EUV high NA anamorphic imagingeuvlsymposium.lbl.gov/pdf/2016/Oral/Wed_S2-3.pdf2016 International...

Preview:

Citation preview

2016 International Symposium on Extreme Ultraviolet Lithography, Hiroshima, Japan

EUV high NA anamorphic imaging

It’s all about the angles Oct-26-2016 Eelco van Setten, Gerardo Bottiglieri, Thorsten Last, Alberto Colina, Jan Lubkoll, Gijsbert Rispens, Jan van Schoot

2016 International Symposium on Extreme Ultraviolet Lithography, Hiroshima, Japan

2016 International Symposium on Extreme Ultraviolet Lithography, Hiroshima, Japan

EUV: it’s all about the angles High-NA comes with large angles

ML reflection MoSi Multilayer

Slide 2

Public

NA=0.55

2016 International Symposium on Extreme Ultraviolet Lithography, Hiroshima, Japan

0%

10%

20%

30%

40%

50%

60%

70%

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18

Mu

ltila

yer

Ref

lect

ivit

y [%

]

Angle of incidence on the mask [deg]

Multilayer

Slide 3

Anamorphic magnification solves the angular spread at the mask Public

Y 0.33NA – Mag 4x

X

Multilayer Reflectivity

2016 International Symposium on Extreme Ultraviolet Lithography, Hiroshima, Japan

0%

10%

20%

30%

40%

50%

60%

70%

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18

Mu

ltila

yer

Ref

lect

ivit

y [%

]

Angle of incidence on the mask [deg]

Multilayer

Slide 4

Anamorphic magnification solves the angular spread at the mask Public

Y 0.33NA – Mag 4x

X

Y - 4x

0.55NA – Mag 4x

X

Multilayer Reflectivity

2016 International Symposium on Extreme Ultraviolet Lithography, Hiroshima, Japan

0%

10%

20%

30%

40%

50%

60%

70%

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18

Mu

ltila

yer

Ref

lect

ivit

y [%

]

Angle of incidence on the mask [deg]

Multilayer

Slide 5

Anamorphic magnification solves the angular spread at the mask Public

Y 0.33NA – Mag 4x

X

Y - 4x

0.55NA – Mag 4x

X

Y - 8x

0.55NA – Mag 4x/8x

Multilayer Reflectivity

2016 International Symposium on Extreme Ultraviolet Lithography, Hiroshima, Japan

0%

10%

20%

30%

40%

50%

60%

70%

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18

Mu

ltila

yer

Ref

lect

ivit

y [%

]Angle of incidence on the mask [deg]

Multilayer

Slide 6

Anamorphic magnification: Circular exit pupil requires elliptical entrance pupil

Anamorphic

MAG 4x/8x

x

y

Public

Entrance pupil

2016 International Symposium on Extreme Ultraviolet Lithography, Hiroshima, Japan

Slide 7

Anamorphic magnification: Circular exit pupil requires elliptical entrance pupil Public

POB

MAG 4x/8x At reticle At wafer

2016 International Symposium on Extreme Ultraviolet Lithography, Hiroshima, Japan

Slide 8

Public

Contents

Mask

Central obscuration

Contrast

Summary & conclusions

2016 International Symposium on Extreme Ultraviolet Lithography, Hiroshima, Japan

Main consequences: Half field imaging, mask pattern is

stretched in scanning direction

x

y

MAG 4x in x

MAG 8x in y

Note: rectangular slit shown for illustration purposes

Slide 9

Public

Anamorphic masks:

• Image field half of current size

• Mask pattern stretched

Feature distortions on mask

• 4x-direction will drive mask

requirements (CD, registration

and defectivity)

2016 International Symposium on Extreme Ultraviolet Lithography, Hiroshima, Japan

Slide 10

• Traditional Mask Error Factor (MEF) is normalized with the magnification:

𝑀𝐸𝐹 =∆𝐶𝐷𝑤𝑎𝑓𝑒𝑟

∆𝐶𝐷𝑚𝑎𝑠𝑘𝑚𝑎𝑔

• Not convenient for anamorphic Same MEF corresponds to different mask

CD tolerance in X and Y

• A new Mask Error Factor, MEF*, is defined:

𝑀𝐸𝐹∗ =∆𝐶𝐷𝑤𝑎𝑓𝑒𝑟

∆𝐶𝐷𝑚𝑎𝑠𝑘

1nm mask CD error gives ‘MEF*’ nm wafer CD error

New MEF definition to handle difference in X,Y-

magnification Public

2016 International Symposium on Extreme Ultraviolet Lithography, Hiroshima, Japan

@ mask @ wafer

1nm

mef* xx

mef* yx

d

2 d

d/4

d/4

Slide 11

2D features experience strong X-Y interaction for mask

CD errors Public

2016 International Symposium on Extreme Ultraviolet Lithography, Hiroshima, Japan

@ mask @ wafer

1nm

mef* xx

mef* yx

d

2 d

d/4

d/4

Slide 12

2D features experience strong X-Y interaction for mask

CD errors Public

@ mask @ wafer

mef* xy

mef* yy

1nm

2016 International Symposium on Extreme Ultraviolet Lithography, Hiroshima, Japan

@ mask @ wafer

1nm

mef* xx

mef* yx

d

2 d

d/4

d/4

Slide 13

2D features experience strong X-Y interaction for mask

CD errors Public

@ mask @ wafer

mef* xy

mef* yy

1nm

@ mask @ wafer

MEF* X

( mef* xx + mef* xy )

MEF* Y

( mef* yy + mef* yx )

1nm

1nm

2016 International Symposium on Extreme Ultraviolet Lithography, Hiroshima, Japan

Slide 14

contributors

@ mask @ wafer

1nm

1nm

Mask error in X-direction dominates error on wafer in

both directions

MEF* Y

( mef* yy + mef* yx )

MEF* X

( mef* xx + mef* xy )

Public

Mask error in X-direction

2016 International Symposium on Extreme Ultraviolet Lithography, Hiroshima, Japan

Slide 15

Public

Contents

Mask

Central obscuration

Contrast

Summary & conclusions

2016 International Symposium on Extreme Ultraviolet Lithography, Hiroshima, Japan

Standard EUV coatings cannot

handle these large angles

Substrate

Slide 16

Public

Multi-layers set limits to angles & angular spread Angles must be reduced for high-NA optics

We have to limit the

angles on the mirror

we need another trick!

2016 International Symposium on Extreme Ultraviolet Lithography, Hiroshima, Japan

Angles and

angular spread

decrease

ob

scu

red

Slide 17

Public

There is a solution: We drill a hole into the mirror. Smaller angles enable transmission gain vs non-obscured NA 0.33

un

ob

scu

red

And even better:

The smaller angular

range increases the

transmission

2016 International Symposium on Extreme Ultraviolet Lithography, Hiroshima, Japan

Slide 18

Confidential

Obscuration may result in:

- application dependent

contrast loss

- Non-telecentricity

Obscuration radius limited to

20% of the pupil radius

( = 4% of the pupil area )

Obscuration comes at expense of blocking parts of

diffraction orders

2016 International Symposium on Extreme Ultraviolet Lithography, Hiroshima, Japan

0

5

10

15

20

25

30

8 12 16 20 24 28 32 36 40 44 48

EL %

Half Pitch [nm]

with_obscurationwithout_obscuration

Slide 19

Public

Contrast loss for semi-dense pitches using Dipole

illumination large part diffraction order blocked

• Worst point has acceptable Exposure Latitude

• Mitigation by means of SMO and SRAFs (for 16nm HP and up)

Horizontal L/S through pitch

DipoleY (20% pupil fill ratio)

2016 International Symposium on Extreme Ultraviolet Lithography, Hiroshima, Japan

0

5

10

15

20

25

30

8 12 16 20 24 28 32 36 40 44 48

EL %

Half Pitch [nm]

with_obscurationwithout_obscuration

Slide 20

Public

Contrast loss for semi-dense pitches using Dipole

illumination large part diffraction order blocked

• Worst point has acceptable Exposure Latitude

• Mitigation by means of SMO and SRAFs (for 16nm HP and up)

• Contrast unobscured pitches very similar to 0.33NA for identical k1 values

0

5

10

15

20

25

30

0.25 0.5 0.75 1 1.25 1.5 1.75 2

EL %

k1 value [-]

with_obscurationwithout_obscuration0.33NA

Horizontal L/S through pitch

DipoleY (20% pupil fill ratio)

2016 International Symposium on Extreme Ultraviolet Lithography, Hiroshima, Japan

0

5

10

15

20

25

30

20 30 40 50 60 70 80 90 100

EL %

Pitch [nm]

with_obscuration

without_obscuration

Slide 21

Public

Negligible contrast loss for illumination shapes of which small

part diffraction orders is blocked

12nm horizontal spaces through pitch

Small annular (20% pupil fill ratio)

2016 International Symposium on Extreme Ultraviolet Lithography, Hiroshima, Japan

0

5

10

15

20

25

30

20 30 40 50 60 70 80 90 100

EL %

Pitch [nm]

with_obscuration

without_obscuration

Slide 22

Public

Negligible contrast loss for illumination shapes of which small

part diffraction orders is blocked

• Contrast very similar to 0.33NA for identical k1 values

0

5

10

15

20

25

30

0.25 0.5 0.75 1 1.25 1.5 1.75 2

EL %

k1 value [-]

with_obscuration

without_obscuration

0.33NA

12nm horizontal spaces through pitch

Small annular (20% pupil fill ratio)

