151,99 €
Mehr erfahren.
- Herausgeber: Wiley-VCH
- Kategorie: Fachliteratur
- Sprache: Englisch
Edited by the leaders in the fi eld, with chapters from highly renowned international researchers, this is the fi rst coherent overview of the latest in silicon nanomembrane research.
As such, it focuses on the fundamental and applied aspects of silicon nanomembranes, ranging from synthesis and manipulation to manufacturing, device integration and system level applications, including uses in bio-integrated electronics, three-dimensional integrated photonics, solar cells, and transient electronics.
The first part describes in detail the fundamental physics and materials science involved, as well as synthetic approaches and assembly and manufacturing strategies, while the second covers the wide range of device applications and system level demonstrators already achieved, with examples taken from electronics and photonics and from biomedicine and energy.
Sie lesen das E-Book in den Legimi-Apps auf:
Seitenzahl: 690
Veröffentlichungsjahr: 2016
Ähnliche
Table of Contents
Cover
Title Page
Copyright
List of Contributors
Part I: Materials and Processes
Chapter 1: Synthesis, Assembly, and Applications of Semiconductor Nanomembranes
1.1 Introduction
1.2 Strategies for Forming Silicon Nanomembranes
1.3 Transfer Printing for Deterministic Assembly
1.4 Compressive Buckling for Deterministic Assembly
1.5 Functional Devices Made from Silicon Nanomembranes
1.6 Conclusions and Outlook
References
Chapter 2: Models of Reactive Diffusion for Resorbable Electronics
2.1 Introduction
2.2 Hydrolysis of Silicon Nanomembranes
2.3 Material-Level Dissolution
2.4 Dissolution of Device with Layered Structures
2.5 Discussion and Conclusion
Acknowledgments
References
Part II: Applications in Bio-Integrated and Flexible Electronics
Chapter 3: Transparent and Foldable Electronics Enabled by Si Nanomembranes
3.1 Introduction
3.2 Fabrication
3.3 Characterization
3.4 Configurations of Transparent and Foldable Electronic Devices
3.5 Concluding Remarks
References
Chapter 4: High-Performance Flexible Electronic and Optoelectronic Devices by Mechanical Exfoliation from a Brittle Substrate
4.1 Introduction
4.2 Steady-State Substrate Cracking Parallel to the Interface in a Bilayers
4.3 Spalling-Mode Fracture for Layer Transfer
4.4 High-Performance Flexible Electronics by Controlled Spalling
4.5 Future Directions
References
Chapter 5: High-Speed, Flexible Electronics by Use of Si Nanomembranes
5.1 Introduction
5.2 Fabrication of High-Speed MOSFETs
5.3 Design and Performance of High-Speed MOSFET
5.4 High-Speed MOSFET Using Strained Si NMs
5.5 High-Speed Diodes and Switches with Si NMs
5.6 Outlook toward Future Fast Electronics
References
Chapter 6: Flexible and Stretchable Biointegrated Electronics Using Silicon Nanomembranes
6.1 Introduction
6.2 Overview of Flexible Organic/NW-Based Biomedical Devices
6.3 Flexible/Stretchable Single-Crystal Silicon-Based Biomedical Devices
6.4 Actively Multiplexed Electrophysiology Using Si NM Transistors
6.5 Conclusions
Acknowledgments
References
Chapter 7: Flexible Si Nanomembrane Sensor for Human–Machine Interface
7.1 Introduction
7.2 Si NM-Based Tactile Sensor
7.3 Si NM-Based Flexible Sensor for Biological Signal Monitoring and Motion Sensing of Humans
7.4 Summary
Acknowledgments
References
Chapter 8: Flexible and Transparent Solar Cells Using Si Nanomembranes
8.1 Introduction
8.2 Fabrication of Ultrathin Monocrystalline Silicon Solar Cells
8.3 Transfer Printing of Ultrathin Silicon Solar Cells
8.4 Photovoltaic Performance of Printed Silicon Microcells
8.5 Unconventional Module Designs of Ultrathin Silicon Microcells
8.6 Photon Management for Ultrathin Silicon Microcells
8.7 Conclusion
Acknowledgment
References
Chapter 9: Silicon Nanomembranes for Efficient and Precise Molecular Separations
9.1 Introduction
9.2 Separation Science
9.3 Silicon Nanomembranes
9.4 Applications
References
Chapter 10: Nanomembranes as Protein Mass Spectrometers
10.1 Introduction
10.2 Chemical Perspective and Current Limitations
10.3 Protein Mass Sensors
10.4 Nanomembrane Operating Basics
10.5 Field-Emission Mechanics
10.6 Field Emission Cooling
10.7 Summary
Acknowledgments
References
Part III: Applications in Thermoelectrics and Optics
Chapter 11: Semiconductor Nanomembranes for Fano Resonance Photonic Crystal Devices
11.1 Introduction
11.2 Principles of Fano Resonance in Photonic Crystal Slabs
11.3 Fano Resonance Photonic Crystal Filters
11.4 Fano Resonance Photonic Crystal Membrane Reflectors
11.5 Fano Resonance Photonic Crystal Membrane Lasers
11.6 Design of MR-VCSELs
11.7 Fano Resonance Photonic Crystal Field Localization and Absorption Engineering
11.8 Conclusions and Prospects
Acknowledgments
References
Chapter 12: Acoustic Phonons in Ultrathin Free-Standing Silicon Membranes
12.1 Introduction
12.2 Fabrication of Free-Standing Silicon Membranes
12.3 Dispersion Relations
12.4 Phonon Transport Regimes
12.5 Contactless Thermal Conductivity Measurements
12.6 Discussion
12.7 Conclusions
Acknowledgments
References
Chapter 13: Thermoelectric Performance of Silicon Nanostructures
13.1 Introduction
13.2 Basics of Seebeck and Peltier Effect
13.3 Enhance Thermoelectric Performance by Reducing Thermal Conductivity
13.4 Enhance Thermoelectric Performance by Enhancing Thermoelectric Power Factor
13.5 Conclusion
References
Index
End User License Agreement
Pages
xiii
xiv
xv
xvi
xvii
xviii
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
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
57
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
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
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
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
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
230
231
232
233
234
235
236
237
238
239
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
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
Guide
Cover
Table of Contents
Part I: Materials and Processes
Begin Reading
List of Illustrations
Chapter 1: Synthesis, Assembly, and Applications of Semiconductor Nanomembranes
Figure 1.1 Representative routes for making single-crystal silicon NMs. (a) Release of a silicon NM from an SOI wafer by selective removal of SiO
2
in hydrofluoric acid. The optical image on right shows a single-crystal silicon NM (∼50 nm in thickness) formed by this manner. After transfer onto a new host, the wrinkles in the silicon NM can flatten. (Reproduced from Rogers
et al.
[13], with permission of Nature Publishing.) (b) Fabrication of silicon NMs by anisotropic wet-chemistry etching of bulk silicon (111) wafers. The frame on the top right is an SEM image of partially undercut silicon NMs. The frame on the bottom right is an SEM image of released silicon NMs. Left frame: Reproduced from Baca
et al.
[15], with permission of Wiley. Right frame: Reproduced from Mack
et al.
[16], with permission of AIP. (c) Generating multilayer stacks of silicon NMs from bulk silicon (111) wafers by anisotropic etching. The processes exploit trenches with sculpted sidewalls, angled electron beam evaporation of gold, and anisotropic wet chemical etching. The cross-sectional SEM images on the right show two intermediate stages of forming stacks of silicon NMs (∼100 nm in thickness) in this manner.
