Extension of 193 Immersion Lithography - Sokudo
Extension of 193 Immersion Lithography - Sokudo
Extension of 193 Immersion Lithography - Sokudo
You also want an ePaper? Increase the reach of your titles
YUMPU automatically turns print PDFs into web optimized ePapers that Google loves.
NIKON PRECISION INC.<br />
<strong>Extension</strong> <strong>of</strong> <strong>193</strong> <strong>Immersion</strong><br />
<strong>Lithography</strong><br />
Steve Renwick<br />
Senior Principal Engineer, NPI
Overview<br />
• EUV Status<br />
• Bridging to EUV<br />
• Scanner Requirements for DP<br />
• NSR-S620D Performance<br />
SOKUDO Breakfast Forum July 14, 2010 2
<strong>Lithography</strong> Technology Roadmap<br />
65<br />
45 nm HP<br />
45<br />
32<br />
22<br />
16<br />
32 nm HP<br />
Water Water <strong>Immersion</strong> <strong>Immersion</strong><br />
S609B<br />
NA 1.07<br />
S610C<br />
NA 1.30<br />
22 nm HP<br />
16 nm HP<br />
Double Double Patterning Patterning<br />
S620D<br />
NA 1.35<br />
ITRS DRAM ½ Pitch (nm)<br />
EUVL EUVL<br />
11<br />
2007 2010 2013 2016 2019 2022<br />
SOKUDO Breakfast Forum July 14, 2010 3
EUV Infrastructure Schedule<br />
Mask infrastructure<br />
development begins<br />
Mask tool availability:<br />
Chan, et al., SPIE 2010, February, 2010<br />
ITRS<br />
DRAM<br />
½ Pitch (nm)<br />
65<br />
45<br />
Aggressive<br />
DRAM<br />
32<br />
22<br />
Aggressive<br />
Logic<br />
Aggressive<br />
NAND<br />
1-2 years required for defect<br />
learning (immersion experience)<br />
HVM Process Development can<br />
begin here.<br />
16<br />
Mask tools available,<br />
defect learning begins<br />
11<br />
2007 2010 2013 2016 2019 2022<br />
Delays in infrastructure continue to push out adoption <strong>of</strong> EUV<br />
increased cost and longer ROI for equipment makers<br />
SOKUDO Breakfast Forum July 14, 2010 4
<strong>Extension</strong> <strong>of</strong> <strong>193</strong>i Before EUV Arrives<br />
• Extend single patterning:<br />
– Source mask optimization and custom illumination<br />
• Enable 32 nm half pitch and beyond<br />
– Spacer double patterning<br />
– Pitch splitting double patterning (LELE, LFLE, etc.)<br />
– Line cutting lithography<br />
Tool makers need to support these schemes<br />
SOKUDO Breakfast Forum July 14, 2010 5
Line Cutting Litho Concept<br />
44 nm HP<br />
Patterning<br />
Spacer<br />
Pitch Doubling<br />
Target<br />
22 nm SRAM<br />
Gate Cell<br />
Cut Hole<br />
Patterning<br />
<br />
Hole Chemical<br />
Shrink<br />
After Etch<br />
<br />
<br />
SOKUDO Breakfast Forum July 14, 2010 6
Line Cutting <strong>Lithography</strong> Concept<br />
+<br />
<br />
1 st patterning<br />
(LS formation)<br />
2 nd patterning<br />
(Cut)<br />
Result<br />
C. Bencher, et al., SPIE 72740G (2009)<br />
Down to 19 nm half pitch can be achieved<br />
SOKUDO Breakfast Forum July 14, 2010 7
S620D – Enabling the Next Generation<br />
• Enabling Superior Yield:<br />
– Overlay accuracy to enable DP<br />
– CD uniformity<br />
• Enabling Affordable <strong>Lithography</strong>:<br />
– Reduced wafer overhead time<br />
– Maximum throughput<br />
– Low CoO via multi-generational use <strong>of</strong> the tool<br />
• Enabling Rapid Production Ramps:<br />
– Faster installation<br />
– Optimal uptime<br />
– Platform to enable reuse<br />
SOKUDO Breakfast Forum July 14, 2010 8
The NSR-S620D Challenge<br />
NA 1.35<br />
Projection<br />
Lens<br />
Streamlign Platform<br />
for 2 nm<br />
self overlay<br />
Bird's Eye<br />
Control<br />
for 200 wph<br />
throughput<br />
Stream<br />
Alignment<br />
for 20 day<br />
installation<br />
Modular 2<br />
Structure<br />
SOKUDO Breakfast Forum July 14, 2010 9
Bird’s Eye Control<br />
Enabling Superior Yield<br />
Bird’s Eye Control<br />
• Hybrid system uses laser<br />
