Ultraflexible Transparent Oxide/Metal/Oxide Stack...
1 downloads
156 Views
1MB Size
Subscriber access provided by University of Sussex Library
Article
An Ultra-Flexible Transparent Oxide/Metal/Oxide Stack Electrode with Low Sheet Resistance for Electrophysiological Measurements Yasutoshi Jimbo, Naoji Matsuhisa, Wonryung Lee, Peter Zalar, Hiroaki Jinno, Tomoyuki Yokota, Masaki Sekino, and Takao Someya ACS Appl. Mater. Interfaces, Just Accepted Manuscript • DOI: 10.1021/acsami.7b12802 • Publication Date (Web): 21 Sep 2017 Downloaded from http://pubs.acs.org on September 22, 2017
Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.
ACS Applied Materials & Interfaces is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.
Page 1 of 29
1 2 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 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
ACS Applied Materials & Interfaces
1
An Ultra-Flexible Transparent Oxide/Metal/Oxide
2
Stack Electrode with Low Sheet Resistance for
3
Electrophysiological Measurements
4
Yasutoshi Jimbo,1 Naoji Matsuhisa,1 Wonryung Lee, 1 Peter Zalar, 1,2 Hiroaki Jinno, 1 Tomoyuki
5
Yokota, 1,2 Masaki Sekino, 1,2 and Takao Someya*1,2
6
*1
7
Bunkyo-ku, Tokyo 113-8656, Japan; *2Exploratory Research for Advanced Technology, Japan
8
Science and Technology Agency, Bunkyo-ku, Tokyo 113-8656, Japan
9
KEYWORDs: transparent electrode, oxide/metal/oxide, neural recording, optogenetics,
10
Electrical Engineering and Information Systems, The University of Tokyo 7-3-1 Hongo,
implantable device, indium tin oxide, ultra-thin metal, flexibility
11
ACS Paragon Plus Environment
1
ACS Applied Materials & Interfaces
1 2 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 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Page 2 of 29
12
ABSTRACT: Flexible, transparent electrodes are a crucial component for future implantable
13
and wearable systems. For practical applications, conductivity and flexibility should be further
14
improved to prevent signal attenuation, heat generation, and disconnection. Herein, we fabricate
15
an ultra-flexible transparent electrode with low sheet resistance (8.6 Ω/sq) using an indium-tin-
16
oxide/Au/indium-tin-oxide multilayer on a 1 µm-thick parylene substrate. The electrodes were
17
foldable, and when compared to pristine ITO displayed greater mechanical robustness.
18
Applicability for large area applications was confirmed through electrochemical impedance
19
measurements, and the compatibility of electrode arrays for in vivo applications was
20
demonstrated with an optogenetic experiment. As a result of the ultra-flexible transparent
21
electrode’s excellent conformity to soft tissue, voltage signals induced by light stimulation
22
directly below the electrode were successfully recorded on the moving muscle.
23
１．Introduction １．
24
Transparent electrodes are a crucial component in flexible optoelectronic applications such as
25
displays,1 imagers,2 and sensors.3 Highly conductive electrodes are preferable for high resolution,
26
large area matrices to suppress energy loss, heat generation, and unevenness of pixel
27
characteristics caused by wire resistance.3 Therefore, owing to their conductivity and
28
transparency, oxides such as indium-tin-oxide (ITO) have been widely used. Nowadays, aiming
29
at broad applications, especially for wearable or implantable devices, various characteristics
30
including flexibility and straightforward processing methods are required for transparent
31
electrodes.
32
Following the trend, significant research effort has been dedicated towards the development of
33
new types of materials for transparent electrodes, such as conducting polymers,4 metal nanowire
ACS Paragon Plus Environment
2
Page 3 of 29
1 2 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 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
ACS Applied Materials & Interfaces
34
based hybrid composites,5 and graphene.6 One of the promising candidates is a kind of
35
multilayered structure, which is often categorized as an oxide/metal/oxide (OMO) or
36
dielectric/metal/dielectric (DMD) structure. It have been widely used for displays or
37
photovoltaics, and attracted great interest due to a number of desirable properties: low sheet
38
resistance,7,8 controlled cavity effect,9 tunability of work function,10 mild processing
39
conditions,11 and mechanical robustness.12,13 Although their wearable3 and in vitro applications
40
as plasmonic biosensors14,15 have already been shown, in vivo applications have yet to be
41
demonstrated.
