Atomic Layer Deposition in Energy Conversion Applications E-Book

• Herausgeber: Wiley-VCH Verlag GmbH & Co. KGaA
• Kategorie: Fachliteratur
• Sprache: Englisch

Seitenzahl: 638

Veröffentlichungsjahr: 2017

Beschreibung

Combining the two topics for the first time, this book begins with an introduction to the recent challenges in energy conversion devices from a materials preparation perspective and how they can be overcome by using atomic layer deposition (ALD). By bridging these subjects it helps ALD specialists to understand the requirements within the energy conversion field, and researchers in energy conversion to become acquainted with the opportunities offered by ALD. With its main focus on applications of ALD for photovoltaics, electrochemical energy storage, and photo- and electrochemical devices, this is important reading for materials scientists, surface chemists, electrochemists, electrotechnicians, physicists, and those working in the semiconductor industry.

Leseprobe

Table of Contents

Cover

Title Page

Copyright

Preface

The Past of Energy Conversion

The Future of Energy Conversion

Technical Ingredients Needed

Scope of This Book

Photovoltaics: Strategies, Length Scales, and ALD

Electrochemical Energy Storage: Principles, Chemistries, and ALD

Other Energy Conversion Strategies Based on Interfaces

References

List of Contributors

Part I: Introduction to Atomic Layer Deposition

Chapter 1: Basics of Atomic Layer Deposition: Growth Characteristics and Conformality

1.1 Atomic Layer Deposition

1.2

In Situ

Characterization for Studying ALD Processes

1.3 Conformality of ALD Processes

References

Part II: Atomic Layer Deposition in Photovoltaic Devices

Chapter 2: Atomic Layer Deposition for High-Efficiency Crystalline Silicon Solar Cells

2.1 Introduction to High-Efficiency Crystalline Silicon Solar Cells

2.2 Nanolayers for Surface Passivation of Si Homojunction Solar Cells

2.3 Transparent Conductive Oxides for Si Heterojunction Solar Cells

2.4 Prospects for ALD in Passivating Contacts

2.5 Conclusions and Outlook

References

Chapter 3: ALD for Light Absorption

3.1 Introduction to Solar Light Absorption

3.2 Why ALD for Solar Light Absorbers?

3.3 ALD Processes for Visible and NIR Light Absorbers

3.4 Prospects and Future Challenges

References

Chapter 4: Atomic Layer Deposition for Surface and Interface Engineering in Nanostructured Photovoltaic Devices

4.1 Introduction

4.2 ALD for Improved Nanostructured Solar Cells

4.3 ALD for Photoelectrochemical Devices for Water Splitting

4.4 Prospects and Conclusions

References

Part III: ALD toward Electrochemical Energy Storage

Chapter 5: Atomic Layer Deposition of Electrocatalysts for Use in Fuel Cells and Electrolyzers

5.1 Introduction

5.2 ALD of Pt-Group Metal and Alloy Electrocatalysts

5.3 ALD of Transition Metal Oxide Electrocatalysts

5.4 Summary and Outlook

Acknowledgment

References

Chapter 6: Atomic Layer Deposition for Thin-Film Lithium-Ion Batteries

6.1 Introduction

6.2 Coated Powder Battery Materials by ALD

6.3 Li Chemistry for ALD

6.4 Thin-Film Batteries

6.5 ALD for Solid-State Electrolytes

6.6 ALD for Cathode Materials

6.7 ALD for Anode Materials

6.8 Outlook

Acknowledgments

References

Chapter 7: ALD-Processed Oxides for High-Temperature Fuel Cells

7.1 Brief Description of High-Temperature Fuel Cells

7.2 Thin Layers in SOFC and MCFC Devices

7.3 ALD for SOFC Materials

7.4 Coatings for MCFC Cathodes and Bipolar Plates

7.5 Conclusion and Emerging Topics

References

Part IV: ALD in Photoelectrochemical and Thermoelectric Energy Conversion

Chapter 8: ALD for Photoelectrochemical Water Splitting

8.1 Introduction

8.2 Photoelectrochemical Cell: Principle, Materials, and Improvements

8.3 Interest of ALD for PEC

8.4 Conclusion and Outlook

References

Chapter 9: Atomic Layer Deposition of Thermoelectric Materials

9.1 Introduction

9.2 ALD Processes for Thermoelectrics

9.3 Superlattices for Enhanced Thermoelectric Performance

9.4 Prospects and Future Challenges

References

Index

End User License Agreement

Pages

xi

xii

xiii

xiv

xv

xvi

xvii

xviii

xix

xx

xxi

xxiii

xxiv

xxv

1

3

4

5

6

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

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

84

85

86

87

88

89

90

91

92

93

94

95

96

97

98

99

101

102

103

104

105

106s

107

108

109

110

111

112

113

114

115

116

117

118

119

120

121

122

1123

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

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

183

184

185

186

187

188

189

190

191

192

193

194

195

196

197

198

199

200

201

202

203

204

205

206

209

210

211

212

213

214

215

216

217

218

219

220

221

223

225

226

227

228

229

230

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

259

260

261

262

263

264

265

266

267

268

269

270

271

272

273

274

275

276

277

278

280

281

Guide

Cover

Table of Contents

Preface

Begin Reading

List of Illustrations

Preface

Figure 1 Diagram illustrating the principle of energy harvesting from fossil fuels and highlighting the massive losses inherent to the heat engine.

Figure 2 Diagram illustrating energy harvesting (green), storage (blue), and exploitation (red) in a renewable energy economy. The forms of energy harvesting and storage displayed are treated in this book.

Figure 3 Diagram illustrating the inherent relationship between nanostructuring and the optimization of exchange phenomena as they occur in energy conversion. (a) A nanostructured interface between two phases allows one to engineer a large volumetric density of exchange sites (black dots) at the interface while maintaining short transport distances within each phase (arrow). (b) Bulk phases exhibit lower exchange site densities. (c) The nanostructures cannot be elongated beyond a certain length, because transport distances increase correspondingly and may become limiting.

Figure 4 Diagram illustrating the morphology (in cross section) of the main three “generations” of solar cells and the corresponding requirements on material properties (light absorption coefficient α and carrier diffusion length ℓ) and interface geometry. The requirements on the materials are relaxed in the third-generation approaches by the large specific surface area.

Figure 5 Diagram of the historical development of solar cell efficiency records in the different categories of cells, as compiled by the NREL [3]. Copyright 2016, The National Renewable Energy Laboratory, Golden, CO.

Chapter 1: Basics of Atomic Layer Deposition: Growth Characteristics and Conformality

Figure 1.1 Principle of ALD, illustrated by the process for deposition of Al

2

O

3

using TMA and H

2

O. (Detavernier

et al

. 2011 [5]. Reproduced with permission of Royal Society of Chemistry.)

Figure 1.2 Overview of materials grown by ALD. Growth of pure elements as well as compounds with oxygen, nitrogen, sulfur, selenium, tellurium, fluorine, and other compounds grouped together is indicated through shadings of different types at different positions. (Miikkulainen

et al

. 2013 [4]. American Institute of Physics.)

Figure 1.3 Linear increase of the amount of deposited material as a function of the number of ALD cycles: principle (a: [43]),

in situ

ellipsometry data for Al

2

O

3

ALD (b), and film thickness data for a MeCpPtMe

3

/O

3

process at 150 °C with 2D growth (on a sputtered Pt surface (c)) versus island-type growth on a SiO

2

surface (d) [44], illustrating that GPC values should be interpreted as “apparent” values that depend on the surface conditions (the scale bar on the electron microscopy image is 100 nm). (Reproduced with kind permission of Annelies Delabie.)