Figure 1.2 (a) Schematic illustration of the process for a transfer-printing approach that uses a rubber stamp to retrieve solid nano-/microstructures of silicon NMs, which can be regarded as “inks,” and to deliver them onto device substrates, as a form of deterministic materials assembly. (b) Schematic diagram of critical energy release rates for the ink/receiver interface and for the stamp/ink interface. The horizontal lines at the top and the bottom represent weak and strong ink/receiver interfaces, respectively, corresponding to conditions for which only retrieval or printing can be realized.
Figure 1.3 (a) Colorized SEM images of elastomeric stamps (blue) bearing soft, pyramidal relief features. Control of the applied pressure provides strong adhesion in a collapsed state (ON, top panel) and weak adhesion in a retracted state (OFF, bottom panel). (b) Measured adhesion strength as a function of peeling velocity in the ON and OFF states.
Figure 1.4 (a) Schematic illustration of transfer printing for the deterministic assembly of GaAs NMs (gray) onto a target substrate. The right frames provide optical images of GaAs NMs (small black squares) printed onto a glass substrate (main image) and a bent sheet of plastic (inset). Left frame: Reproduced from Park
et al.
[34], with permission of AAAS. Right frame: Reproduced from Rogers
et al.
[13], with permission of Nature Publishing. (b) The left panel is an optical image of printed patterned patches of silicon NMs on a PET sheet. The right panel is a cross-sectional SEM image of an eight-layer stack of silicon NMs (∼340 nm in thickness) separated by transparent layers of polymer. The inset is a schematic illustration; the cross-section shown in the main image is outlined by the red box. Left frame: Reproduced from Lee
et al.
[25], with permission of Wiley. Right frame: Reproduced from Rogers
et al.
[13], with permission of Nature Publishing. (c) Picture of a high-throughput, automated transfer-printing tool [29].
Figure 1.5 Schematic illustration of the steps involved in fabricating “wavy” silicon layouts by controlled uniaxial buckling (left frames). Optical (top), SEM (middle), and AFM (bottom) images of wavy silicon structures (right frames). Left frame: Reproduced from Khang
et al.
[37], with permission of AAAS. Right frame:
Figure 1.6 (a) Schematic illustration of controlled biaxial buckling of silicon NMs (left frames). Buckled, stretchable silicon NMs (∼100 nm in thickness) in a wavy herringbone layout bonded to an underlying PDMS substrate, presented in optical (top) and AFM (bottom) images (right frames). Left frame: Reproduced from Choi
et al.
[43], with permission of American Chemical Society. Right frame: Reproduced from Rogers
et al.
[38], with permission of AAAS. (b) Fully stretchable integrated silicon circuit in a wavy geometry, compressed at its center by a glass capillary tube (main) and wavy logic gate built with two transistors (top right inset) [38].
Figure 1.7 (a) Schematic illustration for delamination buckling of silicon on an elastomeric substrate with lithographically patterned surface adhesion sites (left). SEM image of the buckled silicon with variations in wavelengths and amplitudes (right). (Reproduced from Sun
et al.
[45], with permission of Nature Publishing.) (b) Moderate (left) and high-resolution (right) SEM images of silicon NMs (∼100 nm in thickness) patterned into an open-mesh, arch-shaped, buckled layouts onto a rubber substrate. (Reproduced from Rogers
et al.
[38], with permission of AAAS.) (c) Stretchable silicon circuits with an open-mesh design, wrapped onto a Figure tip model, demonstrated in low (left), moderate (middle), and high (right) magnification optical images.
Figure 1.8 (a) Schematic illustration of a 2D silicon precursor with bonding sites (red; top), SEM image of a single helical coil formed from this precursor (left), and corresponding FEA prediction (right). Similar results for a double helical coil (b), and a nested, coaxial pair of connected helical coils (c). (d) SEM image with overlaid FEA prediction of helical coils with right- and left-handed chirality, on the left and right sides of the dashed red line, respectively. (e) SEM image with overlaid FEA prediction of structures whose chirality changes abruptly at the location defined by the dashed red line. (f) SEM images and FEA prediction of a complex 3D mesostructure formed from a 2D precursor that consists of closed-loop circular filamentary serpentines and radially oriented ribbons, selectively bonded to a biaxially stretched silicone substrate. In all cases, the color in the FEA results corresponds to the maximum principal strains. The scale bars in all images are 400 µm.
Figure 1.9 (a) Distributed 3D mesoscale network comprising an 8 × 8 network of double-floor helices, FEA predication (left) and optical image (right). (b) 3D mesostructures composed of a mixed array of tents and tables. The scale bars in (a) and (b) are 400 µm. (c) 3D mesostructures of silicon with lateral dimensions and thicknesses in the submicron regime, with overlaid FEA predictions. The scale bars in (c) are 5 µm.
Figure 1.10 (a) Image of a transient electronic system that incorporates inductors, capacitors, resistors, diodes, transistors, and interconnects all on a silk substrate (top left). Tilted exploded view of the device showing structures of individual components (top right). The inset shows the top view of the overall device. The bottom frames show the time sequence of the dissolution in DI water. (Reproduced from Hwang
et al.
[63], with permission of AAAS.) (b) Exploded view of an experimental setting to study the
in vitro
cytotoxicity and dissolution of silicon NMs (top left). Plot of average height of silicon NMs as a function of elapsed time, showing the constant dissolution of silicon (top right). Fluorescent images of stained cells after 1, 5, and 10 days of culturing used for live and dead cell counts (bottom left). Plot of live (green squares) and dead (red triangles) cell counts as a function of elapsed time showing the growth of population (bottom right). The inset shows a fairly consistent cell viability after 1, 5, and 10 days.
Figure 1.11 (a) Exploded view of an injectable optoelectronic device. From top to bottom: a Pt microelectrode for electrophysiological measurements, a silicon μ-IPD for optical measurements, a μ-ILED array for optical stimulation, and a Pt serpentine wire for temperature measurements and thermal stimulations. The system is mounted on a microneedle for injection. (b) Process of device injection. Cerebrospinal fluid dissolves the binding silk fibroin, releasing the device from microneedle. (c) Photograph of freely moving mice wearing lightweight, flexible wireless modules. (d) Y-maze setup to study the effectiveness of the device. The red arrow indicates the presence of a mouse wearing the device. (e) Heat map of activity showing the absence of location preference in untrained mice (left frame). Map showing the development of robust location preference of trained mice by optical stimulation (right frame). Panel (d): Reproduced from Rogers [66], with permission of MRS. Panels (a), (c), and (e):
Figure 1.12 (a) Anterior view of a device mounted on a rabbit heart showing various functional components. The inset shows the posterior view. The scale bars are 6 mm. (b) Enlarged views of the metrical components. The scale bars are 500 µm. (c) Various mapping results of key cardiac parameters. Top-left and top-right: a 3D map of the heart derived from electrical signals. Bottom-left: a 2D pH map of the heart after 10 min of no-flow ischemia. Bottom right: 2D temperature monitoring of the heart during cold perfusion and subsequent recovery.
Figure 1.13 (a) Photograph of a compound, hemispherical imaging system, with the inset showing top magnified view of one pair of aligned microlens and photodetector. (b) Photograph of the completed apposition camera. (c) Micro X-ray computed tomography (XCT) and finite element modeling (FEM) results showing strain distribution after full development into hemispherical shape. The inset illustrates that maximum strains developed in critical components such as the microlens (∼2.0%) and the photodetectors (∼0.4%) are much smaller than those in the PDMS base (∼50%). (d) Pictures of a soccer ball captured at three polar angles relative to the center of the camera: −50° (left), 0° (center), and 50° (right), to illustrate the wide-angle field of view (top). Pictures of a circle and a triangle, where the circle is placed at different distances to demonstrate the infinite depth of field (bottom). Panel (c): Reproduced from Rogers [66] with permission of MRS. Panels (a), (b), and (d):
Figure 1.14 (a) Exploded view of an adaptive camouflage skin. Layer components from top to bottom are the artificial chromatophore (ac) made of microencapsulated dye, the artificial leucodye (al) made of silver deposition, the artificial muscles (am) made of silicon actuator arrays, the PDMS intermediate layer, and the artificial opsins (ao) made of silicon photodetectors. (b) Photograph of the bent and stretched device. (c) Screenshots from a movie (extracted time in the lower right) indicating adaptive patterns subject to a continuously varying background.
