Atomic Layer Processing E-Book

Thorsten Lill

• Herausgeber: Wiley-VCH GmbH
• Kategorie: Fachliteratur
• Sprache: Englisch

Seitenzahl: 573

Veröffentlichungsjahr: 2021

Beschreibung

Learn about fundamental and advanced topics in etching with this practical guide

Atomic Layer Processing: Semiconductor Dry Etching Technology delivers a hands-on, one-stop resource for understanding etching technologies and their applications. The distinguished scientist, executive, and author offers readers in-depth information on the various etching technologies used in the semiconductor industry, including thermal, isotropic atomic layer, radical, ion-assisted, and reactive ion etching.

The book begins with a brief history of etching technology and the role it has played in the information technology revolution, along with a collection of commonly used terminology in the industry. It then moves on to discuss a variety of different etching techniques, before concluding with discussions of the fundamentals of etching reactor design and newly emerging topics in the field such as the role played by artificial intelligence in the technology.

Atomic Layer Processing includes a wide variety of other topics as well, all of which contribute to the author's goal of providing the reader with an atomic-level understanding of dry etching technology sufficient to develop specific solutions for existing and emerging semiconductor technologies. Readers will benefit from:

• A complete discussion of the fundamentals of how to remove atoms from various surfaces
• An examination of emerging etching technologies, including laser and electron beam assisted etching
• A treatment of process control in etching technology and the role played by artificial intelligence
• Analyses of a wide variety of etching methods, including thermal or vapor etching, isotropic atomic layer etching, radical etching, directional atomic layer etching, and more

Perfect for materials scientists, semiconductor physicists, and surface chemists, Atomic Layer Processing will also earn a place in the libraries of engineering scientists in industry and academia, as well as anyone involved with the manufacture of semiconductor technology. The author's close involvement with corporate research & development and academic research allows the book to offer a uniquely multifaceted approach to the subject.

Leseprobe

Table of Contents

Cover

Title Page

Copyright

List of Abbreviations

1 Introduction

References

2 Fundamentals

2.1 Important Performance Metrics of Etching Processes

2.2 Physisorption and Chemisorption

2.3 Desorption

2.4 Surface Reactions

2.5 Sputtering

2.6 Implantation

2.7 Diffusion

2.8 Transport Phenomena in 3D Features

2.9 Classification of Etching Technologies

References

3 Thermal Etching

3.1 Mechanism and Performance Metrics of Thermal Etching

3.2 Applications Examples

References

4 Thermal Isotropic ALE

4.1 Mechanism of Thermal Isotropic ALE

4.2 Performance Metrics

4.3 Plasma‐Assisted Thermal Isotropic ALE

4.4 Applications Examples

References

5 Radical Etching

5.1 Mechanism of Radical Etching

5.2 Performance Metrics

5.3 Applications Examples

References

6 Directional ALE

6.1 Mechanism of Directional ALE

6.2 Performance Metrics

6.3 Applications Examples

References

7 Reactive Ion Etching

7.1 Reactive Ion Etching Mechanisms

7.2 Performance Metrics

7.3 Applications Examples

References

8 Ion Beam Etching

8.1 Mechanism and Performance Metrics of Ion Beam Etching

8.2 Applications Examples

References

9 Etching Species Generation

9.1 Introduction of Low‐Temperature Plasmas

9.2 Capacitively Coupled Plasmas

9.3 Inductively Coupled Plasmas

9.4 Ion Energy Distribution Modulation

9.5 Plasma Pulsing

9.6 Grid Sources

References

10 Emerging Etching Technologies

10.1 Electron‐Assisted Chemical Etching

10.2 Photon‐Assisted Chemical Etching

References

Index

End User License Agreement

List of Tables

Chapter 2

Table 2.1 Sputtering thresholds for Ar

+

ion bombardment of materials impo...

Table 2.2 Projected ranges and straggle for systems and energies relevant for...

Table 2.3 Classification of continuous etching technologies.

Table 2.4 Classification of atomic layer etching technologies.

Chapter 3

Table 3.1 Silicon etching rates and reaction probabilities of interhalogens. ...

Chapter 4

Table 4.1 Potential conversion ALE reactions based on thermodynamics calculat...

Table 4.2 Thermochemistry of a variety of potential oxidation/fluorination AL...

Table 4.3 Thermal isotropic etching reactions with ligand exchange reaction.

Table 4.4 Thermal isotropic etching reactions: chelation, conversion, oxidati...

Chapter 7

Table 7.1 Emerging memory devices.

Table 7.2 Boiling points of halides and hydrides of elements used in phase ch...

Chapter 8

Table 8.1 Boiling points of selected MRAM materials.

Chapter 9

Table 9.1 Characteristics of nonequilibrium plasmas.

List of Illustrations

Chapter 2

Figure 2.1 Schematic representation of a surface with adsorbed atoms with th...

Figure 2.2 Classification of the most common etching profiles.

Figure 2.3 Lennard‐Jones potential for adsorbent–adsorbate system.

Figure 2.4 Lennard‐Jones diagram for dissociative adsorption.

Figure 2.5 Lennard‐Jones diagram for thermal etching.

Figure 2.6 Illustration of Langmuir–Hinshelwood (a) and Eley–Rideal (b) surf...

Figure 2.7 Left: Series of collision processes leading to sputtering of atom...

Figure 2.8 Effective indium concentration of sputtered neutral (triangles) a...

Figure 2.9 Sputtering rates for 500 eV Ar

+

ion bombardment of CoFe and

W

Figure 2.10 Schematic illustration of oxidation depth vs time according to D...

Figure 2.11 Thickness of oxidized aluminum as a function of time showing inv...

Figure 2.12 Experimental and modeling results for oxidation of various metal...

Figure 2.13 Hole and trench microstructures in advanced DRAM and NAND device...

Figure 2.14 Normalized etch rate as a function of aspect ratio for a fluorin...

Figure 2.15 Angular distributions of neutral beams. Curve a: Angular distrib...

Figure 2.16 Sputtering rate (a) and reflection coefficient (b) for silicon a...

Figure 2.17 (a) Schematic illustration of a blocking or scattering cone. (b)...

Chapter 3

Figure 3.1 Schematic illustration of the temperature process window for ther...

Figure 3.2 Structural formulas of the keto tautomers of Hhfac (a) and Hacac ...

Chapter 4

Figure 4.1 Schematic illustration of the temperature process window for ther...

Figure 4.2 Schematic illustration of the temperature process window for ther...

Figure 4.3 Saturation curves for thermal isotropic ALE of Cu via O

2

/Hhfac li...

Figure 4.4 Depth of fluorination as a function of HF pressure measured with ...

Figure 4.5 Standard Gibbs free energies at 200 °C for fluorination of Al

2

O

3

,...

Figure 4.6 Schematic of proposed reaction mechanism for Al

2

O

3

ALE showing (a...

Figure 4.7 EPC as a function of pressure for thermal isotropic ALE of SiO

2

v...

Figure 4.8 Temperature dependence of EPC for ZnO etched with TMA/HF conversi...

Figure 4.9 EPC as a function of surface temperature for ligand exchange reac...

Figure 4.10 Correlation between EPC for thermal ALE of Al

2

O

3

with HF/TMA and...

Figure 4.11 Root causes for etch rate nonuniformity for ALE processes.

Figure 4.12 Selective etching of various materials with HF/DMAC thermal isot...

