ORGELE 2039 No.

of Pages 7, Model 3G
13 March 2013
Organic Electronics xxx (2013) xxx–xxx
1

Contents lists available at SciVerse ScienceDirect

Organic Electronics
journal homepage: www.elsevier.com/locate/orgel

5
6

3 Origin of mechanical strain sensitivity of pentacene thin-film

4 transistors
7 Q1 V. Scenev a,⇑, P. Cosseddu b,c, A. Bonfiglio b,c, I. Salzmann a, N. Severin a, M. Oehzelt d, N. Koch a,d,
8 J.P. Rabe a,⇑
9 a
Humboldt-Universität zu Berlin, Institut für Physik, Newtonstr. 15D-12489 Berlin, Germany
10 b
University of Cagliari, Dep. of Electrical and Electronic Engineering, Piazza d’Armi, 09123 Cagliari, Italy
11 c
CNR–Institute of Nanoscience, Centre S3 via Campi 213A, I-41100 Modena, Italy
12 d
Helmholtz Zentrum Berlin für Materialien und Energie–BESSY II, D-12489 Berlin, Germany

14
13
15
a r t i c l e i n f o a b s t r a c t
1
3 7
0
18 Article history: We report on bending strain-induced changes of the charge carrier mobility in pentacene 31
19 Received 20 November 2012 organic thin-film transistors employing a combined investigation of morphological, struc- 32
20 Received in revised form 4 February 2013 tural, and electrical properties. The observed drain current variations are reversible if the 33
21 Accepted 22 February 2013
deformation is below 2%. The morphology and structure of the active pentacene layer is 34
22 Available online xxxx
investigated by scanning force microscopy and specular synchrotron X-ray diffraction, 35
which show that bending-stress causes morphological rather than structural changes, 36
23 Q3 Keywords:
modifying essentially the lateral spacing between individual pentacene crystallites. In 37
24 Pentacene
25 Strain
addition, for deformations >2% the rupture of source and drain gold electrodes is observed. 38
26 OFET In contrast to the metal electrodes, the modification of the organic layer remains reversible 39
27 Bending experiment for deformations up to 10%, which suggests the use of soft and flexible electrodes such as 40
28 Morphology graphene or conducting polymers to be beneficial for future strain sensing devices. 41
29 Crown Copyright Ó 2013 Published by Elsevier B.V. All rights reserved. 42

43
44
45 1. Introduction types of sensors: they can be processed under ambient 59
conditions and are generally inexpensive to fabricate 60
46 Thin films of conjugated organic molecules are subject [4–11]. For sensing strain, reversible changes in the electri- 61
47 of intense research due to their applicability in novel cal characteristics of OTFTs were employed, including 62
48 (opto-)electronic devices, including organic thin-film tran- drain current, charge carrier mobility, threshold voltage, 63
49 sistors (OTFTs). Pentacene (PEN) is the prototypical hole- and contact resistance for deformations up to 1–2% 64
50 conducting material with notably high charge carrier [5,6,9,12,13]. In particular, the reported changes in mobil- 65
51 mobilities of up to 5.5 cm2/V s [1–3] in p-type OTFTs. ity were proposed to be due to morphological changes of 66
52 One key advantage of organic electronic devices is the pos- the PEN layers under mechanical strain and/or the activa- 67
53 sibility to produce flexible all-organic OTFTs with the func- tion of trap states in the PEN/electrode interface region 68
54 tional organic semiconductor films deposited on flexible [5,6,9,14,15]. However, up to now there has been no direct 69
55 plastic foils like MylarÒ or polyethylenetherephthalate experimental evidence to support these suggestions. In or- 70
56 (PET) as substrate [4]. Through their intrinsic flexibility, der to enable targeted research for improving current 71
57 OTFTs can be applied as mechanical strain-sensing devices OTFTs for future reliable sensing applications, further 72
58 [5,6] exhibiting significant advantages over conventional experimental work is needed to complete the microscopic 73
picture of strain-impact on OTFT performance. Here we re- 74
port a comprehensive electrical characterization of flexible 75
Q2 ⇑ Corresponding authors. Tel.: +49 1786905049 (V. Scenev). OTFT devices under applied bending-stress, relating the 76
E-mail addresses: vscenev@aol.com (V. Scenev), rabe@physik. device characteristics to morphological and structural 77
hu-berlin.de (J.P. Rabe).