Chapter 2: Models of Reactive Diffusion for Resorbable Electronics
Figure 2.1 Demonstration platform for resorbable electronics, with key materials and device structures. (a) Image of a device that includes transistors, diodes, inductors, capacitors, and resistors, with interconnects and interlayer dielectrics, all on a thin silk substrate. (b) Exploded view schematic illustration, with a top view in the lower right inset. (c) Images showing the time sequence of dissolution in deionized (DI) water.
Figure 2.2 Theoretical (
T
, lines) and experimental (
E
, symbols) dissolution of silicon nanomembranes (Si NMs) (3 µm × 3 µm × 70 nm) in buffer solutions at different pH levels (pH 6, black; pH 7, red; pH 8, blue; pH 10, purple), at (a) room temperature (25 °C) and (b) physiological temperature (37 °C). (Reproduced from Hwang [5], with permission of Wiley.) Calculated (lines, black) and measured (stars, red) dissolution rates as a function of dopant concentration for (c) phosphorus and (d) boron.
Figure 2.3 Distribution of water concentration (a) in the Mg layer without encapsulation layer and (b) in both Mg and MgO layers.
Figure 2.4 Experimental measurements and analytical predictions of remaining thicknesses for SiO
2
with initial thickness of 35, 70, and 100 nm.
Figure 2.5 Experimental and analytical results of electric resistance of Mg, Mg with MgO encapsulation layers and two different silk overcoats with
φ
= 45% and 90%, respectively.
Chapter 3: Transparent and Foldable Electronics Enabled by Si Nanomembranes
Figure 3.1 A representative example to enhance mechanical flexibility and optical transparency through thickness control: bulk graphite to single-atom-thick graphene.
Figure 3.2 (a) Representative SOI wafers with various Si/SiO
2
thicknesses and corresponding target applications. (b) The schematic illustration of a transparent and flexible silicon nanomembrane from a conventional SOI wafer.
Figure 3.3 (a) The schematic illustration of the mechanism for thickness control through oxidation and selective etching. (b) The schematic illustration of oxygen flux during the oxidation process. (c) The calculated diffusivity of oxygen through SiO
2
as a function of temperature.
Figure 3.4 (a) The Si thickness controlled by thermal oxidation (left) at 1000 °C, and a repeated UVO thinning process at room temperature (right). (b) The cross-sectional TEM images of Si NMs after thermal oxidation (left), and the UV-ozone thinning process (right). The inset shows a diffraction pattern from the ⟨110⟩ zone axis. (c) High-resolution TEM image of Si NMs after the thinning process. The inset shows a diffraction pattern from the ⟨100⟩ zone axis.
Figure 3.5 (a) Schematic illustration of bending strain control of a film using thickness control. (b) Calculated critical bending radius of Si NMs as a function of thickness. SEM images showing (c) a Si wafer with a thickness of 600 mm and (d) 340-nm-thick Si ribbons. (Reproduced from Menard
et al.
[42], with permission of AIP.) (e) A 7-nm-thick Si NM.
Figure 3.6 (a) Schematic illustration of bending stiffness control of a film using thickness control. (b) Optical microscope images of wrinkled Si NMs on pre-strained PDMS substrates with various thicknesses (7, 100, 500 nm, and 1.5 µm). (Reproduced from Jang [38], with permission of American Chemical Society.) (c) Calculated wavelength of wrinkled Si NMs (dotted line) caused by compressive strain as a function of Si thickness. Experimentally measured wavelength values are marked as red circles at 7, 100, 500 nm, and 1.5 µm.
Figure 3.7 (a) Schematic diagram of light transmission through a Si NM. (b) Simulated transmittance gradation of a Si NM transferred onto an epoxy/PET substrate as a function of wavelength and thickness of Si NM. (c) Simulated transmittance curves of Si NMs with various thicknesses (5, 10, 15, 20, 25, 30, 35, 40, 45, and 50 nm) as a function of wavelength. (d) Simulated (dotted line) and measured (solid line) transmittance values of Si NMs with thicknesses of 7 nm (black) and 10 nm (red) transferred onto an epoxy/PET substrate. The inset shows an optical image of a 7-nm-thick Si NM transferred onto an epoxy/PET substrate.
Figure 3.8 (a) Schematic illustration of the measurement of the piezoresistance of Si NMs. (b) Normalized variations in resistance in Si NMs with various thicknesses from 15 to 100 nm as a function of time. (c) PZR coefficients of Si NMs with thicknesses ranging from 16 to 100 nm under a 0.3% tensile strain. The inset shows a pulse-like resistance acquired from the Si NM after generating the strain, with a constant excitation current of 1 nA. (d) The normalized variation in resistance according to a pulsed strain in 30-nm-thick Si NMs with different surface states.
Figure 3.9 (a) Schematic mechanism of carrier-density changes in the band structure of Si NM via an optical pump and THz probe with and without strain. (b) Decay constant of the monoexponential fitting from time-resolved data as a function of strain for 100-nm Si (black) and 30-nm Si (red). (c) Amplitudes of THz signals as a function of field delay. (d) Carrier density extracted from frequency-dependent complex conductivity changes at different strains versus applied strain at Δ
t
= 0 ps (black) and Δ
t
= 500 ps (red).
Figure 3.10 Schematic structure of top-gated field-effect transistors. Tables of representative materials for ultrathin dielectric layers and transparent and flexible electrodes consisting of ultrathin top-gated field-effect transistors.
Figure 3.11 (a) Cross section of (a) a U-TFT and (b) a TEM image showing a Si NM (semiconductor), a SAM (dielectric), and graphene layers (electrodes). (c) Transfer characteristics of a U-TFT in linear (black) and logarithmic scales (blue) at a drain voltage of 0.1 V. Channel length and width of the U-TFT are 25 and 200 µm, respectively. (d) Current–voltage characteristics of a U-TFT at several gate voltages from 1.2 to 2 V.
Figure 3.12 (a) Transparent U-TFTs transferred onto a PET substrate. The inset image shows the magnified view of the red box in the main image. (b) Optical transmittance of various parts in a U-TFT as a function of wavelength. Transmittance spectra through the PET, Si NM/epoxy/PET, and graphene/Si NM/epoxy/PET are marked as black, red, and blue lines, respectively.
Figure 3.13 (a) Schematic illustration of the relationship between the bending stiffness of the device and the thickness of the structure. (b) Optical microscope images of wrinkled SU-8 films on pre-strained elastomers of various thicknesses (70, 220, 430 nm, and 1.4 mm). (Reproduced from Park
et al.
[75], with permission of American Chemical Society.) (c) Evolution of calculated stiffness of Si NMs with three different thicknesses (10, 500 nm, and 1.5 µm) as a function of SU-8 substrate thickness.
Figure 3.14 (a) Optical images of folded U-TFT arrays fabricated on a 300-nm-thick SU-8 substrate. A magnified view of the U-TFT arrays at the edge of a 25-mm-thick PET (bottom left) and at a hair of 50 mm in diameter (bottom right). The bending radii at the edge of the PET and the hair are 12.5 and 25 mm, respectively. The devices are wrapped without any cracks. (b) Transfer characteristics of a U-TFT with 200 mm in channel width and 20 mm in channel length at a drain voltage of 0.1 V in linear (solid line) and logarithmic scales (dashed line) before (black) and after folding (blue). (c) Graph of normalized drain current caused by bending or folding versus the maximum bending curvature of the flexible devices based on various materials.