Figure 4.13 Schematic illustrations of perfectly isotropic profiles for a li...

Figure 4.14 Thermal isotropic ALE of Al

2

O

3

with HF/DMAC and a Si

3

N

4

hole mas...

Figure 4.15 Schematic illustration of reactant fluxes in Al

2

O

3

ALE and AlF

3

...

Figure 4.16 ARDE Al

2

O

3

ALE with HF/TMA in a high aspect ratio via feature....

Figure 4.17 Definition of selectivity for deposition processes.

Figure 4.18 Definition of selectivity for etching processes.

Figure 4.19 Schematic illustration of selective ALD using selective ALE as c...

Figure 4.20 Schematic illustration of the formation of horizontal devices us...

Figure 4.21 TEM cross sections of a 4 nm × 50 nm InGaAs nanowire encapsulate...

Chapter 5

Figure 5.1 Lennard‐Jones diagram for radical etching.

Figure 5.2 Schematic illustration of the temperature process window for radi...

Figure 5.3 Fabrication of stacked silicon nanowires for GAA devices.

Chapter 6

Figure 6.1 Classification of ALE by directionality.

Figure 6.2 Illustration of a collision cascade for an ion impinging a silico...

Figure 6.3 Schematic illustration of the process window of directional ALE a...

Figure 6.4 Seminal experiment by Coburn and Winters demonstrating neutral – ...

Figure 6.5 Sputtering yield as a function of ion energy for sputtering of pl...

Figure 6.6 Origin of the ideal window of directional ALE as a function of io...

Figure 6.7 Measured ALE window for silicon with chlorine radical modificatio...

Figure 6.8 Characteristic features of the ideal window for directional ALE....

Figure 6.9 Calculated saturation curves of an ion removal step in directiona...

Figure 6.10 Characteristics of directional ALE as a function of bulk binding...

Figure 6.11 Ion energy scan for carbon ALE with O

2

/Ar

+

for (a) EPC and (...

Figure 6.12 Ideal ALE window (a) and time‐ and energy‐dependent synergy (b) ...

Figure 6.13 Simulated ALE window for Ne, Ar, and Xe ALE of titanium with bro...

Figure 6.14 Top: Schematic illustration of directional ALE with modification...

Figure 6.15 EPC for SiO

2

ALE with C

4

F

8

/Ar

+

for different reactive layer ...

Figure 6.16 EPC with and without C

4

F

8

in removal step and resulting ALE syne...

Figure 6.17 Etch rate and ion flux to the wafer as a function of radius for ...

Figure 6.18 (a) Schematic of EPC for material A (e.g. silicon) and material ...

Figure 6.19 Selectivity of single crystal silicon to thermal silicon oxide a...

Figure 6.20 TEM micrographs of gate oxide before (a) and after (b) ALE etch ...

Figure 6.21 Comparison of gate oxide loss for Cl

2

/Xe ALE and conventional HB...

Figure 6.22 Calculated thickness evolution for ALE of SiO

2

and Si

3

N

4

.

Figure 6.23 Calculated ALE window as a function of incident ion angle for Cl

Figure 6.24 Calculated 2D simulation of 200 cycles of a Cl

2

/Ar ALE of silico...

Figure 6.25 Germanium ALE on patterned test wafer, showing flat etch front a...

Figure 6.26 Surface before and after directional ALE in (a) tilt scanning el...

Figure 6.27 Modeled amorphous layer thickness vs argon ion fluence with chan...

Figure 6.28 EPC and surface roughness as a function of chlorine pressure for...

Figure 6.29 MD simulation results for 200 eV argon ion bombardment of a (111...

Figure 6.30 (a) Schematic of nitride spacer etching directional ALE with H2 ...

Figure 6.31 Profiles resulting from etching the gate structure with a (a–c) ...

Figure 6.32 Self‐aligned contact etch profiles for continuous etching and AL...

Chapter 7

Figure 7.1 Schematic illustration of synchronous and asynchronous plasma pul...

Figure 7.2 Total sputtering yield of silicon by reactive and nonreactive ion...

Figure 7.3 Etching yield of polysilicon as a function of ion incident angle....

Figure 7.4 Etching and sputtering yields of SiO

2

as a function of angle of i...

Figure 7.5 Calculated thicknesses of the reaction layer as a function of inc...

Figure 7.6 Calculated product yields of SiCl

x

(

x

 = 0, 1, 2, 3, 4) species as...

Figure 7.7 Ion‐enhanced polysilicon etching by chlorine atoms and ions (a) a...

Figure 7.8 Illustration of neutral and ion‐limited process regimes.

Figure 7.9 Graphical illustration of the ion‐enhanced polysilicon etching wi...

Figure 7.10 Etching rates of n‐doped and undoped polysilicon for HBr, Cl

2

, N...

Figure 7.11 Silicon oxide etch rates in CHF

3

and C

2

F

4

plasmas as a function ...

Figure 7.12 Etch rates of SiO

2

, Si

3

N

4

, and silicon samples plotted vs the th...

Figure 7.13 Schematic illustration of the origin and effect of local surface...

Figure 7.14 Effect of trench aspect ratio on percentage of ions that impact ...

Figure 7.15 Conceptual etching selectivity framework.

Figure 7.16 Lifetime of gate oxide in an ICP plasma as a function of silicon...

Figure 7.17 Sidewall passivation mechanisms in RIE.

Figure 7.18 Sidewall passivation for silicon etching with Cl

2

/O

2

and HBr/O

2

....

Figure 7.19 Sidewall passivation for silicon etching with HBr/Cl

2

/O

2

compari...

Figure 7.20 Sidewall passivation for silicon etching with HBr/Cl

2

/O

2

and HBr...

Figure 7.21 Illustration of the role of different halogens on RIE profiles....

Figure 7.22 (a) Temperature dependence of the etching rate of silicon and ph...

Figure 7.23 Schematic illustration of the origin of CD bias with and without...

Figure 7.24 CD bias and microloading trends with and without sidewall passiv...

Figure 7.25 CD microloading for polysilicon gate etch with HBr/Cl

2

/O

2

and HB...

Figure 7.26 Silicon surface roughness measured AFM for directional ALE with ...

Figure 7.27 LER/LWR features and frequency of patterned photoresist.

Figure 7.28 CD microloading/LWR trade‐off can be overcome by separating phot...

Figure 7.29 Out‐of‐plane deflection

w

of a buckled TiN‐capped ILD fin over h...

Figure 7.30 Contours of the minimum critical buckling stress Σ

TiN

from 0.5 t...

Figure 7.31 Number of etching steps needed to generate the final mask for de...

Figure 7.32 Schematic illustration of self‐aligned double patterning with po...

Figure 7.33 Schematic illustration of self‐aligned double patterning with ne...

Figure 7.34 Schematic illustration of cross‐SADP for hole structures. (a) Sq...

Figure 7.35 Schematic illustration of positive tone self‐aligned quadruple p...

Figure 7.36 Modeling of silicon etching in a chlorine plasma: (a) convective...

Figure 7.37 Schematic illustrations of planar, FinFET, and GAA transistors....

Figure 7.38 Number of argon and chlorine atoms in the fin after fin etching ...

Figure 7.39 Implantation of silicon with 500 eV argon (A) and 300 eV hydroge...

Figure 7.40 Etching of HfO

2

at room temperature with a BCl

3

plasma.

Figure 7.41 Schematic illustration of the evolution of transistor contacts....