Figure 1.4 Saturation during the TMA and H

2

O exposures for ALD of Al

2

O

3

. (Puurunen 2005 [3]. Reproduced with permission of American Institute of Physics.)

Figure 1.5 Effect of substrate temperature on ALD: (a) schematic [43] and (b) a compilation of data for the TMA/H

2

O process by Puruunen [3], illustrating that the GPC may vary as a function of temperature within the “ALD window.” (Puurunen 2005 [3]. Reproduced with permission of American Institute of Physics.)

Figure 1.6 Plasma configurations used for PE-ALD. (a) Radical-enhanced ALD. (b) Direct plasma. (c) Remote plasma. (d) Direct plasma in remote configuration. (Profijt

et al

. 2011 [46]. Reproduced with permission of American Institute of Physics.)

Figure 1.7 Illustration of some of the advantages of (remote) PE-ALD as compared to thermal ALD. (a) Growth per cycle (GPC) as a function of deposition temperature for the growth of Al

2

O

3

using 2 s TMA and 5 s H

2

O or H

2

O plasma. (b) GPC of V

2

O

5

as a function of H

2

O exposure time for thermal and PE-ALD using vanadyl triisopropoxide (VTIP) as V precursor (data taken from [54]). (c) X-ray photoelectron spectroscopy profiles for AlN films grown by thermal and PE-ALD using 2 s TMA and 5 s NH

3

or NH

3

plasma at 250 °C. The plasma power was in all cases 300 W.

Figure 1.8 QCM measurement during TMA/H

2

O ALD, illustrating the impact of temperature effects on the data. Only the “TUNED” data reflects the actual deposition onto the crystal. For the “COLD” and “HOT” traces, apparent mass changes were recorded due to temperature fluctuations during gas pulsing. (Rocklein and George 2003 [58]. Reproduced with permission of American Chemical Society.)

Figure 1.9 Schematic representation of a basic ellipsometer system. (Fujiware 2007 [64]. Reproduced with permission of John Wiley and Sons.)

Figure 1.10 Difference FTIR spectra acquired during the initial cycles of a TMA/H

2

O ALD process.

Figure 1.11 (a) Cross-sectional SEM image of a 300 nm-thick Al

2

O

3

film deposited on a Si wafer with trench structures (Ritala

et al

. 1999 [89]. Reproduced with permission of Wiley.). (b) Cross-sectional TEM of a conformal Ru ALD coating on a trench-patterned substrate (Kim

et al

. 2009 [18]. Reproduced with permission of Wiley.). (c) Cross-sectional Field Emission Scanning Electron Microscopy (FESEM) images of a conformal PtO

x

film deposited on a Si substrate with trench structures. (Hämäläinen

et al

. 2008 [90]. Reproduced with permission for Elsevier.)

Figure 1.12 Film thickness/coverage profiles obtained for ALD in AAO nanopores. (a) Increasing the Zn-precursor exposure time results in an improved ZnO coverage in AAO pores with a diameter of 65 nm (Elam

et al

. 2003 [22]. Reproduced with permission of American Chemical Society.). (b) Wall thickness measured along the length of a HfO

2

nanotube obtained by ALD of HfO

2

into an AAO template followed by dissolution of the template. (Perez

et al

. 2008 [23]. Reproduced with permission of Wiley.)

Figure 1.15 Mean free path of TMA molecules at 200 °C as a function of pressure in the ALD reactor. At all relevant pressures and for all relevant nano- and microstructures, precursor transport is governed by molecular flow. Different approaches have been developed for quantifying the conformality of ALD processes in different pressure regimes: (a) deposition into nano-sized features, followed by cross-sectional electron microscopy, (b) dedicated macroscopic lateral trench structures, as proposed by Dendooven

et al

. for low-pressure processes, and (c) dedicated micrometer-sized lateral trench structures as recently proposed by Puruunen

et al

. for high-pressure processes.

Figure 1.13 Macroscopic test structures that allow for visual inspection of the penetration depth of the ALD material. (a) WN coating inside a fused-silica capillary tube having an inner diameter of 20 µm. (Becker

et al

. 2003 [96]. Reproduced with permission of American Chemical Society.) (b) Schematic representation of a macroscopic hole with rectangular cross section (blue area) that can be disassembled into planar pieces of SiO

2

wafer after ALD. (c) The conformal deposition of TiN on the interior surface of a hole with an AR of 200 : 1 (defined as depth:width). In (d), the yellow coating outside of the cover is due to Ru deposition, while there is no coating inside the marked hole region, illustrating the lack of conformality for this Ru ALD process.

Figure 1.14 Film thickness profiles obtained for Al

2

O

3

ALD in a macroscopic rectangular hole with an AR of 200 : 1. Increasing the TMA exposure time causes the Al

2

O

3

coating to penetrate deeper into the test structure.

Figure 1.16 Aspect ratio (AR) and generalized aspect ratio (

a

) for a cylindrical hole (a) and a trench (b).

Figure 1.17 (a) Schematic representation of the idea behind the model proposed by Gordon

et al

. The resulting coverage profile is characterized by a step. (b) Schematic representation of the model proposed by Dendooven

et al

. The resulting coverage profile is characterized by a slope of decreasing thickness. (c) TMA coverage as a function of depth for a cylindrical hole with an AR of 100 : 1 for simulations with an initial sticking probability

s

0

= 1, 0.1, and 0.001 (

P

= 0.3 Pa,

t

= 5 s,

K

max

= 4.7 × 10

18

m

−2

).

Figure 1.18 Simulated pressure (a) and coverage (b) profiles calculated for a TMA pulse in a rectangular hole with an AR of 66 : 1 (

P

= 0.3 Pa,

t

= 1 s,

K

max

= 4.7 × 10

18

m

−2

). For panel (a), the initial sticking probability was 10%, resulting in a diffusion-limited deposition behavior. For panel (b), the initial sticking probability was 0.1%, resulting in a reaction-limited deposition behavior.

Figure 1.19 (a) SEM image of freestanding TiO

2

/Ni/TiO

2

nanotubes obtained via ALD in an AAO template. (Daub

et al

. 2007 [121]. Reproduced with permission of American Institute of Physics.) (b) Top view SEM image of triple coaxial HfO

2

nanotubes obtained by using AAO and Al

2

O

3

ALD layers as template and spacer layers, respectively. (Gu 2010 [122]. Reproduced with permission of American Chemical Society.)

Figure 1.20 (a) SEM image of an opal film. (Karuturi

et al

. 2010 [138]. Reproduced with permission of American Chemical Society.) (b) SEM image of a TiO

2

inverse opal synthesized by ALD. (King

et al

. 2005 [132]. Reproduced with permission of Wiley.)

Figure 1.21 TiO

2

ALD in mesoporous silica thin films. (a) Schematic representation of the nanoslab-based mesoporous films. (b) Ti XRF intensity as a function of the TDMAT exposure time on a 115-nm-thick film with about 6.5 nm pores and about 75% porosity. (c) Pore radius distribution calculated from

in situ

EP data measured every 10 ALD cycles on a 150-nm thick film with about 18 nm pores and about 80% porosity. (d) Ti XRF intensity against the number of ALD cycles on a 120-nm film with about 7.5 nm pores and about 75% porosity and on a planar SiO

2

substrate. (e) Electron tomography study of the TiO

2

-coated film in (d): out-of-plane orthoslice through the 3D reconstruction of a micropillar sample. Dark gray, silica; light gray, TiO

2

; and black (arrows), voids. (f) TEM image of a cross-sectional sample of the TiO

2

-coated silica film (i), Si energy-filtered TEM map of this region (ii), and Ti energy-filtered TEM map of this region (iii).