Figure 3.15 (a) Schematic illustration shows changes in bending stiffness due to thickness control of active materials. (b) Optical microscope images of a wrinkled Si inverter on pre-strained (5%) elastomers with various thicknesses of Si (10, 500 nm, and 1.5 mm). (c) Calculated stiffness of Si inverters on a 300-nm SU-8 substrate as a function of Si thickness. (d) Output voltages (solid line) and gain profiles (dashed line) at
V
DD
= 1.5 V of a wrinkled Si inverter with a 10-nm-thick Si NM. The inset shows the corresponding circuit diagram of the NMOS inverter constructed using load (
W
C
= 30 mm and
L
C
= 15 mm) and drive TFTs (
W
C
= 240 mm and
L
C
= 10 mm).
Figure 3.16 (a) Fundamental structure of a stretchable inverter. (b) Schematic illustration of stiffness control of the devices according to the dimension of the films. (c) Geometry of a stretchable inverter. (d) Height profiles of wavy bridges from the flat state (0%) to stretched states (up to 5%). (e) FEM simulation of strain distribution in a Si inverter at the 5% stretched state. (f) Simulated strains applied to a bridge (red) and an island (blue) in a Si inverter at the 5% stretched state as a function of the distance from a reference point. (Panels (d) to (f): Reproduced from Lee
et al
. [93], with permission of John Wiley & Sons, Inc.]
Figure 3.17 (a) Optical microscope images of devices at different stretching levels of 0% (left), 5% (center), and 10% (right), showing changes in length of islands and bridges. (b) Normalized variation in resistance in a stretchable graphene interconnect with SU-8 encapsulation layers in terms of strain applied to the interconnect. (c) Schematic circuit diagram of the corresponding NMOS inverter including a stretching-induced variable resistor. (d) Output characteristics and gain profiles of stretchable Si inverters at
V
DD
= 10 V from the 0% to 10% stretched state and returned state. (e) Normalized gain and
V
M
values of inverters according to strain and cycle number. (Panels (d) and (e): Reproduced from Lee
et al
. [93], with permission of John Wiley & Sons, Inc.]
Chapter 4: High-Performance Flexible Electronic and Optoelectronic Devices by Mechanical Exfoliation from a Brittle Substrate
Figure 4.1 (a) Reduced problem for a spalling mode fracture in a substrate. (b) Simplified spalling problem in the absence of elastic mismatch between the film and an infinitely thick substrate.
Figure 4.2 Equation (4.5) together with the knowledge of
K
Ic
for a substrate provides a guide for identifying the critical combination of film stress and thickness that will result in germanium spalling (dashed line). Below the dashed line (Region I) spalling is avoided, while above the dashed line in Region II spalling will take place.
Figure 4.3 The combination of the film stress and thickness results in three distinct regions in the case of the controlled spalling [30]. Region (I) indicates the combination of the film stress and thickness for which the spalling will not occur. Region III indicates the condition for which spalling occurs spontaneously. The data points in Region II represent the combination of film stress and thickness that permit controlled spalling once the handle layer is applied.
Figure 4.4 (a) Evolution of nanoscale silicon CMOS technology. (b) Comparison of the power-performance for the bulk and fully depleted silicon devices, indicating the advantage of fully depleted devices over their bulk counterparts for realizing low-power high-speed integrated circuits.
Figure 4.5 Schematic illustration of the controlled spalling process. (a) Prefabricated devices and circuits on the rigid silicon handle wafer. (b) Nickel stressor is deposited to a critical thickness before the spontaneous spalling occurs. (c) Polyimide tape is then applied on the surface of the nickel. (d) A single crack is initiated at the edge of the sample and mechanically guided across the sample.
Figure 4.6 (a) The postspalling process involving the selective removal of the residual Si underneath the BOX to enhance the mechanical flexibility of the samples. (b) Optical microscope image of an integrated circuit from the backside of the sample through the BOX. (c) Photograph of a 100 mm UTB-SOI flexible circuit after removing the excess silicon layer.
Figure 4.7 Representative cross-sectional TEM image of nanoscale UTBSOI integrated circuit, confirming the structural integrity of the flexible devices.
Figure 4.8 Illustration of a second layer transfer process to improve the mechanical flexibility of the samples involving (a) transfer of the thinfilm sample in Figure 4.6c onto a flexible plastic substrate, followed by (b) the removal of the polyimide tape and the thick Ni layer. (c) Photograph of a 100 mm flexible UTB-SOI circuit prepared using this strategy, illustrating the high level of the mechanical flexibility of the circuits.
Figure 4.9 Optical images illustrating (a,b) probed and (c) nonprobed contact pads on a flexible sample. Different depth of focus between the pads and the device area indicates the plastic deformation of the probed regions. (d) Transfer characteristics of the same n-FET and p-FET, before and after the layer transfer process, indicating significant degradation of the drive current for the flexible p-FET.
Figure 4.10 Comparison of the n- and p-FET device performance before and after the layer transfer, confirming that the performance degradation of p-FETs on the flexible sample, shown in Figure 4.9d, is a measurement artifact and caused by the probe pressure.
Figure 4.11 (a) Butterfly curves of a nanoscale flexible SRAM cell exhibiting good symmetry down to
V
DD
= 0.6 V. (b) Stage delay characteristics of flexible ring oscillators, exhibiting a stage delay as low as 16 ps.
Figure 4.12 The average residual strain in the flexible samples was determined by measuring the displacement of the alignment marks (Δ
D
=
D
2
−
D
1
) with respect to the original spacing of the marks (
D
1
).
Figure 4.13 (a) Representative transfer characteristics of an n-FET under different tensile bending conditions. (b) The tensile bending strain causes the sixfold symmetry of the silicon conduction band to break, resulting in a shift in the threshold voltage of the n-FETs. (c) Theoretical and empirical results for the threshold voltage shift of n-FETs under uniaxial bending strain [56], consistent with the observed threshold voltage shift in (a).
Figure 4.14 Schematic illustration of the inverted dual-junction InGaP/(In)GaAs tandem solar cell grown on Ge substrates.
Figure 4.15 Schematic illustration of the key process steps for making the flexible III–V solar cells involving (a) controlled spalling to separate the solar cell structure from the growth wafer, (b) selective removal of the residual Ge and the buffer layer to expose the solar cell structure, (c) front grid and anti-reflection coating (ARC) depositions and cell isolation. (d) Photograph of the final 100 mm diameter flexible InGaP/(In)GaAs solar cell.
Figure 4.16 Light
J–V
characteristics of a 1 cm × 1 cm flexible solar cell, exhibiting a conversion efficiency of 28.1% at one sun intensity.
Chapter 5: High-Speed, Flexible Electronics by Use of Si Nanomembranes
Figure 5.1 SIMS results of phosphorus doping profiles of two implantation conditions: 12 keV/1 × 10
16
cm
−3
(a) before and (b) after annealing; 150 keV/4 × 10
15
cm
−3
(c) before and (d) after annealing. A thin layer of screen oxide was grown before ion implantations. The nitrogen environment of the anneal furnace contains 5% oxygen, which caused further growth of top oxide layer, as indicated by the oxygen trace.