42
In vivo sensors, which employed graphene or ITO to enable optical stimulation and imaging
43
while sensing, have been previously reported.16–18 They are fabricated on flexible plastic films to
44
reduce the mechanical damage and inflammation of the tissue.19 Moreover, thinner substrates
45
provide more conformal contact.20 The advantage of graphene is due to its flexibility and broad
46
transmittance spectrum.16 The feasibility of simultaneous imaging, stimulation, and
47
electrochemical measurement has been demonstrated using graphene electrode arrays.17 This
48
multilayered graphene showed a sheet resistance of 76 Ω/sq and a transmittance of ~90%. On the
49
other hand, ITO provides higher conductivity. One technique to avoid the high temperature
50
process, which is not compatible with plastic substrates, is DC sputtering with a pre-oxidation
51
step. It results in a low resistance of 10-50 Ω/sq and a transmittance of 80-90%18,21 without
52
annealing. This method is preferable for removing the effect of wire resistance in sensor arrays.
53
The feasibility of in vivo interface was demonstrated using transparent neural electrode arrays
54
vertically integrated with LED chips.18 This research also revealed the environmental stability of
55
parylene/ITO electrode through a 25-day soak test in saline solution.
ACS Paragon Plus Environment
3
ACS Applied Materials & Interfaces
1 2 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 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Page 4 of 29
56
However, it is still difficult to realize transparent electrode with both enough conductivity and
57
mechanical durability for large scale (~10 cm) transparent sensing arrays. Under certain
58
condition, ITO can be disconnected irreversibly under strains below 1%,22 so that long readout
59
made from ITO can be broken while sensing or processing.23 Graphene is stable up to ~10%
60
strain and can recover from severe bending stress,6 but its sheet resistance is significantly higher
61
than that of ITO (several tens to hundreds of ohms per square).17 Therefore, opaque metal wiring
62
is usually employed to connect each cell.18,23 It covers a large portion of the area except for the
63
contact via so that optical stimulation and imaging can still be performed, although in a limited
64
manner.
65
In this report, we fabricate an ITO/Au/ITO stack on a 1 µm-thick parylene substrate as an ultra-
66
flexible transparent electrode with lower sheet resistance (8.6 Ω/sq) and better mechanical
67
robustness than that of pristine ITO. ITO, Au, and parylene are known as stable materials in
68
saline solution, and they have been used in the previous reports.18,24 Owing to the ultra-thin
69
parylene substrate and encapsulation, this electrode was highly flexible and even foldable. The
70
feasibility of large area, high resolution electrode arrays compatible with in vivo conditions was
71
demonstrated by an electrochemical impedance measurement and an in vivo optogenetic
72
experiment: neural signals from a mouse’s gracilis muscle induced by light stimulation directly
73
under the electrode were successfully recorded.
74
2. Experimental
75
2.1 Device Structure and Fabrication
ACS Paragon Plus Environment
4
Page 5 of 29
1 2 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 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
ACS Applied Materials & Interfaces
76
Figure 1a shows a schematic of the structure of the multilayered electrode, consisting of an
77
ITO/Au/ITO multilayer stack deposited on a 1 µm-thick parylene substrate. For materials
78
selection, we took conductivity, transmittance, stability in saline solution, and processability into
79
account. Parylene provides a smooth surface on supporting substrates, and it has been used for
80
numerous in vivo devices because of its flexibility, low permeability, and biocompatibility.18,25,26
81
ITO is also known as a promising transparent conductor because it has been used for versatile in
82
vivo experiments owing to its high conductivity and transparency. Also, its behavior when
83
interfaced with bio tissues has been well studied.18,24 Au is employed as an ultra-thin metal layer
84
in order to obtain high conductivity. Whereas other highly conductive metals such as Ag or Cu
85
are usually not stable in saline solution14,27 and sometimes cytotoxic,28,29 Au itself is a stable
86
material. Because the thickness of the Au layer was only 14 nm, the multilayer electrode
87
appeared transparent (Figure 1b).
88
The fabrication process starts with a parylene dix-SR (Daisan Kasei Co. Ltd.) deposition using
89
chemical vapor deposition (CVD) by a LABCOATER PDS2010 (Specialty Coating Systems) on
90
a supporting glass substrate. The glass was cleaned and coated by spincoating (2000 rpm) with a
91
fluorinated polymer, a mixture of NOVEC (3M Company) 1700 and 7100 in 1:6 volume ratio,
92
which acts as a release layer for the device. Both the top and bottom ITO layers were deposited
93
using RF sputtering by a SH-250-T04 (ULVAC Co. Ltd.) without heating the substrate.