encoders w/interferometers<br />
• Dramatically improves<br />
accuracy and stability<br />
• Targeting 2 nm overlay<br />
capabilities<br />
• Superior focus control<br />
Enabling Affordable <strong>Lithography</strong><br />
Enabling Rapid Production Ramps<br />
Stream Alignment<br />
Modular 2 Structure<br />
SOKUDO Breakfast Forum July 14, 2010 10
Stream Alignment<br />
Enabling Affordable <strong>Lithography</strong><br />
Stream Alignment<br />
• Five-Eye FIA<br />
• Straight Line Aut<strong>of</strong>ocus<br />
• Greatly reduced wafer<br />
overhead time<br />
• Targeting throughput up to<br />
200 wph<br />
Enabling Superior Yield<br />
Enabling Rapid Production Ramps<br />
Bird’s Eye Control<br />
Modular 2 Structure<br />
SOKUDO Breakfast Forum July 14, 2010 11
Modular 2 Structure<br />
Enabling Rapid Production Ramps<br />
• Faster installation - 20 day<br />
target<br />
• Simplified maintenance<br />
• Optimal uptime<br />
• Extendible platform to<br />
enable reuse<br />
Modular 2 Structure<br />
Enabling Affordable <strong>Lithography</strong><br />
Enabling Superior Yield<br />
Stream Alignment<br />
Bird’s Eye Control<br />
SOKUDO Breakfast Forum July 14, 2010 12
Scanner Requirements for DP<br />
32 nm hp DP Budget<br />
Budget<br />
Spec<br />
Line<br />
CDU<br />
Space CDU<br />
L1 − L 2<br />
CDU (3σ<br />
)<br />
1.0 nm<br />
2.4<br />
2.9 nm<br />
OL<br />
m −<br />
1<br />
m<br />
2<br />
0.5<br />
3.3 nm<br />
OL<br />
(3σ<br />
)<br />
2.4<br />
CD control and overlay are critical for DP<br />
SOKUDO Breakfast Forum July 14, 2010 13
S620D Overlay Stability<br />
Overlay average within lot<br />
Total 3σ<br />
3σX<br />
1.94<br />
3σY<br />
1.80<br />
[nm]<br />
ave.[nm]<br />
5<br />
4<br />
3<br />
1 2<br />
-1 0<br />
-2<br />
-3<br />
-4<br />
-5<br />
1 3 5 7 9 11 13 15 17 19<br />
Ave.X<br />
Ave.Y<br />
• 20 wafers continuous exp.<br />
• Common linear terms<br />
removed<br />
3sigma[nm]<br />
5<br />
4<br />
3<br />
2<br />
1<br />
wafer count<br />
Overlay 3sigma within lot<br />
3sigma.X<br />
3sigma.Y<br />
0<br />
1 3 5 7 9 11 13 15 17 19<br />
wafer count<br />
S620D meets overlay requirements for 32 nm hp DP<br />
SOKUDO Breakfast Forum July 14, 2010 14
Focus Uniformity<br />
14.3 nm, including edge die<br />
3σ [nm]<br />
20<br />
10<br />
14.8 15.0 14.9<br />
0<br />
1 2 3<br />
wafer #<br />
-50 nm<br />
50 nm<br />
SOKUDO Breakfast Forum July 14, 2010 15
Budget vs. S620D Data<br />
32 nm hp DP Budget and Actual Data<br />
L1 − L 2<br />
∆CD<br />
m1 − m 2<br />
∆OL<br />
Budget Spec<br />
1.0 nm 1.1<br />
2.4 2.1<br />
0.5 0.7<br />
2.4 1.9<br />
Line<br />
CDU<br />
2.9 nm<br />
2.5<br />
Space CDU<br />
3.3 nm<br />
3.3<br />
S620D data meet the budget requirement<br />
SOKUDO Breakfast Forum July 14, 2010 16
S620D Overall Performance<br />
Line CDU (3σ)<br />
Space CDU (3σ)<br />
2.5 nm<br />
3.3 nm<br />
22 nm L/S<br />
Line 1<br />
Line 2<br />
Space 1<br />
Space 2<br />
S620D enables pitch splitting DP<br />
SOKUDO Breakfast Forum July 14, 2010 17
Extendible Platform for ArF <strong>Immersion</strong><br />
Modular 2 Structure allows multigenerational use<br />
SOKUDO Breakfast Forum July 14, 2010 18
Summary<br />
• The industry needs an interim solution for the<br />
32 nm and 22 nm nodes, prior to the HVM<br />
development <strong>of</strong> EUV<br />
• <strong>193</strong> immersion lithography will be extended by:<br />
– Source optimization and computational lithography<br />
– Double patterning<br />
– Multiple patterning and cutting lithography<br />
• This places severe new requirements on a scanner<br />
for overlay and CD uniformity<br />
• The Nikon S620D enables superior yield, affordable<br />
lithography, and rapid production ramps for 32 nm -<br />
with extendibility to 22 nm<br />
SOKUDO Breakfast Forum July 14, 2010 19