Figure 5.2 Generic process for Si NM release from SOI and transfer. (a) Use SOI as starting material; (b) pattern top Si template layer into strips or meshed NM and partially expose BOX; (c) immerse SOI in aqueous HF to undercut BOX; (d) Si template layer falls down as BOX is fully undercut and gets registered on the handling substrate. Two transfer routes exist: direct flip transfer (e,f) and stamp-assisted transfer (g)–(i). (e) Flexible substrate with adhesive coating is attached to Si NM. (f) Peel off the plastic substrate with NM transferred. (g) Si NM is first picked up by the elastomeric stamp. (h) Bring the stamp into contact with (adhesive-coated) a new host substrate. (i) The stamp is slowly peeled off, leaving NM being attached/transferred to the new host.
Figure 5.3 Evolvement of the structure of flexible RF MOSFET for better performance by reducing S/D resistance and channel length. (a) Device A, gate length/channel length = 2.0/3.0 µm. (b) Device B, gate length/channel length = 2.5/1.5 µm. (c) Device C, gate length/channel length = 2.0/1.0 µm. (d) Planar optical microscopic view of device C. (e)–(g) RF Figure of merits corresponding to device structures device A–device C, respectively. (h) Gate-width-normalized drain current and transconductance versus gate bias.
Figure 5.4 (a) Atomic lattice schematic showing the strain sharing principle. Optical images show the strained NM during release and after finishing release. (b) Process flow to implement the strain sharing principle. (c) Critical and metastable thicknesses of SiGe that can be coherently grown on Si. (d) Si strain as a function of Ge fraction in SiGe used to guide the design and epitaxial growth of Si/SiGe/Si trilayer. (e) X-ray diffraction measurement verifying strain sharing. On-axis line scan around the (004) reflection before and after the release of the trilayer NM. (f) Off-axis reciprocal-space map (RSM) around the (044) reflection for the as-grown trilayer structure.
Figure 5.5 (a) Ion implantation and anneal processes are applied to the as-grown Si/SiGe/Si trilayer. (b) (i) Schematic of patterned strained NM for RF device fabrication. (ii) 3D microscopic images of released strained trilayer, showing the un-flat topology. Note that the undoped region of the trilayer NM stays flat. (c) Doping of strained trilayer structure. (i) Unstrained Si NM was ion implanted and annealed. (ii) Si NM was thinned down. (iii) SiGe and Si epi-growth on thinned Si NM to form symmetric trilayer. (iv) Release of Si/SiGe/Si trilayer. (d) Images of doped strained NM at different processing stages. (i) Before release. (ii) After finishing undercut, sitting on Si handling substrate. (iii) After being transferred to a plastic substrate. (e) X-ray diffraction of as-grown ion-implanted trilayer NM. Off-axis reciprocal-space map (RSM) around the (044) reflection. (f) On-axis line scans around the (004) reflection comparing the doped and undoped as-grown trilayer NMs.
Figure 5.6 Process flow for device fabrication. (a) Released strained NM sitting on handling substrate. (b, c) NM was flip transferred to a PET substrate. (d) Gate stack formation using lift-off techniques. (e) Source/drain and interconnect metalization. (f) Cross-sectional illustration of strained NM device dimensions. (g) Optical image of a fabricated strained device. (h) Cross-sectional illustration of unstrained device dimensions. (i) Optical image of a fabricated unstrained device. For both strained and unstrained devices, the gate length (
L
g
=
L
ch
+
L
sg
+
L
dg
) is 2.5 µm and gate width is 40 µm. Having a gate overlap distance with source/drain regions (
L
SG
and
L
DG
) of 0.5 µm, the effective channel lengths (
L
CH
) are 1.5 µm.
Figure 5.7 Device DC and RF response characteristics of unstrained and strained devices. (a) Transfer curves and calculated transconductance (
g
m
) curves of unstrained and strained devices (
V
ds
= 50 mV). (b)
g
m
is plotted as a function of drain current. Point
A
indicates the peak
g
m
where peaks
f
T
and
f
max
were measured. Point
B
is where 3.5 GHz
f
max
can be obtained. The drain current at point
B
is roughly two orders lower than that at point
A
. (c) Current gain (
H
21
) and power gain (
G
max
) as a function of frequency of unstrained and strained devices (
V
g
= 4 V,
V
ds
= 5 V). (d) and (e)
f
T
and
f
max
of strained devices as a function of gate bias under fixed drain bias (
V
ds
= 5 V) and that as a function of drain bias under fixed gate bias (
V
g
= 4 V), respectively. (f) Calculated mobility values from measured transconductance as a function of bending induced strain for both unstrained and strained devices. (g)
f
T
and
f
max
of both unstrained and strained devices as a function of bending induced external strain.
Figure 5.8 Fabrication process illustration for series-shunt PIN diode RF switches. (a) Selective ion implantation for P
+
and N
+
regions was performed in two photolithography steps on an SOI substrate, followed by thermal annealing. (b) The ion implanted top Si layer was then patterned into strips in another photolithography step followed by plasma dry etching. The buried oxide layer was exposed. (c) The SOI was immersed into HF to release the top Si layer as Si NM. Si NM settled down and became weakly bonded with the handling substrate (“in-place bonding”). (d) A PET substrate with SU-8 spun on was brought face-to-face with the Si substrate to pick up the Si NM strips by the stronger bonding force of SU-8 with Si NM, completing the flip transfer of Si NM. (e) The PET substrate is flipped with the original bottom side of Si NM on top. (f) Metalization with e-beam metal evaporation to complete the fabrication of the PIN diodes and RF switches.
Figure 5.9 (a) Optical-microscope image of a finished shunt-series PIN diode SPST switch. The diode area of
D
1
and
D
2
is 240 and 40 µm
2
, respectively. The I-region width of all the PIN diodes is 2 µm. The top inset shows the cross-section of a lateral Si NM PIN diode on flexible PET substrate and the bottom inset shows the circuit schematic of the RF switch. (b) Optical image of finished PIN diode and switch arrays on a bent PET substrate. (c) Measured typical current–voltage (
I–V
) curves of an 80 µm
2
lateral Si NM PIN diode on flexible PET substrate. (d) Measured insertion loss (ON state) and isolation (OFF state) and (e) return loss of a shunt-series Si NM PIN SPST switch. The series and shunt PIN diodes have areas of 240 and 40 µm
2
, respectively. The ON state is biased at
I
f
= 10, 20, and 30 mA. The OFF state is biased at zero bias on
D
1
and
I
2
= 10 mA. The two Figure have the same
x
-axis scale.
Figure 5.10 (a) Schematic of the mechanical bending for the diode/switch. (b) Bending fixture (strain = 0.31% as an example) and the measurement setup. Measured small-signal results of the flexible diode under (c) ON and (d) OFF state with various strains. Measured small-signal results of the flexible switch under (e) ON and (f) OFF state with various strains.
Figure 5.11 Comparison of measured (solid lines) and calculated (dashed lines) small-signal results of the bended flexible (a) diode and (b) switch under ON state (
I
f
= 10 mA), with 0.42% strain as an example. The ON state dominant parameters reduction (in percentage) depend on tensile bending strains for (c) series intrinsic and parasitic resistance and (d) series parasitic inductance for diodes and switches with various structures.
Chapter 6: Flexible and Stretchable Biointegrated Electronics Using Silicon Nanomembranes
Figure 6.1 (a) Cross-sectional schematic of a flexible organic TFT on a PI substrate. (b) Image of a thin catheter that measures the spatial distribution of pressure using an organic TFT-based actively multiplexed sensor layer. A cross-sectional illustration of the thin catheter is also shown; (inset) the sensor array are tightly wrapped around the catheter in deformation. (c) Organic TFT active matrix array with tactile sensors on an ultrathin PEN substrate. (d) An applied form of the organic TFT active matrix array with tactile sensors on a human upper jaw model. (e) (Left) Optical image of the PEDOT:PSS organic TFT probe and surface electrode probe array conforming to a curvilinear surface; (inset) its connected form with the ZIF connector for external data transport; (right) magnified view of the organic TFT probe and surface electrode probe. (f) (Left) Optical image of the probe array attached on the somatosensory cortex of a Wistar rat; (right) bicuculline-induced epileptiform spike recorded from the (pink) organic TFT probe, (blue) surface electrode probe, and (black) penetrating electrodes.