94
Sputtering was performed in a mixture of gas flow (Ar: 5 sccm, O2: 0.05 sccm). Before the
95
introduction of gas, the pressure inside the chamber was reduced below 2 × 10-3 Pa. Au was
96
thermally evaporated by an EX-200 (ULVAC Co. Ltd.) under vacuum conditions (< 5 × 10-4 Pa)
97
and the deposition rate was 0.1 ± 0.005 nm/s. It is important to note that properties of ultra-thin
98
Au is strongly affected by deposition conditions, especially by the deposition rate.30 Some
ACS Paragon Plus Environment
5
ACS Applied Materials & Interfaces
1 2 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 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Page 6 of 29
99
samples, such as ones used for animal experiments, had a 1-µm thick parylene encapsulation
100
layer deposited by the same process used for the substrate. Access to the electrode was prepared
101
using a RIE-10NR (Samco Inc.) with O2 gas and a polyimide shadow mask to remove the
102
parylene above the sensing area (Figure 1a). No specific damage to the ultra-thin substrate was
103
observed with these process conditions. After all layers were deposited, the completed device
104
was peeled off from the supporting substrate without any change in resistance. To test
105
mechanical durability, several samples were fabricated on 125 µm-thick polyethylene
106
naphthalate (PEN) films or laminated onto soft elastomers. The structures of the devices used for
107
each experiment are all listed in Table S1.
108 109
2.2 Device Characterization Equipment
110
The sheet resistance measurement was performed with a four-point probe measurement system
111
(Model Σ-5+, NPS Inc.). Transmittance spectra were measured with a spectrometer (ARM
112
500N, JASCO Co.). An x-ray diffraction system (Smart Lab, Rigaku Co.), atomic force
113
microscope (AFM, NanoScope IIIa, Veeco Instruments), and scanning electron microscope
114
(SEM, S4800, Hitachi High-Technologies Co.) were all employed to closely examine film
115
condition. Electrochemical impedance was measured using a material characterization cell (EC
116
Frontier Co., Ltd.) and a precision LCR meter (4284A, Agilent Technologies). The thickness of
117
ITO was controlled by deposition time; that of parylene was controlled by the amount of loaded
118
material; and that of Au was controlled by the quartz crystal microbalance attached to the
119
evaporator. All were calibrated with the use of a stylus surface profiler (Dektak XT, Bruker Co.).
120 121
2.3 Mechanical Durability Tests
ACS Paragon Plus Environment
6
Page 7 of 29
1 2 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 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
ACS Applied Materials & Interfaces
122
To test the intrinsic mechanical durability of the ITO/Au/ITO multilayer stack, simple bending
123
tests were performed using 125 µm-thick PEN substrates instead of the 1 µm-thick parylene
124
substrates in order to apply precise tensile strain on the electrodes. They were patterned in 300
125
µm × 300 µm, and 70 nm-thick Au was employed as a readout. While it was bent from 10 mm to
126
5 mm of bending radius, the profile remained circular and no irreversible deformation of PEN
127
substrate itself was observed.
128
Due to the difficulty of fitting a 1 µm-thick film electrode to a circle with ~1 mm or less
129
bending radius, we performed crumple tests of the ultra-thin film electrodes to compare the
130
mechanical durability of each structure against severe stress. The samples fabricated on 1 µm-
131
thick parylene substrates were laminated onto 1 mm-thick elastomers (Ecoflex 00-30, Smooth-
132
On Inc.) by adhesive tape. The elastomer was pre-stretched and fixed to the stage, and then it
133
was released slowly after the ultra-thin film was laminated on it. While it was being released the
134
laminated film was crumpled, resulting in many tiny wrinkles and severe mechanical stress on
135
the film. The samples were patterned with a length of 23 mm and a width of 3 mm, and 70 nm-
136
thick Au was also employed as interconnects. During the cyclic test, the stretching and relaxation
137
speed were kept at 5 %/s and a 5 second rest interval was observed after each motion (both
138
stretching and relaxation) using a high precision mechanical system (AG-X, SHIMADZU Co.).
