ASML NXE:3100 PRE-PRODUCTION EUV SCANNER …euvlsymposium.lbl.gov/pdf/2013/pres/S1-2_EHendrickx.pdfimec 2013 asml nxe:3100 pre-production euv scanner performance at imec eric hendrickx,

IMEC 2013

ASML NXE:3100 PRE-PRODUCTION EUV SCANNER PERFORMANCE AT IMEC

ERIC HENDRICKX, JAN HERMANS, GIAN LORUSSO, PHILIPPE FOUBERT, MIEKE GOETHALS, RIK JONCKHEERE, GEERT VANDENBERGHE, KURT RONSE

IMEC 2013

CONTENTS

Introduction

NXE:3100 stability

Outlook to 3300

Conclusions

2013 EUVL SYMPOSIUM - HENDRICKX 2

IMEC 2013

IMEC EUV LITHOGRAPHY TOOL ROADMAP

2006 - 2011 2011 - now 2014

ASML Alpha-Demo tool

40nm � 27nm LS0.25 NA

ASML NXE:3100 –pre production

27nm LS, 22nm LS0.25 NA

ASML NXE:3300 –production

22, 18nm LS0.33 NA


IMEC 2013

NXE:3100

Main specifications

▸ Field size: 26x33mm2

▸ NA=0.25 and σ=0.81

▸ 6 off-axis illumination conditions available

▸ MMO vs NXT:1950i < 7nm

Interfaced to TEL LITHIUS™ Pro - EUVDischarge Produced Plasma source


SUSS MicroTecMaskTrack Pro

EUV TechnologiesOutgassing tool

IMEC 2013

CONTENTS

Introduction

NXE:3100 stability

Monitoring scheme

Productivity

CD control

Overlay control

Out of band radiation

Outlook to 3300

Conclusions


IMEC 2013

MONITORING PROCEDURE ON NXE:3100


FEM

CD

U/o

verl

ay

= NXE:3100

Test\logging Target Parameters

Power Productivity Power at IF, RS, WS

FEM wafer CD control 27nm LS BE, BF, EL, DOF

Uniform wafer CD control 27nm LS CDU

Overlay control

Grid, IF residuals

Ilias Lens stability Lensaberration terms

Al mask exposure

Out of band radiationstability

% OoB

IMEC 2013

NXE:3100 PRODUCTIVITYCUMULATIVE WAFERCOUNT

▸ 24/7 operation

▸ DPP source 2013 average power at IF 4.9W

▸ Average power at waferstage 310 microWatt/mm

▸ Average 2013 throughput 2-3 full wafers per hour

▸ Average system uptime 2013 ~52%

Cumulative wafercount of exposed wafers now exceeds 6000 wafers on NXE:3100

2013 EUVL SYMPOSIUM - HENDRICKX

0

1000

2000

3000

4000

5000

6000

7000

41

2011

49 5

2012

13 21 29 37 45 1

2013

9 17 25 33

Tota

l w

afe

r co

un

t

7

IMEC 2013

NXE:3100 PRODUCTIVITYCOLLECTOR LIFETIME AND IMPACT ON POWER

▸ NXE:3100 DPP source was operated using the same swap flange (collector mirror + debris mitigation) for 9 months

▸ Due to low power on system, it was then decided to replace the swap flange (collector + debris mitigation system)

▸ Both power at IF and power at waferstage were recovered, improving productivity

▸ Post-mortem confirmed collector erosion and Snaccumulation in debris mitigation system


0.0

1.0

2.0

3.0

4.0

5.0

6.0

7.0

0

50

100

150

200

250

300

350

400

450

500

Mar-13 Apr-13

IF power (W

)

Slit Integrated power

(microWatt/m

m) WS power

IF power

DPP collector mirror lifetime exceeded 6 months under normal operation conditions

Power at IF and at waferstageReplacement of swap flange

IMEC 2013

NXE:3100 PRODUCTIVITYLENS TRANSMISSION

▸ Lens transmission can be estimated by power measurement at waferstage and at reticle stage

▸ After 1 year of use, cycles of lens cleaning were started using on-board cleaning and lens transmission improved by ~16%


Cleaning cycle

WS/

RS

po

wer

rat

io

Lens transmission only showed limited impact of one year system use First demonstration of scanner on-board lens cleaning in the field

96%

100%

104%

108%

112%

116%

120%

IMEC 2013

CD CONTROL OVER 21 MONTHSBEST ENERGY 27NM H AND V LS FROM 12/2011 – 08/2013


10.0

11.0

12.0

13.0

14.0

15.0

16.0

17.0

0 20 40 60 80 100 120

Best energy (mJ/cm2)

