2D Graphene Oxide Nanosheets as an Adhesive Over-Coating Layer for Flexible Transparent Conductive Electrodes
In Kyu Moon,1† Jae Il Kim,2† HanLeem Lee,3† Kangheon Hur,2 Woon Chun Kim,2* and Hyoyoung Lee1,3*
1
Center for Smart Molecular Memory, Department of Chemistry, Sungkyunkwan University,
Suwon-si 440-746, Republic of Korea 2
Corporate R&D Institute, Samsung Electro-Mechanics, Suwon-si 443-743, Republic of
Korea 3
Center for Smart Molecular Memory, Department of Energy Science, Sungkyunkwan
University, Suwon-si 440-746, Republic of Korea * E-mail:
[email protected],
[email protected] †These authors contributed equally to this work.
Correspondence
and
requests
for
materials
[email protected],
[email protected].
should
be
addressed
to
Figure S1 | Photo-image of conductive AgNW/PET and GO/AgNW/PET film after coating in atmosphere at room temperature. Photograph of AgNW and GO/AgNW films with T > 92% on the same side of the PET substrate (d ≈ 125 μm). Right-bottom inset: the reflection photo-image of AgNW/PET and GO/AgNW/PET films on mobile phone under day-light.
Figure S2 | Sheet resistances of 1000 × 500 mm2 GO/AgNW/PET film prepared by spray coating at several different places (Supplementary Movie S1).
50
Count
40
30
20
10
Total 121 points
on 50 mm x 50 mm GO/AgNW/PET film
60
0 21
22
23
24
25
26
27
28
Sheet Resistance (Ω/sq) Figure S3 | Histogram of the sheet resistances of Fig. 1d obtained at 121 points with the 5 mm distance between each points in a 50×50 mm2 area.
To evaluate the quality of GO/AgNW/PET film for practical applications, the spatial distribution of Rsh on a 50×50 mm2 sheet was tested (Fig. 1d). As a result, the sheet resistances ranged from 22 to 28 Ω/sq over 50×50 mm2 areas. The distribution histogram of the corresponding sheet resistance is shown in Fig. S3. Typically, most of the sheet resistances were distributed between 24 and 26 Ω/sq, depending on the coating method, annealing temperature and percolation density between AgNWs. Consequently, our proposedmethod shows a narrowly dispersed sheet resistance over the large-area conductive film
prepared by wet process under low temperature.
100 PET/AgNW/GO (PET side)
96
GO/AgNW/PET (GO side) PET/AgNW (PET side)
92
L*
88
AgNW/PET (AgNW side) 84
805 2.4 0 2.5
.4
-0 44 . -0 45 . 6 -0
b*
0 2.6
0 .4 -0 41 . -0 42 . 3 -0
5 2.5
.4
7 -0 .4 -0 8 .4 -0
5 2.6
a*
9 .4 -0 0 .5 -0
0 2.7
Figure S4 | Plot of CIELAB coordinates in the color space for GO/AgNW/PET and AgNW/PET films.
Table S1 | Color coordinates, haze and the effect of the illuminated-side on film for both GO/AgNW/PET and AgNW/PET filmsa L*(D65)
a*(D65)
b*(D65)
Ttotalb
Tparallelb
Hazec
[%]
[%]
[%]
GO/AgNW/PET 95.16
-0.47
2.47
89.62
86.77
3.18
95.17
-0.46
2.51
89.42
86.62
3.13
95.07
-0.44
2.68
89.26
86.16
3.47
95.1
-0.43
2.69
89.30
86.29
3.01
(PET side) GO/AgNW/PET (GO side) AgNW/PET (PET side) AgNW/PET (AgNW side) a
The color parameters were calculated according to a CIE LAB equation, using a blank as a standard. L* is
lightness, with 100 means white. A positive a* means red color, while a negative a* indicates green color. A positive b* means yellow color, while a negative b* implies blue color. bThe total transmitted light (Ttotal) and the parallel transmitted light (Tparalle) were measured using a haze meter. cHaze [%] was calculated from (Ttotal/Tparallel) ×100; Ttotal=diffused transmitted light+ Tparallel.
