Fundamentals of the ReactionFundamentals of the Reaction--Diffusion Diffusion Process in Model EUV PhotoresistsProcess in Model EUV Photoresists

Bryan D. Vogt, Kristopher A. Lavery, Vivek M. Prabhu, Eric K. Lin, Wen-li Wu, Sushil K. Satijaa

Polymers Division, aCenter for Neutron Research, National Institute of Standards and Technology, Gaithersburg, MD

Kwang-Woo Choi*, Michael J. Leesonb, Heidi B.Caob, Hai DengGeorge Thompson,

Intel Corporation, Santa Clara, CAbIntel Corporation, Hillsboro, OR

*Intel Assignee to NIST

CRADA 1893

Fourth International Extreme Ultra Violet Lithography (EUVL) Symposium07-09 November 2005San Diego, California

Adapted from : Melissa Shell Sematech EUV perspectives Meeting 5/10/05

Motivation and Industry needsMotivation and Industry needs

•Simultaneously meeting resolution, sensitivity, and line width roughness

Current ChallengesCurrent Challenges

1. Photoacid distribution at line-edge – dose related

• Few photons → statistical origin to LER?Higher PAG loadings → improve sensitivity

2. Acid diffusion

• How much is good / bad?Sharp latent image interfaces → Beneficial to LER

Identify the resist characteristics that lead to sharp latent image profiles

Lithography Process Lithography Process

Post-Exposure• PEB conditions • Acid diffusion

DevelopmentOptical

EU

V

How to measure the interface formation without optical effects?

Ideal StepIdeal Step--Exposure Profile:Exposure Profile: Bilayer Geometry Bilayer Geometry

• Multiple spin coat steps1. Acid feeder layer2. Resist layer

• Currently – using broadband DUV (248-nm) for experimentation

Si Si SiSi

PAB PEB

Si

λ = DUV

1. 2. 3. 4. 5.

PAB PAB PEB Develop

PAB

Sample Preparation

Model Latent Image LineModel Latent Image Line--EdgeEdge

Expose

Si Si SiSi

PABPAB PEB

Si

λ = DUV

1. 2. 3. 4. 5.

PAB PAB PEB Develop

PABPAB

Sample Preparation

Model Latent Image LineModel Latent Image Line--EdgeEdge

Expose

substrate

film

x-raysor

neutrons

ρD

roughness

-5

-4

-3

-2

-1

0

log

(R)

0.200.150.100.05Q (Å-1)

QQcc22 ⇒ ρ⇒ ρ

DD ≅ 2π/Δ≅ 2π/ΔQQ

fringe persistence fringe persistence ⇒⇒ surface roughnesssurface roughness

Information:• thickness DD• density profile,QQcc

22 , ρρ(z)(z)• interfacial roughness

Q

Air or D2O

Measurement: Neutron Reflectivity Measurement: Neutron Reflectivity

Neutron scattering length densitiesNeutron scattering length densities

•Contrast largest between hydrogen and deuterium – very sensitive measurements

•Introduce isotopic substitution (H,D) to measure resist latent image profiles

1H 2D 12C 16O 28Si14N 19F 27Al 64Cu

6.646.67-3.74 9.36 5.08 5.64 3.45 4.15 7.72

Reaction Front EvolutionReaction Front Evolution

1.0

0.8

0.6

0.4

0.2

0.0d-

PB

OC

St V

olum

e Fr

actio

n

10008006004002000z from silicon surface (Å)

0 s, PEB 15 s, PEB 20 s, PEB 30 s, PEB

-10

-8

-6

-4

-2

0

log(

Ref

lect

ivity

)

0.160.140.120.100.080.060.040.02q (Å

-1)

Si

PBOCST

PHS+

PAG

PEB

•Deuterium-labeled protection group provides neutron reflectivity contrast

Neutron Reflectivity Results in situ protection profiles

CH2 CH

OC

O

O

C

CD3

CD3

CD3

dPBOCSt

CH2 CH

OH

CCD2 CD3CO2H+

Δ

CD3

+ +

PHOSt

CH2 CH

OC

O

O

C

CD3

CD3

CD3

dPBOCSt

CH2 CH

OH

CCD2 CD3CO2H+

Δ

CD3

+ +

PHOSt

CH2 CH

OC

O

O

C

CD3

CD3

CD3

dPBOCSt

CH2 CH

OH

CCD2 CD3CO2H+

Δ

CD3

+ +

PHOSt

•Ideal Line-Edge leads to broadening (110°C PEB)•Wide-range of protection levels at reaction-front•Lin et al., Science 2002, 297, 372

