Modulation of mechanical properties and stable light energy harvesting by poling in polymer integrated perovskite films: a wide range, linear and highly sensitive tactile sensor

Abstract

The integration of polymer chains leads to modulation of mechanical properties in organolead halide perovskites, MAPbI₃, making their films softer. As a result these films have a lower modulus and are able to dissipate the applied mechanical stimuli. This effect is used to make a self-powered tactile sensor that has a very wide operating range (up to 450 kPa) and linear response and high sensitivity over the whole range in a single structure. Further the films have an energy harvesting density of 1.1 W m⁻² due to stable poling effects, which is comparable to those of the best reported triboelectric harvesters. Both continuous energy harvesting and continuous tactile detection are possible with these films due to their semiconducting nature and polarization effects.

---

1. Introduction

Integration of polymers with inorganic crystals and materials leads to versatile behavior in many natural systems such as bones and mollusk shells.^1–3 The flexibility of polymer chains with the ability to dissipate energy, the regular presence of functional groups and their versatile nature in having hydrophobic or hydrophilic groups, and the ability to have limited or high electronic and ionic conductivity are the properties that make their integration into such systems complementary to the characteristics of the inorganic components.^4–6 This concept has therefore been actively researched to develop materials with high mechanical strength and toughness, high electrochemical performance, and sensitivity for the detection of stimuli.^7–11 Here we show polystyrene (PS) coupled organolead halide perovskite (MAPbI₃) films whose mechanical properties can be tuned by varying their polymer content. Specifically, the modulus of the films can be varied from ∼23 GPa to ∼15 GPa based on the PS content. Further due to the restriction of ion-migration in PS–MAPbI₃, these films are stable for more than 1000 h under ambient conditions and can be poled at high voltages leading to a continuous and stable power generation density of 1.1 W m⁻² on illumination. The combination of softness and stability in the PS–MAPbI₃ films allows them to be applied as a pressure sensor that has a linear response with high sensitivity (up to 20 kPa⁻¹) over a very wide and tunable operating range (up to 450 kPa) and a minimum detection limit of 4 Pa in a single structure. Further they can be self-powered by just light illumination. This combination of characteristics is crucial for monitoring diverse stimuli ranging from a low pressure (<1 kPa) to a high pressure range (>100 kPa). Moreover, as the pressure sensor operates in a linear regime, the user can obtain accurate information from its output without the need for any additional signal processor thus meeting the increasing demand for device miniaturization and low power consumption. Although several nano–micro structuring approaches have been adopted in order to attain high sensitivity, achieving it in combination with a wide operating range and linear response still remains a challenge, limiting their practical use. This challenge is addressed in a simple monolithic pressure sensor fabricated by integrating an organic polymer (PS) with an organolead halide perovskite (MAPbI₃).

The hydrophobic (organic) PS interacts with PbI₂ (due to its Lewis acid characteristics) and MA⁺ cations (due to the π electrons of PS), resulting in these stable PS–MAPbI₃ films. Poling generates an internal polarization in the films and on illumination the generated charge carriers are separated and collected by electrodes.¹² These films can hence harvest energy from ambient light. A maximum power density of 1.1 W m⁻² is obtained on illumination with 0.1 sun for devices poled for 5 min. They continuously generate power for more than 24 h and can then be repoled to regain their energy harvesting efficiency. Although triboelectric and piezoelectric nanogenerators have been actively researched for the harvesting of mechanical energy for powering devices, they cannot continuously produce energy from static pressure due to the conceptual limitation.^13–15 The semiconducting nature of the perovskite combined with its polarization effects and integration with PS presents a method to complement the capabilities of triboelectric and piezoelectric generators for use in a variety of energy harvesting devices for broader application.^16–18 This capability can also be integrated with the sensing of stimuli as the active layer in the device is a semiconducting perovskite.

2. Results and discussion

The plain MAPbI₃ and PS–MAPbI₃ films are made by the standard solution casting and solvent annealing methods.¹⁹ For the PS–MAPbI₃ films, the wt% of PS is controlled in the precursor solution and the time for crosslinking is kept constant for all the films. The field emission scanning electron microscopy (FESEM) images of the plain MAPbI₃ and 1 wt%, 3 wt%, and 7 wt% PS–MAPbI₃ films are shown in Fig. 1a–d, and uniform crystalline films are observed in all cases. The corresponding X-ray diffraction and Raman spectra are presented in the ESI (Fig. S1 and S2†). Specifically, we observe a Raman shift in the Pb-I mode from 84 cm⁻¹ in plain MAPbI₃ films to 92 cm⁻¹ in 7 wt% PS–MAPbI₃ films, which is attributed to the interaction between PS and PbI₂ (Fig. 1e).¹⁹ We also observe that the MA⁺ libration mode shifts from 143 cm⁻¹ in plain MAPbI₃ films to progressively higher energy as the PS content is increased, reaching 150 cm⁻¹ for 7 wt% PS–MAPbI₃ films (Fig. 1f). Similarly, the MA⁺ torsional mode also shifts from ∼247 cm⁻¹ in plain MAPbI₃ films to ∼258 cm⁻¹ in 7 wt% PS–MAPbI₃ films (Fig. 1g). The MA⁺ libration and torsional shifts signify the interaction between the pi-electrons of PS and the MA⁺ cations, which directly impacts its local motion in the perovskite lattice.¹⁹ Recently, we have reported the cross-linking of the PS chains due to the Lewis acid nature of PbI₂ by gel permeation chromatography.¹⁹