1566-1199/$ - see front matter Crown Copyright Ó 2013 Published by Elsevier B.V. All rights reserved.
http://dx.doi.org/10.1016/j.orgel.2013.02.030

Please cite this article in press as: V. Scenev et al., Origin of mechanical strain sensitivity of pentacene thin-film transistors, Org. Electron.
(2013), http://dx.doi.org/10.1016/j.orgel.2013.02.030
 ORGELE 2039 No. of Pages 7, Model 3G
13 March 2013
2 V. Scenev et al. / Organic Electronics xxx (2013) xxx–xxx

78 properties of the active layer, as determined by a combina- measurements. The electrical and morphological charac- 108
79 tion of scanning force microscopy (SFM) and specular X- terization of the PEN layers under deformation was carried 109
80 ray diffraction (XRD). We identify the deterioration of the out in situ under ambient conditions. For electrical charac- 110
81 metal electrodes to be responsible for irreversibilities in terization a Keithley 2636A sourcemeter was used. The 111
82 the device characteristics, while the observed modifica- surface topography was imaged with SFM (Nano Wizard, 112
83 tions of the active organic layer remain reversible for JPK instruments). The PEN film strain was calculated with 113
84 deformations up to 10%, which suggests using soft organic the model of beam buckling (Fig. 1b), which allows to cal- 114
85 electrodes for future improved sensing applications. culate the film strain, e, directly from the bending radius 115
[16], as demonstrated before [9,17]. The bending radius 116
was calculated taking into account the homogeneous 117
86 2. Experimental bending of the PET foil. This setup allowed to strain PEN 118
films up to 10%, with the upper limit defined by the small- 119
87 OTFTs were fabricated on 200 lm thin and flexible est accessible radius of curvature and the thickness of the 120
88 polyethylenetherephthalate (PET) substrates. A gold layer PET substrate. All electrical device characterization, XRD 121
89 was vacuum-deposited on the substrates as gate electrode. and SFM measurements, were carried out under ambient 122
90 Polyvinylalcohol (PVA) with ammonium dichromate (AD) conditions. 123
91 salt (cross-linking agent) was deposited from water solu-
92 tion as gate dielectric by spin-coating and subsequent
93 UV-curing for cross-linking. Gold source and drain elec- 3. Results 124
94 trodes (Fig. 1a) were patterned by thermal deposition
95 through a shadow mask. Pentacene (PEN) was used as or- Transfer curves (i.e., drain current (ID) versus gate volt- 125
96 ganic semiconductor. PEN films were evaporated at a base age (VG) characteristics) of the devices were recorded for 126
97 pressure <107 mbar; the film thickness and the deposition different degrees of strain. The bending-stress was released 127
98 rate (0.5 nm/min) were monitored by a quartz crystal after acquiring transfer characteristic for a particular strain 128
99 microbalance placed next to the sample; the device is in order to verify the reversibility of the process (i.e., if the 129
100 sketched in Fig. 1a. A special apparatus for defined bending current recovers to the initial value). We find, as expected 130
101 of the OTFT (Fig. 1c) was developed to be suitable for use in [9], a decrease in drain current as function of device defor- 131
102 both SFM and specular XRD experiments. XRD measure- mation (Fig. 2a). For deformations of up to 1.7%, straining 132
103 ments were carried out at the beamline W1 at DESY– of the device causes reversible variations of the transfer 133
104 HASYLAB (Hamburg, Germany) with a primary beam en- curves. Starting from 1.7% ID does not recover to the initial 134
105 ergy of 10.5 keV, using the bending apparatus (Fig. 1c) as value of the pristine device upon bending-stress relief. 135
106 sample holder for in situ experiments, and a standard setup Transfer characteristics exhibit a pronounced hysteresis 136
107 (PVA samples fixed planar on a Si wafer) for reference with the hysteresis loop-area increasing with increasing 137