Figure 7.42 Simplified process flow illustrating (a) “via‐first” and (b) “tr...

Figure 7.43 Fluxes and powers to the etch front as a function of aspect rati...

Figure 7.44 Angular distribution of scattered ions.

Figure 7.45 Energy distribution of scattered ions.

Figure 7.46 Decreasing etch rate with increasing aspect ratio for high aspec...

Figure 7.47 Simulated etch profiles on the top of high aspect ratio SiO

2

fea...

Figure 7.48 Effect of neutral depositor flux upon simulated etch profiles. C...

Figure 7.49 Necking and bowing of the simulated etch profiles vs variations ...

Figure 7.50 Erosion of carbon and silicon mask for 1 keV argon ion bombardme...

Figure 7.51 Erosion of carbon and silicon mask for 1 keV argon ion bombardme...

Figure 7.52 Definition of hole twisting and distortion.

Figure 7.53 High aspect ratio bottom hole distortion and twisting.

Figure 7.54 Calculated ion flux distributions at the hole bottom for axisymm...

Figure 7.55 Twisting starting depth

d

as a function of minimum necking width...

Figure 7.56 AFM measurements of high aspect ratio trench sidewall roughness ...

Figure 7.57 Twisting depth as a function of necking width and ion energy....

Figure 7.58 Normalized O

2

+

and CF

+

ion currents as a function of asp...

Figure 7.59 Current across contact holes for a 20 V voltage with and without...

Figure 7.60 Horizontal slices through the final etching profiles with overet...

Figure 7.61 3D NAND memory architecture and critical processing steps.

Figure 7.62 Effect of layer to layer selectivity on 3D NAND channel hole etc...

Figure 7.63 Schematic illustration of programming and reading of PCM cells....

Figure 7.64 Cross‐point implementation of PCM.

Figure 7.65 Fractional crystallization vs laser pulse heating time for nitro...

Figure 7.66 Cross‐sectional images and the TEM/EDX images of the etched side...

Figure 7.67 Oxygen concentration at the surface of a cleaned GST sample as a...

Figure 7.68 GST oxidation in air as function of humidity during the first 30...

Figure 7.69 Resistance as a function of temperature in 100 nm thick Ge

2

Sb

2

Te

Chapter 8

Figure 8.1 Comparison of MRAM profiles etched with RIE (a) and IBE (b).

Figure 8.2 Simulated elemental mapping of intermixing after IBE for 1000 eV ...

Chapter 9

Figure 9.1 Electron energy distribution function (EEDF) for a nonequilibrium...

Figure 9.2 Sheath formation in a bounded plasma. (a) Electrons escape the pl...

Figure 9.3 Plasma chamber with external RF voltage. (a) Most negative part o...

Figure 9.4 Time‐dependent sheath voltage for resistive and capacitive sheath...

Figure 9.5 Energy distributions of ions extracted through the ground plane f...

Figure 9.6 PIC simulation of ion energy distributions for a capacitively cou...

Figure 9.7 Energy distributions of H

3

+

, H

2

O

+

, and Eu

+

for 13.56 ...

Figure 9.8 Measured ion energy distribution for a capacitively coupled CF

4

p...

Figure 9.9 Full width ion energy dispersions as a function of the atomic mas...

Figure 9.10 Dependence of the ion energy dispersion

ΔE

i

on RF frequency....

Figure 9.11 Generalized sheath diagram for time‐dependent sheaths.

Figure 9.12 Equivalent circuit in capacitively coupled two‐electrode RF plas...

Figure 9.13 Time‐averaged potentials on the powered and grounded electrodes ...

Figure 9.14 Cathode and anode areas for a generic CCP RIE reactor.

Figure 9.15 Bias voltage as a function of RF frequency for a capacitively co...

Figure 9.16 Electron density dependence on RF frequency for the experiments ...

Figure 9.17 Electron density dependence on RF frequency for pressures of 50,...

Figure 9.18 Calculated IEDs as a function of frequency mixing for an argon p...

Figure 9.19 Schematic illustration of (a) ohmic and (b) stochastic heating....

Figure 9.20 Schematic illustration of electric and magnetic fields in of ICP...

Figure 9.21 Schematic illustration of typical ICP/TCP reactors. (a) Reactor ...

Figure 9.22 Schematic illustration of the effect of tailored bias waveforms ...

Figure 9.23 Two‐peak tailored bias waveforms (a) and corresponding IEDs (b)....

Figure 9.24 Schematic illustration of IED for combined bias pulsing and TWB....

Figure 9.25 Evolution of species wall fluxes predicted by a model of pulsed ...

Figure 9.26 Ion energy distribution functions in an argon microwave plasma w...

Figure 9.27 SEM cross‐sectional polysilicon gate profiles etched in continuo...

Figure 9.28 Comparison of an RIE or directional ALE reactor (a), an ion beam...

Figure 9.29 Schematic illustration of an ion beam source with three grids.

Chapter 10

Figure 10.1 Electron‐assisted etching of Si

3

N

4

(a) and SiO

2

(b) using 1500 eV...

Figure 10.2 Etched depth of GaAs vs number of cycles for ALE with surface mo...

Figure 10.3 EPC of GaAs for ALE with chlorine gas and 248 nm excimer laser r...

Guide

Cover

Table of Contents

Begin Reading

Pages

iii

iv

ix

x

xi

xii

xiii

xiv

1

2

3

4

5

7

8

9

10

11

12

13

14

15

16

17

18

19

20

21

22

23

24

25

26

27

28

29

30

31

32

33

34

35

36

37

38

39

40

41

42

43

44

45

46

47

48

49

50

51

52

53

54

55

56

57

58

59

60

61

62

63

64

65

66

67

68

69

70

71

72

73

74

75

76

77

78

79

80

81

82

83

85

86

87

88

89

91

92

93

94

95

96

97

98

99

100

101

102

103

104

105

106

107

108

109

110

111

112

113

114

115

116

117

118

119

120

121

122

123

124

125

126

127

128

129

130

131

132

133

134

135

136

137

138

139

140

141

142

143

144

145

146

147

148

149

150

151

152

153

154

155

156

157

158

159

160

161

162

163

164

165

166

167

168

169

170

171

172

173

174

175

176

177

178

179

180

181

182

183

184

185

186

187

188

189

190

191

192

193

194

195

196

197

198

199

200

201

202

203

204

205

206

207

208

209

210

211

212

213

214

215

216

217

218

219

220

221

222

223

224

225

226

227

228

229

231

232

233

234

235

236

237

238

239

240

241

242

243

244

245

246

247

248

249

250

251

252

253

254

255

256

257

258

259

260

261

262

263

264

265

266

267

268

269

271

272

273

274

275

276

277

278

279

280

281

282

283

284

285

Atomic Layer Processing

Semiconductor Dry Etching Technology

Thorsten Lill

Author

Dr. Thorsten Lill

Lam Research Corporation

VP Emerging Etch Technologies & Systems

4400 Cushing Parkway

94538 Freemont, CA

United States

All books published by Wiley-VCH are carefully produced. Nevertheless, authors, editors, and publisher do not warrant the information contained in these books, including this book, to be free of errors. Readers are advised to keep in mind that statements, data, illustrations, procedural details or other items may inadvertently be inaccurate.

Library of Congress Card No.:

applied for

British Library Cataloguing-in-Publication Data

A catalogue record for this book is available from the British Library.

