaN RF epitaxy.
Production grade, sometimes called prime grade, is intended for processes where wafer defects and variation directly influence device yield, reliability and manufacturing consistency.
These wafers normally have the tightest available controls for:
Production grade SiC wafers are commonly selected for:
A production-grade RFQ may include limits for:
Production grade does not automatically mean “zero defects.” If zero micropipes or another specific defect limit is required, it must be stated separately.
Defects originating in the substrate can propagate into the epitaxial layer or create new surface defects during growth. Basal-plane dislocations, stacking faults, pits and other extended defects may affect leakage, forward-voltage stability, breakdown behavior and long-term reliability.
A 2025 review in the Journal of Applied Physics explains how multiple SiC epitaxial defects can contribute to device degradation and failure.
Production grade is therefore usually justified when the cost of epitaxy, device processing and failed dies is much higher than the additional substrate cost.
Research grade SiC wafers provide a balance between functional material quality and purchasing cost.
They may have the correct:
However, they generally allow more crystal defects, surface imperfections or geometry variation than production grade material.
Research grade wafers may be appropriate for:
Some research grade wafers are supplied with an epi-ready CMP surface. Others may contain polishing marks, scratches or subsurface damage that make them unsuitable for high-quality epitaxy.
Before ordering, confirm:
A low-cost research wafer with a poorly prepared surface may cause false conclusions during epitaxial development.
Research grade is often the most economical choice when:
For comparative experiments, all wafers should ideally come from a consistent grade and lot. Mixing different grades can make it difficult to determine whether performance changes come from the experimental process or the substrate.
Dummy grade SiC wafers are primarily intended for mechanical, equipment or non-device process testing.
They may have the correct nominal diameter and thickness but relaxed requirements for:
Dummy grade wafers are commonly used for:
Unless the supplier provides additional guarantees, dummy grade should not normally be used for:
A dummy wafer may survive the handling process but still contain defects that make meaningful electrical evaluation impossible.
| Item | Production Grade | Research Grade | Dummy Grade |
|---|---|---|---|
| Main purpose | Commercial devices and qualified epitaxy | R&D and prototype development | Equipment and mechanical testing |
| Crystal defect control | Tightest | Moderate | Relaxed or not fully specified |
| Micropipe/BPD/TSD limits | Normally required | May be relaxed | May not be guaranteed |
| Resistivity control | Tight range and uniformity | Functional but wider variation | May not be guaranteed |
| Surface finish | Normally epi-ready CMP | Epi-ready or standard polish | Lapped, polished or cosmetic |
| Surface scratches | Strictly limited | Some defects may be accepted | More defects may be allowed |
| TTV, bow and warp | Tight | Moderate | Relaxed |
| Usable area | High | Moderate | Not always specified |
| Inspection report | Detailed or available | Basic or optional | Limited |
| Lot traceability | Normally required | Often available | May be limited |
| Relative price | Highest | Medium | Lowest |
| Suitable for device production | Yes | Only after qualification | Normally no |
| Suitable for destructive testing | Possible but expensive | Yes | Yes |
| Suitable for equipment calibration | Unnecessarily expensive | Possible | Recommended |
The exact limits must be confirmed with the supplier because grade terminology can vary.
Micropipes are hollow-core defects that can severely affect high-voltage devices. Modern production wafers may have very low or zero specified micropipe density.
Research and dummy grades may permit a higher density.
Ask for:
BPDs may propagate into the epitaxial layer or contribute to stacking-fault formation in bipolar devices.
Extended defects in SiC epitaxial layers have been linked to reduced device yield and reliability, making BPD limits important for high-value production wafers. Study of SiC defects affecting yield
TSDs can be associated with surface pits, leakage paths and local device failures. Production material normally requires a measurable maximum density.
A 4H-SiC wafer should maintain the required 4H polytype throughout the usable area. Local 3C or 6H inclusions may affect epitaxy and device performance.
Carbon-rich inclusions originating during crystal growth may produce surface defects after slicing and polishing or influence epitaxial defect formation.
A surface may look mirror-polished while still containing polishing damage below the surface. These defects can become visible after hydrogen etching or epitaxial growth.
For epitaxy, ask whether the wafer is:
TTV, bow and warp influence:
Even a dummy wafer requires adequate geometry if it will be used to qualify automated handling equipment.
A dummy grade wafer may be manufactured as new material but classified with relaxed defect limits.
A reclaimed wafer has already been used and then processed again through steps such as:
Reclaimed wafers may have reduced thickness, altered surface history or unknown contamination risks.
When purchasing low-cost wafers, ask whether the material is:
These categories should not be assumed to be equivalent.
