PROGRESS IN SPECTROSCOPIC ELLIPSOMETRY FOR THE IN … · Faculty of Electrical and Computer...

Faculty of Electrical and Computer Engineering Institute of Semiconductors and Microsystems

Dresden – March 10, 2014

PROGRESS IN SPECTROSCOPIC ELLIPSOMETRY

FOR THE IN-SITU REAL-TIME INVESTIGATION

OF ATOMIC LAYER DEPOSITIONS

8th Workshop Ellipsometry by the Arbeitskreis Ellipsometrie (AKE) - Paul Drude e. V.

Marcel Junige, V. Sharma, D. Schmidt, M. Albert, M. Schubert, J. W. Bartha

phone: +49 351 463-37967 | - 34382

mobile: +49 163 5809233

fax: +49 351 436-37172

e-mail: [email protected]


OUTLINE

1. Introduction

2. Experimental setup

3. Results

4. Conclusion

Metrology issues

film thickness & quality

ALD growth behavior

reaction mechanisms

ALD benefits for nanoscale

film manufacturing

atomic-level thickness control

conformality in 3D structures

uniformity on large substrates

MOTIVATION

1

20

µm

M. Knaut et al.: JVST A 30, 01A151 (2012).

M. Knaut et al.: Microelectron. Eng. 107, 80 (2013).

D. Schmidt, E. Schubert, and M. Schubert:

Appl. Phys. Lett 100, 11912 (2012).

Ch. Wenger et al.:

18th WoDiM (Kinsale Co Cork, 2014).

– submitted

http://dx.doi.org/10.1109/SCD.2011.6068739


precursor A

reactant B

reaction byproducts

precursor A adsorbate

functional surface group

functionalized film surface

ATOMIC LAYER DEPOSITION

M. Leskelä , M. Ritala , and O. Nilsen: MRS Bulletin 36, 877 (2011).

G. N. Parsons, S. M. George, and M. Knez: MRS Bulletin 36, 865 (2011).

M. Ritala, and M. Leskelä; in: “Deposition and processing of thin films” 1, 103 (Academic Press, 2002).

half-reaction A

half-reaction B

purge or evacuation

purge or evacuation

2

material amount

time

ATOMIC LAYER DEPOSITION

V. Miikkulainen et al.: J. Appl. Phys. 113, 21301 (2013).

S. Elliott, and M. Shirazi: AVS 59th International Symposium & Exhibition (AVS, Tampa, 2012). 3

adsorbed

precursor

~ 1 − 𝑒−𝒌𝑨∙𝑡

~𝑒−𝒌𝑩∙𝑡

eliminated

ligands

condensed /

deposited

film


OUTLINE

1. Introduction


3. Results

4. Conclusion

Industry-oriented ALD

class 10 cleanroom

capability for 300 mm wafers

rapid thermal annealing (RTA)

up to 1000 °C in vacuo

current processes:

Al2O3; Ta2O5;

TaNx(O,C); Ru(C); RuO2

4

transfer

chamber

ALD

metals

SE, QMS

PES SPM

handler

ALD

dielectrics

load-lock

Ar

QMS

SESE

MFC Ar

O2

MFC

MFC

ArMFC

ECPR

ArMFC

TBTDET

TMA

NH3

MFC

O3

QCM

Industry-oriented ALD

class 10 cleanroom

capability for 300 mm wafers

rapid thermal annealing (RTA)

up to 1000 °C in vacuo

current processes:

Al2O3; Ta2O5;

TaNx(O,C); Ru(C); RuO2

In-situ, on-site &

In-vacuo metrology

Spectroscopic Ellipsometry

Quadrupole Mass Spectrometry

Photoelectron Spectroscopy (X-ray, Ultra Violet)

Scanning Probe Microscopy (Atomic Force Microscopy,

Scanning Tunneling Microscopy)

Quartz Crystal Microbalancing

4


OUTLINE

1. Introduction


3. Results

a) in-situ real-time SE algorithm

b) ALD growth behavior of Al2O3

c) ALD growth behavior of TaNx

4. Conclusion

120

130

140

150

160

170

180

10

15

20

25

30

35

0 500 1000 1500 2000Δ

(°)

Ψ (

°)

wavelength (nm)