Figure 1.22 TiO

2

ALD in a microporous silica thin film with a porosity of about 40% and a thickness of about 80 nm. (a) Size of the TDMAT molecule as compared to the average pore size. (b) Ti XRF intensity against the number of ALD cycles on the microporous film and a planar SiO

2

substrate.

Chapter 2: Atomic Layer Deposition for High-Efficiency Crystalline Silicon Solar Cells

Figure 2.1 Various silicon solar cell concepts and their current-record efficiencies (January 2016). Note that these efficiencies might be several percent absolutely lower when produced industrially. The first class of cells is based on homojunctions, the second class on heterojunctions. The functional thin films, which could be prepared by ALD, are indicated in bold.

Figure 2.2 Schematic band diagram of the passivated silicon surface, indicating the generation of excess carriers by the absorption of light and the recombination thereof, which can take place directly, via a third carrier (in the Auger process), or via defect states (SRH recombination). At the surface, a high density of interface (defect) states

D

it

exists. A passivation layer reduces the

D

it

, and (in this example) reduces the electron concentration at the surface via a negative fixed charge density

Q

f

, which induces a space-charge region and upward band bending.

Figure 2.3 Schematic overview of typical interface defect densities and fixed charge densities of passivation schemes, adapted from Ref. [30]. Note that the actual interface properties might strongly depend on the processing conditions of the passivation layer.

Figure 2.4 The effects of the fixed charge density and the surface doping concentration on the surface saturation current density for a fixed level of chemical passivation. The results are evaluated using Equation (2.2) for a single defect, with

S

n0

=

S

p0

= 5000 cm s

−1

. The carrier densities are derived from the Girisch algorithm [31] using Fermi–Dirac statistics,

n

i

= 9.65 × 10

9

cm

−3

[32], a base injection level of Δ

n

= 1 × 10

15

cm

−3

, and base doping level of

N

base

= 1 × 10

15

cm

−3

. For the band-gap narrowing of

n

- and

p

-type Si, the empirical models of Yan and Cuevas are used [33, 34].

Figure 2.5 Interface properties as determined by corona oxide characterization of semiconductors (COCOS) of

n

-type Si (100) passivated by Al

2

O

3

films prepared by plasma (i.e., using O

2

plasma) or thermal ALD (i.e., using H

2

O) in as-deposited state and after postdeposition annealing at 425 °C in N

2

ambient. (a) After annealing, the

D

it

is significantly reduced for both deposition methods. (b) The

D

it

of the Si/Al

2

O

3

interface is particularly reduced near mid-gap. Nandakumar of National University of Singapore is acknowledged for the COCOS measurements.

Figure 2.6 High-resolution TEM image of the annealed Al

2

O

3

layer prepared by plasma ALD on Si, showing the amorphous character of the Al

2

O

3

layer. The SiO

2

interface layer has a thickness of 1.5 nm. (Hoex

et al

. 2006 [42]. Reproduced with permission of American Institute of Physics.)

Figure 2.7 Optical microscopy image of blisters formed in an Al

2

O

3

/SiN

x

film after firing at 850 °C. The film stack passivates a Si surface with a random-pyramid texture. (Bordihn

et al

. 2014 [57]. Reproduced with permission of American Institute of Physics.)

Figure 2.8 Schematic representation of batch ALD (a) and spatial ALD (b). (Delft

et al

. 2012 [78]. Reproduced with permission of Institute of Physics.)

Figure 2.9 Comparison between the passivation qualities in terms of upper limit of

S

eff

or SRV (where a low

S

eff

is preferred) of Al

2

O

3

passivation layers prepared by spatial ALD, PECVD, and sputtering, before and after firing at 800 °C. (Schmidt

et al

. 2010 [82]. Reproduced with permission of EU PVSEC. Acknowledged by Katrin Aust.)

Figure 2.10 Schematic display of the SiO

2

/Al

2

O

3

passivation scheme (a). Passivation results obtained using SiO

2

/Al

2

O

3

stacks on an

n

+

Si surface (having

N

s

= 2 × 10

20

cm

−3

) on a symmetrical lifetime sample (b). (Van De Loo

et al

. 2015 [11]. Reproduced with permission of Elsevier.)

Figure 2.11 SEM images of different approaches that are used to enhance the light absorption in Si solar cells. In (a), the current industrial standard method for monocrystalline Si is used, where random pyramids are created by wet chemical etching of the Si. On the RP-textured surface, subsequently, the passivation layer and ARC are deposited. In (b), a “black Si” texture is shown, which is formed by reactive ion etching and subsequently is passivated by a 30-nm Al

2

O

3

film prepared by plasma ALD (the authors would like to acknowledge Ingenito from Delft University of Technology for the black Si synthesis). (c) A cross section of a TiO

2

resonator is shown. The TiO

2

is deposited by electron beam evaporation on a 30 nm Al

2

O

3

film prepared by plasma ALD, which enables the surface passivation. (Panel (c): Spinelli

et al

. 2013 [88]. Reproduced with permission of American Institute of Physics.)

Figure 2.12 Bright-field TEM image of black Si pillars shown in Figure 2.11b. A conformal Al

2

O

3

layer of 30 nm thickness is deposited by plasma ALD.

Figure 2.13 Contour plots of the simulated

J

sc

(in mA cm

−2

) of SHJ solar cells featuring a ZnO-based (a) or In

2

O

3

-based (b) TCO as a function of carrier density and mobility. Dashed lines are isolines of constant resistivity. The simulated cell structure consists of a textured wafer with 5 nm of intrinsic a-Si:H, 10 nm of

p

-type a-Si:H and 75 nm of TCO on the front side of the solar cell. Photon currents were simulated using OPAL2 [108], and the optical constants of In

2

O

3

and ZnO were taken from ellipsometry measurements [106, 109]. In order to calculate the effect of free-carrier absorption for varying carrier density and mobility, the contribution of the Drude oscillator ϵ

Drude

to the modeled dielectric function was varied accordingly. The interband absorption was assumed to remain constant, together with a constant

m*

of 0.23

m

e

for In

2

O

3

and 0.4

m

e

for ZnO, respectively. The thick solid line is the mobility limit as a function of the carrier density. For ZnO, this was calculated using the Masetti

et al.

model [110] with the most recent parameters of [111]. The mobility limit for In

2

O

3

was calculated using the mobility limits due to both phonon and ionized impurity scattering [109, 112]. For ZnO, literature values (found in Table 2.5) of ALD films with various dopant atoms are displayed, as well as ZnO:B obtained by low-pressure CVD and ZnO:Al by expanding thermal plasma CVD [113]. For In

2

O

3

, values for amorphous (a-In

2

O

3

:H) and crystallized indium oxide(c-In

2

O

3

:H) (prepared by ALD [109]), ITO [106], IMO [114], and amorphous IZO [115] (prepared by sputtering) and IWO (prepared by reactive plasma deposition (RPD) [116] and pulsed laser deposition (PLD) [117]) are shown. ALD processes are denoted in bold [118].

