Deformable field-effect transistors (FETs) are expected to facilitate new technologies like stretchable displays, conformal devices, and electronic skins. We previously demonstrated stretchable FETs based on buckled thin films of polyfluorene-wrapped semiconducting single-walled carbon nanotubes as the channel, buckled metal films as electrodes, and unbuckled flexible ion gel films as the dielectric. The FETs were stretchable up to 50% without appreciable degradation in performance before failure of the ion gel film. Here, we show that by buckling the ion gel, the integrity and performance of the nanotube FETs are extended to nearly 90% elongation, limited by the stretchability of the elastomer substrate. The FETs maintain an on/off ratio of >104 and a field-effect mobility of 5 cm2 V−1 s−1 under elongation and demonstrate invariant performance over 1000 stretching cycles.

Electronics that are not simply flexible but stretchable as well have the potential to enable novel and unconventional applications1,2 such as biosensors that are conformal,3 wearable devices,4 and stretchable displays.5 One critically important but challenging element to implement in a stretchable manner is the field-effect transistor (FET)6–12—the fundamental basis for complex circuits. Thin-film percolating networks of electronic-type controlled semiconducting carbon nanotubes are highly intriguing options for the active channel of stretchable FETs due to the excellent mechanical resilience of individual nanotubes,13 their possibility to accommodate large strain in thin-film form via nanotube–nanotube sliding and buckling,14 and their exceptional charge transport properties.15 

We previously demonstrated stretchable FETs based on buckled thin films of polyfluorene-wrapped semiconducting single-walled carbon nanotubes as the channel, buckled metal films as electrodes, and unbuckled flexible ion gel films as the dielectric.16 Ion gels are attractive for stretchable FETs because of their excellent ionic conductivity, high specific capacitance, printability, and flexibility.17–20 The high double layer capacitance of ion gels also allows FETs to operate at low gate biases of less than 2 V. Our previous nanotube FETs incorporating unbuckled ion gel dielectrics were stretchable up to 50% without appreciable degradation in performance before fracture of the ion gel film.16 Here, we show that by buckling the ion gel, the integrity and performance of the FETs are extended to nearly 90% elongation, limited only by the degree of pre-strain that can be accommodated by the elastomeric substrate during fabrication. The FETs maintain an on/off ratio of >104 and a field-effect mobility of 5 cm2 V−1 s−1 under elongation and demonstrate invariant performance over 1000 stretching cycles.

The procedures used to prepare the polymer wrapped carbon nanotubes and ion gels and to electrically characterize the FETs are similar to those used in our previous report16 and are described in detail in the supplementary material.21 Figure 1(a) depicts the new FET fabrication approach. First, a rectangular Polydimethylsiloxane (PDMS) substrate is held on two ends and elongated by a pre-strain of 80%. Next, unlike in our previous work, the nanotubes, electrodes, and ion gel are all deposited onto the stretched PDMS, to define the FET, before it is released. The nanotube network is deposited via doctor-blade casting a 1,2-dichlorobenzene solution of (7, 5) enriched semiconducting single-walled carbon nanotubes wrapped by poly(9,9-dioctylfluorenyl-2,7-diyl) (PFO), at 130 °C in a nitrogen glove box. The electrodes Cr/Au/Cr (4 nm/26 nm/4 nm) are deposited via thermal evaporation through a shadow mask, defining a channel of length and width of 0.1 mm and 1 mm, respectively. The length of the channel (from source to drain) is parallel to the direction of strain. The ion gel is deposited by drop-casting a solution of 1-ethyl-3-methylimidazolium in ethyl acetate, followed by post-annealing at 65 °C for 30 min and then 105 °C for 60 min. Finally, the PDMS substrate is released manually over a duration of roughly 30 s to its original size.

FIG. 1.

(a) Schematic of device fabrication. (b) SEM image of buckled nanotube film (scale bar = 500 nm). (c) SEM image of buckled metal electrode film (scale bar = 2 μm). (d) Optical image of buckled ion gel film (scale bar = 100 μm).