Fig. 1. (a) Schematic representation of the top-contact OTFT architecture. (b) Model used to calculate PEN film strain e; R denotes the radius of curvature, t
the device thickness. Red and dashed green lines indicate the upper and neutral plane, respectively. Strain of PEN film is assumed to be equal to the strain of
the beam surface. (c) The bending apparatus with the OTFT device in place. (d) Scheme of the device under bending stress with the SFM-tip; electrical
contacts are indicated for the drain-current sensitivity measurements in the deformed OTFT and for the in situ morphological investigations on both
electrodes and active layer by SFM. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)

Please cite this article in press as: V. Scenev et al., Origin of mechanical strain sensitivity of pentacene thin-film transistors, Org. Electron.
(2013), http://dx.doi.org/10.1016/j.orgel.2013.02.030
 ORGELE 2039 No. of Pages 7, Model 3G
13 March 2013
V. Scenev et al. / Organic Electronics xxx (2013) xxx–xxx 3

Fig. 2. Electrical characteristics of a representative OTFT device. (a) Transfer curves recorded during successive device bending and relief of bending stress
after each deformation (indicated with e = 0%) with VD = 60 V. (b) Transfer curves representing the increase of the hysteresis loop area with increasing
deformation (with VD = 60 V). (c) Dependency of the hysteresis loop area versus bending deformation. (d) Device strain sensitivity of the drain current at
60 V gate voltage versus deformation; the red line denotes the linear fit within the range, where the device sensitivity is reversible; the green line indicates
the transition from reversible to irreversible variation of drain current. (For interpretation of the references to color in this figure legend, the reader is
referred to the web version of this article.)

138 deformation (Fig. 2b and c). This is attributed to charge car- film morphology (Fig. 3b–d). Due to the distortions of the 166
139 rier trapping enhanced by the strain. The effect of bending SFM images, which are inherent to the SFM instrumenta- 167
140 strain induced activation of trap states in the active layer tion including piezo creep, the strain of 1.3% could not be 168
141 and its relationship to the hysteresis in transfer character- detected directly. 169
142 istics of OTFTs has already been studied and attributed to The PEN film morphology on PVA substrates remained 170
143 the intrinsic properties of the organic semiconductor qualitatively alike for strains beyond 2%, and the impact 171
144 [6,15]. The sensitivity of ID to the strain is determined by of bending was clearly reflected in the topography images 172
145 plotting the difference between pristine and bended device (Fig. 4b), revealing an increased distance between remote 173
146 currents (recorded with 60 V drain voltage) normalized morphological features for higher values of induced sur- 174
147 to the pristine value versus strain (Fig. 2d). face strain (Fig. 4c). Moreover, the radius of curvature 175
148 Our electrical measurements reveal that the strain- could be directly estimated from the cross-section analysis 176
149 sensitivity of PEN OTFT devices scales almost linearly with of the SFM micrograph (see Supporting Information, 177
150 strain for deformations up to 1.7% (Fig. 2a). Beyond this va- Fig. S1) to (1.1 ± 0.1) mm. Since the bending radius was 178
151 lue we find that the device becomes less sensitive to strain measured [16,19] to be 1 mm (see Supporting Information, 179
152 (Fig. 2b). With our experimental set up it was possible to Fig. S2), the value from cross-section analysis is in very 180
153 obtain the device sensitivity up to 7.5% mechanical stress. good agreement. However, the resolution of the SFM imag- 181
154 The finding that after a certain degree of deformation ing was not sufficient to clearly reveal whether individual 182
155 the current does not recover to its initial value is clear evi- grains became strained or whether the distance between 183
156 dence that irreversible processes occurred in the device grains (along grain boundaries) increased. Therefore we 184
157 structure. To shed light on this issue we performed mor- performed specular XRD measurements to verify straining 185
158 phological and structural investigations of the deformed of PEN crystallites as described further below. 186
159 device, with particular attention on the active layer and In contrast to PEN films on PVA, we detected the forma- 187
160 the electrodes. tion of cracks within the gold electrodes on top of PEN film 188
161 SFM height images of unstrained devices reveal uni- for strains larger than 2% (Fig. 4d and e). The cracks per- 189
162 formly distributed grains with lateral grain sizes of few sisted after release of strain. 190
163 hundreds of nm (Fig. 3a–d), typical for PEN films grown To investigate the impact of strain on the PEN film 191
164 on PVA by vacuum deposition [18]. Straining the device structure, we performed specular XRD measurements of 192
165 up to 1.3% did not cause any noticeable variation in PEN PEN films in situ deformed up to 3.3% (Fig. 4f). For the un- 193