Figure 2.14 (a) Absorption coefficient of Al-doped ZnO layers of varying carrier density prepared by thermal ALD at 200 °C using DEZ and DMAI as Zn and Al source, respectively. The increasing Drude contribution and Burstein–Moss (BM) shift with doping level are indicated. (b) Schematic band diagram of an unintentionally doped TCO (a) and an intentionally doped TCO (b). The electron affinity χ denotes the energetic distance from the conduction band edge to the vacuum level, whereas the work function ϕ is the distance from the Fermi level

E

F

to the vacuum.

Figure 2.15 Schematic representation of the ALD supercycle principle. In ALD supercycles used for doping of ZnO,

n

cycles of the ZnO ALD process are followed by 1 cycle of a process containing a dopant element (Al, B, Ga, etc.). This supercycle is repeated until the desired film thickness is reached. (Adapted from Ref. [46].)

Figure 2.16 (a) Schematic representation of the superstructure obtained when using ALD supercycles. The cycle ratio

n

controls the

vertical

spacing of the dopants. (b) Demonstration of the accurate control over the carrier density in ZnO that is possible using the supercycle approach. The process used employs DMAI as dopant precursor and DEZ as Zn precursor, at a deposition temperature of 200 °C.

Figure 2.17 Schematic of the correlation between dopant clustering and doping efficiency. One approach for reduced dopant clustering is the use of dopant precursors with bulky ligands that lead to steric hindrance, as shown for TMA and DMAI in the figure.

Figure 2.18 Schematic band diagrams of various approaches for obtaining carrier selective contacts. All band diagrams refer to the situation under illumination and are not drawn to scale. (a) A conventional electron-selective contact made by

n

+

-type doping. (b) An electron-selective contact obtained by the TOPCon concept, consisting of a tunnel oxide and a thin (partially) crystalline

n

+

-Si film. (c) An electron-selective contact that is realized through

band alignment

of the Si with a metal oxide film. (d) A hole-selective contact that is realized through

induced band bending

by a high-work-function metal oxide film. Often concepts (c) and (d) also employ separate ultrathin passivation layers, but these were not drawn for the sake of simplicity.

Figure 2.19 Schematic of the band offsets of a selection of oxides with Si. Offsets are denoted in electronvolt. (Adapted from Refs [168] and [169].)

Figure 2.20 Contour plot of the calculated upper limit of efficiency of a solar cell featuring a full-area passivating contact, as a function of

J

0

and ρ

c

. The calculation is done similarly as in Refs [173] and [174] and assumes no other recombination channels (surface nor bulk), no shunting, and no optical losses (i.e., a

J

sc

of 44 mA cm

−2

). For comparison, data points for various reported structures/cells are shown, along with the efficiencies of the full devices. These include an SHJ cell of Kaneka (2015, private communication) [14], the TOPCon concept of F-ISE [3, 175], the SiO

2

/ITO stack of NREL [173], the TiO

2

cell of ANU [176], the UNSW PERL [177], a

p

-type PERC cell [177], the IMEC nPERT [178, 179], the ECN nPasha (2015, private communication) [178], and a

p

-type Si/MoO

x

contact [180]. Hole-selective contacts are denoted by star-shaped symbols, whereas electron-selective contacts are denoted by circular symbols. Concepts employing a full-area rear contact are noted in bold. For the PERL cell, the

J

0

was estimated using the reported surface recombination velocity in Ref. [177] and case 3 in Ref. [24]. For solar cell concepts that use a partial rear contact, the

J

0

and ρ

c

values have been corrected for the contact area fraction

f

.

Figure 2.21 Schematic showing the current flow pattern in (a) a locally contacted cell and (b) a solar cell with a full-area rear contact. (Adapted from Ref. [3].)

Figure 2.22 False-colored cross-sectional TEM image of a stack of amorphous silicon, ALD MoO

x

, and crystallized ALD hydrogen-doped indium oxide (c-In

2

O

3

:H). (Adapted from Ref. [20].)

Chapter 3: ALD for Light Absorption

Figure 3.1 The spectral power of our Sun incident upon the Earth at 37° latitude (red). The approximate solar power absorbed by a 100-micron-thick silicon wafer after accounting for reflectivity and absorption coefficient is overlaid (green). The power extracted as electricity (gray) is estimated from perfect photon-to-electron efficiency but is reduced by the modest voltage of Si photovoltaics (0.64 V) relative to the absorbed photon energy (4.5–1.1 eV).

Figure 3.2 Shockley–Queisser single-junction efficiency limit, with record photovoltaic efficiency plotted against band gap. Materials for which ALD processes have been reported are highlighted in blue. (Record efficiencies compiled from Green, Dasgupta, and references therein.)

Figure 3.3 Band-gap variation of ALD PbS grown on SiO

2

. Band gaps calculated by an effective mass model are shown as a dotted line. By tuning the size of quantum-confined nanoparticles, the wavelength of light absorbed may be precisely tuned. (Dasgupta 2009. Reproduced with permission of American Chemical Society.)

Figure 3.4 The volume of absorber in a projected area may be readily increased without increasing film thickness by performing ALD over a transparent template with large aspect ratio. (a) A “folded junction” geometry may increase the volume of absorber per projected area without increasing film thickness. (b) The same ALD Fe

2

O

3

film on flat glass and folded into anodic aluminum oxide membranes (aspect ratio >100).

Figure 3.5 Worldwide production of some elements used in solar absorbers plotted on a logarithmic scale.

Figure 3.6 Predicted carrier concentration and conductivity of Cu

2−

x

S as a function of Cu deficiency compared to literature data. ALD of Cu

2

S using a Cu(I) precursor produces a highly stoichiometric semiconductor as evidenced by its relatively low carrier concentration. (Martinson

et al

. [13]. Reproduced with permission of Royal Society of Chemistry.)

Figure 3.7 (a) High-resolution cross-sectional TEM image showing the β-Fe

2

O

3

(001)∥ITO(001) interface produced by the ALD of Fe

2

O

3

. (b–d) Electron diffraction patterns along the [100] axes of β-Fe

2

O

3

, ITO, and YSZ, respectively. (Emery

et al

. [19]. Reproduced with permission of American Chemical Society.)

Figure 3.8 Graphical representation of band gaps for several ALD-grown sulfides relevant to solar absorption. The height of each bar corresponds to the theoretical single-gap efficiency limit for photovoltaics.

Figure 3.9 Transmission electron micrographs of ALD-grown Sb

2

S

3

on nanocrystalline TiO

2

on conductive glass (FTO). The conformal nature of ALD coating is clearly visible. (Wedemeyer

et al

. [5]. Reproduced with permission of Royal Society of Chemistry.)

Chapter 4: Atomic Layer Deposition for Surface and Interface Engineering in Nanostructured Photovoltaic Devices

Figure 4.1 Electron transport and possible recombination processes in a DSSC before (a) and after (b) ALD. (1) Injection of a photoelectron to the TiO

2

from the excited dye. (2) The recombination of an injected electron in TiO

2

with an oxidized dye molecule. (3) The recombination of an electron in TiO

2

with the HTM. (4) The recombination of a collected electron in the transparent conductive oxide (TCO) with a hole in the HTM. The deposition of ultrathin compact and blocking layers via ALD can suppress these types of recombination processes.