Bibliographic information published by the Deutsche Nationalbibliothek

The Deutsche Nationalbibliothek lists this publication in the Deutsche Nationalbibliografie; detailed bibliographic data are available on the Internet at <http://dnb.d-nb.de>.

© 2021 WILEY-VCH GmbH, Boschstr. 12, 69469 Weinheim, Germany

All rights reserved (including those of translation into other languages). No part of this book may be reproduced in any form – by photoprinting, microfilm, or any other means – nor transmitted or translated into a machine language without written permission from the publishers. Registered names, trademarks, etc. used in this book, even when not specifically marked as such, are not to be considered unprotected by law.

Print ISBN: 978-3-527-34668-4

ePDF ISBN: 978-3-527-82418-2

ePub ISBN: 978-3-527-82420-5

oBook ISBN: 978-3-527-82419-9

Printed on acid-free paper

List of Abbreviations

Symbols

A

surface area and parameter in DG model

AR

aspect ratio

B

parameter in DG model

c

concentration

C

capacitance

CD

critical dimension

D

diffusion coefficient

d

depth, thickness

DC

direct current

dc

duty cycle

E

energy and Young’s modulus

electric field

EPC

etch per cycle

EPE

edge placement error

ER

etching rate

ERNU

etching rate non‐uniformity

G

°

standard Gibbs free energy

GPC

growth per cycle

magnetic field

H

°

standard enthalpy

h

height

h

G

gas phase transport coefficient

I

current

J

particle flux

K

transmission probability

k

constant, coefficient: for instance reaction rate or sputtering coefficient

M

atomic mass

N

number. For instance: number of molecules, number of adsorbed surface sites, etc.

n

density (for gases)

r

distance between atoms or radius

R

reaction rate

RIE

reactive ion etching

R

p

,

Δ

R

p

projected range and straggle

S

etching synergy

S

°

standard entropy

s

sticking coefficient

SR

sputtering rate

T

temperature

t

time

V

voltage or electric potential

v

velocity

V

LJ

Lennard–Jones potential energy

w

width

X

reactance

Z

atomic number

Greek Symbols

θ

angle with respect to surface normal

σ

cross section

ω

circular frequency

τ

characteristic time

κ

dielectric constant

Δ

difference

ε

energy difference, for instance depth of potential energy well

α

,

β

etch amount in step A and B of an ALE process

Σ

film stress

ΔΦ

Mott

Mott potential (eV)

Γ

sputtering yield

Θ

surface coverage

ν

volume

λ

wavelength

Subscripts

0

denotes initial value

a

activation

A

adsorption, adsorbate

b

bottom

B

bias

c

capacitive

ca

cathode

col

collision

D

desorption

DC

direct current

dense

dense features

diff

diffusion

diss

dissociation

e

electron

G

gas

i

ion

iso

isolated features

im

impact

in

incoming

iz

ionization

kin

kinetic

M

maximum

m

minimum

n

neutrals

ox

oxidation

p

plasma

RF

radiofrequency

S

surface

sol

solution

sh

sheath

sp

sputtering

sw

sidewall

t

top

th

threshold

out

outgoing

ox

oxide, oxidation

w

wall

Acronyms

AC

alternating current

AFM

atomic force microscopy

ALE

atomic layer etching

ALD

atomic layer deposition

AR

aspect ratio

ARDE

aspect ratio dependent etching

BARC

bottom antireflective coating

BCA

binary collision approximation

BEOL

back end of line

BPS

bounded plasma system

BPSG

boron phosphorous silicon glass

BST

barium strontium titanate: Ba

1−

x

Sr

x

TiO

3

CAIBE

chemically assisted ion beam etching

CBRAM

conductive bridge random access memory

CCP

capacitively coupled plasma

CD

critical dimension

CDE

chemical downstream etching

CFSTR

continuous flow stirred tank reactor

CM

Cabrera–Mott oxidation model

CMOS

complementary metal–oxide–semiconductor MOSFET fabrication process

CMP

chemical mechanical polishing

CVD

chemical vapor deposition

DARC

dielectric antireflective coating

DC

direct current

DFT

density functional theory

DG

Deal–Grove oxidation model

DMAC

dimethyl aluminum chloride

DRAM

dynamic random access memory

ECP

electro copper plating

ECR

electron cyclotron resonance

ESC

electrostatic chuck

FEOL

front end of line

FeRAM

ferroelectric random access memory

FET

field effect transistor

FG

floating gate flash device

FinFET

fin field effect transistor

FTIR

Fourier transform infrared spectroscopy

GAA

gate‐all‐around (transistors)

GST

phase change material comprised of germanium, antimonium, and tellurium

HPEM

hybrid plasma equipment model

IAD

ion angular distribution

IBE

ion beam etching

ICP

inductively coupled plasma

IED

ion energy distribution

IIP

ion‐ion plasma

ILD

inter‐layer dielectric

LEIS

low energy ion spectroscopy

LELE

Litho–Etch–Litho–Etch multipatterning

LER

line edge roughness

LWR

line width roughness

LSS

Lindhard, Scharff, and Schiott theory

MD

molecular dynamics

MEMS

micro‐electromechanical systems

MEOL

mid end of line

MMP

mixed Mode Pulsing

MRAM

magnetic random access memory

MOSFET

metal oxide semiconductor field effect transistor

NAND

logic gate with “false” output if all inputs are “true.” This type of logic gates is used in flash memory devices. 3D NAND is an implementation of flash memory devices where the gates are stacked in the third dimension inside tall vertical channels

ONON

oxide–nitride–oxide–nitride 3D NAND

OPOP

oxide–polysilicon–oxide–polysilicon 3D NAND

OxRAM

metal oxide resistive random access memory

PIC

particle‐in‐cell plasma model

PVD

physical vapor deposition

PCM

phase change memory

PSD

power spectral density

PZT

lead zirconate titanate: Pb(Zr

x

Ti

1−

x

)O

3

QCM

quartz crystal microbalance

ReRAM

resistive random access memory

RF

radio frequency

RG

replacement gate flash

RIBE

reactive ion beam etching

RIE

reactive ion etching

SADP

self‐aligned double patterning

SAQP

self‐aligned quadruple patterning

SCM

storage class memory

SE

spectroscopic ellipsometry

SEM

scanning electron microscopy

SIMS

secondary ion mass spectrometry

SIT

sidewall image transfer

SRIM

“stopping and range of ions in matter” program

SOS

spacer‐on‐spacer implementation of self‐aligned quadruple patterning

STI

shallow trench isolation

TCP

transformer coupled plasma

TEM

transmission electron microscopy

TMA

trimethylaluminum

TPD

temperature programmed desorption

TRIM

“transport of ions in matter” program

TSV

though silicon via

TWB

tailored waveform bias

UHV

ultra‐high vacuum

VUV

vacuum ultraviolet light

ZBL

Ziegler, Biersack, and Littmark model

Constants

e

elementary charge: 1.602 176 62 × 10

−19

 C

ε

0

vacuum permittivity: 8.854 187 812 8(13) × 10

−12

 F/m

ε

dielectric constant

k

B

Boltzmann constant

R

universal gas constant

N

A

Avogadro constant

Units

Å

angstrom (length)

C

coulomb (charge)

°C

centigrade (temperature)

deg, °

degree (angle)

eV

electron volt (energy)

F

Farad (standard unit of capacitance in the SI system)

h

hour (time)

K

Kelvin (temperature)

L

Langmuir (surface coverage)

m

meter (length)

min

minute (time)

Pa

pascal (pressure)

s

second (time)

Torr

pressure

1Introduction

People have been scratching, engraving, and carving stone, wood, bones, and other materials since the dawn of time to record information and to create art. These early forms of material removal can possibly be viewed as the origins of etching technology.