Recommended starting point:
Production grade 4H-N SiC
Specify:
Recommended starting point:
Production grade high-purity semi-insulating 4H-SiC
Specify:
Recommended starting point:
Research grade with a guaranteed epi-ready surface
Use production grade reference wafers alongside research grade material when a reliable benchmark is required.
Recommended starting point:
Research grade wafer or diced SiC coupons
Confirm that defects within the test area will not invalidate the experiment.
Recommended starting point:
Dummy grade
Prioritize:
Electrical and crystal-defect specifications may be unnecessary.
Recommended starting point:
Dummy or research grade
Choose dummy grade for machine setup and research grade if the experiment must reflect realistic single-crystal behavior or surface quality.
A lower wafer price does not always reduce total project cost.
Potential hidden costs include:
If a dummy wafer causes the loss of a high-value epitaxial run, the substrate saving becomes insignificant.
Buyers should compare total experiment or production cost rather than only the price per wafer.
| RFQ Item | Information to Specify |
| Application | Device production, epitaxy, R&D or equipment test |
| Grade | Production, research or dummy |
| Material condition | New, reclaimed or recycled |
| Polytype | 4H, 6H or other |
| Conductivity | N-type, P-type or semi-insulating |
| Dopant | Nitrogen, vanadium, high-purity undoped or other |
| Diameter | 2, 3, 4, 6, 8 inch or custom |
| Thickness | Nominal value and tolerance |
| Orientation | On-axis or off-axis |
| Offcut | Angle, direction and tolerance |
| Wafer face | Si-face or C-face |
| Resistivity | Range or minimum value |
| Surface | SSP, DSP, CMP, lapped or epi-ready |
| Roughness | Maximum Ra |
| MPD | Maximum density |
| BPD | Maximum density |
| TSD/TED | Maximum density |
| Inclusions | Carbon and polytype limits |
| TTV | Maximum value |
| Bow and warp | Maximum values |
| Edge quality | Bevel, chips and cracks |
| Usable area | Minimum percentage |
| Inspection | Defect map, XRD, AFM and geometry report |
| Packaging | Single-wafer box or cassette |
| Quantity | Qualification and production volume |
Application: SiC MOSFET production
Grade: Production grade
Material: 4H-N SiC
Diameter: 150 mm
Orientation: 4° off-axis toward ⟨11-20⟩
Surface: Si-face epi-ready CMP
Resistivity: Defined production range
Defects: MPD, BPD, TSD and TED limits required
Geometry: TTV, bow and warp limits required
Inspection: Full wafer defect and geometry report
Quantity: Qualification lot followed by monthly production
Application: CVD epitaxial process development
Grade: Research grade
Material: 4H-N SiC
Diameter: 100 mm
Surface: Si-face epi-ready CMP
Accepted Defects: Relaxed crystal-defect limits
Critical Requirement: No deep scratches in the central test area
Inspection: Basic surface and geometry report
Quantity: 10 pieces
Application: Wafer-handling robot calibration
Grade: New dummy grade
Diameter: 150 mm
Thickness: Matched to production wafers
Critical Requirements: Diameter, thickness, bow, warp and edge profile
Surface: Standard polish sufficient
Electrical Properties: Not required
Quantity: 25 pieces
Yes, provided the wafer has an epi-ready surface and its defect levels are acceptable for the experiment. Production-level yield should not be assumed.
Dummy wafers are normally intended for handling, equipment or destructive process tests. Device fabrication is not recommended unless the supplier provides suitable electrical, crystal and surface guarantees.
Not always. Zero-MPD may be a higher or separately defined grade. The purchase order should state the actual micropipe-density limit.
Yes. It is often the most cost-effective choice for material studies, process development and small-area prototypes.
Yes, when they are used to test robot handling, cassettes, load ports, vacuum chucks or equipment clearance.
No. Dummy material may be a new wafer with relaxed specifications. Always confirm whether the wafers are new, reclaimed or recycled.
Production grade high-purity semi-insulating 4H-SiC is normally the safest starting point for qualified RF epitaxy. Research grade may be suitable for early development.
Production, research and dummy grade SiC wafers serve different purposes.
Production grade provides the tightest control of defects, surface condition, geometry and electrical uniformity. It is the appropriate choice for commercial epitaxy, device manufacturing and high-reliability applications.
Research grade offers a practical balance between quality and cost for process development, laboratory testing and prototype fabrication.
Dummy grade is the most economical choice for equipment qualification, robot calibration, dicing trials and other mechanical or sacrificial processes.
The correct purchasing decision should be based on total process risk—not simply the lowest wafer price.
When requesting a quotation, buyers should provide:
A complete specification allows the supplier to recommend the lowest-cost grade that can still perform reliably in the intended process.
english
français
Deutsch
Italiano
Русский
Español
português
Nederlandse
ελληνικά
日本語
한국
العربية
Türkçe
polski