Figure 4.2 (a) Schematic illustration of the resistance measurements of compact TiO

2

layers (deposited by different techniques) after coating with Ag paste and vacuum-evaporated Ag. (b) The average resistance between Ag contacts on the compact TiO

2

layers prepared by different methods (the thickness is reported to be ∼50 nm for each sample). (Wu

et al

. 2014 [29]. Reproduced with permission of American Chemical Society.)

Figure 4.3 Morphology and photovoltaic properties. (a) Cross-sectional FESEM image of the ALD TiO

2

at TiO

2

nanorods. (b) Schematic illustration of the device in (a) representing the pore-filling mechanism. (c)

J–V

measurements of the devices fabricated with different ALD TiO

2

thicknesses on TiO

2

nanorods measured under 100 mW cm

−2

of illumination. (d) Respective IPCE spectra recorded in the 300−900 nm wavelength range. (Mali

et al

. 2015 [41]. Reproduced with permission of American Chemical Society.)

Figure 4.4 (a) Schematic illustration of ALD TiO

2

layers postassembly and (b) HR-TEM image of the posttreated TiO

2

:OrgD with 10 cycles of ALD TiO

2

. (Son

et al

. 2013 [68]. Reproduced with permission of American Chemical Society.)

Figure 4.5 Schematic illustration of the different recombination barrier layer configurations in the QDSSC: (a) TiO

2

/Al

2

O

3

/QD and (b) TiO

2

/QD/Al

2

O

3

, from the deposition of the ALD Al

2

O

3

layer before and after the addition of CdS QDs. Spiro-OMeTAD is the hole-transporting material. Arrows indicate the undesirable recombination pathways. Pathways that may be blocked by the Al

2

O

3

barrier layer are shown by dashed arrows. (Roelofs

et al

. 2013 [85]. Reproduced with permission of American Chemical Society.)

Figure 4.6 Schematic illustrations of a DSSC photoanode consisting of (a) a conventional planar TCO and a TiO

2

nanoparticle film. (b) The proposed 3-D TCO-TiO

2

core–shell nanowire array. (Noh

et al

. 2011 [89]. Reproduced with permission of John Wiley and Sons.)

Figure 4.7 (a) Schematic of the TiO

2

double-shell layer. (b and c) TEM images of the TiO

2

double-shell layer on ITO nanowire. (d)

J–V

measurements of a device with the TiO

2

double-shell layer. (Han

et al

. 2013 [43]. Reproduced with permission of Royal Society of Chemistry.)

Figure 4.8 Schematic illustration of conformally coated CNT networks with ALD TiO

2

as electron transport layer. (a) In the inverted organic solar cells, the active materials are PCBM as the light absorber and P3HT as the HTM. (Jin

et al

. 2012 [109]. Reproduced with permission of Elsevier.) In (b), the CNTs are transferred to an FTO substrate and then conformally coated with ALD TiO

2

to use as a stratified photoanode in DSSC. (Yazdani

et al

. 2014 [110]. Reproduced with permission of American Chemical Society.)

Figure 4.9 (a) Photocurrent density versus potential as a function of the length of n-Si NW/TiO

2

arrays: 20 µm (blue line), 10 µm (green line), 5 µm (red line) long NW arrays and n-Si planar/TiO

2

(black line). (b) Relationship between photocurrent density and the length of n-Si NW/TiO

2

, illustrating that longer wire arrays have higher photocurrent. The axis to the right is the current density normalized by the length of the nanowire. (Hwang

et al

. 2009 [111]. Reproduced with permission of American Chemical Society.)

Figure 4.10 Schematic diagram illustrating the light-trapping and charge carrier separation mechanisms for the TiO

2

/nanoFTO architecture. A thin film of ALD TiO

2

(orange) is coated over the nanoFTO network (green). On the left, a region is shown where the TiO

2

film causes some pores within the network to become closed, thus preventing an electrochemically active semiconductor–liquid junction from being within reach of photogenerated holes. (Cordova

et al

. 2015 [119]. Reproduced with permission of Royal Society of Chemistry.)

Chapter 5: Atomic Layer Deposition of Electrocatalysts for Use in Fuel Cells and Electrolyzers

Figure 5.1 Schematic illustration of (a) a proton-exchange membrane fuel cell and (b) a water electrolyzer.

Figure 5.2 Morphology of ALD-derived Pt NPs on different substrates: carbon cloth without acid treatment (a) and with acid treatment for 6 h (b), and CNTs acid-treated for 6 h mixed with 0.06 wt% PTFE (c) and 0.0006 wt% PTFE (d). The cycle numbers were 200 and 300 for carbon cloth and CNTs, respectively. (Liu

et al

. [11]. Reproduced with permission of John Wiley and Sons.) TEM image of ALD-derived Pt NPs on CNTs (e). The inset is a high-resolution TEM image of a Pt NP. (Hsueh

et al

. [16]. Reproduced with permission of Elsevier.)

Figure 5.3 (a) Schematic diagram showing the difference in CNTs with and without O

2

plasma treatment after a complete ALD cycle. (Hsueh

et al

. [15]. Reproduced with permission of Institute of Physics.) (b) A stitched TEM image showing uniform deposition of Pt NPs over the entire length of a 25-µm-long O

2

plasma-treated CNT. The deposition was carried out over a 25-µm-thick vertically aligned CNT array for 200 ALD cycles. (Dameron

et al

. [12]. Reproduced with permission of Elsevier.)

Figure 5.4 Polarization curves obtained with MEAs made of (a) ALD-Pt and (b) commercial Pt (Johnson Matthey) anodes at 40, 60, and 80 °C. (c) Constant-voltage durability test conducted with MEAs made of ALD-Pt and commercial Pt at 0.7 V and 60 °C for 50 h. (Shu

et al

. [13]. Reproduced with permission of Elsevier.)

Figure 5.5 Performance of single PEM fuel cells with both anode and cathode made by ALD at various ALD cycle numbers (a and b) and various amount of CNTs (c and d). The cell made with commercial E-Tek electrodes is included for comparison. (Hsueh

et al

. [16]. Reproduced with permission of Elsevier.)

Figure 5.6 HAADF-STEM images of Pt/GNS samples. (A, B, C) present the results with 50, 100, and 150 ALD cycles, respectively, and (A′, B′, C′) show the corresponding magnified images. Inset in each Figure shows the corresponding histogram of Pt clusters on GNS. (D) Cyclic voltammograms (CVs) of methanol oxidation on various Pt catalysts. The inset is the enlarged CV curves in the onset potential region of methanol oxidation. (E) Chronoamperometry (CA) curves of various Pt catalysts recorded at a constant potential of 0.6 V versus RHE for 20 min. The electrocatalysis tests were conducted at room temperature in an Ar-saturated aqueous solution containing 1 M MeOH and 0.5 M H

2

SO

4

. Label of samples: (a) ALD50Pt/GNS, (b) ALD100Pt/GNS, (c) ALD150Pt/GNS, and (d) Pt/C. (Sun

et al

. [17]. Reproduced with permission of Nature Publishing Group.)