The importance of etching throughout history can be illustrated with a few remarkable examples. Hammurabi’s code of law was inscribed into a stone stele at around 1754 BCE and is one of the earliest and influential legal tests. Carved woodblocks were applied to print paper money during the Tang dynasty in China during the second half of the first millennium CE. Michelangelo’s statue of David is an embodiment of the European renaissance. All these etching techniques use physical energy to remove material.

Chemical etching techniques using acids evolved in medieval Europe to decorate armor with greater detail. Selected areas of a surface were covered by soft “maskants,” which could be easily removed with sharp objects and the exposed areas were removed by “etchants.” One of the greatest etchers of all times was Rembrandt who created around 290 prints. Many of his etching plates still survive.

John Senebier discovered in 1782 that certain resins lost their solubility to turpentine after exposure to light. This allowed to create early forms of photomasks and ultimately led to the development of photographic methods. Paul Eisler invented the printed and etched circuit board in 1936. Etching was also instrumental for the realization of the first integrated circuits by Jack Kilby and Robert Noyce in 1958. The words “etch” and “etching” figure 11 times in Kilby’s seminal US patent 3 138 743 “Miniaturized electronic circuits” (Kilby 1959).

Originally, integrated circuits were etched with wet chemical methods using photoresists as a mask. While these methods can be directional for some single crystal materials and selected etchants, removal of amorphous materials etch proceeds in all directions roughly with the same rate. This kind of etching is also called isotropic. It works only for features where the lateral dimension is much larger than the thickness of the material to be etched. This property is obviously an obstacle for device shrinking. Another drawback of wet etching is the creation of large amounts of toxic waste.

To overcome these challenges, dry plasma etching methods were introduced into the manufacturing of integrated semiconductor devices in the 1980s. When a plasma is in contact with a solid surface a phenomenon called sputtering occurs, which causes material removal. Sputtering was discovered by W.R. Grove in 1852. Physical sputtering with noble gas plasmas was used in the 1960s in the electronics industry. When the wafer is placed on a radio frequency (RF) powered electrode, ions are accelerated, and the sputter rate can be increased to make the method more productive (Coburn and Kay 1972). However, physical sputtering is still too slow to be useful in the manufacturing of semiconductor devices. It also critically lacks the selectivity to mask and stop materials.

Chemistry provided the necessary performance boost. The development of chemical plasma etching started with stripping of photoresists in oxygen RF plasmas (Irving et al. 1971). Soon, fluorine and chlorine plasma were tested to etch a wide range of materials. An increase of the silicon etching rate by a factor of 10–20 was observed when replacing argon with fluoro‐chloro‐hydrocarbon gases (Hosokawa et al. 1974). The term “reactive ion etching” (RIE) was coined in the mid 1970s for etching technologies involving chemically reactive plasmas where the wafer is placed on an RF‐powered electrode. Initially, the mechanism of the etch rate enhancement was not understood even though the benefits were clearly demonstrated in experiments (Bondur 1976). Coburn and Winters found that “the magnitude of the etch rates which are observed are such that the enhancement caused by ion bombardment cannot be easily explained by simply superimposing a physical sputtering process onto the chemical etching process” (Coburn and Winters 1979). Their seminal experiments demonstrated the existence of synergy between the ion and neutral fluxes. Synergy is also a key concept in atomic layer etching (ALE) with atomic layer fidelity. We will use this concept throughout this book.

Production‐worthy etching reactors took hold in the semiconductor industry with the introduction of batch RIE reactors based on developments at Bell Labs. An overview of the evolution plasma etching equipment can be found in a review article by Donnelly and Kornblit (2013). The 1990s saw the introduction of single wafer etching reactors, which improved wafer to wafer repeatability and overall process control. This decade was also the time of search for the best source technology for the large number of rapidly emerging applications. The first single wafer etching reactors were simple parallel plate reactors with RF power applied to the wafer pedestal. Some embodiments featured etch rate enhancing magnetic fields.

High‐density plasmas powered by transformer‐coupled plasma (TCP) or inductively coupled plasma (ICP) established themselves as the tools of choice for silicon and metal etching. Medium‐density capacitively coupled plasma (CCP) sources proved superior for etching of silicon oxide and other dielectric materials. CCP reactors found widespread application with the introduction of damascene metallization in the end of the 1990s, which created a large market for etching of materials with low relative dielectric constant, the so‐called low‐k materials.

The 2000s were the decade of continuous improvement of uniformity across the wafer by means of radial uniformity tuning knobs for ion flux, neutral flux, and temperature. This was driven by the transition from 200 to 300 mm wafers and escalating uniformity requirements to satisfy Moore’s Law. The last decade was characterized by a strong focus on within die and feature scale performance. This is caused by the transition from traditional Moore’s Law scaling to vertical device scaling, which drives devices with increasingly high aspect ratios such as 3D NAND flash and fin field effect transistors (FinFET’s).

One of the solutions to within die performance challenges is “time domain processing,” for instance plasma pulsing, and mixed mode pulsing (MMP) where RF power and gas flows are pulsed. Time domain processing necessitates that all subsystems operate repeatably on second timescales and faster. This is an enormous engineering challenge considering all the parameters that need to be controlled with the large number of process parameters including the radial tuning knobs. Model‐based process controllers and machine learning process development algorithms are being introduced.

As semiconductor devices are shrinking to sub‐10 nm dimension, etching technologies with atomic‐scale fidelity are required. Here fidelity refers to the degree of matching to the intent of design engineers in shape and composition (Kanarik et al. 2015). ALE, which has been studied in laboratories for 30 years, promises to deliver this level of performance. The first report on ALE was published in Yoder’s US patent 4 756 794 entitled “Atomic layer etching” (Yoder 1988). After a first wave of research during the 1990s, a second wave of interest and development started in the mid‐2000s driven by the need for etching technologies with infinite selectivity and the ability to remove controlled amounts of material down to sub‐monolayer resolution.

A variety of etching technologies were discussed under the umbrella of “ALE” including very slow RIE processes, radical and vapor etching. This lack of common understanding and terminology in the etching community slowed the development of true ALE. A definition of ALE as an etching process comprising of at least two self‐limited steps was adopted during a Sematech workshop on ALE in April 2014. This definition is in analogy to its counterpart of atomic layer deposition (ALD). Many of the established concepts in ALD were adopted in ALE. The separation of the etching process into self‐limiting steps breaks the trade‐offs caused in RIE by simultaneous ion and neutral fluxes. The result is improved uniformity across the wafer, across features with different critical dimension called aspect ratio dependent etching (ARDE), and surface smoothness (Kanarik et al. 2015). It also greatly simplifies the process and makes ALE accessible to a rigorous fundamental understanding.

This book covers the latest research and developments of directional and isotropic ALE and puts them into the context of established dry etching technologies for semiconductor devices. In this book, we will introduce etching technologies in the order of increasing complexity. We will begin with critical elementary surface processes, followed by single species etching technologies (thermal etching and radical etching), sequential multi‐species etching (ALE), and multi‐species continuous processing (RIE). Finally, we will review plasmas and other methods to produce the species we discussed in the first half of the book.