139 140
2.4 Electrophysiological Measurements
141
A rat (W-Tg(Thy1-COP4-YFP*)4Jfhy, the National Bio Resource Project of the Rat in Japan,
142
male, 12-weeks-old) was genetically modified to express channelrodopsin-2 (ChR2) in its motor
143
nerves and their terminals. It was anesthetized using 2%–2.5% isoflurane mixed with air, and the
144
skin was incised to expose the gracilis muscle of its left hind limb. Local muscle movement was
ACS Paragon Plus Environment
7
ACS Applied Materials & Interfaces
1 2 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 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Page 8 of 29
145
induced by light stimulation using a blue laser (wavelength: 473 nm, frequency: 2 Hz, duration:
146
5 ms, diameter: 500 µm, intensity: 40 mW), whose position was guided by an optical fiber. The
147
stimulation target on the motor nerve bundle was identified by optical stimulation before
148
laminating the film device. The readout wire of the electrode laminated on the muscle was
149
connected to an amplifier system marketed for medical use (Neuropack µ, MEB-9104, Nihon
150
Kohden Co.) to measure the voltage signal. A ground electrode (V-040M4, Nihon Kohden Co.)
151
was attached to the skin on the contralateral limb.26,31 All animal-based experiments were
152
approved by the Ethics Committee of the University of Tokyo (KA12-1-1). (a)
(b)
Electrode
5mm
Figure 1. (a) Schematic structure of the ITO/Au/ITO multilayered electrode. (b) An image of the ITO/Au/ITO multilayered electrode patterned on parylene substrate, showing its visual transparency. The thickness of each layer is the same for the device used in the optogenetic experiment. 153 154 155
ACS Paragon Plus Environment
8
Page 9 of 29
1 2 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 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
ACS Applied Materials & Interfaces
156
3. Results and Discussions
157
3.1 Electrical and Optical Characteristics
158
Figure 2a shows the relationship between the thickness of Au and the sheet resistance of the
159
ITO/Au/ITO electrodes when the thickness of the top/bottom ITO was fixed to 48 nm. The
160
electrodes with a 14 nm thick Au layer showed a sheet resistance of 8.6 Ω/sq, although that of 96
161
nm thick pristine ITO was 60 Ω/sq. The sheet resistance of the electrodes showed an inverse
162
relationship with the thickness of Au. When the thickness is less than 8 nm, the sheet resistance
163
increased drastically. This is due to the formation of island-like structures when it is thinner than
164
8 nm.8,30,32 This island-like thin film growth was confirmed by scanning electron microscope
165
(SEM) images (Figure S1) and x-ray diffraction (XRD) (Figures S2). The sheet resistance of ITO
166
(48 nm) / Au (≤ 6 nm) / ITO (48 nm) became higher than that of a single ITO layer (96 nm). This
167
result is consistent with previous reports.8,33 On the other hand, the thickness of ITO did not
168
affect the sheet resistance as much as Au did because of the high conductivity of Au. The sheet
169
resistances of ITO / Au (14 nm) / ITO were 8.9 ± 0.5 Ω/sq and 7.9 ± 0.3 Ω/sq when the ITO
170
thicknesses were 16 and 64 nm, respectively. The surface roughness of ITO (48 nm) / Au (14
171
nm) / ITO (48 nm) (Ra = 4.1 nm) was rather smaller than that of the parylene substrate (Ra = 5.7
172
nm). Atomic force microscopy (AFM) images are available in Figure S3.
173
Figure 2b shows transmittance spectra of both ITO (48 nm) / Au (14 nm) / ITO (48 nm)
174
(without encapsulation) electrodes fabricated on a parylene substrate as well as ITO on a
175
parylene substrate. The periodic wavy shapes appeared due to light interference with the thin
176
parylene substrate. In the long wavelength range (>500 nm), the transmittance of ITO/Au/ITO is
177
compatible with that of ITO. Indeed, the transmittance for shorter wavelengths becomes lower,
178
yet the transmittance at a wavelength of 473 nm is as high as 69%. This value is sufficient in the
ACS Paragon Plus Environment
9
ACS Applied Materials & Interfaces
179
latter optogenetic applications. Figure 2c shows further investigation of the transmittance as a
180
function of Au thickness. The thickness of each ITO layer was fixed to 48 nm. By increasing the
181
thickness of Au from 6 nm to 10 nm, the transmission peak shifted towards longer wavelengths
182
because of absorption due to the localized surface plasmon resonance.30 When it reached 14 nm,
183
transmittance was slightly decreased in the entire visible range because of the absorption and
184
reflectance by the Au layer. In addition, the thicknesses of the top and bottom ITO layers were
185
varied (Figure S4). The thickness of the bottom ITO did not affect optical transmittance except
186
for the shift of the periodic interference peak. On the other hand, the top ITO layer affected the
187
transparency, and the highest average transparency was obtained when the thickness was 48 nm.