FIG. 1.

(a) Schematic of device fabrication. (b) SEM image of buckled nanotube film (scale bar = 500 nm). (c) SEM image of buckled metal electrode film (scale bar = 2 μm). (d) Optical image of buckled ion gel film (scale bar = 100 μm).

Close modal

The relaxation of the PDMS substrate to its original length induces buckling of all the components. The characteristic length-scale for buckling depends on the thickness of the layers and their mechanical properties, including Young's modulus and Poisson ratio. The nanotube network alone, the electrodes alone, and the ion gel on top of the nanotube network buckle with characteristic wavelength-scales of 0.1–0.2 μm, 1–5 μm, and 10–100 μm, respectively (Figs. 1(b), 1(c), and 1(d), respectively). Hobbie et al. have shown that the wrinkling of nanotube films can be described by treating them as networks of “rod-like” particles with weakly interacting junctions.22 Instead of sinusoidal wrinkles, “cusp-like” folds are observed due to mesoscopic thickness and modulus variation within the networks.23 The amplitude of the wrinkles in the ion gel film varies from 2 to 4 μm (Fig. S1). Small cracks are noted in the buckled metal electrodes along with the wrinkles (Fig. 1(c)), which have been observed by Akogwu et al. for similar metal films.24 These cracks decrease the electrode conductivity; however, the electrode conductivity is never a limiting factor at any strain, as shown below.

The electrical behaviors of the released FETs are characterized in Figs. 2(a) and 2(b). Fig. 2(a) presents representative output curves (drain current-drain voltage, IDVD), which show typical linear and saturated characteristics. Fig. 2(b) presents representative transfer curves (drain current-gate voltage, IDVG), indicating p-type FET behavior. The high double layer capacitance of the ion gel allows the device to turn on at low gate biases, for example, ID reaches more than 0.5 μA μm−1 at VG = −2 V and VD = −1 V. The released FETs have a typical on/off ratio of ∼104. As determined from the linear regime of the transfer curves at VD = −0.1 V (inset of Fig. 2(b)), a field-effect mobility of 5 cm2 V−1 s−1 is calculated using a measured ion gel capacitance of 4.4 μF cm−2 (Fig. S2). A slightly higher field-effect mobility of 6.1 cm2 V−1 s−1 is obtained if also accounting for the carbon nanotube quantum capacitance.21,25,26 The observed hysteresis in the IDVG curve is typical of nanotubes that are solution processed in air without encapsulation or surface treatment.27 

FIG. 2.

Typical output (a) and transfer (b) current-voltage characteristics of nanotube FETs (inset shows the transfer curve on linear scale at VD = −0.1 V).

FIG. 2.

Typical output (a) and transfer (b) current-voltage characteristics of nanotube FETs (inset shows the transfer curve on linear scale at VD = −0.1 V).

Close modal

After release, the substrates are held at the ends and stretched, and the electrical behaviors of the FETs are measured as a function of the substrate elongation. Figs. 3(a) and 3(b) show the transfer curves on log and linear scales, respectively, as a function of the elongation. The FETs exhibit stable performance up to an elongation of 88% before decaying at higher elongation. The mobility and on/off current are maintained up to an elongation of 80% (Fig. 3(c)). The experiments are repeated for three times to assess reproducibility. At zero elongation, the mobility and on/off ratio are 4.7 ± 1.1 cm2 V−1 s−1 and 104.6±0.06, respectively. At 80% elongation, these parameters are 4.6 ± 1.1 cm2 V−1 s−1 and 104.6±0.05, respectively. Some variance in the initial mobility is observed and can be attributed to the hand-casting process of depositing the nanotube and ion gel films, leading to variation in film thickness. More reproducible casting techniques, for example, spray printing, will likely allow for better control of film thickness in future implementations.28,29 Nonetheless, all the FETs demonstrate relatively stable performance up to 80% strain.