This structure does not consider the chronological order of discovery or the size of the market of the various etching technologies. Novel ALE will be studied before classical RIE. Directional ALE is introduced as a simplified embodiment of RIE, which is amiable to a rigorous treatment. Salient RIE properties will be presented as the result of a lack of self‐limitation of continuous processing where all species fluxes are on all the time. The goal is to understand RIE on an atomic level as rigorously as possible to illuminate the “black box” that RIE still is today (Winters et al. 1977; Gottscho et al. 1999).

Specific etching applications such as gate etching, contact etching, or 3D NAND channel hole etching will be introduced as examples for the mechanisms discussed without attempting to give a comprehensive description of the process challenges and solutions. The emergence and evolution of semiconductor devices and the corresponding etching applications is simply too fast paced, and such an attempt would be outdated within a few years. Rather, the intend of this book is to provide an atomic level understanding of all dry etching technologies, which will hopefully help to develop specific solutions for existing and emerging semiconductor devices.

Plasmas are the method of choice to generate ions and radicals used in dry etching. In this book, the plasma and source technologies are covered to a level of detail sufficient enough to understand how they impact the species fluxes to the etching surface. For deeper understanding, we refer to the seminal monography on plasma technology and materials processing is Liebermann’s monography (Lieberman and Lichtenberg 2005).

References

1976 Bondur, J.A. (1976). Dry process technology (reactive ion etching).

J. Vac. Sci. Technol.

13: 1023–1029.

1972 Coburn, J.W. and Kay, E. (1972). Positive‐ion bombardment of substrates in rf diode glow discharge sputtering.

J. Appl. Phys.

43: 4965–4971.

1979 Coburn, J.W. and Winters, H.F. (1979). Ion‐ and electron‐assisted gas‐surface chemistry – an important effect in plasma etching.

J. Appl. Phys.

50: 3189–3196.

2013 Donnelly, V.M. and Kornblit, A. (2013). Plasma etching: yesterday, today, and tomorrow.

J. Vac. Sci. Technol., A

31: 050825 1–48.

1999 Gottscho, R.A., Cooperberg, D., and Vahedi, V. (1999). The black box illuminated.

Workshop on Frontiers in Low Temperature Plasma Diagnostics III (LTPD)

, Saillon, Switzerland.

1974 Hosokawa, N., Matsuzaki, R., and Asamaki, T. (1974). RF sputter‐etching by fluoro‐chloro‐hydrocarbon gases.

Jpn. J. Appl. Phys. Suppl.

2: 435–438.

1971 Irving, S.M., Lemons, K.E., and Bobos, G.E. (1971). Gas plasma vapor etching process. US Patent 3,615,956.

2015 Kanarik, K.J., Lill, T., Hudson, E.A. et al. (2015). Overview of atomic layer etching in the semiconductor industry.

J. Vac. Sci. Technol., A

33: 020802 1–14.

1959 Kilby, J.S. (1959). Miniaturized electronic circuits. US Patent 3,138,743.

2005 Lieberman, M.A. and Lichtenberg, A.J. (2005).

Principles of Plasma Discharges and Materials Processing

, 2e. Wiley.

1977 Winters, H.F., Coburn, J.W., and Kay, E. (1977). Plasma etching – a “pseudo‐black‐box” approach.

J. Appl. Phys.

48: 4973–4983.

1988 Yoder, M.N. (1988). Atomic layer etching. US Patent 4,756,794.

2Fundamentals

2.1 Important Performance Metrics of Etching Processes

Etching is a process of material removal and requires breaking of bonds and removing of atoms from the surface of a solid material. Except for physical sputtering, dry etching processes for semiconductor devices deploy chemical reactions. To understand how the technologies work, it is important to understand the mechanisms and elementary steps. In dry etching, chemically active species are delivered to the surface through the gas phase and adsorb on the surface. It is also possible to bombard the surface with chemically active ions (see Section 7.1.2). The role of the adsorbents is to weaken the bonds of the surface atoms to the bulk material. Energy is provided simultaneously or sequentially to break the bonds and to remove the reaction products. This can be achieved by thermal energy or kinetic energy of ions and atoms as well as by other means such as photons or electrons. Sufficiently high thermal or kinetic energy also leads to pure physical removal such as evaporation and sputtering, respectively. Evaporation and physical sputtering are, however, of limited usefulness in the etching of semiconductor devices.

Figure 2.1 illustrates the effect of adsorption of chemically active species with a schematic representation of a solid surface with adsorbed atoms. The bond energy between adsorbate and surface atoms is denoted with EA, the bond energy between surface and bulk atoms with ES, and between bulk atoms with EO. Adsorbed atoms form bonds with the surface atoms with an energy EA. This weakens the bonds of the surface atoms to the bulk characterized by ES relative to the bonds in the bulk, EO. When sufficient energy is provided, the weakest bonds will break. If these are the bonds characterized by ES, the result is etching. Breaking of the other bonds will result in desorption (in case EA is weakest) or sputtering/evaporation of the bulk material (if EO is the weakest bond).

Figure 2.1 Schematic representation of a surface with adsorbed atoms with the bond energy between adsorbate and surface atoms EA, between surface and bulk atoms ES, and between bulk atoms EO. Condition for etching: ES < EO and EA.

Source: Lill et al. (1994).

This is a greatly simplified, phenomenological representation, which ignores among others the actual structure of the surface and the kinetics of the reactions and activation barriers. The surface may also be amorphous or damaged and intermixed with chemically active species. Reactive layers can extend beyond the first atomic layer of the etching material. In that case, the entire reactive layer is characterized by weak chemical bonds.

The energy that is applied to break bonds can be thermal energy, kinetic energy of ions, chemical energy of atoms, molecules, and radicals, as well as chemical or thermal energy of photons, and electrons. Even the mechanical energy of the tip of an atomic force microscope has been reported to induce the etching of a chemically modified surface (Chen et al. 2018).

In this framework for etching, one or more fluxes of particles and energy sources interact with the surface simultaneously or in sequence to modify the surface and to break bonds of the material to be removed. Depending on what particles and energy sources are used and whether they interact with the surface simultaneously or in sequence, dry etching technologies can be classified into thermal etching, thermal isotropic atomic layer etching (ALE), radical etching, directional or ion‐assisted ALE, reactive ion etching (RIE), and ion beam etching (IBE). We will introduce a classification of etching technologies in Section 2.9.

Etching processes of semiconductor devices form 3D structures that can consist of several different materials. Extensive, application‐specific tables of requirements must be met when etching advanced semiconductor devices. Here is a list of common important etching requirements.

2.1.1 Etching Rate (ER)

This is the most important etching performance parameter. It is expressed as the change in thickness of the etching film as a function of time and is usually measured in nm/min. For a sputtering process, this rate can also be expressed as a function of the sputtering yield Г, which is the ratio of ejected atoms and ions to impinging ions. For chemically enhanced processes such as RIE, the dependence of the etching rate (ER) on the incoming fluxes is much more complex (see Section 7.1).

For ALE, the term etching rate is replaced by the term “etching per cycle” (EPC) expressed in nm or Å because the process is cyclic and removes a well‐defined amount of material every cycle. This approach is analogous to the nomenclature in thin film deposition where continuous technologies such as chemical vapor deposition (CVD) are characterized by a deposition rate while atomic layer deposition (ALD) features “growth per cycle” (GPC).