Please cite this article in press as: V. Scenev et al., Origin of mechanical strain sensitivity of pentacene thin-film transistors, Org. Electron.
(2013), http://dx.doi.org/10.1016/j.orgel.2013.02.030
 ORGELE 2039 No. of Pages 7, Model 3G
13 March 2013
4 V. Scenev et al. / Organic Electronics xxx (2013) xxx–xxx

Fig. 3. (a) SFM topography of PEN film. (c) Zoom-in before straining; (d) an in situ 1.3% strained PEN film. (b) Provides a cross-section analysis of the SFM
height images carried out on PEN films under deformations of up to 1.3% to allow for a comparison of the lateral distance between two distinctive positions
(marked with arrows in (c)). Arrows in (c) correspond to arrows in the cross-sections (b), respectively. Also for deformations of up to 1.3% no significant
changes in PEN films were found compared to the pristine device. (For interpretation of the references to color in this figure legend, the reader is referred to
the web version of this article.)

Fig. 4. Representative SFM micrographs of (a) a pristine PEN layer, and (b) an in situ 10% strained PEN film, together with (c) the corresponding cross-
sections; arrows in (a) and (b) correspond to arrows in the cross-sections. (d) SFM micrograph of the top gold electrode, demonstrating crack formation
upon deformation of the gold electrode; (e) transition region (interface) of electrode (left)/pentacene (right). (f) Specular XRD data in the range of the (001)-
reflection of the PEN thin-film phase (lattice
0
spacing d (001) = 1.54 nm) recorded as function of in situ applied bending-stress e; qz denotes the vertical
momentum transfer; the peak at 0.46 A Å1 (on the black curve) labels a reflection assigned to the substrate; the peak intensity decreases with the degree of
deformation due to geometrical reasons in the experiment (i.e., less area being illuminated by the X-ray beam). Insets: (left) sketch of the (inclined)
molecular orientation in PEN thin-film phase with respect to the PVA substrate (illustrated by yellow triangles); (right) specular XRD data (ex situ) for a
pristine sample and the same sample after releasing a stress of 4.6%. (For interpretation of the references to color in this figure legend, the reader is referred
to the web version of this article.)

194 strained device (Fig. 4f, black curve) the strong 0peak at a (see yellow triangles in the left inset of Fig. 4f), and that 200
195 A1 corre-
value of vertical momentum transfer (qz) of 0.41 Å the molecules are arranged in a herringbone motif typical 201
196 sponds to the (0 0 1) reflection of the PEN thin-film poly- for this specific polymorph [20]. Recently, bending-stress- 202
197 morph [20]. This indicates that the PEN molecules adopt driven phase transitions were suggested upon flattening 203
198 a standing orientation within the film, however, slightly in- of inward and outward bent PEN devices [21]. To observe 204
199 clined by ca. 6° with respect to the PVA substrate normal such phase transitions upon deformation of the present 205

Please cite this article in press as: V. Scenev et al., Origin of mechanical strain sensitivity of pentacene thin-film transistors, Org. Electron.
(2013), http://dx.doi.org/10.1016/j.orgel.2013.02.030
 ORGELE 2039 No. of Pages 7, Model 3G
13 March 2013
V. Scenev et al. / Organic Electronics xxx (2013) xxx–xxx 5