FIG. 3.

The device performance as a function of applied strain. Typical transfer current-voltage characteristics of nanotube FETs on a log (a) and linear (b) scale (VD = −0.1 V). (c) On and off currents and mobility as a function of applied strain.

FIG. 3.

The device performance as a function of applied strain. Typical transfer current-voltage characteristics of nanotube FETs on a log (a) and linear (b) scale (VD = −0.1 V). (c) On and off currents and mobility as a function of applied strain.

Close modal

As the FETs are elongated, the wrinkles in the nanotube network (Fig. 4(a)), electrodes, and ion gel (Fig. S3) begin to change morphology, as the films flatten. For example, after release, the nanotube network is buckled perpendicular to the pre-strain direction within a periodicity of 150 nm. As the substrate is stretched, the perpendicular wrinkles are reduced and mostly disappear at 80% elongation (corresponding to the pre-strain elongation during device fabrication). At an elongation of 100%, wrinkles that run parallel to the direction of elongation form due to the compression of the substrate in the transverse direction due to the Poisson ratio of the PDMS. This observation is similar to that observed in our previous work; however, the transition between perpendicular and parallel wrinkles now occurs at an elongation of more than 80% compared to 50%, corresponding to the pre-strain of the PDMS substrate during nanotube film casting in each case.

FIG. 4.

(a) SEM images of buckled nanotube film as a function of elongation (scale bar = 1 μm). (b) On and off currents and mobility as a function of cycle for an elongation range of 0%–70%.

FIG. 4.

(a) SEM images of buckled nanotube film as a function of elongation (scale bar = 1 μm). (b) On and off currents and mobility as a function of cycle for an elongation range of 0%–70%.

Close modal

This reconfiguration of the wrinkles is reversible after the FETs are relaxed. In order to quantify the extent of the reversibility, we stretch a FET to 70% elongation and relax it repeatedly over 1000 cycles, at a rate of about 0.2 cycles s−1 (Fig. 4(b)). The on-current (at VD = −0.1 V) increases from 0.028 to 0.035 μA μm−1 and the mobility is enhanced from 4.2 to 5.5 cm2 V−1 s−1 over the first 200 stretch-release cycles and are then invariant, except for measurement to measurement fluctuations, for the next 800 cycles. The off-current is invariant over all 1000 cycles. The initial increase in on-current and mobility may indicate that the nanotube network reorganizes during the first 200 cycles.

The failure of the FETs when they are elongated over 90% coincides with a delamination of the ion gel films from the PDMS substrates (Fig. S4). We hypothesize that the 90% point of failure is determined by the magnitude of the initial pre-strain used during device fabrication (here, 80%). Pre-strain of the PDMS exceeding 80% was not possible at the 130 °C used for casting the nanotube films.

To show that each individual component of the FETs maintains integrity during stretching, we separately study the electrical characteristics and morphology of the nanotube network, ion gel, and metal electrodes as a function of elongation. In the absence of the ion gel, the nanotube network conductance at VD = −20 V is invariant up to an elongation of 95% (Fig. S5), and the film morphology is still excellent up to elongation of 100% (Fig. 4(a)).

To measure the conductance of the metal electrodes, alone, as a function of elongation, we fabricate a buckled metal stripe, of length 3 mm and width 0.36 mm, on an 80% elongated PDMS substrate, and then release the substrate to buckle the metal film in the same way that the FET electrodes are buckled. Upon subsequent stretching, the conductance of the stripe remains relatively stable up to an elongation of 50% and then sharply increases as the elongation is further increased up to 100% (Fig. S6(a)). This increase may indicate closure of the cracks previously noted as the metal film unwrinkles. For elongation over 100%, the conductance decreases because additional cracks perpendicular to the elongation direction are formed,30 as evidenced in the optical micrograph in Fig. S6(b). It is worth noting that even at an elongation of 105%, when the conductance of the metal electrodes is significantly reduced, the conductance of the electrodes is still 10 times higher than the measured on-conductance of the FETs. These results indicate that the metal electrodes are not the source of the failure of the FETs when they are elongated over 90%.