188
It can be attributed to the different scale of reflections on the ITO/Air interface and the
189
parylene/ITO interface. (b)
(c)
120
100
100
100 80
ITO (96 nm)
60 40 20 0
4
6 8 10 12 14 Au Thickness [nm]
80 60
Parylene ITO 40 ITO/Au/ITO (Front Side) 20 ITO/Au/ITO (Back Side) 0 400 500 600 700 800 Wavelength [nm]
Transmittance [%]
(a)
Transmittance [%]
Sheet Resistance [Ω/sq]
1 2 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 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Page 10 of 29
80 60 40 20 0
Au
6 nm 10 nm 14 nm
400 500 600 700 800 Wavelength [nm]
Figure 2. Resistivity and transparency of ITO/Au/ITO electrodes. (a) Sheet resistance of ITO (48 nm) /Au/ ITO (48 nm) electrodes with Au layers of various thicknesses.
(b) Transmittance of a bare
parylene substrate, ITO (96 nm), and ITO /Au / ITO (48/14/48 nm) electrode. The spectrum was identical for light entering from the front side (ITO) and the back side (parylene substrate). (c) Transmittance spectrum of ITO (48 nm)/Au/ITO (48 nm) electrodes with various Au layer thicknesses.
190 191
ACS Paragon Plus Environment
10
Page 11 of 29
1 2 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 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
ACS Applied Materials & Interfaces
192
3.2 Mechanical Characteristics
193
The high conductivity of ITO/Au/ITO multilayers under tensile strain was demonstrated by a
194
bending durability test. Figure 3a shows the measurement setup. 125-µm thick PEN substrates
195
were used for precisely applying tensile strain to the multilayer. For comparison, a single ITO
196
layer and ITO/Au/ITO with different thicknesses were also tested. Strain applied to the electrode
197
is given by ε = ⁄2 , where and stand for the thickness of film and bending radii,
198
respectively,34 assuming the effect of ITO and Au is negligible because they are far thinner than
199
PEN substrate. Figure 3b shows the sheet resistance-bending radius (tensile strain)
200
characteristics. ITO (48 nm) /Au (14 nm) /ITO (48 nm) multilayers showed a lower sheet
201
resistance than that of 96-nm thick ITO against various tensile strains. The wire resistance of
202
both the ITO and the ITO/Au/ITO electrode started increasing drastically at a certain strain, and
203
finally reached a highly resistive state. It is known that this critical strain corresponds to crack
204
initiation and propagation on the film.12,22 Here we define the critical strain as the strain where
205
resistance goes above twice that of its original value. The critical strain of each electrode is listed
206
in Table S2. The critical strain of the multilayer and ITO were similar: 0.93% for ITO/Au/ITO
207
and 0.90% for ITO, respectively. Even lower resistance and higher critical strain (1.03%) were
208
obtained in ITO (16 nm) / Au (14 nm) / ITO (16 nm), and the improvement in critical strain was
209
also observed in ITO with a thickness of 32 nm (1.11%). The thickness dependence of crack
210
propagation in ITO can possibly be attributed to this trend.22 After it reached a highly resistive
211
state, the resistance remained stable for both ITO and ITO/Au/ITO during this bending test. At
212
bending radii less than 5 mm the PEN substrate deforms irreversibly, causing either the ITO or
213
the ITO/Au/ITO to disconnect. In addition, resistance recovery after bending with a 5 mm
214
bending radius was significantly improved by the presence of Au layers (Table S3). ITO/Au/ITO
ACS Paragon Plus Environment
11
ACS Applied Materials & Interfaces
1 2 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 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Page 12 of 29
215
multilayer samples returned to their initial sheet resistance regardless of the thickness of the ITO
216
layers. In contrast, ITO remains in the highly resistive state. This could be attributed to the
217
difference in crack size and formation due to either the ductility of Au or weak adhesion between
218
Au and ITO (SEM images are available in Figure S5).
219
The improved mechanical robustness of an ITO/Au/ITO multilayer was also confirmed when
220
fabricated on a 1 µm-thick parylene substrate. Severe mechanical stress was applied to the
221
samples by stretching crumpled samples on elastomers. The crumples were made by first
222
laminating onto pre-stretched elastomers and then releasing the strain (Figure 3c). The
223
deformation of samples on elastomers was already studied and modeled.35 Samples with an ITO
224
(48 nm) /Au (14 nm) /ITO (48 nm) multilayer and ITO (96 nm) were laminated on 15 % pre-
225
stretched elastomers and the strain was subsequently released, resulting in a resistance change.