For this calculation CIELAB values, L* (lightness axis), a* (red-green axis) and b* (yellowblue axis) represent the average values for each of the films. The plot of the threedimensional CIELAB color space and the summarized-CIELAB values for GO/AgNW/PET and AgNW/PET films are shown in Fig. S4 and Table S1. These CIELAB values are important for most flexible display applications, in which neutral color is desired. All films showed a lower b* value (a yellowness index) in contrast to the corresponding the indium-tinoxide.1 The total transmitted light, the parallel transmitted light and haze ratio were measured
by using a haze meter and are presented in Table S1. The b* intensity and the haze ratio of the direct-illuminated-side on AgNW/ side before and after GO coating are both slightly decreased, indicating that the AgNWs are more closely packed between AgNWs junctions by GO nanosheets, meaning that GO nanosheets would effectively decrease the scattering of incident white-light.
Figure S5 | Magnified optical images of only AgNW (a) and only GO on glass substrate (b). The morphology of thin GO sheets is clearly observed. The yellow-scale bar is 10 μm.
50.1nm(CS)
Figure S6 | SEM of GO/AgNW film on glass. a. A slice of the GO/AgNW film/glass substrate cut by focused ion beam (FIB) lithography to obtain a high-resolution crosssectional SEM image, clearly showing a distribution of the two components, GO and AgNW. The GO coating layers serve as a mechanically robust protection layer as well as an anti-
oxidant agent for the AgNWs. b. The extensive coverage of GO on the AgNW network.
The SEM images exhibit that the prepared GO film is uniform throughout the surface on the AgNW networks. Fig. S6 shows high-resolution SEM images of GO/AgNW film. Fig. S6a shows an SEM image of a GO/AgNW film sliced by a focused ion beam (FIB). AgNW sliced to a film thickness of ~50 nm is encapsulated between GO lamellar film and PET substrate. As seen in the top-view SEM image of the GO/AgNW/PET film depicted in Fig. S6b, the GO nanosheets have mostly a lamellar structure on the AgNW networking film. Considering that GO nanosheets have strong adhesion to PET substrate without delamination, the lamellar GO structure can effectively serve as a protecting layer against oxidation of AgNW networks.
Figure S7 | AFM (tapping mode) images of AgNW film (a), GO film sprayed 3 times on AgNW film (Top) and on SiO2 (Bottom) (b), and GO/AgNW film (c) on PET substrate.
AFM images of AgNW film (Fig. S7a), GO film (Fig. S7b, Top) and GO/AgNW film (Fig. S7c) on the PET substrate revealed uniform and dense coverage of GO nanosheets on the AgNW macronetwork films. The surface roughness (the average roughness, Ra) of AgNW film was 14.297 nm in the 3D-height AFM image (Fig. S7a). With the GO layer, the surface roughness of the GO/AgNW film decreased sharply to Ra = 4.848 nm, indicating uniform and smooth surface by GO layers. It is assumed that the strong adhesion of GO nanosheets to the hydrophilic pretreated PET surface allows for the reduction of the surface roughness. As seen in the highly magnified AFM image shown in Fig. S7b, the average thickness of GO layers sprayed 3 times on the GO/AgNW/PET substrate is estimated to be less than 3-4 layers. As shown in Fig. S7b (bottom), the average GO thickness on SiO2 substrate used as a control experiment had approximately 2-4 nm. An interesting feature of the GO nanosheets is that they are quite tightly distributed on the AgNW macro-networks.
500 GO/AgNW/glass AgNW/glass
450
Intensity (a.u.)
400 350 300
(111) (220)
250 200 (111)
150 100
(220)
50 0 20
40
60
Degrees 2-Theta
80
100
Figure S8 | XRD diffractograms of AgNW/glass and GO/AgNW/glass films.