2.1.2 Etching Rate Nonuniformity (ERNU)

Advanced semiconductor devices are manufactured on wafers with 300 mm diameter. Etching rate uniformity requirements are very stringent. This parameter is expressed as the ratio of the etching rate difference across the wafer divided by the average etching rate as percentage or as the standard deviation. Typically, etching rate nonuniformity (ERNU) is expressed in % of 1 sigma of the standard deviation. ERNU can be introduced by nonuniformities of any of the incoming species fluxes across the wafer or at the extreme edge of the wafer.

2.1.3 Selectivity

Very few etching applications are limited to etching one material only. Masks made of slower etching materials are needed to etch 3D structures such as trenches or holes. Selectivity of material 1 to material 2 is expressed as the ratio of their etching rates ER1/ER2 or their etching amount per cycle EPC1/EPC2. Etching technologies are characterized by an “intrinsic” selectivity. This is the selectivity without deposition sub‐reactions. The intrinsic selectivity is a function of “excess” removal energy. For instance, thermal processes deploy neutral molecules at thermal energies with just about enough energy to break the bonds of the etching material. Because chemical pathways are very different for different materials, etching processes with infinite selectivity can be designed. In contrast, RIE uses ions with energies of several hundred electron‐volt (eV). This is much more than the energy needed to break the critical bonds of the etching material. Bonds of materials that the process should be selective to will also be broken. The result is a finite selectivity based on the etching rate difference. Therefore, depositing species are added to RIE processes to deposit a protective layer where etching is not desired while etching in locations where it is desired.

2.1.4 Profile

The use of ions provides the etching process with directionality or anisotropy. In combination with a selective etching mask, this allows to create vertical or nearly vertical features. The resulting shape of the vertical cross section is called an etching profile. Ideally, this cross section is a square, but various artifacts can lead to profile deformations.

One of the reasons for nonideal profiles is the effect of neutral species that can etch in vertical and horizontal directions. The latter is called isotropic etching and can be suppressed among others by means of reducing the temperature, choosing less reactive neutral species, and by adding inhibiting species that suppress isotropic etching (Coburn 1994). Figure 2.2 shows common conventions for how to characterize an etching profile. These terms are not rigorously defined in the etching community and can vary from organization to organization.

Figure 2.2 Classification of the most common etching profiles.

2.1.5 Critical Dimension (CD)

The term critical dimension (CD) is applied to the space of a trench, the width of a line, or the diameter of a hole. Depending on where the measurement is taken, top and bottom CDs can be distinguished. If the profile is bowed, the largest extent of this profile deformation is called bow CD. Many more specific CD‐related terms can be introduced in a similar fashion. The term CD can also be applied to the incoming resist mask. The difference between the lithography CD and etch CD is called CD bias or ΔCD. The uniformity of the CD bias across the wafer is directly related to ERNU. The repeatability of the CD is a function of the wafer‐to‐wafer and chamber‐to‐chamber repeatability.

2.1.6 Line Width and Edge Roughness (LWR and LER)

Line edge roughness (LER) is the roughness of only one edge and considers the contributions from line bending. Line width roughness (LWR) is defined as the variance of the width of a line. This parameter considers the roughness of both edges of the line. Because line bending impacts both edges of a line, LWR typically does not include it.

2.1.7 Edge Placement Error (EPE)

Edge placement error (EPE) is defined as the relative displacement of the edges of two features from their intended target position. CD bias, LER, and LWR are components of EPE, which also considers lithography overlay and other contributions.

2.1.8 Aspect Ratio‐Dependent Etching (ARDE)

This parameter measures the change in etching rate as the feature evolves and the aspect ratio increases. The aspect ratio (AR) for a feature with vertical sidewall is defined as the ratio of the feature width and depth (AR = w/d). For non‐vertical profiles, the average or minimum width is also used. The root cause of aspect ratio‐dependent etching (ARDE) is the transport of etching species to the etching surface inside the feature. For etching processes with one type of etching species, for instance for low‐pressure radical etching, the etching rate slows down as the aspect ratio increases because transport becomes the limiting step. RIE uses the combined effect of ions and neutrals for etching. Ions and neutral species have different angular distributions and hence the fluxes to the etch front are attenuated to different degrees as the aspect ratio of the feature evolves (Gottscho et al. 1992). We will investigate the root causes for ARDE for each etching technology in Sections, 3.1, 4.2, 5.2, 6.2, 7.2, and 8.1.

2.2 Physisorption and Chemisorption

A chemically modified surface layer is a precondition for chemically enhanced etching (see Figure 2.1). In dry etching, this modified layer is formed by adsorption, specifically chemisorption, diffusion, and ion implantation.

Molecules, atoms, or radicals approach the surface with thermal energy where they bounce off (scatter) or adhere to form bonds and thus convert kinetic into potential energy. When incoming particles form weak bonds, for instance van der Waals and hydrogen bonds, with the surface the process is called physisorption. The adsorption process is classified as chemisorption when covalent bonds are formed. Adsorption processes can be illustrated by drawing the potential energy as a function of the distance between the surface atoms (adsorbent) and the incoming atom or molecule (adsorbate). The potential energy is the sum of attractive and repulsive forces between the adsorbate and adsorbent and can be mathematically expressed, for instance, by the Lennard‐Jones or 12–6 potential:

(2.1)

Here, ɛ is the depth of the potential well, r is the distance between the atoms, and rm the distance at which the potential reaches its minimum. Figure 2.3 shows the potential energy as a function of the distance r between the adsorbate and adsorbent. As the adsorbate molecule is brought closer to the surface, the interaction is first attractive until repulsive forces prevail. The equilibrium distance is reached when the net potential energy is at a minimum. This representation is valid for the formation of chemical bonds in general.

Figure 2.3 Lennard‐Jones potential for adsorbent–adsorbate system.

This concept can be applied to adsorption of chemical species in dry etching: molecules and radicals. The diagram in Figure 2.4 is called Lennard‐Jones diagram for dissociative adsorption of molecules (see Figure 5.1 for adsorption of radicals). Because chemically stable molecules have saturated bonds, they first interact with the surface via long‐ranging van der Waals forces with an equilibrium distance r1. When the molecule is moved closer to the surface, the atoms that form the molecule start to form individual bonds with the surface and as a result, the molecule dissociates. A new equilibrium distance r2 with a lower potential energy characterizes these chemisorbed atoms.

Figure 2.4 Lennard‐Jones diagram for dissociative adsorption.

An energy barrier Ea,A is formed by the overlapping L–J curves for the molecule and the atom. This energy barrier is the activation energy for adsorption. It separates the equilibrium states for physisorption and chemisorption and is the activation barrier for dissociation. With sufficient kinetic energy, which means at high enough temperature, this barrier can be overcome. When the temperature of the surface is high enough, some species can desorb from the surface. Because the molecule is dissociated into atoms, the species that leave the surface leave as atoms.

Frequently used gases in dry etching are, for instance, Cl2, H2, and O2. These gases dissociate into radicals, which are atoms or molecule fragments with unpaired electrons that make them highly reactive. The existence of an activation barrier for dissociative adsorption means that the process is temperature dependent and can be described by an Arrhenius equation:

(2.2)

Here, RA is the adsorption rate, k0 is a rate constant, Ea,A the activation energy for adsorption, R the universal gas constant, and θA the adsorption surface coverage. The universal gas constant equals the Boltzmann constant multiplied by the Avogadro constant: R = kBNA.