226
The resistance of the ITO devices increased from 0.46 ± 0.04 kΩ to 4.7 ± 2.2 kΩ. On the other
227
hand, that of ITO/Au/ITO devices showed a much smaller change from 66 ± 2 Ω to 106 ± 43 Ω.
228
These changes are in the same order with the above mentioned resistance recovery after severe
229
bending stress (bending radii = 5 mm) with PEN substrates.
230
The mechanical durability of the crumpled samples were further investigated by cyclical 10%
231
strain tests (Figure 3d). For both ITO and ITO/Au/ITO, the resistance of multiple samples was
232
evaluated after 100 stretch cycles. For both structures, the graph displays the sample that showed
233
the lowest resistance after the 100th cycle. The resistance went up when the samples were
234
stretched, even though the crumples on the film device disappeared. This implies that not all
235
strain was consumed to remove the wrinkles on the film, but on stretching of the film itself. This
236
could be caused by uneven wrinkle formations. ITO devices showed a larger resistance change
237
than ITO/Au/ITO multilayer devices. The baseline resistance slowly increased after the
ACS Paragon Plus Environment
12
Page 13 of 29
1 2 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 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
ACS Applied Materials & Interfaces
238
completion of each full stretch/relax cycle. This effect may be caused by the viscoelasticity of
239
the elastomer layer. At the 100th cycle, the relative resistance change between the stretched state
240
and the relaxed state (
⁄ ) was 6.6 for ITO, whereas it was 1.4 for the
241
ITO/Au/ITO multilayer.
242
In spite of the improved flexibility of the ITO/Au/ITO electrode, it no longer shows high
243
conductivity after lamination onto a 25 % pre-stretched elastomer (Figure S6). It is known that a
244
larger pre-stretch results in both wrinkles with smaller bending radius, and more severe
245
mechanical stress.36,37 The radius of curvature which may contribute to the crack formation is
246
estimated at several tens of micrometers based on behavior observed on PEN substrates. The
247
histogram in Figure S7 indicates that the radius of curvature decreased for samples on elastomers
248
with a higher degree of pre-stretching. Higher mechanical durability is achievable by employing
249
a rational encapsulation strategy which places the ITO/Au/ITO layer on the neutral mechanical
250
plane position in a film.38 By encapsulating the ITO/Au/ITO electrode with parylene of the same
251
thickness as used for the substrate, it can endure stress on a 100 % pre-stretched elastomer. Even
252
when laminated on a 200 % pre-stretched elastomer, the resistance increased only by 17%.
ACS Paragon Plus Environment
13
ACS Applied Materials & Interfaces
1 2 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 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
(a)
(b)
(c)
(d)
Page 14 of 29
Figure 3. (a) Photograph of a bending test. A PEN film was pinched by parallel stages and deformed along the red circle. (b) Resistance change of ITO and ITO/Au/ITO electrodes during bending tests described in (a). Each data point and error bar indicates the geometric mean and geometric standard deviation of the samples. (c) Schematic illustration of the process for laminating a film onto an elastomer. (d) 10 % cyclic stretching test for an electrode laminated on a 15 % pre-stretched elastomer as described in (c). The samples were patterned with a length of 23 mm and a width of 3 mm. The samples which showed the lowest resistance after 100 cycles were plotted for both ITO and ITO/Au/ITO. 253
ACS Paragon Plus Environment
14
Page 15 of 29
1 2 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 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
ACS Applied Materials & Interfaces
254
3.3 Electrophysiological Measurements
255
The practical feasibility of ITO/Au/ITO electrodes was successfully demonstrated by
256
electromyogram (EMG) measurements using optogenetic stimulation. For this demonstration, an
257
ITO/Au/ITO electrode was encapsulated with a 1 µm-thick parylene layer (except for a sensing
258
area with a diameter of 500 µm). The low sheet resistance of the ITO/Au/ITO layer enabled
259
sensing independent from the wire resistance. Figure 4a shows the electrochemical impedance of
260
an ITO/Au/ITO electrode and ITO in saline solution. The effect of wire resistance on
261
electrochemical impedance was investigated by preparing samples with four different wire
262
lengths (L) and fixed width (W) (300 µm). The detailed design is available in Table S1 and
263
Figure S8. At low frequency (10 kHz), the impedance was dominated by the wire resistance.
266
Because of the low sheet resistance, the impedance for EMG (
