DF claims all-diamond MOSFETs with 50x faster switching
DF, the publisher of the df.com page (the abbreviation is not expanded in the text), presents what it calls all-diamond MOSFETs. The page says the company is developing, in partnership with Stanford University, transistors that use diamond as the semiconductor. The headline claim is "50x faster switching". The page gives no baseline for that figure: it does not say faster than what, or under what conditions.
The technical idea is surface transfer doping. Instead of putting dopant atoms into the diamond crystal, the transistors induce charge carriers without introducing substitutional dopant atoms into the bulk. The page explains the mechanism in three steps. First, the surface of a single-crystal diamond wafer is terminated with an atomic layer of hydrogen, which makes its electron affinity negative. Second, that lets electrons move from the diamond valence band to suitable acceptor species at the surface, often adsorbates or engineered surface layers. Third, the transfer leaves behind a high-density two-dimensional hole gas (2DHG), confined within a few nanometers of the surface. This gives p-type conductivity while, according to DF, preserving the intrinsic purity, thermal conductivity and breakdown strength of the diamond bulk.
DF argues that because no ion implantation or lattice substitution is needed, the method avoids the defect generation and carrier scattering associated with conventional doping. It calls this particularly attractive for high-power, high-frequency and radiation-hard electronic device platforms.
The second performance claim is that "DF devices now match state-of-the-art 3 nm silicon in subthreshold swing." No numeric swing value is given, only the comparison to 3 nm silicon. The page also points to a DF paper in IEEE titled "On the Conduction Mechanism in Diamond 2-Dimensional Hole Gas Field Effect Transistors", with no authors or other publication details stated.
The page also has a section on R&D partnerships. DF says it supplies diamond substrates to R&D groups that are building their own diamond transistors. A news list at the bottom of the page includes the headlines "DF adds $2.77B fab in Spain" and "WSJ: Diamond Key to Faster Microchips"; the page gives no detail on either beyond the headline.
Key facts
- DF says it is developing all-diamond MOSFETs in partnership with Stanford University and headlines the page with "50x faster switching", with no baseline or conditions given.
- The devices use surface transfer doping: a hydrogen-terminated single-crystal diamond surface forms a two-dimensional hole gas within a few nanometers of the surface, with no substitutional dopant atoms in the bulk crystal.
- DF claims its devices now match state-of-the-art 3 nm silicon in subthreshold swing, but states no numeric swing value.
- The page references a DF paper in IEEE on the conduction mechanism in diamond 2D hole gas field effect transistors.
- DF supplies diamond substrates to R&D groups building their own diamond transistors.
Why it matters
Diamond is described on the page as a semiconductor with intrinsic purity, high thermal conductivity and high breakdown strength, which is why a working all-diamond transistor is of interest for high-power, high-frequency and radiation-hard electronics. The notable technical step claimed here is doping without putting dopant atoms into the crystal: surface transfer doping creates a hole gas at the surface instead. If the claims hold, DF says its devices already match the subthreshold swing of state-of-the-art 3 nm silicon, a benchmark from mainstream chipmaking.
Who it affects
The page speaks most directly to research groups building diamond transistors, to whom DF says it supplies diamond substrates. It also points to the high-power, high-frequency and radiation-hard device platforms as the target uses. The page names Stanford University as the development partner.
How to use it
This is a product and technology page, not a release a reader can adopt. The practical route it offers is for R&D groups: DF supplies diamond substrates to groups developing diamond transistors. No commercial availability, production volume, pricing or timeline for diamond transistors is given.
How solid is it
Thin. Everything comes from the company's own marketing page. The 50x switching claim has no baseline, no device comparison and no measurement conditions. The subthreshold swing claim has no number. The page references a DF paper in IEEE on the conduction mechanism, but gives no authors, and no independent verification or third-party measurements are cited. The physics description of surface transfer doping is stated by DF as fact, and the performance figures should be read as vendor claims.
Risks and caveats
Without a stated baseline, "50x faster switching" cannot be compared to any silicon or other device. Matching 3 nm silicon in subthreshold swing is a single metric and says nothing on its own about drive current, scaling or manufacturability. The page does not say when, or whether, diamond transistors will be commercially available. The news list mentions a $2.77B fab in Spain as a headline only, with no location, timeline or purpose given, so it should not be read as a manufacturing plan for transistors.
“DF devices now match state-of-the-art 3 nm silicon in subthreshold swing.”
— DF, Diamond Transistor product page (df.com)