The capacitance of a gold electrode/ion gel interface is measured using a LCR meter (20 Hz), as a function of elongation, to probe the integrity of the ion gel films. The capacitance increases slightly from 3.7 to 4.7 μF cm−2 as the strain increases from 0% to 90%. When the applied strain reaches over 90%, the capacitance starts to decrease (Fig. S2), coinciding with the point at which the ion gel films begin to delaminate. Fig. S3 shows optical micrographs of an ion gel film at 0%, 25%, 50%, and 100% elongation. With increasing elongation, the buckling period increases and the film flattens. Interestingly, unlike the nanotube film, when the elongation reaches 100%, some local buckling can still be seen, potentially due to the film thickness inhomogeneity.

It should be noted that in this work, the FETs are not tested under elongation that is transverse to the original direction of the pre-strain. However, previous nanotube FETs fabricated with unbuckled ion gels have been able to accommodate a 10% elongation in the transverse direction.16 Ultimately, devices that can operate under much higher transverse elongation should be possible by applying biaxial pre-strain and buckling strategies.

In summary, a simple buckling method has been demonstrated to fabricate highly stretchable carbon nanotube FETs on elastomeric substrates using ion gel films as dielectrics. The stretchable FETs achieve a high on/off ratio of ∼104 and a mobility of ∼5 cm2 V−1 s−1, without degradation of performance, at up to 90% substrate elongation, and performance is invariant for at least 1000 cycles. Analysis indicates that the individual components (the nanotubes, metal electrodes, and ion gel) are stable beyond 90% elongation. However, the FETs fail when the ion gel films delaminate from the substrates at elongation larger than the pre-strain used during fabrication (here, 80%). These results point to the possibility of achieving even higher performance at more extreme elongation by using elastomeric substrates with higher yield strain. This work is expected to enable new, unconventional, and next-generation applications in electronics and optoelectronics like implantable conformal biosensors, wearable electronic devices, and stretchable displays.

This work was primarily supported by the University of Wisconsin-Madison Center of Excellence for Materials Research and Innovation (DMR-1121288) (M.-Y.W.) and U.S. Army Research Office, W911NF-12-1-0025 (J.Z., F.X., and M.S.A). This work was partially supported via the Department of Energy (DOE) Office of Science Early Career Research Program (Grant No. DE-SC0006414) through the Office of Basic Energy Sciences (R.M.J.), for microstructural characterization of the buckled ion gel films. R.M.J. also acknowledges support from the Department of Defense (DOD) through the National Defense Science and Engineering Graduate Fellowship (NDSEG) Program. This work was also partially supported by the Air Force Office of Scientific Research (AFOSR) under Grant No. FA9550-09-1-0482 (T.-H.C. and Z.M.), for analysis of ion gel failure under elongation.