Wafer Number

H

V

Best dose of 27nm H and V LS showed a gradual drift over the first 10 months of operation – was solved by more frequent calibration of NXE:3100 WS sensors to reference

IMEC 2013

CD CONTROL27NM V LS CD STABILITY OVER 10 MONTHS


07/09/2013

10/08/2013

08/07/2013

0

1

2

3

4

5

6

7

8

9

10

20

21

22

23

24

25

26

27

28

29

30

1 11 21 31 41 51 61

CD 3s (nm)

Average CD (nm)

CD

CD 3s

CD3s - IF

After repair of sensor drift, average CD, intrafield signature,and across wafer CD signature are stable over time

IMEC 2013 2013 EUVL SYMPOSIUM - HENDRICKX

OVERLAY CONTROLINTRAFIELD RESIDUALS FROM DECEMBER 2011 –AUGUST 2013

12

November 2011 September 2013

Intrafield distortion signature is stable over 22 months

First layer: XT:1450Second layer: NXE:3100After removal of 10 par OVL correction model

IMEC 2013

1 11 21 31 41 51 61 71 81 91 101 111 121 131

IF R

esi

du

als

C1

(n

m)

Wafer Count

x

In 22 months, only 5 events when a particle was present on the reticle clamp during monitor. In the last 4 events, the particle could be removed without breaking vacuum


y

OVERLAY CONTROLINTRAFIELD RESIDUALS FROM DECEMBER 2011 –AUGUST 2013


Rik Jonckheere, session 2

IMEC 2013 2013 EUVL SYMPOSIUM - HENDRICKX 14

OVERLAY CONTROLGRID RESIDUALS FROM 12/2011 – 09/2013

5

6

7

8

9

10

11

12

13

14

15

1 11 21 31 41 51 61 71 81 91 101 111 121

Grid residual 3s (nm)

3s Grid Residual Y (nm)

3s Grid Residual X (nm)

Wafer grid residual 3s is stable to within +- 1nm and has a stable signature


Wafer BS particle

November 2011

September 2013

IMEC 2013

0

1000

2000

3000

4000

5000

6000

7000

-20

-18

-16

-14

-12

-10 -8 -6 -4 -2 0 2 4 6 8

10

12

14

16

18

20

Fre

qu

en

cy

Overlay (nm)

X

Y

NXE:3100 BEST ACHIEVABLE MEASUREDOVERLAY – CPE AND iHOPCAPPLIED

X: |Mean|+3σ: 6.0nmY: |Mean|+3σ: 5.6nm

Reference grid from NXE:3100, second layer on XT:1900i1 wafer, 83 fields, 26x33mm2, 17x22 pts/field

Applying 10-parameter, CPE and iHOPC corrections, brings measured overlay down to 6nm |Mean|+3σ

2013 EUVL SYMPOSIUM - HENDRICKX 15J. Hermans 2012 SPIE 8332

IMEC 2013

NXE:3100 LENS ABERRATIONSRMS ERROR FROM 5 MEASUREMENTS OVER 10 MONTHS


0

0.2

0.4

0.6

0.8

1

1.2

Spherical Coma Astigmatism 3-foil Z5-Z37

rms lens aberration (nm)

Lens aberration measurement over 9-month timeframe shows that aberration signature remains well in specification

UncorrectedCorrected

IMEC 2013 2013 EUVL SYMPOSIUM - HENDRICKX

y = -1.3482x + 247.36

0

10

20

30

40

50

0 100 200 300 400Thickness (nm)

Energy (mJ/cm2)

AL - SEVR140

Eo = 183.47

y = -38.478x + 166.41

0

10

20

30

40

50

2 2.5 3 3.5 4 4.5 5

Thickness (nm)

Energy (mJ/cm2)

ML - SEVR140

Eo = 4.32

OoB (%) in SEVR140 resist on NXE:3100 (DPP) = 2.4%

%100(%)maskAlwithcleartoDose

maskMLwithcleartoDoseOoB ⋅=

MEASUREMENT OF OUT-OF-BAND COMPONENTDOSE-TO-CLEAR OF AL COATED MASK AND ML MASK

17

IMEC 2013

0.0

0.5

1.0

1.5

2.0

2.5

3.0

3.5

4.0

01/04/2012 10/07/2012 18/10/2012 26/01/2013 06/05/2013 14/08/2013

OoB (%)