2016 International Symposium on Extreme Ultraviolet Lithography, Hiroshima, Japan

0

5

10

15

20

25

30

35

10 12 14 16 18 20 22 24 26 28 30

EL [

%]

Half Pitch [nm]

with_obscurationwithout_obscuration

0

5

10

15

20

25

30

35

10 12 14 16 18 20 22 24 26 28 30

EL [

%]

Half Pitch [nm]

with_obscurationwithout_obscuration

Slide 23

Public

High contrast through pitch maintained for 2D features

Dense contact holes through pitch

Small annular (20% pupil fill ratio)

Dense contact holes through pitch

Quasar (20% pupil fill ratio)

2016 International Symposium on Extreme Ultraviolet Lithography, Hiroshima, Japan

Slide 24

Public

Obscuration has limited impact on non-telecentricity for 1D

features driven by Mask3D effects

• Compensation by SMO (asymmetric illumination) or mask stack optimization

-25

-20

-15

-10

-5

0

5

8 12 16 20 24 28 32 36 40 44 48

Tele

cen

tric

ity

[mra

d]

Half Pitch [nm]

with_obscurationwithout_obscuration

Horizontal L/S through pitch

DipoleY (20% PFR)

-25

-20

-15

-10

-5

0

5

20 30 40 50 60 70 80 90 100

Tele

cen

tric

ity

[mra

d]

Pitch [nm]

with_obscuration

without_obscuration

12nm horizontal spaces through pitch

Small annular (20% PFR)

2016 International Symposium on Extreme Ultraviolet Lithography, Hiroshima, Japan

Slide 25

Public

Obscuration has limited impact on non-telecentricity for 1D

features driven by Mask3D effects

• Compensation by SMO (asymmetric illumination) or mask stack optimization

-25

-20

-15

-10

-5

0

5

8 12 16 20 24 28 32 36 40 44 48

Tele

cen

tric

ity

[mra

d]

Half Pitch [nm]

with_obscurationwithout_obscuration

-25

-20

-15

-10

-5

0

5

0.25 0.5 0.75 1 1.25 1.5 1.75 2

Tele

cen

tric

ity

[mra

d]

k1 value[-]

with_obscuration

without_obscuration

0.33NA

Horizontal L/S through pitch

DipoleY (20% PFR)

-25

-20

-15

-10

-5

0

5

20 30 40 50 60 70 80 90 100

Tele

cen

tric

ity

[mra

d]

Pitch [nm]

with_obscuration

without_obscuration

-25

-20

-15

-10

-5

0

5

0.25 0.5 0.75 1 1.25 1.5 1.75 2

Tele

cen

tric

ity

[mra

d]

k1 value [-]

with_obscuration

without_obscuration

0.33NA

12nm horizontal spaces through pitch

Small annular (20% PFR)

2016 International Symposium on Extreme Ultraviolet Lithography, Hiroshima, Japan

Slide 26

Public

Obscuration has limited impact on non-telecentricity for 2D

features driven by Mask3D effects

• Small compensation effect observed for semi-dense pitches

-20

-15

-10

-5

0

10 12 14 16 18 20 22 24 26 28 30Tele

cen

tric

ity-

Y [

mra

d]

Half Pitch [nm]

with_obscurationwithout_obscuration

Dense contact holes through pitch

Quasar (20% pupil fill ratio)

-20

-15

-10

-5

0

10 12 14 16 18 20 22 24 26 28 30Tele

cen

tric

ity-

Y [

mra

d]

Half Pitch [nm]

with_obscurationwithout_obscuration

Dense contact holes through pitch

Small annular (20% pupil fill ratio)

Anamorphic Placement aware SMO mitigates non-

telecentricity Confidential

Slide 27

Pattern shift through focus Flat-Mask SMO source

Placement aware SMO source

Multi-pitch L/S pattern

18nm min. pitch

S1

S2

S3

S4

S5

S6

S7

-0.5

-0.3

-0.1

0.1

0.3

0.5

-15 -5 5 15

Pat

tern

Sh

ift

[nm

]

Defocus [nm]

S1

S2

S3

S4

S5

S6

S7

-0.5

-0.3

-0.1

0.1

0.3

0.5

-15 -5 5 15

Pat

tern

Sh

ift

[nm

]

Defocus [nm]