206 PEN/PVA films, we performed both in situ specular XRD on ing cross-section analyses (Fig. 5) comparing the morphol- 228
207 films deformed up to 3.3% (Fig. 4f), and ex situ specular ogy of the same PEN area before (Fig. 5a and c) and after 229
208 XRD on a film after the release of strain as high as 4.6% (Fig. 5b and d) stress release, where no significant changes 230
209 (Fig. 4f, right inset). Most notably, in no case a shift of in morphology are observed. Note that the identity of the 231
210 the (0 0 1)-peak maximum occurred, which would, how- investigated areas in the two micrographs is clearly con- 232
211 ever, be expected if a phase transition to any other PEN firmed by characteristic morphological features present 233
212 polymorph [22–25] would take place (expected change in in both the micrographs (Fig. 5a and b) and in the respec- 234
213 vertical momentum
0
transfer of the (0 0 1) reflection: tive cross-sections (Fig. 5c and d). Moreover, both micro- 235
214 Dqz P 0.02 A Å1), and if, therefore, the lateral intermolecu- graphs exhibit the same morphological features, which 236
215 lar distance would accordingly change. Note that, apart demonstrates that the PEN film completely recovers to 237
216 from phase transitions, relevant stress-related changes of its initial state, even after the present severe mechanical 238
217 the intermolecular distance can be excluded within this deformation. 239
218 deformation range, as they typically occur upon pressures
219 in the GPa region, which is reported in related high-pres- 4. Discussion 240
220 sure investigations [26]. Note, that the elastic modulus of
221 PEN film is 15 GPa and this large value is due to the poly- The electrical measurements agree well with previous 241
222 crystalline nature of PEN film [27]. In particular, already a reports, where both spherical [5,6] and cylindrical [9,28] 242
223 change in lattice spacing by 2.5% would correspond to deformations showed similar dependencies of the electri- 243
224 Dqz  0.01 Å1 and would, thus, be well captured by our cal response on mechanical strain, i.e., a reversible varia- 244
225 experiment. tion of transfer curves for small strain values and 245
226 This is also true for even higher deformations of up to irreversible changes for large strain. Our present study re- 246
227 10%, as indicated by SFM micrographs and the correspond- veals two distinct deformation regimes of the flexible 247

Fig. 5. SFM micrographs of (a) a pristine PEN film and (b) a flattened PEN film after deformation to 10%. Corresponding cross-sections taken along white line
between two positions marked with arrows in the SFM micrographs of (c) a pristine and (d) flattened PEN film after deformation to 10%. (For interpretation
of the references to color in this figure legend, the reader is referred to the web version of this article.)

Please cite this article in press as: V. Scenev et al., Origin of mechanical strain sensitivity of pentacene thin-film transistors, Org. Electron.
(2013), http://dx.doi.org/10.1016/j.orgel.2013.02.030
 ORGELE 2039 No. of Pages 7, Model 3G
13 March 2013
6 V. Scenev et al. / Organic Electronics xxx (2013) xxx–xxx