Figure 5.7 (a) Cyclic voltammograms (CVs) of ALD-Pt catalysts with an average particle size of 2.9 nm, recorded in 0.5 M H

2

SO

4

+ 0.5 M ethanol at various scan rates. (Juang

et al

. [19]. Reproduced with permission of Elsevier.) (b) CVs of Pt clusters supported on TiO

2

nanotubes carried out in 1 M KOH and 1 M KOH + 1 M ethanol. Scan rate: 50 mV s

−1

. (Reproduced from Ref. [21] with permission from the Royal Society of Chemistry.) (c) CV profiles of ALD-Pt catalysts measured in 0.5 M H

2

SO

4

+ 0.5 M HCOOH at 10 mV s

−1

. (Hsieh

et al

. [14]. Reproduced with permission of Elsevier.) (d) CVs of different ALD-Pt catalysts measured in 0.5 M H

2

SO

4

+ 0.5 M HCOOH at 50 mV s

−1

. The inset shows CV curves within the potential range of 0–1 V versus Ag/AgCl. (Hsieh

et al

. [20]. Reproduced with permission of Elsevier.)

Figure 5.8 (a) Top-view TEM image of the ALD-Pt/TiSi

2

heteronanostructures after a typical 50-cycle ALD growth; (b) polarization curves of ALD-Pt/TiSi

2

measured in 0.1 M KOH at a scan rate of 10 mV S

−1

at varying rotation rates. (Xie

et al

. [50]. Reproduced with permission of American Chemical Society.) (c, d) TEM images of the ALD-Pt/ZrC catalysts. (e) Polarization curves of electrodes made from ALD-Pt/ZrC, chemically reduced (CW)-Pt/ZrC composites and E-Tek Pt/C catalyst in an O

2

-saturated 0.5 M H

2

SO

4

solution at room temperature (1600 rpm, scan rate: 10 mV s

−1

). (f) Specific and mass activity at 0.9 V versus RHE for these catalysts. (Cheng

et al

. [51]. Reproduced with permission of Royal Society of Chemistry.)

Figure 5.9 (a) SEM image showing ALD-Pt NTs with a thin Nafion® coating layer. Inset: optical photograph of the ALD-Pt NT membrane electrodes. (b) Polarization curves of the ALD-Pt NT (red triangles) and Pt/C (blue squares) electrodes measured in 0.5 M H

2

SO

4

in O

2

-saturated atmosphere. (Galbiati

et al

. [18]. Reproduced with permission of Elsevier.)

Figure 5.10 TEM images of (a) commercial Pd/C and (b) ALD-Pd/C catalysts. The inset shows the histogram of NP sizes. CV curves of (c) ethanol and (d) isopropanol oxidation measured in the electrolyte containing 0.1 M NaOH and 1 M alcohol at 10 mV s

−1

and 1800 rpm. The third CV scan is presented. (e) Chronoamperometric curves measured at 0.7 V versus RHE in 0.1 M NaOH + 1 M alcohol. (Rikkinen

et al

. [24]. Reproduced with permission of American Chemical Society.)

Figure 5.11 (a, b) HAADF STEM images of a 1.8 µm long N-CNT decorated with PtRu catalyst NPs. The arrow points at the CNT growth catalyst (Ni). (c) Anodic scans (1st CV cycle) of the ALD-Pt, -PtRu, and -Ru catalyst measured in 0.5 M H

2

SO

4

+ 1 M MeOH at 10 mV s

−1

. (d) Chronoamperograms of the catalysts measured at 0.4 V versus NHE in the same electrolyte. (Johansson

et al

. [32]. Reproduced with permission of Elsevier.)

Figure 5.12 (a) HAADF-STEM image and (b) EDX mapping of Pd/Pt core/shell NPs (150 cycle Pd; 50 cycle Pt) grown on an Al

2

O

3

-covered Si

3

N

4

TEM window. (c) HAADF-STEM image showing Pt/Pd core/shell NPs deposited on Al

2

O

3

-coated GaP nanowires. The large particle at the top of the nanowire is a gold particle, which is used to grow the nanowires in the VLS process. (Weber

et al

. [77]. Reproduced with permission of Institute of Physics.) (d) Schematic illustration for fabricating the core/shell NPs through area-selective ALD on ODTS modified substrates. (e) HAADF-STEM line scan over a typical Pt/Pd core/shell NP. Inset: the HAADF-STEM image. (Adapted from [78]. http://www.nature.com/articles/srep08470 Used under creative commons license: https://creativecommons.org/licenses/by/4.0/.)

Figure 5.13 TEM images of (A) pristine CNTs and (B) CNT-NiO hybrids obtained after 400 ALD cycles. Inset: SAED pattern. (C) Cyclic voltammograms of CNT-NiO catalysts measured in 0.5 M methanol + 1 M KOH solution at a scan rate of 50 mV s

−1

. (D) Chronoamperometric curves of CNT-NiO catalysts with different loadings recorded at the potential of 0.45 V versus Ag/AgCl. (Tong

et al

. [81]. Reproduced with permission of Wiley.)

Figure 5.14 SEM images showing the morphology of the ALD-deposited MnO

x

catalysts. (a) As-deposited MnO. (b) Mn

2

O

3

obtained by annealing MnO at 480 °C in air for 10 h. (c) ORR performance of the MnO

x

catalysts tested in O

2

-saturated 0.1 M KOH at 1600 rpm. (d) OER performance of the MnO

x

catalysts tested in O

2

-saturated 0.1 M KOH solution. (Pickrahn

et al

. [82]. Reproduced with permission of John Wiley and Sons.) (e) Digital photograph of a porous Fe

2

O

3

electrode. (f) SEM image of the porous Fe

2

O

3

electrode. (g) Cyclic voltammogram of a freshly prepared nanostructured electrode (green curve), compared with a flat electrode of the same macroscopic area (0.30 cm

−2

) prepared with ALD of Fe

2

O

3

on an ITO film sputtered on Si (red dotted curve). Scan rate: 20 mV s

−1

. pH = 7. (Gemmer

et al

. [83]. Reproduced with permission of Elsevier.)

Chapter 6: Atomic Layer Deposition for Thin-Film Lithium-Ion Batteries

Figure 6.1 Schematic energy diagram of a battery with SEI layers toward both the cathode and anode. Gray and colored areas represent vacant and occupied energy levels, respectively. The positions of the LUMO for the SEI layer toward the cathode and the HOMO for the SEI layer toward the anode is largely arbitrary.

V

oc

is the open-circuit potential and

E

g

is the band gap of the electrolyte.

Figure 6.2 Electrochemical performance for ALD-coated natural graphite composite electrodes. (a) Cycle performance at 50 °C, (b) schematic representation of transport in composite electrodes prepared by ALD on powder and by ALD directly on the electrode. (Jung

et al

. 2010 [16]. Reproduced with permission of John Wiley and Sons.)

Figure 6.3 The three most common precursors for deposition of Li-based materials.

Figure 6.4 Lithium content of the films, as measured by TOF-ERDA, as a function of number of subsequent lithium subcycles

n

in the pulsing scheme of 400 × (1 × TiO

2

+ 3 × La

2

O

3

+

n

× Li

2

O). (Aaltonen

et al

. 2010 [50]. Reproduced with permission of Royal Society of Chemistry.)

Figure 6.5 Li–Al–O growth rate measured by ellipsometry as a function of % LiOH ALD cycles. Gray-shaded area designates the region of stable growth with constant, linear growth as a function of ALD cycles. (Comstock and Elam 2013 [35]. Reproduced with permission of American Chemical Society.)

Figure 6.6 Ionic conductivity of ALD LiPON films plotted as a function of N content along with linear fit to the data. (Kozen

et al

. 2015 [65]. Reproduced with permission of American Chemical Society.)