S1

S2

S3

S4

S5

S6

S7

Anamorphic Placement aware SMO mitigates non-

telecentricity Confidential

Slide 28

• Max non-telecentricity reduces from 13 to 4mrad

Flat-Mask SMO source

Placement aware SMO source

Multi-pitch L/S pattern

18nm min. pitch

S1 S2 S3

S4 S5 S6

S7

Feature dependent non-telecentricity

-15

-10

-5

0

5

10

S1 S2 S3 S4 S5 S6 S7Te

lece

ntr

icit

y [m

rad

] Multi-pitch spaces

Flat mask SMO

M3D SMO

Anamorphic Placement aware SMO restores

overlapping Process window Confidential

Slide 29

Exposure Latitude vs DoF Flat-Mask SMO source

Placement aware SMO source

Multi-pitch L/S pattern

18nm min. pitch

S1

S2

S3

S4

S5

S6

S7

0

5

10

15

20

25

30

0 30 60 90 120

Exp

osu

re L

atit

ud

e [

%]

Defocus [nm]

S1

S2

S3

S4

S5

S6

S7

0

5

10

15

20

25

30

0 30 60 90 120

Exp

osu

re L

atit

ud

e [

%]

Defocus [nm]

S1

S2

S3

S4

S5

S6

S7

Anamorphic Placement aware SMO restores

overlapping Process window Confidential

Slide 30

• Overlapping PW @ 10% EL increases from 36 to 64nm

Flat-Mask SMO source

Placement aware SMO source

Multi-pitch L/S pattern

18nm min. pitch

S1 S2 S3

S4 S5 S6

S7

Overlapping Process window

0

5

10

15

0 20 40 60 80Ex

po

sure

Lat

itu

de

[%

]

Depth of Focus [nm]

Flat mask SMO

Placement aware SMO

2016 International Symposium on Extreme Ultraviolet Lithography, Hiroshima, Japan

Slide 31

Public

Contents

Mask

Central obscuration

Contrast

Summary & conclusions

2016 International Symposium on Extreme Ultraviolet Lithography, Hiroshima, Japan

Slide 32

Public

0.55NA gives high image contrast down to 8nm L/S

and 10nm contact holes

NILS vs resolution for 1D L/S NILS vs resolution for Contact holes

0

2

4

6

8

5 10 15 20 25

NIL

S [-

]

Half pitch [nm]

0.33NA

0.55NA

0

2

4

6

8

5 10 15 20 25 30

NIL

S [-

]

Half pitch [nm]

0.33NA

0.55NA

2016 International Symposium on Extreme Ultraviolet Lithography, Hiroshima, Japan

Slide 33

Public

Local CDU dominant contributor to EPE – Mitigation

via image contrast enhancement

2016 International Symposium on Extreme Ultraviolet Lithography, Hiroshima, Japan

Slide 34

Public

Local CDU dominant contributor to EPE – Mitigation

via image contrast enhancement

Jo Finders, SPIE 2016 and Fraunhofer IISB Litho simulation workshop 2016

2016 International Symposium on Extreme Ultraviolet Lithography, Hiroshima, Japan

0

1

2

3

4

5

6

10 15 20 25 30

req

uir

ed

NIL

S fo

r 1

5%

LC

DU

Half Pitch (nm)

17mJ 35mJ 70mJ

Slide 35

Confidential

High NA most powerful knob to control Local CDU and

dose requirements

Courtesy: Sander Wuister, Gijsbert Rispens

0

1

2

3

4

5

6

10 15 20 25 30

req

uir

ed

NIL

S fo

r 1

5%

LC

DU

Half Pitch (nm)

17mJ 35mJ

70mJ NILS 0.33NA

NILS 0.55NA0

10

20

30

40

50

60

70

80

1 2 3 4 5 6

Do

se [

mJ/

cm2

]

NILS [-]

Dose to meet 15% LCDU requirement

for 18nm dense holes

0.33NA

0.55NA

2016 International Symposium on Extreme Ultraviolet Lithography, Hiroshima, Japan

Slide 36

Public

Contents

Mask

Central obscuration

Contrast

Summary & conclusions

2016 International Symposium on Extreme Ultraviolet Lithography, Hiroshima, Japan

Slide 37

Confidential

Summary and conclusions

It’s all about the angles • Anamorphic imaging solves the angular spread at the mask

• M3D effects similar to 0.33NA

• Central obscuration enables high throughput with good imaging performance

• Anamorphic SMO does the rest

EUV High NA • Increased resolution enabling 8nm HP imaging

• Above all… superior contrast

• Key for reducing LCDU and EPE budget for maintaining shrink roadmap

2016 International Symposium on Extreme Ultraviolet Lithography, Hiroshima, Japan

Slide 38

Public

The authors would like to thank:

• Sander Wuister

• Jo Finders

• Stephen Hsu (Brion)

• Matthias Rösch

• Jens Timo Neumann

• Paul Graüpner

• Jörg Zimmerman

• Bernard Kneer

• Sacha Migura

2016 International Symposium on Extreme Ultraviolet Lithography, Hiroshima, Japan

Thank you for your

attention !

Recommended