248 OTFTs: reversible deformation up to 10% occurs for PEN on structural investigations. Our results indicate that the 306
249 PVA substrates, while an irreversible deterioration of the bending-strain sensitivity of PEN OTFTs is due to changes 307
250 gold electrodes is observed starting at a strain value of in the film morphology and that the correlated increase 308
251 2%. Note that, in full agreement with this observation, thin in spacing between PEN crystallites is responsible for the 309
252 gold electrodes have been previously reported to deterio- observed decrease in drain current, which is found to be 310
253 rate for strains exceeding 2% [29]. As it has been studied reversible for small deformations. From our study, the irre- 311
254 elsewhere [29,30], the electrical resistance of thin gold versibility in electrical characteristics for deformations lar- 312
255 films decreases with mechanical strain. Moreover, very re- ger than 2% is recognized to be due to crack formation in 313
256 cently, A.N. Sokolov and coworkers reported about irre- gold source- and drain electrodes. In contrast to irrevers- 314
257 versible changes observed in OTFTs upon mechanical ible changes in the metal layers, PEN itself is found to be 315
258 deformation. In particular, it was found that for surface fully reversibly modified even for extreme deformations 316
259 strain higher than 1.3%, source and drain electrodes buck- of up to 10%, which identifies the organic/metal interface 317
260 ling and cracking starts to appear. This issue typically led as bottleneck of the investigated OTFTs and suggests the 318
261 to a decrease of the electrodes conductivity that induced use of soft flexible electrodes for future strain-sensing 319
262 a permanent decrease of the OTFT output current [31]. devices. 320
263 Therefore, we propose that the observed irreversible varia-
264 tion of transfer curves is mainly due to the deterioration of Acknowledgements 321
265 the gold electrodes. The reversibility of both the transfer
266 curve variation and the film topography until gold elec- We thank Wolfgang Caliebe (DESY–HASYLAB), Tatjana 322
267 trode deterioration starts infer that the PEN film itself sus- Djuric and Armin Moser (both TU-Graz) for experimental 323
268 tains large strains of up to, at least, 10%. Most importantly, support. Financing through the DFG (Germany) and the 324
269 our study reveals that PEN films might be used in devices Austrian Science Fund (FWF) project P21094-N20 is grate- 325
270 to sense substantially larger strains than previously re- fully acknowledged. 326
271 ported. From these findings, hence, a design strategy for fu-
272 ture improved organic sensors clearly arises: using soft,
273 flexible and mechanically stable materials like graphene Appendix A. Supplementary material 327

274 or conductive polymers might enable the full reversibility

275 of strain-related measures in reliable future sensor devices. Supplementary data associated with this article can be 328

276 Furthermore, our study allows proposing a well found, in the online version, at http://dx.doi.org/10.1016/ 329

277 founded model for the reason of the observed reversibility j.orgel.2013.02.030. 330