1.
J. A.
Rogers
,
T.
Someya
, and
Y. G.
Huang
,
Science
327
(
5973
),
1603
(
2010
).
2.
T.
Sekitani
and
T.
Someya
,
Adv. Mater.
22
(
20
),
2228
(
2010
).
3.
S. W.
Hwang
,
H.
Tao
,
D. H.
Kim
,
H. Y.
Cheng
,
J. K.
Song
,
E.
Rill
,
M. A.
Brenckle
,
B.
Panilaitis
,
S. M.
Won
,
Y. S.
Kim
,
Y. M.
Song
,
K. J.
Yu
,
A.
Ameen
,
R.
Li
,
Y. W.
Su
,
M. M.
Yang
,
D. L.
Kaplan
,
M. R.
Zakin
,
M. J.
Slepian
,
Y. G.
Huang
,
F. G.
Omenetto
, and
J. A.
Rogers
,
Science
337
(
6102
),
1640
(
2012
).
4.
D. H.
Kim
,
N. S.
Lu
,
R.
Ma
,
Y. S.
Kim
,
R. H.
Kim
,
S. D.
Wang
,
J.
Wu
,
S. M.
Won
,
H.
Tao
,
A.
Islam
,
K. J.
Yu
,
T. I.
Kim
,
R.
Chowdhury
,
M.
Ying
,
L. Z.
Xu
,
M.
Li
,
H. J.
Chung
,
H.
Keum
,
M.
McCormick
,
P.
Liu
,
Y. W.
Zhang
,
F. G.
Omenetto
,
Y. G.
Huang
,
T.
Coleman
, and
J. A.
Rogers
,
Science
333
(
6044
),
838
(
2011
).
5.
T.
Sekitani
,
H.
Nakajima
,
H.
Maeda
,
T.
Fukushima
,
T.
Aida
,
K.
Hata
, and
T.
Someya
,
Nat. Mater.
8
(
6
),
494
(
2009
).
6.
D. Y.
Khang
,
H. Q.
Jiang
,
Y.
Huang
, and
J. A.
Rogers
,
Science
311
(
5758
),
208
(
2006
).
7.
D. H.
Kim
,
J. H.
Ahn
,
W. M.
Choi
,
H. S.
Kim
,
T. H.
Kim
,
J. Z.
Song
,
Y. G. Y.
Huang
,
Z. J.
Liu
,
C.
Lu
, and
J. A.
Rogers
,
Science
320
(
5875
),
507
(
2008
).
8.
S. K.
Lee
,
B. J.
Kim
,
H.
Jang
,
S. C.
Yoon
,
C.
Lee
,
B. H.
Hong
,
J. A.
Rogers
,
J. H.
Cho
, and
J. H.
Ahn
,
Nano Lett.
11
(
11
),
4642
(
2011
).
9.
S. H.
Chae
,
W. J.
Yu
,
J. J.
Bae
,
D. L.
Duong
,
D.
Perello
,
H. Y.
Jeong
,
Q. H.
Ta
,
T. H.
Ly
,
Q. A.
Vu
,
M.
Yun
,
X.
Duan
, and
Y. H.
Lee
,
Nat. Mater.
12
(
5
),
403
(
2013
).
10.
M.
Kaltenbrunner
,
T.
Sekitani
,
J.
Reeder
,
T.
Yokota
,
K.
Kuribara
,
T.
Tokuhara
,
M.
Drack
,
R.
Schwodiauer
,
I.
Graz
,
S.
Bauer-Gogonea
,
S.
Bauer
, and
T.
Someya
,
Nature
499
(
7459
),
458
(
2013
).
11.
H. S.
Wu
,
S.
Kustra
,
E. M.
Gates
, and
C. J.
Bettinger
,
Org. Electron.
14
(
6
),
1636
(
2013
).
12.
A.
Chortos
,
J.
Lim
,
J. W. F.
To
,
M.
Vosgueritchian
,
T. J.
Dusseault
,
T.-H.
Kim
,
S.
Hwang
, and
Z.
Bao
,
Adv. Mater.
26
(
25
),
4253
(
2014
).
13.
M. F.
Yu
,
O.
Lourie
,
M. J.
Dyer
,
K.
Moloni
,
T. F.
Kelly
, and
R. S.
Ruoff
,
Science
287
(
5453
),
637
(
2000
).
14.
D. J.
Lipomi
,
M.
Vosgueritchian
,
B. C. K.
Tee
,
S. L.
Hellstrom
,
J. A.
Lee
,
C. H.
Fox
, and
Z. N.
Bao
,
Nat. Nanotechnol.
6
(
