Fin patterns with varying spacing without fin cut
Methods of forming semiconductor fins include forming first spacers on a first sidewall of each of a plurality of mandrels using an angled deposition process. A second sidewall of one or more of the plurality of mandrels is masked. Second spacers are formed on a second sidewall of all unmasked mandrels. The second sidewall of the one or more of the plurality of mandrels is unmasked. The mandrels are etched away. Fins are formed from a substrate using the first and second spacers as a mask.
1. A method of forming semiconductor fins, comprising:
forming first spacers on a first sidewall of each of a plurality of mandrels using a first oxygen gas cluster ion beam that is angled between about 10° and about 80°, the plurality of mandrels being amorphous silicon;
masking a finless region by forming a mask directly on a second sidewall of one or more of the plurality of mandrels;
isotropically forming second spacers on a second sidewall of all unmasked mandrels from the plurality of mandrels using a second oxygen gas cluster ion beam;
unmasking the finless region;
anisotropically etching spacer material from horizontal surfaces of each of the plurality of mandrels after unmasking the finless region;
etching away each of the plurality of mandrels; and
forming said semiconductor fins from a substrate, comprising a nitride pad layer directly on a semiconductor layer, using the first and second spacers as a pattern mask, such that no semiconductor fins are formed in the finless region.
2. The method of claim 1 , wherein masking the finless region further comprises forming the mask to directly contact a first spacer on an adjacent mandrel.
3. The method of claim 1 , wherein no semiconductor fin is removed after formation of the semiconductor fins.
4. The method of claim 1 , wherein the nitride pad layer includes silicon nitride.
5. The method of claim 1 , wherein the second oxygen gas cluster ion beam includes an angled deposition process.
6. The method of claim 1 , wherein unmasking the finless region includes unmasking the second sidewall of the one or more of the plurality of mandrels.
7. The method of claim 1 , further comprising patterning a mandrel material using a hard mask to form the plurality of mandrels prior to forming the first spacers.
8. The method of claim 1 , wherein the first spacers on the first sidewall include silicon dioxide.
9. The method of claim 1 , wherein masking the finless region includes forming the mask in regions where the semiconductor fins are not needed.
10. The method of claim 1 , wherein forming the second spacers encapsulates all the unmasked mandrels in spacer material from the first and second spacers.
11. The method of claim 1 , wherein unmasking the finless region exposes the second sidewall on all masked mandrels from the plurality of mandrels.
12. A method of forming semiconductor fins, comprising:
forming first spacers on a first sidewall of each of a plurality of mandrels using a first oxygen gas cluster ion beam that is angled between about 10° and about 80°, the plurality of mandrels being amorphous silicon;
masking a finless region by forming a mask directly on a second sidewall of one or more of the plurality of mandrels;
isotropically forming second spacers on a second sidewall of all unmasked mandrels from the plurality of mandrels using a second oxygen gas cluster ion beam;
unmasking the finless region to expose the second sidewall on all masked mandrels from the plurality of mandrels;
anisotropically etching spacer material from horizontal surfaces of each of the plurality of mandrels after unmasking the finless region;
etching away each of the plurality of mandrels; and
forming said semiconductor fins from a substrate, comprising a silicon nitride pad layer directly on a semiconductor layer, using the first and second spacers as a pattern mask, such that no semiconductor fins are formed in the finless region.
13. The method of claim 12 , wherein masking the finless region further comprises forming the mask to directly contact a first spacer on an adjacent mandrel.
14. The method of claim 12 , wherein no semiconductor fin is removed after formation of the semiconductor fins.
15. The method of claim 12 , wherein the second oxygen gas cluster ion beam includes an angled deposition process.
16. The method of claim 12 , wherein unmasking the finless region includes unmasking the second sidewall of the one or more of the plurality of mandrels.
17. The method of claim 12 , further comprising patterning a mandrel material using a hard mask to form the plurality of mandrels prior to forming the first spacers.
18. The method of claim 12 , wherein the first spacers on the first sidewall include silicon dioxide.
19. The method of claim 12 , wherein masking the finless region includes forming the mask in regions where the semiconductor fins are not needed.
20. The method of claim 12 , wherein forming the second spacers encapsulates all the unmasked mandrels in spacer material from the first and second spacers.