278 of transfer curves variations. One may speculate of tensile

279 stress having two distinct effects on the PEN thin film: (i), References 331
280 the lateral size of the pentacene crystals might become
[1] T. Hasegawa, J. Takeya, Science and Technology of Advanced 332
281 stretched, therefore modifying the PEN microstructure, Materials 10 (2009) 024314. 333
282 which might be detrimental to charge transport, or (ii), [2] Y. Yamashita, Science and Technology of Advanced Materials 10 334
283 the distance between individual neighboring PEN crystal- (2009) 024313. 335
[3] N. Koch, ChemPhysChem 8 (2007) 1438. 336
284 lites might be increased, therefore reducing the tunneling [4] P. Cosseddu, A. Bonfiglio, I. Salzmann, J.P. Rabe, N. Koch, Organic 337
285 probability of charge carriers in the process of charge Electronics 9 (2008) 191. 338
286 transport through the grain boundaries. From the SFM [5] I. Manunza, A. Sulis, A. Bonfiglio, Applied Physics Letters 89 (2006) 339
143502. 340
287 and XRD data in our present study the related fundamental 341
[6] I. Manunza, A. Bonfiglio, Biosensors & Bioelectronics 22 (2007) 2775.
288 process can now be clearly identified as (ii): tensile stress [7] G. Darlinski et al., Journal of Applied Physics 97 (2005) 093708. 342
289 leads to an increase in inter-grain distance while leaving [8] H. Kawaguchi, T. Someya, T. Sekitani, T. Sakurai, Ieee Journal of Solid- 343
State Circuits 40 (2005) 177. 344
290 the crystal structure unaffected, as no phase transitions 345
[9] T. Sekitani et al., Applied Physics Letters 86 (2005) 073511.
291 were observed. It is well established that hopping trans- [10] H. Sirringhaus, N. Tessler, R.H. Friend, Science 280 (1998) 1741. 346
292 port, in which a conducting path from the source to the [11] T. Someya et al., Proceedings of the National Academy of Sciences of 347
the United States of America 102 (2005) 12321. 348
293 drain electrode is formed by coupled grain chains oriented
[12] P. Cosseddu, A. Piras, A. Bonfiglio, Ieee Transactions on Electron 349
294 in random directions [9] rather than band transport dom- Devices 58 (2011) 3416. 350
295 inates conduction in polycrystalline thin organic films [13] P. Cosseddu, S. Milita, A. Bonfiglio, Ieee Electron Device Letters 33 351
296 [9,32,32]. The increase in distance between crystalline (2012) 113. 352
[14] F.C. Chen, T.D. Chen, B.R. Zeng, Y.W. Chung, Semiconductor Science 353
297 PEN grains translates into an increase in potential barrier and Technology 26 (2011) 034005. 354
298 for (thermally activated) tunneling between grains, which, [15] T. Lindner, G. Paasch, S. Scheinert, Journal of Applied Physics 98 355
299 therefore, reduces the overall charge carrier mobility in (2005) 114505. 356
[16] J.K.F. Yau, N. Savvides, C.C. Sorrell, Physica C 266 (1996) 223. 357
300 bended PEN thin film OTFTs and, thus, fully explains the 358
[17] T. Sekitani et al., Journal of Non-Crystalline Solids 352 (2006) 1769.
301 observed PEN-OTFT strain sensitivity. [18] T.B. Singh et al., Advanced Materials 17 (2005) 2315. 359
[19] A. Jedaa, M. Halik, Applied Physics Letters 95 (2009) 103309. 360
[20] S. Schiefer, M. Huth, A. Dobrinevski, B. Nickel, Journal of the 361
American Chemical Society 129 (2007) 10316. 362
302 5. Conclusions [21] C. Yang et al., Applied Physics Letters 92 (2008) 243305. 363
[22] R.B. Campbell, J. Trotter, J.M. Robertson, Acta Crystallographica 14 364
(1961) 705. 365
303 The underlying mechanism of bending-strain sensitiv-
[23] D. Nabok et al., Physical Review B 76 (2007) 235322. 366
304 ity observed for PEN OTFTs was investigated by combining [24] S.C.B. Mannsfeld, A. Virkar, C. Reese, M.F. Toney, Z.N. Bao, Advanced 367
305 in situ electrical characterization with morphological and Materials 21 (2009) 2294. 368

Please cite this article in press as: V. Scenev et al., Origin of mechanical strain sensitivity of pentacene thin-film transistors, Org. Electron.
(2013), http://dx.doi.org/10.1016/j.orgel.2013.02.030
 ORGELE 2039 No. of Pages 7, Model 3G
13 March 2013
V. Scenev et al. / Organic Electronics xxx (2013) xxx–xxx 7

369 [25] D. Holmes, S. Kumaraswamy, A.J. Matzger, K.P.C. Vollhardt, [29] S.P. Lacour, S. Wagner, Z.Y. Huang, Z. Suo, Applied Physics Letters 82 376
370 Chemistry–A European Journal 5 (1999) 3399. (2003) 2404. 377
371 [26] M. Oehzelt et al., Physical Review B 74 (2006) 104103. [30] T. Li, Z.Y. Huang, Z. Suo, S.P. Lacour, S. Wagner, Applied Physics 378
372 [27] D. Tahk, H.H. Lee, D.Y. Khang, Macromolecules 42 (2009) 7079. Letters 85 (2004) 3435. 379
373 [28] T. Sekitani, I. Shingo, Y. Kato, T. Someya, Japanese Journal of Applied [31] A.N. Sokolov, Y.D. Cao, O.B. Johnson, Z.A. Bao, Advanced Functional 380
374 Physics Part 1-Regular Papers Brief Communications & Review Materials 22 (2012) 175. 381
375 Papers 44 2841 (2005). [32] Y. Takamatsu, T. Sekitani, T. Someya, Applied Physics Letters 90 382
(2007) 133516. 383
384

Please cite this article in press as: V. Scenev et al., Origin of mechanical strain sensitivity of pentacene thin-film transistors, Org. Electron.
(2013), http://dx.doi.org/10.1016/j.orgel.2013.02.030