Figure 6.2 (a) Image of the experimental configuration showing a heart on a NWFET chip. The arrows indicate (red) the heart in the active region, (yellow) Ag/AgCl reference electrode, and (blue) interconnecting wires. (b) Magnified view of the heart on the NWFET chip; (inset) magnified view of the active region indicated by a blue dotted box, showing three pairs of NW devices. The vertical red lines indicate orientation of the NWs. (c) ECG signals simultaneously measured by (black) a glass pipette and (red) a NWFET chip. (d) Schematic of a NW-based electronic cell scaffold incorporating the biological elements into the nanoelectronic networks. The incorporation is designed in biomimetic and bottom-up steps: step A, patterning, metallization, and epoxy passivation for a single NWFET; step B, formation of 3D networks by patterning and combining with extracellular matrices (ECMs); and step C, incorporating the cells and culturing the tissue within the electronic scaffolds. The yellow dots indicate the NW components. The ribbons indicate the wiring interconnects: (blue) metal and epoxy; (green) ECM components. The cells are colorized with pink.
Figure 6.3 Fabrication techniques of silicon integrated circuits on unconventional substrates. (a–d) Integrated circuits are fabricated on the plastic substrate by transfer printing Si NMs onto the substrate and post processing. (e–h) After the fabrication of the integrated circuits on the SOI wafer, the whole devices are transfer printed onto the polymeric substrate.
Figure 6.4 (a) Images of a finger tube where a Si NM strain gage array is integrated. The finger tube effectively conveys the mechanical deformation to the sensor array corresponding to the bending motion of the thumb. (b) Change in the relative resistance of the Si NM strain gage during (black) three bending cycles and (red) side-to-side motion of the thumb. (c) (Black) Theoretically calculated and (red) experimentally measured responses of the (inset) Si NM strain gage to the applied strain along the longitudinal direction. (d) Image of a Si NM strain gage array integrated on a hydrocolloid skin patch; (inset) boron-doped Si NM strain gages positioned along the longitudinal and transverse directions. (e) (Top) Stretched and (bottom) compressed Si NM strain gage array during bending motion of the wrist. (f) (Top) Resistance change of the Si NM strain gage depending on the frequency of the wrist bending; (bottom) corresponding retrieved data from the resistive memory. The stored data are classified into four levels: [00], [01], [10], and [11], which are indicated by the black, orange, red, and blue dotted boxes, respectively.
Figure 6.5 (a) Image of the Si NM strain gage array fabricated on a handle wafer. (b) The magnified view of the red dotted box indicates a strain gage rosette incorporating longitudinal, diagonal, and transverse Si NM strain gages. (c) Image of the Si NM strain gage array transferred to a 0.3-mm-thick silicone substrate. (d) Si NM strain gage array conformally attached to an uniaxially stretched biological tissue. (e) Strain distribution map on the silicone substrate and Si NM strain gages under a 10% strain; (left) almost 10% of the strain is applied in most area of the silicone except for the longitudinally positioned Si NM strain gage; (center) longitudinal and (right) transverse strain distribution maps of the Si NM strain rosette. (f) Resistance changes in the Si NM strain gage attached to a beating rabbit heart under various conditions. (g) Design of the Si NM strain gage array for integration to a 3D-MIM for cardiac measurement. (h) Enlarged image of the Si NM strain rosette indicated by the red dotted box. (i) Optical image of a 3D-MIM with the Si NM strain gage array conformally covering a Langendorff-perfused rabbit heart. (j) (Blue) Percent resistance change of the Si NM strain gage under various conditions of the rabbit heart with (green) simultaneously measured ECG signals.
Figure 6.6 (a) Ultrathin flexible surgical suture with integrated Si NM temperature sensors and metallic heater. (b) (Left) Microscopic image of a Si NM temperature sensor and (right) its electrical characteristics depending on temperature; (inset) A calibration plot obtained by measuring applied voltages inducing specific current at various temperatures. (c) (Left) Images of an 8 × 8 Si NM temperature-sensor array transferred on an elastomeric substrate with the magnified view of (center) a 2 × 2 sensor array and (right) a single sensor. (d) (Left) Image of a heated Cu pattern on an 8 × 8 Si NM temperature-sensor array and (right) measured areal temperature distribution.
Figure 6.7 (a) Stacked structure of a multifunctional prosthetic skin incorporating Si NM strain-, pressure-, and temperature-sensor arrays with a metallic heater and a humidity sensor array. (b) Microscopic images of Si NM (top) strain, (bottom left) pressure, and (bottom right) temperature sensors with (insets) corresponding cross-sectional schematics. (c) SEM image of an intentionally buckled Si NM. The inset shows the magnified view of the buckled region, which displays a large Si NM flexibility. (d) Areal strain distribution changes during clenching motion of the prosthetic hand measured by the Si NM strain gage array. (e) Images of the prosthetic hand (top left) typing a keyboard, (bottom left) grasping a baseball and (right) the corresponding resistance changes in the integrated Si NM pressure sensor. (f) Images of the prosthetic hand touching (top left) hot and (bottom left) cold surfaces of a cup; (right, red line) corresponding current change in the Si NM temperature sensor and (blue dots) simultaneously measured surface temperature by the IR sensor.
Figure 6.8 (a) Schematic exploded view of the flexible, actively multiplexed electrode array. (b) Photograph of the electrode array with 360 channels; (inset) magnified view of the electrode. The size of each electrode is 300 × 300 µm with a 500-µm space between each electrode. (c) Image of the electrode array placed on the visual cortex; (inset) image of the flexible electrode array inserted in the interhemispheric fissure. (d) Folded flat electrode array inserted in the interhemispheric fissure; (inset) folded electrode array before insertion.
Figure 6.9 (a) μECoG signal with five different phases during a seizure measured by an electrode of the array. (b) Spatial–temporal μECoG voltage patterns of each phase during the seizure gathered from all 360 electrodes in the array.
Figure 6.10 (a) Image of the flexible, actively multiplexed electrode array on the brain for auditory-cortex signal recording. (b) Image of the flexible multiplexed electrode array with 196 channels; (inset) its magnified view. (c) (Top) Map of the FRAs obtained from all electrodes in the array. Red colored A, B, and C indicate the three selected FRAs and (bottom) their magnified signal level.
Figure 6.11 (a) Image of the flexible, actively multiplexed electrode array on the porcine heart and (inset) its magnified view at a different angle. (b) Images of the electrode maintaining its conformal contact during the dynamic heart movement. The blue lines indicate the dynamic movement and bending of the heart surface following a cardiac rhythm; (black arrow) conventional pacing electrode is slipped after the first picture. (c) Cardiac electrophysiology signal measured from an electrode of the array; (inset) magnified view of the signal from the black-arrow region showing the signal noise. (d) Maps of the cardiac electrophysiology signals gathered from all electrodes in the array at four different times showing the spatiotemporal waveform pattern of the signal.
Chapter 7: Flexible Si Nanomembrane Sensor for Human–Machine Interface
Figure 7.1 Flexible strain sensing array with Wheatstone bridge configuration: (a) an exploded view of the device showing its structure and materials; (b) a photo of a fabricated device and (b, inset) a circuit diagram of unit cell; (c) voltage output of a Wheatstone bridge consisting of four bar-shaped Si NM strain gages versus applied strain; and (d) stabilities of the voltage output over 1000 cycles at various bending radii.