2.3 Desorption

Desorption is the process of a molecule or atom leaving a surface. Depending on the nature of the adsorbate–adsorbent bond and the form of energy deployed to break it, a multitude of desorption mechanisms exist. These desorption mechanisms are very important for etching.

In the case of thermal desorption, the desorption rate is a function of the surface temperature. The temperature of a solid state material is a measure of the vibrational energy of the atoms in the solid. The energies of the atoms are not all equal, they follow a distribution. The atomic vibrations are related to sound waves and are called phonons in their quantum mechanical particle representation. Energies of non‐interacting quantum mechanical phonons, for instance, are distributed according to the Bose–Einstein statistics. The implication of the existence of a distribution is that for a given temperature, there is a probability that the energy stored in the vibration of a bond will lead to its breakage. The lower the bond energy, the higher the probability for breaking the bond. Atoms that are bonded to the surface by weak van der Waals or hydrogen bonds (bond energy typically in the 10–100 meV rage) will desorb at much lower temperatures than chemisorbed atoms or molecules, which form stronger covalent bonds to the surface (bond energy typically around 1 eV). This is the reason why physisorption occurs at relatively low temperatures. When the vibrational energy of the atoms is increased by increasing the surface temperature, statistically, the weakest bonds will break first. In homogenous materials all bonds are similar and therefore the bulk material will sublimate at high enough a temperature.

Chemisorption weakens the surface‐to‐bulk bonds. This is because the strength of a covalent bond between atoms is proportional to the probability of the shared electrons to be located between these atoms. Hence, adsorbates that bond to a surface atom weaken its bond to the surface by shifting the electron density. This can lead to the condition shown in Figure 2.1. Increasing the temperature will break the bonds between the surface and bulk atoms at a lower temperature than for the homogeneous bulk material. As a result, only the top layer of the solid material is removed. Thus, chemisorption enables chemically assisted etching via thermal desorption as illustrated by the Lennard‐Jones diagram in Figure 2.5. A new L–J potential curve is added as a dashed line to represent the potential of a newly formed molecule containing at least one surface atom and another surface atom.

Figure 2.5 Lennard‐Jones diagram for thermal etching.

The desorption rate can be described by Arrhenius equation:

(2.3)

Here, RD is the desorption or etching rate, and Ea,D the activation energy for desorption or etching. The activation energy Ea,D is related to the adsorption energy EA and surface energy ES for desorption and thermal etching, respectively. Ea,D considers the kinetics of the reaction and the corresponding energy barriers, while EA and ES represent only the enthalpies of formation without considering activation barriers (see Figure 2.4). This is, of course, a very simplified representation.

2.4 Surface Reactions

Dissociative adsorption is a surface‐mediated reaction, but it must not necessarily lead to removal of surface atoms or etching. For etching to occur, the atoms on the surface must rearrange into molecules that contain at least one surface atom. These newly formed molecules must desorb at lower energies than the adsorbed molecules or atoms themselves. For instance, F2 can physisorb on a silicon surface, and dissociate into chemisorbed F atoms that can form SiFx. SiFx has a weaker bond to the silicon than F. Therefore, in this system, heating of the fluorine‐covered silicon sample leads to the removal of SiF2 and SiF4 (Engstrom et al. 1988).

In Figure 2.5, the transition of the reaction coordinate from the system chemisorbed atom/surface to the reaction product/surface is indicated by an arrow. In the simplest case, this is just a change in what bond is represented. In reality, the transition may involve chemical reactions to rearrange bonds and to form new molecules. In this case, the transition between desorption coordinates involves chemical reactions with their own reaction coordinates and energy barriers. If the energy barrier of the reaction is higher than the barrier for desorption, this reaction will be the rate‐limiting step.

Removal of fluorine from a previously fluorine‐etched silicon surface without losing any additional silicon atoms is a challenge. In this case, yet another surface reaction, for instance, with water vapor, must be used to just remove the fluorine in the form of HF. Multiple surface reactions involving two or more molecules or atoms are also the mechanism of thermal ALE and will be covered in Section 4.1.

The potential for surface reactions to proceed can be predicted by calculating the Gibbs free energy, ΔG°:

(2.4)

Here, ΔH° is the change in enthalpy and represents the heat of reaction. ΔS° is the change in entropy for the reaction and reflects the change in order of the system. The superscript denotes standard temperature and pressure, which are 273.15 K and 105 Pa. The Gibbs free energy of a chemical reaction can be calculated using commercially available software programs.

When ΔG° in Eq. (2.4) is negative, the reaction is thermodynamically favorable and is said to be spontaneous. Spontaneous reactions are needed to realize thermal etching and thermal ALE. Thermodynamic calculations are therefore a first test of the feasibility of a thermal etching reaction. The individual values for ΔH° and ΔS° provide guidance for the temperature behavior of an etching reaction. A negative ΔH° means that the reaction is exothermic. The equilibrium will be shifted in favor of etching at lower temperatures. If ΔS° is negative, the reaction is also more favorable at lower temperatures. Even when an etching reaction is thermodynamically favorable, the reaction kinetics may prevent meaningful etching rates. The reaction pathway may include activation energy barriers that have to be overcome. These barriers determine the reaction rate. The calculation of reaction kinetics is much more complicated.

Surface reactions that involve more than one adsorbate can be classified by the underlying kinetics as shown in Figure 2.6. In the Langmuir–Hinshelwood mechanism, two molecules adsorb on neighboring sites and the adsorbed molecules undergo a reaction. For Elay–Rideal reactions, only one of the molecules adsorbs and the other one reacts with it directly from the gas phase, for instance, by direct impact.

Figure 2.6 Illustration of Langmuir–Hinshelwood (a) and Eley–Rideal (b) surface reaction mechanisms.

2.5 Sputtering

Sputtering is the process of ejecting surface atoms by means of bombardment by energetic (several 10 eV) particles such as ions or fast atoms. Besides etching, sputtering is used in deposition (physical vapor deposition, PVD) and in analytical techniques, for instance, secondary mass spectrometry (SIMS). Sputtering was discovered in 1852 but the underlying mechanisms were not understood until about 100 years later. It was not until the discovery of the so‐called Wehner spots that sputtering was attributed to a sequence of atomic collision processes rather than the result of local evaporation (Wehner 1955).

Sputtering is commonly described by collision cascade theory, which was developed originally as a tool to determine the amount of radiation damage generated by fast neutrons (Sigmund 2012). This kind of a cascade can be theoretically treated using the binary collision approximation (BCA) simulation approach, which calculates the energy transfer as binary elastic collisions between two atoms:

(2.5)

The scattering angle θ is a function of the alignment between the two atoms, called impact parameter, and the interatomic potentials. The implication of Eq. (2.5) is that energy transfer is the most efficient for atoms with similar mass. Very light atoms such as hydrogen do not transfer energy efficiently. They can penetrate deep into the material and create damage deep inside the material. This is important, for instance, when choosing plasmas containing hydrogen, such as HBr for etching. An example is damage below the gate oxide for HBr containing RIE overetches (see Section 7.3.2).

In a collision cascade, if momentum is imparted to a surface atom with an energy more than the surface binding energy, that atom can be ejected from the surface depending on the direction of its momentum vector. The energy of the impinging ion is shared among several recoil atoms. The minimum energy for which atoms are sputtered from the surface is larger than the bond energy, typically about 10 times larger. For instance, the bond energy for silicon is 4.7 eV (Yamamura and Tawara 1996