Figure 6.7 (a) SEM images of V

2

O

5

samples obtained by 5000 ALD cycles on silicon substrate, showing plate-like morphology. (b) Simulation of a surface equivalent to the sample deposited using 5000 cycles. (Østreng

et al

. 2014 [47]. http://pubs.rsc.org/en/content/articlehtml/2014/ta/c4ta00694a Used under creative commons license: CC BY SA 3.0 https://creativecommons.org/licenses/by/3.0/.)

Figure 6.8 Discharge rate cycling stability at 120 C conducted directly after the rate performance test to 960 C. The gray band indicates a window with less than 80% capacity loss relative to the initial capacity (55 mAh g

−1

at 120 C). Coulombic efficiency is close to 100%. Inset: charge and discharge curves shown for the 2nd and 2000th cycles. (Østreng

et al

. 2014 [47]. http://pubs.rsc.org/en/content/articlehtml/2014/ta/c4ta00694a Used under creative commons license: CC BY SA 3.0 https://creativecommons.org/licenses/by/3.0/.)

Figure 6.9 Constant current (CC) (dis)charge cycling between 3.0 and 4.1 V (0.35 C-rate) for an ALD-deposited LiCoO

2

(

x

= 4) film on Si/TiO

2

/Pt, using LiClO

4

in ethylene carbonate/diethyl carbonate (EC/DEC 1/1) as liquid electrolyte. The electrochemical storage capacity upon cycling showing data for

x

= 2 and

x

= 4, where

x

denotes the Co/Li pulsing ratio. (Donders

et al

. 2013 [71]. Reproduced with permission of The Electrochemical Society.)

Figure 6.10 Discharge capacities and first-cycle potentiograms (insets) for an 86-nm MnO

2

treated with 200 cycles of LiO

t

Bu + H

2

O, 1000 charge–discharge cycles with 200 μA. (Miikkulainen

et al

. 2013 [38]. Reproduced with permission of American Chemical Society.)

Chapter 7: ALD-Processed Oxides for High-Temperature Fuel Cells

Figure 7.1 Working principle of high-temperature fuel cells SOFC and MCFC.

SOFC

: Oxygen is dissociated at the cathode yielding O

2−

, which migrates at high temperature through the electrolyte and is combined at the anode with hydrogen, to form water and release electrons. The reactions involved are Equations 7.3 and 7.4:

MCFC

: Oxygen in the presence of CO

2

is consumed at the cathode producing carbonate ions (CO

3

2−

), which migrate from the cathode to the anode. The reactions involved are the following:

7.4

7.5

Figure 7.2 Role of thin layers in a high-temperature single cell, in particular SOFC [10].

Figure 7.3 Influence of electrolyte geometry: (a) Scheme of the (100) silicon substrate, (b) SEM cross section of the ALD-processed YSZ electrolyte between two platinum layers. (Su

et al

. 2008 [28]. Reproduced with permission of American Chemical Society.)

Figure 7.4 SEM cross section of the mixed conductivity system ZrO

2

–In

2

O

3

(1030 nm) deposited on an YSZ pellet by ALD at 300 °C. (Brahim

et al

. 2009 [34]. Reproduced with permission of Royal Society of Chemistry.)

Figure 7.5 ALD-processed TiO

2

(300 nm) on porous Ni cathode as-deposited (a, c, e) and after immersion (b, d, f) in molten Li–K carbonates during 230 h at 650 °C. (Meléndez-Ceballos

et al

. 2013 [65]. Reproduced with permission of Elsevier.)

Chapter 8: ALD for Photoelectrochemical Water Splitting

Figure 8.1 Schematic description of the artificial leaf concept. (Nocera 2012 [7]. Reproduced with permission of American Chemical Society.)

Figure 8.2 Energy diagrams of photocatalytic water splitting. (a) Single-excitation process with a single semiconductive photoelectrode, (b) single-excitation process with a combination of a semiconductive photocathode and a metallic anode, (c) single-excitation process with a combination of a semiconductive photoanode and a metallic cathode, and (d) double-excitation process or Z-scheme.

Figure 8.3 Schematic of a water-splitting device concept utilizing structured solar absorbers and a proton-permeable membrane for ion transport. The high aspect ratio structures can improve light absorption for semiconductor materials with short minority carrier diffusion lengths, and the high surface area can enhance catalyst loading. (Warren

et al

. 2014 [62]. Reproduced with permission of American Chemical Society.)

Figure 8.4 SEM and TEM images of CdS QD-sensitized 288-nm-diameter TiO

2

inverse opals. (a) A 20° tilted view at the crack area. (b) TEM image of several QD-coated inverse TiO

2

spheres. (c) HRTEM image of the CdS/TiO

2

interface. (d) Magnified HRTEM image of the CdS QD in (c). The inset is the corresponding Fourier-transform electron diffraction (ED) pattern of the QDs observed in zone axis [0001]. (Cheng

et al

. 2011 [94]. Reproduced with permission of John Wiley and Sons.)

Figure 8.5 (a) Schematic illustration of the design principle, which involves the use of a highly conductive TiSi

2

nanonet as an effective charge collector. The electronic band structure is shown in the enlarged cross-sectional view. Efficient charge collection is achieved when the hematite thickness is smaller than the charge-diffusion distance. (b) Low-magnification TEM image showing the structural complexity of a typical hetero-nanostructure and its TiSi

2

core/hematite shell nature. (c) HRTEM data. A dashed line has been added at the interface as a guide to the eyes. Insets: (left) lattice-fringe-resolved HRTEM image showing the hematite lattice spacings for (110) (0.250 nm) and (0.145 nm); (right) ED pattern of hematite. (Lin

et al

. 2011 [98]. Reproduced with permission of American Chemical Society.)

Figure 8.6 Functional principle of the coaxial nanocylindrical solar cell. (a) Schematic view of the geometry of one individual coaxial p–i–n junction among the large numbers of parallel cylinders constituting the solar cell device. (b) Band diagram of the semiconductors involved. (Wu

et al

. 2015 [56]. Reproduced with permission of The Royal Society of Chemistry.)

Figure 8.7 The differences between Pt nanoparticle catalysts produced by (a) other techniques and by (b) ALD. Electron micrographs of SiNWs decorated with ALD Pt. (c) Cross-sectional SEM showing no obvious catalyst aggregation. (d) Low-magnification TEM showing the distribution and uniformity of ALD Pt. (d) HRTEM showing the crystalline quality of Pt particles grown on the SiNW surface by ALD. The existence of amorphous SiO

2

is indicated by an arrow. (Dai

et al

. 2013 [107]. Reproduced with permission of John Wiley and Sons.)

Figure 8.8 (a and b) Top and (c and d) cross-sectional SEM images of (a and c) TiO

2

NTs and (b and d) Al

2

O

3

-covered TiO

2

NTs (180 cycles). TEM images of (e) TiO

2

NTs and (f) Al

2

O

3

-covered TiO

2

NTs (25 cycles). (g and h) Schematic diagram of the field-effect passivation by Al

2

O

3

coating. (g) Illustration of the structure of the TiO

2

NTs deposited with (powder blue) Al

2

O

3

shell. (h) Energy band diagram of TiO

2

NTs coated with Al

2

O

3

shell. Under UV irradiation, photogenerated holes are trapped at the surface due to the presence of the negative charges located in the Al

2

O

3

film, leaving behind unpaired electrons in the center of tube wall. (Gui

et al

. 2014 [118]. Reproduced with permission of American Chemical Society.)

