Nanosheet with selective dipole diffusion into high-k
A method is presented for attaining different gate threshold voltages across a plurality of field effect transistor (FET) devices without patterning between nanosheet channels. The method includes forming a first set of nanosheet stacks having a first intersheet spacing, forming a second set of nanosheet stacks having a second intersheet spacing, where the first intersheet spacing is greater than the second intersheet spacing, depositing a high-k (HK) layer within the first and second nanosheet stacks, depositing a material stack that, when annealed, creates an amorphous HK layer in the first set of nanosheet stacks and a crystallized HK layer in the second nanosheet stacks, depositing a dipole material, and selectively diffusing the dipole material into the amorphous HK layer of the first set of nanosheet stacks to provide the different gate threshold voltages for the plurality of FET devices.
1. A method for attaining different gate threshold voltages across a plurality of field effect transistor (FET) devices without patterning between nanosheet channels, the method comprising:
forming a first set of nanosheet stacks having a first intersheet spacing;
forming a second set of nanosheet stacks having a second intersheet spacing, where the first intersheet spacing is greater than the second intersheet spacing;
depositing a high-k (HK) layer within the first and second nanosheet stacks;
depositing a material stack that, when annealed, creates an amorphous HK layer in the first set of nanosheet stacks and a crystallized HK layer in the second nanosheet stacks;
depositing a dipole material; and
selectively diffusing the dipole material into the amorphous HK layer of the first set of nanosheet stacks to provide the different gate threshold voltages for the plurality of FET devices.
2. The method of claim 1 , wherein the material stack includes a first titanium nitride (TiN) layer, a silicon oxide (SiO 2 ) layer, and a second TiN layer.
3. The method of claim 2 , wherein the material stack selectively dopes the HK layer with silicon (Si) to increase a crystallization temperature and create the amorphous HK layer in the first set of nanosheet stacks.
4. The method of claim 1 , wherein the dipole material is lanthanum oxide (LaO).
5. The method of claim 1 , wherein a sacrificial layer and a capping layer are deposited after deposition of the dipole material to enable diffusion of the dipole material into the amorphous HK layer of the first set of nanosheet stacks.
6. The method of claim 5 , wherein the sacrificial layer is TiN and the capping layer is amorphous silicon (a-Si).
7. The method of claim 1 , further comprising depositing a gate metal, inner spacers, and source/drain regions.
8. A method for modulating threshold voltages for nanosheet stacks without patterning between nanosheet channels, the method comprising:
forming first nanosheet stacks having a first intersheet spacing;
forming second nanosheet stacks having a second intersheet spacing, where the first intersheet spacing is greater than the second intersheet spacing;
constructing an amorphous high-k (HK) layer within the first nanosheet stacks;
constructing a crystallized high-k (HK) layer within the second nanosheet stacks;
depositing a dipole material; and
selectively diffusing the dipole material into the amorphous HK layer of the first nanosheet stacks to modulate the threshold voltages for the nanosheet stacks.
9. The method of claim 8 , further comprising depositing a material stack that, when annealed, creates the amorphous HK layer in the first nanosheet stacks and the crystallized HK layer in the second nanosheet stacks.
10. The method of claim 9 , wherein the material stack includes a first titanium nitride (TiN) layer, a silicon oxide (SiO 2 ) layer, and a second TiN layer.
11. The method of claim 10 , wherein the material stack selectively dopes the HK layer with silicon (Si) to increase a crystallization temperature and create the amorphous HK layer in the first nanosheet stacks.
12. The method of claim 8 , wherein the dipole material is lanthanum oxide (LaO).
13. The method of claim 8 , wherein a sacrificial layer and a capping layer are deposited after deposition of the dipole material to enable diffusion of the dipole material into the amorphous HK layer of the first nanosheet stacks.
14. The method of claim 13 , wherein the sacrificial layer is TiN and the capping layer is amorphous silicon (a-Si).
15. The method of claim 8 , further comprising depositing a gate metal, inner spacers, and source/drain regions.
16. A semiconductor structure for modulating threshold voltages for nanosheet stacks without patterning between nanosheet channels, the semiconductor structure comprising:
first nanosheet stacks having a first intersheet spacing;
second nanosheet stacks having a second intersheet spacing, where the first intersheet spacing is greater than the second intersheet spacing;
an amorphous high-k (HK) layer disposed within the first nanosheet stacks;
a crystallized high-k (HK) layer disposed within the second nanosheet stacks; and
a dipole material disposed within the first and second nanosheet stacks, wherein the dipole material is selectively diffused into the amorphous HK layer of the first nanosheet stacks to modulate the threshold voltages for the nanosheet stacks.
17. The semiconductor structure of claim 16 , wherein the dipole material is lanthanum oxide (LaO).
18. The semiconductor structure of claim 16 , wherein the amorphous HK layer includes a dopant to increase a crystallization temperature to maintain the amorphous HK layer in its current state.
19. The semiconductor structure of claim 16 , wherein a gate metal directly contacts the amorphous high-k (HK) layer within the first nanosheet stacks and the crystallized high-k (HK) layer in the second nanosheet stacks.
20. The semiconductor structure of claim 16 , wherein the amorphous high-k (HK) layer within the first nanosheet stacks and the crystallized high-k (HK) layer in the second nanosheet stacks directly contact inner spacers formed between silicon (Si) layers of the first and second nanosheet stacks.