Date

OoB measurements shows variability in 1.6-3.5% range (DPP source), which exceeds the reproducibility of the test (0.02% 3s)Not correlated to major interventions on system

MEASUREMENT OF OUT-OF-BAND COMPONENTDOSE-TO-CLEAR OF AL COATED MASK AND ML MASK


IMEC 2013

CONTENTS

Introduction

NXE:3100 stability

Monitoring scheme

Productivity

CD control

Overlay control

Out of band radiation

Outlook to 3300

Conclusions


IMEC 2013

2012 POR TEST ON NXE:330022NM LS PROCESS CD UNIFORMITY

▸ 22nm LS, NXE:3300, conventional illumination

▸ 2012 POR \ NXE:3100 monitor mask (CDU27_1)


2012 POR tested on NXE:3300 with 22nm LS on 3100 mask -showed IF signature (mask) and process limits at 22nm LS

3 x 5 samplingRaw data – includes mask signature

IMEC 2013

NXE:3300 MONITOR MASK ABSORBER WIDTH UNIFORMITY ON MASK


0

0.5

1

1.5

2

2.5

3

3.5

43

2n

m H

32

nm

V

27

nm

H

27

nm

V

27

nm

H

27

nm

V

22

nm

H

22

nm

V

27

nm

H

27

nm

V

22

nm

H

22

nm

V

22

nm

H

22

nm

V

CDU32 CDU27_1 CDU27_2 CDU22

Mask IntrafieldCD 3s (4x, nm)

ADT 3100 3300

Mask intrafield CD uniformity for 22nm LS at <2nm 3s at 4x

44nm pitch CH (1x)44nm pitch LS (1x)

40nm pitch CH (1x)

34nm pitch CH (1x)

IMEC 2013

▸ NXE:3100 exposure

▸ Wafer coated on TEL Lithius Pro

▸ Dipole 60-X illumination, 20.5 mJ/cm2

▸ Full wafer and full field exposure

▸ CD measured in 3 x 5 field positions, including field edges

▸ Raw data reported – split up in IF and across wafer signature

Total1.67nm 3s

Intrafield1.50nm 3s

Intrafield subtracted0.82nm 3s


EUV RESIST LS PERFORMANCE22nm LS CDU – RESIST C

IMEC 2013 2013 EUVL SYMPOSIUM - HENDRICKX 23

NXE:3300 PROCESS NEW CANDIDATE52NM PITCH CH CD UNIFORMITY AFTER LITHO-ETCH

Litho

Etch

Mean CD: 25.34 nmLCDU 3σ : 2.70 nm

Mean CD: 31.16 nm

LCDU 3σ : 3.79 nm

Contact hole uniformity 3s across wafer improves to 2.70nm 3s through resist etch

31nm CH at 52nm pitch printed to 31nm after lithoCD distribution over 32 fields, 75 CH per field, 27.25 mJ/cm2

5.8nm etch bias1.1nm 3s reduction

Ming Mao et al, 2013 MNE

IMEC 2013

CONCLUSIONS

242013 EUVL SYMPOSIUM - HENDRICKX

NXE:3100

Throughput More than 6000 wafers exposed Collector mirror influences power level~6 months collector mirror lifetime

CD control Stable CD and CD map

OVL control Grid and IF residuals are stable, but can be impacted by particles on mask and wafer BSBest demonstrated is ~6nm (matched to XT1900i)

NXE:3100 has sufficient CD and overlay control to enable pre-production device fabrication in wafer batch modeNXE:3300 preparation for track, resist, and mask ongoing

NXE:3300

Resist Candidate resist tested on NXE:3100 – 22nm VLS 1.67nm 3s

Mask New mask fabricated with improved uniformity

Track Accepted TEL Lithius Pro-Z track (09/2013)

IMEC 2013

ACKNOWLEDGEMENTS

ASML▸ Amir Sharomi

▸ Theo Van Den Akker

▸ Jan-Willem Van Der Horst

▸ Andre Van Dijk

▸ Charles Pigneret

▸ Leo Heusschen

▸ Tom Jonckers

▸ Guiseppina Toto

▸ Sjef Haagmans

▸ Oktay Yildrim

▸ Roel Knops

▸ Mircea Dusa

▸ Timon Fliervoet


TEL▸ Kathleen Nafus

▸ Koichi Matsunaga

▸ TEL support team

Documents

ASML NXE:3100 PRE-PRODUCTION EUV SCANNER …euvlsymposium.lbl.gov/pdf/2013/pres/S1-2_EHendrickx.pdfimec 2013 asml nxe:3100 pre-production euv scanner performance at imec eric hendrickx,