Figure 8.9 (a) Schematic view of the functionalized of Cl-terminated p-Si (111) substrates with CH

3

and 3,5-dimethoxyphenyl, followed by successive ALD of TiO

2

and Pt. (b) Corresponding current–voltage curves for CH

3

(black) and 3,5-dimethoxyphenyl (green) in 0.5 M H

2

SO

4

under AM 1.5G 100 mW cm

−2

at 0.1 V s

−1

. (Seo

et al

. 2015 [120]. Reproduced with permission of American Chemical Society.)

Figure 8.10 Potentiostatic stability tests of various n-Si/TiO

2

/Ni samples during water oxidation in 1.0 M KOH(aq.) under simulated 1-sun illumination. The electrodes were held at 1.85 V versus RHE. All films were ∼100 nm thick except for the TTIP-ALD sample, for which the thickness varied within the sample from ∼50 to ∼150 nm. (McDowell

et al

. 2015 [129]. Reproduced with permission of American Chemical Society.)

Figure 8.11 (a) Schematic representation of the electrode structure. (b) Scanning electron micrograph showing a top view of the electrode after ALD of 5 × (4 nm ZnO/0.17 nm Al

2

O

3

)/11 nm TiO

2

followed by electrodeposition of Pt nanoparticles. (Paracchino

et al

. 2011 [135]. Reproduced with permission of American Chemical Society.)

Chapter 9: Atomic Layer Deposition of Thermoelectric Materials

Figure 9.1 (a) Dependencies of the individual factors of the thermoelectric Figure of merit

Z

, that is,

S

, σ, and κ, on the carrier concentration, and (b) the dependence of the TE conversion efficiency on the heat-source/hot-side temperature (cold-side temperature: 25 °C).

Figure 9.2 Thermoelectric figure-of-merit values,

ZT

, at different temperatures for representative thermoelectric material families.

Figure 9.3 Scheme of the blocking of phonon transport while maintaining the electron transport in superlattice thin films consisting of thin organic layers regularly embedded between thicker thermoelectric oxide layers.

Figure 9.4 (a) Schematic illustration of the fabrication of oxide:organic ZnO:HQ superlattices by ALD/MLD; (b) verification of the superlattice structure by XRR technique; (c) atomic-level model of the ZnO:HQ superlattice derived from first-principles calculations; (d) experimental and theoretical IR spectra of ZnO:HQ superlattice.

Figure 9.5 Thermal conductivity as a function of superlattice period for ZnO:HQ (as-deposited), TiO

2

:HQ (as-deposited), and TiO

2

:C (annealed) thin-film series fabricated by ALD/MLD. The data for about 100-nm superlattice periods refer to pure inorganic thin films.

Figure 9.6 Nanolaminate structure of alternating ALD Bi

2

Te

3

and Sb

2

Te

3

double layers exhibiting localized epitaxial growth as revealed by high- resolution TEM cross-sectional analysis [38]. (Nminibapiel

et al

. 2013 [41]. Reproduced with permission from The Electrochemical Society.)

List of Tables

Preface

Table 1 Contrasting features of economies based on fossil fuels and renewable energy sources

Chapter 2: Atomic Layer Deposition for High-Efficiency Crystalline Silicon Solar Cells

Table 2.1 A selection of commonly used parameters to assess surface passivation quality

Table 2.2 Comparison of typical film and equipment requirements by the integrated-circuit (IC) and photovoltaic (PV) industries

Table 2.3 Comparison of industrial aspects for temporal and spatial ALD

Table 2.4 Selection of optimal surface passivation conditions provided by several materials prepared by ALD

Table 2.5 Selection of reported low-temperature ALD processes of doped ZnO using DEZ as the Zn source

Table 2.6 Selection of reported ALD processes for (doped) In

2

O

3

Table 2.7 Selection of reported S-ALD results of both intrinsic and doped ZnO using DEZ as the Zn source

Table 2.8 Selection of ALD processes reported in the literature of potential carrier-selective oxides

Chapter 3: ALD for Light Absorption

Table 3.1 Semiconducting metal chalcogenide solar absorbers grown by ALD. The highest PV device efficiencies that result from ALD and any deposition method are listed

Chapter 4: Atomic Layer Deposition for Surface and Interface Engineering in Nanostructured Photovoltaic Devices

Table 4.1 Barrier recombination (compact and blocking) layers deposited by ALD on different third-generation solar cells

Chapter 5: Atomic Layer Deposition of Electrocatalysts for Use in Fuel Cells and Electrolyzers

Table 5.1 Comparison of electrochemical performance of several PEM fuel cells made with ALD-derived Pt catalysts at the anode

Chapter 6: Atomic Layer Deposition for Thin-Film Lithium-Ion Batteries

Table 6.1 Li compounds deposited by ALD

Chapter 9: Atomic Layer Deposition of Thermoelectric Materials

Table 9.1 ALD process parameters and resultant room-temperature thermoelectric characteristics for ALD-fabricated thermoelectric thin films

Atomic Layer Deposition in Energy Conversion Applications

Julien Bachmann

Editor

Prof. Julien Bachmann

Friedrich-Alexander Universität

FAU Anorganische Chemie

Egerlandstr. 1

91058 Erlangen

Germany

Cover

Photograph in the background - fotolia/danielschoenen

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.

© 2017 Wiley-VCH Verlag GmbH & Co. KGaA, 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: 978-3-527-33912-9

ePDF ISBN: 978-3-527-69481-5

ePub ISBN: 978-3-527-69483-9

Mobi ISBN: 978-3-527-69484-6

oBook ISBN: 978-3-527-69482-2

Cover Design Schulz Grafik-Design, Fußgönheim, Germany

Preface

Julien Bachmann

Friedrich-Alexander University of Erlangen-Nürnberg, Department of Chemistry and Pharmacy, Egerlandstrasse 1, 91058 Erlangen, Germany

The Past of Energy Conversion

The ability to harvest energy from the environment and utilize it is a defining characteristic of life. It also represents an activity of central importance to mankind. In fact, several major (pre)historic events are intimately related to the control of novel energy forms. Mastering fire and subsequently the labor of domestic animals are two most crucial achievements of the Paleolithic and the Neolithic, respectively. Much later, craftsmanship became powered by water or wind. Subsequently, the industrial revolution originated from the invention of the heat engine, with coal and steam featured as the most prominent energy carriers (Figure 1). In the twentieth century, they were replaced with petroleum derivatives in the internal combustion engine, used for decentralized energy conversion. Simultaneously, electrical power was established as the most versatile energy form. However, even to date, its generation still mostly relies on coal and steam in a highly centralized infrastructure based on very large energy-converting units (power plants).

Figure 1 Diagram illustrating the principle of energy harvesting from fossil fuels and highlighting the massive losses inherent to the heat engine.

The Future of Energy Conversion

This hybrid situation, in which electrical power is generated from fossil fuels and complemented by them, especially for mobile applications, might represent an intermediary stage toward a society in which electricity has become the universal energy carrier and fossil fuels are eventually outdated. One can envision that in a not-too-distant future, mankind will harvest solar power, be it directly or via wind, water, and biomass (all of which originate from it) into electrical current without the intermediacy of any heat engine. Given that mankind consumes approximately 18 TW (18 × 1012 J s−1, or roughly the equivalent of 18 000 large “traditional” power plants) whereas sunlight provides 120 000 TW (120 × 1015 J s−1), capturing only a small fraction of the incoming solar energy would suffice [1].

Of