12
),
788
(
2011
).
15.
T.
Durkop
,
S. A.
Getty
,
E.
Cobas
, and
M. S.
Fuhrer
,
Nano Lett.
4
(
1
),
35
(
2004
).
16.
F.
Xu
,
M.-Y.
Wu
,
N. S.
Safron
,
S. S.
Roy
,
R. M.
Jacobberger
,
D. J.
Bindl
,
J.-H.
Seo
,
T.-H.
Chang
,
Z.
Ma
, and
M. S.
Arnold
,
Nano Lett.
14
(
2
),
682
(
2014
).
17.
J. H.
Cho
,
J.
Lee
,
Y.
Xia
,
B.
Kim
,
Y. Y.
He
,
M. J.
Renn
,
T. P.
Lodge
, and
C. D.
Frisbie
,
Nat. Mater.
7
(
11
),
900
(
2008
).
18.
J.
Lee
,
L. G.
Kaake
,
J. H.
Cho
,
X. Y.
Zhu
,
T. P.
Lodge
, and
C. D.
Frisbie
,
J. Phys. Chem. C
113
(
20
),
8972
(
2009
).
19.
M. J.
Ha
,
Y.
Xia
,
A. A.
Green
,
W.
Zhang
,
M. J.
Renn
,
C. H.
Kim
,
M. C.
Hersam
, and
C. D.
Frisbie
,
ACS Nano
4
(
8
),
4388
(
2010
).
20.
J.
Pu
,
Y.
Yomogida
,
K. K.
Liu
,
L. J.
Li
,
Y.
Iwasa
, and
T.
Takenobu
,
Nano Lett.
12
(
8
),
4013
(
2012
).
21.
See supplementary material at http://dx.doi.org/10.1063/1.4928041 for experimental details; nanotube quantum capacitance calculations; characterization of the buckling and delamination of the ion gel films; and failure analysis of the individual components.
22.
E. K.
Hobbie
,
D. O.
Simien
,
J. A.
Fagan
,
J. Y.
Huh
,
J. Y.
Chung
,
S. D.
Hudson
,
J.
Obrzut
,
J. F.
Douglas
, and
C. M.
Stafford
,
Phys. Rev. Lett.
104
,
125505
(
2010
).
23.
J. M.
Harris
,
J. Y.
Huh
,
M. R.
Semler
,
T.
Ihle
,
C. M.
Stafford
,
S. D.
Hudson
,
J. A.
Fagan
, and
E. K.
Hobbie
,
Soft Matter
9
,
11568
(
2013
).
24.
O.
Akogwu
,
D.
Kwabi
,
S.
Midturi
,
M.
Eleruja
,
B.
Babatope
, and
W. O.
Soboyejo
,
Mater. Sci. Eng., B
170
(
1–3
),
32
(
2010
).
25.
V. K.
Sangwan
,
R. P.
Ortiz
,
J. M. P.
Alaboson
,
J. D.
Emery
,
M. J.
Bedzyk
,
L. J.
Lauhon
,
T. J.
Marks
, and
M. C.
Hersam
,
ACS Nano
6
(
8
),
7480
(
2012
).
26.
G. J.
Brady
,
Y.
Joo
,
M.-Y.
Wu
,
M. J.
Shea
,
P.
Gopalan
, and
M. S.
Arnold
,
ACS Nano
8
(
11
),
11614
(
2014
).
27.
W.
Kim
,
A.
Javey
,
O.
Vermesh
,
Q.
Wang
,
Y.
Li
, and
H.
Dai
,
Nano Lett.
3
(
2
),
193
(
2003
).
28.
H.
Numata
,
K.
Ihara
,
T.
Saito
,
H.
Endoh
, and
F.
Nihey
,
Appl. Phys. Express
5
(
5
),
055102
(
2012
).
29.
P.
Chen
,
Y.
Fu
,
R.
Aminirad
,
C.
Wang
,
J.
Zhang
,
K.
Wang
,
K.
Galatsis
, and
C.
Zhou
,
Nano Lett.
11
(
12
),
5301
(
2011
).
30.
J.
Lee
,
W.
Kim
, and
W.
Kim
,
ACS Appl. Mater. Interfaces
6
(
16
),
13578
(
2014
).

Supplementary Material