Ψ, Δ spectrum after 100 Al2O3 ALD cycles at 350 °C

5

IN-SITU REAL-TIME SPECTROSCOPIC ELLIPSOMETRY

spectroscopic

Ψ, Δ = ƒ(wavelength)

in situ

at the place

real-time

Ψ, Δ = ƒ(time)

in-situ real-time Spectroscopic Ellipsometry (irtSE) algorithm

acquire ellipsometric spectra

with the fastest possible sampling rate

calibrate ex-situ and in-situ elements

in the ellipsometer’s light path

evaluate the acquired ellipsometric spectra

time slice by time slice

post-process

the evaluated time-resolved result

⑦ reveal process parameter (inter)dependencies

of characteristic ALD growth values

⑤ fit modeled to experimental data

time slice by time slice

④ construct an optical model

③ acquire ellipsometric spectra

in situ and in real-time

with the fastest possible sampling rate

calibrate additional in-situ elements in the

ellipsometer’s light path at the vacuum reactor

calibrate the rotating compensator ellipsometer’s

basic components on the ex-situ stage

MSE ok? no

yes

uncalibrated or suspicious hardware

① fine correct experiment-specific deviations

② correct inaccuracies

due to the fastest possible data acquisition

⑥ extract the ALD characteristic curve

for the linear growth region

start of ALD processing sequence

end of ALD processing sequence

new sample of interest

6

Δ

Ψ

𝜽

offset in the

angle of incidence

SE FOR THE IN-SITU REAL-TIME INVESTIGATION OF ALD

7

in-plane & out-of-plane

window effects

various

temperature effects

additional in-situ elements in the ellipsometer’s light path at the vacuum reactor

timewise mean absolute deviation of Ψ, Δ at 450 nm

in dependence on effective acquisition time

comparing fast acquisition and high accuracy mode

8

155.2

155.3

155.4

155.5

155.6

155.7

155.8

0.1 10000Δ

at

450 n

m (

°)

time (s)

19.8

19.9

20.0

20.1

20.2

20.3

0.1 10000

Ψ a

t 450 n

m (°)

time (s)

24.0

24.1

0.00

0.01

0.02

0.03

0.04

0.05

0.06

0.07

0.08

0 10 20 30 40 50 60

mea

n a

bso

lute

dev

iati

on

of Ψ

, Δ

(°)

effective acquisition time per ellipsometric spectrum (s)

fast acquisition Psi fast acquisition Delta

high accuracy Psi high accuracy Delta

Ψ, Δ at 450 nm over time

comparing fast acquisition

and high accuracy mode


timewise mean absolute deviation of optical layer thickness

in dependence on effective acquisition time

comparing fast acquisition and high accuracy mode, respectively

9

0.00

0.01

0.02

0 10 20 30 40 50 60

mea

n a

bso

lute

dev

iati

on

of

op

tica

l la

yer

th

ick

nes

s (n

m)


fast acquisition

high accuracy


MSE vs. effective acquisition time

comparing fast and high accuracy mode

10

0

5

10

15

20

25

30

35

40

45

0 10 20 30 40 50 60

mea

n-s

qu

are

d e

rro

r

effective acquisition time per spectrum (s)

fast acquisition

high accuracy

J. A. Woollam et al., in Optical metrology, G. A. Al-Jumaily, Ed.

(SPIE Optical Engineering Press, Bellingham, 1999), pp. 3–28.



timewise mean Ψ, Δ error vs. effective acq. time


MSE vs. effective acquisition time


10

0

5

10

15

20

25

30

35

40

45

0 10 20 30 40 50 60

mea

n-s

qu

are

d e

rro

r


fast acquisition

high accuracy

0.00

0.01

0.02

0.03

0.04

0.05

0 10 20 30 40 50 60

mea

n Ψ

, Δ

err

or

(°)





11

-1.0

-0.5

0.0

0.5

1.0

0 500 1000 1500 2000

Ψ, Δ

dif

fere

nce

(°)

wavelength (nm)


-1.0

-0.5

0.0

0.5

1.0

0 500 1000 1500 2000

Ψ, Δ

dif

fere

nce

(°)

wavelength (nm)


spectral Ψ, Δ difference (experiment – model)

in fast acquisition mode

spectral Ψ, Δ difference (experiment – model)

in high accuracy mode


spectral mean absolute deviation of Ψ, Δ difference (experiment – model)

vs. effective acquisition time comparing fast and high accuracy mode

12

0.00

0.05

0.10

0.15

0.20

0.25

0 10 20 30 40 50 60

mea

n a

bso

lute

dev

iati

on

of

Ψ, Δ

dif

fere

nce

(°)