Figure 7.2 Responses of the developed strain sensing array under compressive, tensile, and a combination of both strain conditions.
Figure 7.3 A schematic view of one of possible tactile sensor designs. An array of sensing elements (e.g., Si strain gages) is sandwiched by elastomeric covers.
Figure 7.4 One of examples of variations in a structural design that incorporates bumps to measure three-axial forces at the same time.
Figure 7.5 Si NM-based artificial electronic skin for a prosthetic hand: (a) a demonstration of the electronic skin covering entire surface of a prosthetic hand. The inset shows the skin is stretchable up to ∼20%. All scale bars correspond to 1 cm. (b) An exploded view of the artificial skin comprising six stacked layers. Each layer has its interconnected wires to convey signals to external instruments. (c) Representative microscopic images of a serpentine-shaped (left frame) pressure sensor made of p-dope Si NM and (right frame) temperature sensor made of p–n junction of Si NM. (d) (Left frame) Microscopic image of a humidity sensor. Scale bar, 2 mm. Bottom right inset shows the magnified view of the central area, showing separate electrodes with identical interspiral gap. Scale bar, 0.5 mm; (right frame). Microscopic image of an electroresistive heater. Scale bar, 4 mm. (c, d) The upper right insets of each Figure show the cross-sectional structure of each device.
Figure 7.6 Variations of the design of Si NM mechanical sensors showing different sensitivities and detection ranges: (a) Sequential images of Si NM strain gages (top frames) under different applied strains (0%, 15%, and 30%) and (bottom frames) corresponding FEA results. Scale bar, 1 mm. (b) The resistance changes for different curvatures of Si NM strain gages, depending on applied strain (left) and temporal resistance changes of different curvature of Si NM strain gages under cyclical stretching (right). (c) Schematics showing the working principle of the Si NM pressure sensor with a cavity (top left) compared with the SiNR pressure sensor without a cavity (bottom left), scanning electron microscope images of the device's cross-section with a cavity (top middle) and without one (bottom middle) and FEA results (top right; bottom right). Scale bar, 500 nm. (d) The resistance changes of a pressure sensor with a cavity (black) and without a cavity (red) with respect to the applied pressure for different design of the Si NM strain gage (
S
1: graph on the left,
S
6: graph on the right).
Figure 7.7 Illustration of the tactile sensor. The dry transfer method was used to fabricate the device in a low temperature. The transferred gages and the transistors were shaped in such a way to include the isolation process, gate dielectric deposition, first metallization, SU-8 passivation, and second metallization. This device was built on the polyimide substrate (25 µm) to achieve mechanical flexibility. The inset on the bottom left gives an optical image of a unit cell consisting of a strain gage and a transistor. The inset on the center shows the electrical circuit including the resistance, electrode lines, and transistors.
Figure 7.8 (a) The transfer characteristic of drain-to-source current from a transistor in logarithmic (left
y
-axis) and linear (right
y
-axis) scales as gate to source voltage (VGS) was swept from −10 to +10 V. (b)
I–V
curve of the switching transistor at gate voltages from 0 V (bottom) to 5 V (top) in 1 V steps. (c) The characteristic curve as a function of time at various pressure levels with 249.7 kPa steps. The inset on bottom center shows the properties in a low range of pressure levels with 12.48 kPa steps. (d) The variation of the output of a single tactile unit for the repeatable loads (cycles up to 1000) with various pressure levels from 0 to 1248 kPa.
Figure 7.9 (a) The tactile sensor attached to a human fingertip. It consists of an 8 × 8 array of Si strain gages and Si TFTs. (b) The voltage output of the tactile sensor in response to localized pressure inputs showing its low crosstalk and spatial resolution. (c) A pressure distribution map acquired during the multitouch test. (d) A series of five frames of the tactile image showing tracking the position of a stylus tip moving fast on the sensing array. The data acquisition electronic system grabbed each frame of the tactile image every 10 ms. Only five frames are shown among the acquired frames.
Figure 7.10 (a) Image of a demonstration platform for multifunctional electronics with physical properties matched to the epidermis. Mounting this device on a sacrificial, water-soluble film of PVA, placing the entire structure against the skin, with electronics facing down, and then dissolving the PVA leaves the device conformally attached to the skin through van der Waals forces alone, in a format that imposes negligible mass or mechanical loading effects on the skin. (b) Multifunctional EES on skin in compressed (left) and stretched (right) state.
Figure 7.11 (a) Simulated video game control by pattern recognition on EMG data. The player icon is moved from an initial position (red) to destination (green). (b) The spectrogram of the alpha rhythm. The first and next 10 s correspond to periods when the eyes were closed and open, respectively. The responses at ∼10 and ∼14 s correspond to eye opening and blinking, respectively.
Figure 7.12 (a) (Upper frame) Optical images of a 1 × 4 array of temperature sensors based on Si nanomembranes (top) and a 1 × 4 array of microheaters based on Au wires in a serpentine layout (bottom), both on a handle wafer. The small total thickness (<3 µm) and the neutral mechanical plane (NMP) design minimize strains induced by deformations. Five metal pads at the end connect either to a data acquisition system (DAQ) or a power supply (PS) through an anisotropic conductive film (ACF). (Lower frame) Optical image of a temperature sensor array based on Pt resistors, on a handle wafer. The two arrays have four sensors each, which are arranged in a bifacial configuration, with an offset to provide an effective 1 × 8 array. (b) Optimized mechanical structure shown in a schematic illustration (left frame) and examples of a device under various deformations, such as folding (center frame) and wrapping (right frame). (c) Temperature monitoring using an instrumented suture strip at the location of an incision on an animal model. Schematic illustration of a suture strip and needle (left frame) and corresponding image after suturing (right frame). Temperature differences inside and outside of the epidermis can be monitored by suturing the incision.
Figure 7.13 Wearable electronic patch composed of data storage modules, diagnostic tools, and therapeutic actuating elements. (a) Wearable memory array consisting of a TiO
2
NM-Au NPs-TiO
2
NM switching layer and Al electrodes (top left inset: layer information). The memory array was transfer-printed on the bottom side of an elastomeric hydrocolloid skin patch. The electroresistive heater/temperature sensor was fabricated on the top-side of the patch, with the Si strain sensor on the opposite side. The m-silica NP array was transfer-printed on the hydrocolloid side of the patch. (b) Corresponding image of (a), showing the wearable biointegrated system. Inset: Wearable 10 × 10 RRAM array on the hydrocolloid side of the patch.
Figure 7.14 Fingertip electronics for electrotactile stimulation and tactile sensing: (a) a thin (500 µm thick), closed-form membrane elastomeric tube on the finger of a model hand; (b) schematic illustration of a multiplexed electrotactile array with serpentine mesh interconnects, with a magnified diagram (right top) and an image (right bottom) of a PIN Si NM diode; (c) turning the tube inside out relocates the array on the inner surface of the finger-tube, shown here at the midway point of this flipping process; (d) voltage required for electrotactile sensation as a function of stimulation frequency; inset: electrotactile array on a human finger during experiments; (e) 2 × 3 array of capacitive pressure sensors consisting of inner electrodes (also used for electrotactile stimulation) (upper frame) and outer electrodes aligned with inner electrodes (lower frame); and (f) measured and analytically calculated change in capacitance of a single sensor with applied pressure and tensile strain.