The low-wide-angle XRD patterns of GO/AgNW/glass and AgNW/glass are shown in Fig. S8. Although the diffraction peaks of (200) and (311) were not very strong, two diffraction peaks were observed at 2θ~38.1° and 64.7°, which correspond to (111) and (220) reflection peaks of AgNWs (JCPDS card No. 87-0720). At 1.14, the intensity ratio (I(111)/I(220)) of the GO/AgNW/glass was much smaller than that of AgNW/glass (I(111)/I(220)=2.15), which indicates that the GO nanosheets were covered on the AgNW film.
Figure S9 | XPS spectra of GO and GO/AgNW films on Si wafer.
Table S2 | The variation of carbon concentration of GO and GO/AgNW film.
GO
GO/AgNW
C 1s
concentration (%)
C=C
27.631
C-O
29.592
C=O
4.573
C=C
13.588
C-O
48.566
C=O
4.552
In this study, C 1s and O 1s core-level spectra are fitted with Gaussian-Lorentizan waveforms. In accordance with previous XPS studies of GO,4 we assign the first peak occurring at ~284 eV to C-C; the second peak at ~286 eV is assigned to hydroxyl and epoxide groups; the third peak at ~288 eV is assigned to carbonyl species. After coating GO onto AgNW, the hydroxyl and expoxide groups of the C 1s peak of GO/AgNW at ~286 eV became much larger than those of GOs, indicating that the plasma treatment of the AgNW’s ligands increased the amount of oxygen groups, including hydroxyl and epoxide groups (Fig. S9 and Table S2) that corresponded to the changes of WCA from 53.86o to 20.65o (Fig. 1a). With the plasma treatment of the AgNW/PET film, surface oxidation of the AgNW’s ligands was responsible for the stronger adhesion between GO and AgNW in Fig. S9.
Mean Sheet Resistance (Ω/sq.)
30 25 20 15 10 5 0
after plasma as-AgNW/PET GO/AgNW/PET treatment of AgNW/PET
Figure S10 | Mean Rsh values of before and after plasma treatment on the AgNW/PET film and the GO/plasma-treated AgNW/PET film.
The typical mean Rsh values of AgNW/PET film before and after plasma treatment for each process step are given in Fig. S10. The mean Rsh value of plasma-treated AgNW/PET film increased slightly from 30.5 to 30.8 Ω/sq. After coating GO onto plasma-treated AgNWs/PET film, the mean Rsh value of the GO/plasma-treated AgNW/PET film decreased sharply.
Figure S11 | Typical conduction processes in a metal-insulator-metal structure. a. Direct tunneling for the ultrathin GO film. b. Fowler-Nordheim tunneling for the thick GO film.
After overcoating GO on AgNW/PET film, the linear current-voltage (I-V) curve was retained (Fig. 3d), although GO is known as an insulator. This result may provide direct evidence of direct tunneling through ultrathin GO nanosheets located between both electrodes of Au-GO-AgNW, like in a metal-insulator-metal model. The direct tunneling process is illustrated in Fig. S11. This phenomenon strongly depends on the thickness of GO films (Fig. 3d); when the quantum state in the electrical thickness of the ultrathin-insulator is independent of temperature, the direct tunneling current is dominant, unlike other mechanisms (Fowler-Nordheim tunneling; Thermoionic emission; Poole-Frenkel emission; Schottky emission; Ohmic hopping conduction, etc.)5,6 This usually occurs with insulating thickness of a few nanometers, like with the monolayer thickness of GO nanosheet (1.1±0.2 nm).7
a
b
AgNWs
PET
Au
AgNWs
Au
GOs
Figure S12 | Fabrication process of Au/GO/AgNW/PET film. Fig. S12 illustrate a fabrication process of Au electrodes on the GO/AgNW/PET.
Figure S13 | Mechanical adhesion test of the conductive AgNW film on PET substrate. a.
Resistance measurement before 3M-Scotch tape peeling b. Resistance measurement after 3M-Scotch tape peeling. The detachment test by Scotch tape removed AgNWs on the PET substrate.
