THE THICKNESS LIMIT

A thinner channel is a better channel

Control A thinner channel puts every carrier close to the gate, so a shorter gate still works

The catch Silicon loses mobility below about 5 nm as carriers scatter off its rough surfaces

A way out Layered materials form crystals with no dangling bonds at their surface, even one atom thick

Candidates Graphene (no gap), MoS₂ and WSe₂ (gaps of 1 to 2 eV), and rolled-up carbon nanotubes

The best channel would be one atom thick and still fast. Silicon cannot be that.

THIN CHANNELS

Short gate, high mobility

L_min ≈ 6λ, λ = √(ε_ch/ε_ox · t_ch · t_ox)

The natural length shrinks with the square root of the channel thickness, so a one-atom layer can in principle be gated at a few nanometers. Silicon's mobility falls off a cliff below 5 nm; a layered semiconductor does not suffer the same surface scattering and keeps its mobility down to a single layer of atoms.

That combination is what 2D semiconductors offer and silicon cannot.

CARBON NANOTUBES

Roll a sheet, choose a gap

E_g ≈ 0.84 ÷ d eV

A carbon nanotube is graphene rolled into a cylinder 1 to 2 nm across. How it is rolled, the chirality (n, m), decides everything: one third of the possibilities are metallic, and the rest are semiconductors whose band gap is inversely proportional to the diameter. A tube about 1.4 nm across has a gap near 0.6 eV, enough to switch off well.

A transistor needs a semiconducting tube. Growth must deliver the right chirality, almost every time.

THE HARD PARTS

Why they are not in your phone yet

Contacts Low-resistance metal contacts to a one-atom layer are very hard to make, and contact resistance dominates small devices

Growth Wafer-scale, single-crystal, defect-free films, and in the case of nanotubes, nearly 100% semiconducting tubes placed where needed

Dielectrics A gate insulator will not stick to a surface with no dangling bonds, so special seeding layers are needed

Progress A working nanotube computer was shown in 2013, and a 16-bit RISC-V processor with about 14,000 nanotube transistors in 2019

The physics works. The manufacturing is the open problem.

CHECK THE CANDIDATES

Graphene, nanotube or MoS₂?

Graphene conducts, nanotubes depend on chirality, and MoS₂ is a true 2D semiconductor.

UNIT 24 STUDY COMPLETE

🧪

Ready for the Fab Challenge?

You've followed the thin-channel argument to its end, and met the materials that could carry it past silicon.