Figure 7.15 Detection of finger motion with arrays of stretchable Si NM strain gages. (a) FEM results of the maximum principal strain for a 1 × 4 array of gages (straight, vertical structures near the top of the serpentine interconnect mesh) due to an overall 10% strain applied along the longitudinal (
y
) direction. The upper inset shows the strains in the gage highlighted by the yellow dashed box. The lower inset provides an image of a fabricated device with a layout that matches that of the FEM results. (b) Experimentally measured and analytically calculated changes in resistance for a representative Si NM strain gage as a function of applied strain along the longitudinal direction. The inset provides an SEM image of a portion of the device, with the Si NM gage located in the dashed box. (c) Images of a strain gage array on a finger-tube mounted on the thumb, in straight (I) and bent (II) positions. (d) Change in resistance of a representative gage during three bending cycles (black) and side-to-side motion (red).
Figure 7.16 A conformal Si NM-based device for mapping cardiac electrophysiology: (a) Photograph of a fabricated device in a slightly bent state. (Inset) magnified view of a pair of unit cells. (b) Optical microscope image of the unit cell (an amplified electrode). (c) A circuit diagram of the unit cell. (d) Photograph of flexible device conforming to the cardiac tissue of a porcine heart via surface tension. (Inset) magnified image at a different viewing angle. (e) Representative single voltage trace without external pacing, recorded from the epicardial surface of the porcine heart with a flexible electrophysiology mapping device. Inset: a magnified view of the system noise. Black arrow: source of the inset data. The SNR of the recorded signal was ∼34 dB. (f) Color map of relative activation times for two different external pacing sites. The activation times are plotted with the color scale shown at the right. Asterisks (*) indicate the relative location of the external pacing electrode. The scale bar illustrates the spacing between electrode locations.
Figure 7.17 Multifunctional inflatable balloon catheters: (a) Various sensors imbedded in balloon catheters to sense EKG, temperature (upper frame), pressure (lower left frame), and flow (lower right frame). (b) Optical image of a stretchable, interconnected passive network mesh integrated on a balloon catheter (deflated) showing the overall construction, including connectors and ACF metal traces on the proximal side of the balloon and its wrapping configuration along the length of the catheter shaft. (c) Optical image of the balloon inflated by 130% relative to its deflated state (inset). (d)
In vivo
epicardial recordings of cardiac electrophysiological, tactile in a rabbit heart, (upper frame) a photograph showing an array of tactile (white dashed boxes) and electrogram sensors positioned direct contact with the surface of the RV. The inset shows a magnified view of a tactile sensor. (Lower frame) Simultaneous recordings of electrical activation and mechanical contact measured on the surface of the beating heart. (e)
In vivo
temperature monitoring before, during, and after RF ablation. The time constants for the temperature rise and the absolute temperatures achieved during ablation are comparable to those associated with conventional cardiac ablation catheters. (f) Optical image of an instrumented surgical glove in close proximity to the beating heart.
Figure 7.18 Transient, biodegradable electronics: (a) (Left frame) Image of the demonstration platform for transient electronics includes transistors, diodes, inductors, capacitors, and resistors, with interconnects and interlayer dielectrics, all on a thin silk substrate. (Right frame) Exploded view schematic illustration, with a top view in the lower right inset. (b) Images showing the time sequence of dissolution in DI water.
Figure 7.19 Materials and procedures for fabricating transient electronic circuits on biodegradable substrates. (a) Schematic illustrations of key processes for fabricating transient complementary metal-oxide-semiconductor (CMOS) on a carrier substrate (upper left). Defining a mesh type structure allows dissolution of the poly (methylmethacrylate) (PMMA) layer in boiling acetone at 90 °C to release the entire device (upper middle). Retrieving the released device onto a PDMS stamp (right, image of the device on PDMS in the inset), allows exposure and removal of the bottom layer of diluted polyimide (D-PI) by reactive ion etching. Transfer printing onto a poly lacticco-glycolic acid (PLGA) substrate (lower middle), and RIE etching of the top D-PI completes the process (lower left). (b) Images showing an array of transient CMOS inverters on a piece of paper coated with PLGA (left), with magnified view (right), and microscope image of an inverter in the inset. (c) Output voltage characteristics of a representative CMOS inverter with
V
dd
= 10 V. The voltage gain is ∼50.
Chapter 8: Flexible and Transparent Solar Cells Using Si Nanomembranes
Figure 8.1 Schematic illustration of fabrication steps to form micro- and nanoscale ribbons, bars, platelets, and membranes of single-crystalline Si from bulk (111) wafers by anisotropic wet chemical etching and top-down fabrication processes.
Figure 8.2 (a) Schematic illustration of steps for fabricating ultrathin silicon microcells from a bulk (111) wafer. SEM images of (b) trenches formed after the ICP RIE, (c) trenches with patterned SiO
2
as the mask layer for selective phosphorous doping, and (d) completed ∼2-µm-thick microcell after the KOH undercut etching.
Figure 8.3 Schematic illustration of doping design of an individual silicon microcell and the interconnected solar module on a foreign substrate by means of transfer printing.
Figure 8.4 (a) The rate-dependent energy release rate (or separation energy),
G
(J/m
2
), as a function of separation speed,
v
(cm/s), calculated from the “rolling-cylinder” experiment where a steel cylinder rolls down an inclined slab of PDMS at a constant speed (
v
), dependent on the angle of inclination (
θ
).
Figure 8.5 (a) Schematic illustration of processing steps for simultaneously printing and planarizing silicon microcells using a flat PDMS stamp. The photocurable polymer (NOA 61) is used as both an adhesive and a planarizing medium. (Reproduced from Yoon
et al.
[23], with permission of Nature Publishing). (b) Optical image of printed and interconnected arrays of ultrathin silicon microcells on a glass substrate. Inset shows optical image of retrieved microcell arrays on a PDMS stamp.
Figure 8.6 (a) Representative current density (J) and voltage (V) curves of an individual silicon microcell with and without a back-side reflector (BSR). (b)
J–V
curves of individual microcells produced in the first, second, and third fabrication cycle from a single source wafer.
Figure 8.7 (a) Schematic illustration of a strain-engineered flexible solar module that consists of arrays of ultrathin silicon microcells and metal interconnect lines embedded in a polymer matrix, where the neutral mechanical plane is positioned near the center of the microcell. (b) Schematic illustration of a corresponding model composite structure with key parameters. (c) Analytically obtained bending strain at the top and bottom surface of silicon microcells bent along the cell length (
y
) direction as a function of top polymer layer thickness. (d) Contour plot of calculated bending strain through the cross-section of a mechanically flexible solar module at a bending radius of 4.9 mm bent along the cell length (
x
) direction.
Figure 8.8 Photographic image of a mechanically flexible solar module using printed assemblies of single-crystalline silicon microcells.
Figure 8.9 Schematic illustration of test geometries for electrical characterization of mechanically flexible microcell module under various bending conditions.
Figure 8.10
J–V
data from a module measured under AM1.5 standard solar illumination in a flat condition and bent along (a) the cell-width and (b) cell-length directions, at bending radii of 12.6, 8.9, 6.3, 4.9 mm, respectively. Plot of efficiency and fill factor at various (c) bending radii and (d) bending cycles.
Figure 8.11 (a) Photographic images of printed microcells on PDMS-coated PET substrate at intercell spacings (
d
) of (a) 26 µm and (b) 397 µm, placed on a piece of paper with texts and logos to illustrate the differences in transparency. (c) Optical images of printed microcells on a PET substrate at various intercell spacings and (d) corresponding normal incidence transmission spectra.
Figure 8.12 (a) Optical image and (b) schematic illustration of micro-optic concentrator solar module that combines molded lenticular lens arrays and printed microcells. (c)
JV
curves of this module with and without lenticular lens arrays under AM1.5 simulated illumination. (d) Normalized maximum output power (
P
max
) from the concentrator module as a function of incidence angle (
θ
) for tilt along
x
and
y
axes.