Experimental : Model EUV resist Experimental : Model EUV resist

H

HOST TBA

CD3

CD3CD3

H+

•Deuterium labeling to directly measure the reaction-diffusion front•Model system resist copolymer with photoacid generator

++ side reactions

+2

3

3CD2 CD

3

CD3

+ side reactions

(volatile)

50/50 copolymer*

Photoacid generator

*

HOST d-TBA

Neutron Reflectivity : CopolymerNeutron Reflectivity : Copolymer

10-4

10-3

10-2

10-1

100

Ref

lect

ivity

0.100.080.060.040.02q (Å-1)

Silicon

+Polymer Feeder

z

•No PEB, interface between acid feeder and copolymer is < 2 nm•Follow interface evolution with processing (varied dose & Fixed PEB 130 °C/30s)

150

100

50

0

Qc2

(Å-2

x10-6

)

10008006004002000Distance (Å)

CD3

CD3CD3

CD3

CD3CD3

CD3

CD3CD3

Results Results -- Reaction Front Reaction Front

10-8

10-7

10-6

10-5

10-4

10-3

10-2

10-1

100

Ref

lect

ivity

80x10-3604020q (Å-1)

UnexposedE02.5 E04.0 E05.0 E0Fits

10-8

10-7

10-6

10-5

10-4

10-3

10-2

10-1

100

Ref

lect

ivity

80x10-3604020q (Å-1)

10-8

10-7

10-6

10-5

10-4

10-3

10-2

10-1

100

Ref

lect

ivity

80x10-3604020q (Å-1)

UnexposedE02.5 E04.0 E05.0 E0Fits

Data vertically offset for clarityEo : Dose-to-clear for bilayer

•Neutron reflectivity indicates bilayer structure

Results Results -- Reaction Front Reaction Front 0.25

0.20

0.15

0.10

0.05

0.00

Dep

rote

ctio

n E

xten

t

200150100500Distance from Interface (Å)

UnexposedE0

2.5 x E0

σ

σ

6% PAG Loading in Feeder LayerPEB 30 sec @ 130°C

0.25

0.20

0.15

0.10

0.05

0.00

Dep

rote

ctio

n E

xten

t

200150100500Distance from Interface (Å)

UnexposedE0

2.5 x E0

0.25

0.20

0.15

0.10

0.05

0.00

Dep

rote

ctio

n E

xten

t

200150100500Distance from Interface (Å)

0.25

0.20

0.15

0.10

0.05

0.00

Dep

rote

ctio

n E

xten

t

200150100500Distance from Interface (Å)

UnexposedE0

2.5 x E0

σ

σ

6% PAG Loading in Feeder LayerPEB 30 sec @ 130°C

•Deprotection profile : Length-scale (σ) is (248 nm) dose sensitive•Acid profile : Unknown – initially sharp

Higher dose → more acid →increased reaction-rateAcid reaction-diffusion models to determine mechanism of reaction length scale – many factors

Complexity of acid diffusion in HOST-TBA copolymer – F.A. Houle et al. (2000)Potential for auto-accelerated reaction and thermal deprotection - Wallraff et al. (1994)

Results Results –– PAG Loading PAG Loading

Increasing PAG loading

5 % 10% 15 % 20 % 25 %2 %0 %

TBA homopolymerTBA homopolymer

+

•Segregation within homopolymer increases with loading•Fixed PAB temperature and time

(90°C PAB)

PAG loading with HOSTPAG loading with HOST--TBA resistTBA resist

20 % 30 %

+

60 / 40 HOST / TBA60 / 40 HOST / TBA

•PAG segregation dependent upon resist HOST composition•Chemical force microscopy See poster (2-ME-09) by JT Woodward

ConclusionsConclusions

• Reaction front profile width dependent upon dose

• High PAG loading leads to phase separation in TBA homopolymer

• PAG segregation at high loadings not observed for TBA-HOST copolymers

• Copolymers confer miscibility

AcknowledgementsAcknowledgements

• Shuhui Kang – NIST• Melissa Shell – Intel Hillsboro• Christof Krautschik – Intel Santa Clara• Jim Sounik and Michael Sheehan – DuPont Electronic Technologies

Funding• NIST MSEL Laboratory Director – Leslie Smith (Ret.)• NIST Office of Microelectronics Programs – Steven Knight, Director

Official U.S. Government Contribution

• Certain equipment, instruments or materials are identified in this presentation in order to adequately specify the experimental details. Such identification does not imply recommendation by the National Institute of Standards and Technology nor does it imply the materials are necessarily the best available for the purpose.

• Official contribution of the National Institute of Standards and Technology; not subject to copyright in the United States.