euv actinic blank inspection tool developmenteuvlsymposium.lbl.gov/pdf/2012/pres/k.takehisa.pdf ·...

1 2012 International Symposium on Extreme Ultraviolet Lithography

EUV Actinic Blank Inspection Tool Development

EUVL Symposium 2012

Kiwamu Takehisa 1, Hiroki Miyai 1, Tomohiro Suzuki 1,

Haruhiko Kusunose 1, Takeshi Yamane 2, Tsuneo Terasawa 2,

Hidehiro Watanabe 2, Soichi Inoue 2, Ichiro Mori 2

1 Lasertec Corporation 2 EUVL Infrastructure Development Center


Contents

1. Introduction

2. Illumination efficiency improvement

3. Solutions to use a new EUV filter

4. High magnification review optics

5. Summary


Contents

1. Introduction

2. Illumination efficiency improvement



5. Summary


design

EIDEC – Lasertec Blank Inspection Project

2010 2011 2012 2013 2014 2015

parts fabrication

assembling / adjustment

evaluation / development

improvement

Prototype tool for HVM 16nm hp HVM 11nm hp

development

EIDEC BI program started

-- Assembly of the ABI HVM tool has been completed.

-- Adjustment and evaluation are in progress on schedule.

ABI HVM Tool Development Schedule


ABI HVM Tool


Actinic mask blank inspection sensitivity roadmap

15 30

60

40

Sen

sit

ivit

y (

nm

)

30

50

45 60

Inspection time (min.)

300

MIRAI/Selete

W:55nm H: 1.0nm

HVM for 16nm hp W:50nm H: 1.0nm

HVM for 11nm hp W: 35nm H: 0.7nm


1. Introduction

2. Illumination optics improvement



5. Summary


Illumination optics of the ABI tool

Second ellipsoidal mirror

Xe Plasma EUV DPP source

Mirror chamber

(illumination Optics)

EUV source and

mirror chamber

First ellipsoidal mirror

Main

chamber


Illumination optics for the HVM tool

Two times higher transmission can be expected

by improving the EUV filter and the vane trap

1. Previous design

EUV filter

EUV

DPP

Source Vane trap

2. HVM design

New EUV filter

Improved

EUV

Source

New vane trap


Anticipated problems using a new EUV filter

Debris

Previous design

Debris

HVM design

・Xe leak

・OoB radiation

Filter membrane Filter membrane


1. Introduction




5. Summary


13.5nm radiation transmission in various gas

Xe diffusion into the mirror chamber need to be suppressed

H2

He

Ar

N2

Xe


Xe diffusion into the mirror chamber can be stopped

Xe diffusion can be stopped in the vane trap by transparent gas

Vane trap

Xe H2, He, Ar, N2

Gas diffusion simulation results

Xe

100% H2

100%

EUV filter

EUV Source

Xe

H2, He, Ar or N2

Vane trap

Small opening

(Differential pumping)

Main

chamber Mirror

chamber

CCD/TDI


Out of band radiation does not affect sensitivity

EUV mask blank

CCD/TDI

Dark field

illumination

0

2

4

6

8

10

12

14

W60 W70 W78 W85 W100 W110 W120 W130 W140 W160 W190 W210 W250

H1.5 H1.8 H2 H2.2 H2.5 H2.5 H2.5 H2.5 H2.5 H2.5 H2.5 H2.5 H2.5

SB

R

Zr

Si

Almost equal sensitivity was obtained between Zr and Si.

Sensitivity test using a Si filter

which greatly transmits visible light.

Data was taken by MIRAI/Selete tool.


1. Introduction




5. Summary


Pixel size on Mask

460nm

Low magnification

: 26X (Inspection)

Difficult to specify

a defect position

within a pixel

Defect position

can be accurately

defined.

High magnification

: 1200X (Review)

Pixel size on Mask

10nm

Purpose of the high magnification review optics

1200X magnification review mode :

a basis for the accurate defect positioning


CCD

camera

Switch

mirror 2nd concave

mirror

Mask blank

Schwarzschild

objective (SO)

26x

~ 45 min

Inspection mode (Mask whole area)

M1 =

High-Speed Inspection & High Resolution Review

implemented in one system

Review mode integration

CCD/TDI

M2 = b／a

= 46.15

2nd concave

mirror

M1・M2 = 1200x

~ 5 min

for 10 defects

Review mode (Area where defects are detected)

CCD


□50nm

□20nm

□10nm

~35nm

~22nm

~19nm

1/e of

the peak

Simulation results of the 1200X projection

Re～k1 λ

NA

λ=13.5nm

NA=0.27

k1=0.25

= 12.5nm

Simulation results are given

by Dr. Toyoda of Tohoku Univ.


Spot diagrams in the 0.47x0.47mm area for the 26X SO

24um at TDI sensor

Small aberration can be obtained at the center

Inspection mode Review mode

Refocus

Concave: Ellipsoidal Concave: Aspherical

An improved SO using

an aspherical concave mirror

is also being developed.


1200X

0μm 0.5μm

Deviation

1μm 5μm

Concave mirror

Convex mirror

2μm 3μm 4μm

Projected image considering the pixels of the sensor at 1200X

Review image degradation by the imperfection of SO

Deviation of the coaxiality of the convex and the concave mirror

Mask blank

200μm at sensor

SO

Although 5μm deviation is small at 26X,

less than 0.5μm deviation is required at 1200X.

SO with a fine adjustment mechanism is also being developed.


1. Introduction




5. Summary


1. ABI HVM tool has a target of < 45min inspection time

by improving the illumination optics as well as EUV source.

2. In order to increase the illumination optics efficiency,

a new EUV filter and a new vane trap are employed.

3. Xe diffusion can be suppressed by transparent gas.

Also it has been demonstrated that the OoB radiation does

not affect the inspection sensitivity.

4. 1200X magnification review optics are design based on the

26X SO. Also an improved SO is being developed.

Summary


We would like to thank:

-- all the members of the EIDEC BI program

-- Prof. Kinoshita and Dr. Harada of Hyogo Univ.

-- Dr. Toyoda of Tohoku Univ.

This work was supported by New Energy and Industrial

Technology Development Organization (NEDO) and

Japan Ministry of Economy, Trade and Industry (METI).

Acknowledgement

euv actinic blank inspection tool developmenteuvlsymposium.lbl.gov/pdf/2012/pres/k.takehisa.pdf ·...

Documents