Figure S14 | Mechanical adhesion test of the conductive GO/AgNW film on PET substrate. a. Sheet resistance GO/AgNW/PET film. b. Detachment process of 3M-Scotch tape from GO/AgNW/PET film. c. Sheet resistance of 3M-Scotch tape after detachment of 3M-Scotch tape. d. Sheet resistance of the GO/AgNW/PET film remained after detachment of 3M-Scotch tape. The 3M-Scotch tape did not remove AgNW from the GO/AgNW/PET film, but the 3M-Scotch tape easily removed AgNWs from AgNW/PET film.
Mechanical adhesion tests demonstrated in Figs. S13 and S14 clearly showed that the GO layer used as the over-coating layer is essential to attach AgNWs.
Original position of AgNWs macronetworks on plastic substrate
In-situ bending
bending bending
relaxation
relaxation
2) delamination
Individual AgNWs moved from the original position through slipping and then delamination between AgNW junctions Figure S15 | Proposed fatigue of AgNW’s macronetworks on plastic substrate.
Stable electrical and electrical-mechanical fatigue properties of GO/AgNW/PET hybrid films are a critical challenge for flexible flat-panels, and were measured as shown in Figs. 3d and 4d, respectively. Unlike PEDOT:PSS conducting polymer films, the Rsh changes of GO/AgNW/PET film for both in-situ tension and compression modes were increased at initial bending (Supplementary Movies S2 and S3.). When Rsh changes at initial bending, the initial change can be easily overcome with optical clear film prepared in the fabrication process of capacitive TSP. If AgNWs on plastic substrate are placed without a protecting layer in air, the Rsh will drastically increase due to easy oxidation of the AgNWs. When a polymer is used as an overcoating (protecting) layer on AgNW film, the Rsh is also very high, mainly due to the use of thick film in comparison with GO film. The electro-mechanical fatigue properties of the GO thin film on plastic substrate have been reported.8 Very stable electromechanical properties in in-situ bending process were exhibited. Some of the main problems may originate from AgNW networks. As we already proposed, the increased Rsh value in the initial-bending state would be simply resolved by slipping and/or delamination in two crossed AgNWs.9 As illustrated in Fig. S15, the Rsh value could then be saturated to an equilibrium state of the crossed AgNWs. Unfortunately, however, these phenomena during the bending process were not observed using SEM, TEM and AFM. On the basis of the above discussion, it is demonstrated that a lamellar structure of GO nanosheets can effectively encapsulate AgNW networks between GO nanosheets and PET substrate, which can be seen in Fig. S6, S7 and S15.
References: 1. D.S. Hecht, et. al. Carbon-nanotube film on plastic as transparent electrode for resistive touch screens, Journal of the SID, 17, 941-946 (2009).
2. S. Bae, et. al., Roll-to-roll production of 30-inch graphene films for transparent electrodes, Nature Nanotechnology, 5, 574-578 (2010). 3. J. Wang, et. al. Rod-coating: towards large-area fabrication of uniform reduced graphene oxide films for flexible touch screens, Adv. Mater. 24, 2874-2878 (2012); Doi:10.1002/adma.201200055 4. D. Yang. et. al. Chemical analysis of graphene oxide films after heat and chemical treatments by X-ray photoelectron and Micro-Raman spectroscopy, Carbon 47, 145-152 (2009). 5. D. A. Muller et. al. The electronic structure at the atomic scale of ultrathin oxides. Nature 399, 758-761 (1999). 6. Hori, T. Gate dielectrics and MOS ULSIs. (Springer Verlag, Berlin, 1997). 7. H.C. Schniepp, et. al. Functionalized single graphene sheets derived from splitting graphite oxide. J. Phys. Chem. B. 110, 8535-8539 (2006). 8. H. Chang. et. al. A transparent, flexible, low-temperature, and solution-processible graphene composite electrode, Adv. Funct. Mater., 20, 2893-2902 (2010). 9. S.-I. Part, et. al. Theoretical and experimental studies of bending of inorganic electronic materials on plastic substrates. Adv. Funct. Mater. 18, 2673-2684 (2008).
Supplementary Movie S1. Supplementary Movie S2. Supplementary Movie S3.