0

5

10

275 285 295 305AL

D c

ycl

e in

dex

ALD process time (min)

8.5

9.0

9.5

10.0o

pti

cal

lay

er

thic

kn

ess

(nm

)

1st derivative

increment by 1

whenever

a threshold value

is exceeded

13

-5

0

5

gro

wth

ra

te

(nm

/ s)


TM

A

O3

TM

A

O3

14


Al2O3 optical layer thickness

after averaging over 10 cycles

Al2O3 optical layer thickness

in the course of one ALD cycle before averaging

0

1

2

3

0 60 120 180

op

tica

l la

yer

th

ick

nes

s (Å

)

time (s)

0

1

2

3

0 60 120 180

op

tica

l la

yer

th

ick

nes

s (Å

)

time (s)

timewise mean absolute deviation of averaged optical layer thickness

in dependence on the number of ALD cycles involved into averaging

15


0.0

0.1

0.2

0 1 2 3 4 5 6 7 8 9 10

mea

n a

bso

lute

dev

iati

on

of

av

era

ged

op

tica

l la

yer

th

ick

nes

s (Å

)

number of ALD cycles involved into averaging


OUTLINE

1. Introduction


3. Results

a) in-situ real-time SE algorithm

b) ALD growth behavior of Al2O3

c) ALD growth behavior of TaNx

4. Conclusion

TM

A

O3

averaged optical layer thickness in the course of one Al2O3 ALD cycle

at varied deposition temperatures

16

ALD GROWTH BEHAVIOR OF AL2O3

0

1

2

3

0 60 120 180

op

tica

l la

yer

th

ick

nes

s (Å

)

time (s)

500 °C

400 °C

300 °C

200 °C

100 °C

V. Sharma: Student Research Project (Technische Universität Dresden, Dresden, 2014). – manuscript


extracted ALD characteristic values

in dependence on the actual deposition temperature

17 V. Sharma: Student Research Project (Technische Universität Dresden, Dresden, 2014). – manuscript

0

1

2

3

0 100 200 300 400 500

(a. u

.)

actual Si surface temperature (°C)

growth per cycle (Å)

adsorbed precursor

removed ligands

loss during purge 2

ALD GROWTH BEHAVIOR OF AL2O3


TB

TD

ET

NH

3

averaged optical layer thickness in the course of one TaNx ALD cycle

at varied deposition temperatures

18

0.0

0.1

0.2

0.3

0.4

0 50 100 150

op

tica

l la

yer

th

ick

nes

s (n

m)

time (s)

400 °C

250 °C

200 °C

150 °C

ALD GROWTH BEHAVIOR OF TANX

T. F. Walther: Diplomarbeit (Technische Universität Dresden, Dresden, 2014). – manuscript



OUTLINE

1. Introduction


3. Results

4. Conclusion

in-situ real-time Spectroscopic Ellipsometry

with a sampling rate of 0.8 s per spectrum

and a thickness MAD below 0.01 nm

ALD characteristic curve

in the homogeneous growth region

containing both saturation curves

detailed insight into the

ALD growth behavior of Al2O3 and TaNx

in dependence on deposition temperature

CONCLUSION

assistance at scientific work, especially

Marion Geidel, Martin Knaut, Christoph Hossbach, Christoph Kubasch,

Tino Hoffmann, Keith B. Rodenhausen, Stefan Schöche,

Daniela Seiffert (geb. Schmidt), Steffen Strehle,

Thomas Wagner, Greg Pribil,

Eugene Irene, Maria Losurdo, Kurt Hingerl, Thomas Hingst,

Tillmann F. Walther, Ralf Tanner

financial support

The PhD project of Marcel Junige is funded by the European Social Fund (ESF) and the

Free State of Saxony of the Federal Republic of Germany 2011-2014 (contract number: 100077335).

ACKNOWLEDGMENT

PROGRESS IN SPECTROSCOPIC ELLIPSOMETRY FOR THE IN … · Faculty of Electrical and Computer...

Documents

Transcript of PROGRESS IN SPECTROSCOPIC ELLIPSOMETRY FOR THE IN … · Faculty of Electrical and Computer...