Fig. 1 Characterization of plain and polystyrene-incorporated perovskite films. (a–d) FESEM images of (a) plain MAPbI₃, (b) 1 wt% PS–MAPbI₃, (c) 3 wt% PS–MAPbI₃, and (d) 7 wt% PS–MAPbI₃ films. (e–g) Raman spectra of plain MAPbI₃ and 1 wt%, 3 wt%, and 7 wt% PS–MAPbI₃ films at (e) 60–130 cm⁻¹, (f) 135–160 cm⁻¹, and (g) 180–360 cm⁻¹ wavenumbers show a shift towards higher wavenumbers with PS content.

The short-circuit current density (J_sc) and open-circuit voltage (V_oc) characteristics of plain MAPbI₃ and PS–MAPbI₃ films with varying amount of PS in the precursor solution after poling at electric fields of 2.5 V μm⁻¹ (applied for 5 min, details in the ESI†) are shown in Fig. 2a and b, respectively. We observe that all films show J_sc and V_oc generation, however, the 1 wt% PS films have the highest response. These films are therefore studied in greater detail. Further, ion migration in the films is recorded from their current response to a constant bias in the dark; the observed decay in current is a direct measure of the extent of ion migration. As seen in Fig. 2c the presence of PS significantly reduces the ion migration current, which allows these PS films to be poled at higher field strengths compared to the plain MAPbI₃ films. The 1 wt% PS–MAPbI₃ films show a monotonous increase in power density with poling fields (Fig. 2d), consistent with the expectation that higher fields will increase internal polarization. The maximum power density of the 1 wt% PS–MAPbI₃ films is recorded as 215 mW m⁻² using an external load resistor as shown in Fig. 2e, after poling at 5 V μm⁻¹. Perovskite films are good hole conductors but their electron conductivity is limited,²⁰ and hence to further improve the performance of energy harvesting a top layer of ZnO nanosheets (morphology shown in Fig. S3†) is interfaced with the PS–MAPbI₃ films (schematic in Fig. 2f, with a static load of 100 kPa) for more efficient extraction of electrons. As a result, the power density increases to 1.1 W m⁻² (Fig. 2e). The band diagram of the device (Fig. S4†) shows efficient electron extraction with the use of the ZnO layer. Continuous and stable power generation is observed in these films (with the ZnO layer) for more than 24 h (Fig. 2g), and subsequently, on repoling the efficiency is recovered. The loss is hence attributed to the depolarization of the films and not to any structural degradation. 1 wt% PS–MAPbI₃ films without the ZnO interface also show a similar behavior (see Fig. S5†). Since continuous power generation is observed with light illumination and under constant interfacing (static load) with the ZnO layer, the observed effect is attributed to the polarization effects in the perovskite layer.

Fig. 2 Device performance, structure, and energy-harvesting capability. (a) J_sc and (b) V_oc cycles of plain MAPbI₃ and 1 wt%, 3 wt%, and 7 wt% PS–MAPbI₃ devices after 2.5 V μm⁻¹ poling for 5 min in air under 0.1 sun illumination. (c) Dark current (and ion migration current) response from the perovskite devices at a constant bias of 3 V. (d) Power density dependence on the strength of the poling electric fields for the 1 wt% PS–MAPbI₃ device without the ZnO layer. (e) Output current and power density as a function of different external resistances for the 1 wt% PS–MAPbI₃ device (after 5 V μm⁻¹ poling) with and without the ZnO layer. (f) Schematic diagram of the self-powered PS–MAPbI₃ pressure sensor where ZnO nanosheets are interfaced with the PS–MAPbI₃ film. (g) Operational stability and continuous power generation from the 1 wt% PS–MAPbI₃ device (after 5 V μm⁻¹ poling) interfaced with ZnO nanosheets, examined at a maximum power point with a constant resistance of 10 kΩ and a static load of 100 kPa under continuous 0.1 sun illumination in air.

Plain MAPbI₃ films have a reported elastic modulus of ∼22 GPa.²¹ The integration of softer PS (with a reported modulus of ∼3–4 GPa) should affect the mechanical properties of the PS–MAPbI₃ films. Measured by nano-indentation, we see (Fig. 3a) that as the wt% of PS is increased in the precursor solution the films become softer. The elastic modulus for plain MAPbI₃ films measured at an indentation depth of ∼75 nm is recorded as ∼23 GPa, and this reduces to 19.2 GPa and 15.4 GPa as the PS content in the precursor solution is increased to 1 wt% and 7 wt%, respectively. The corresponding hardness values as a function of indentation depths are shown in Fig. S6.† The ability to modulate the mechanical properties of these films has direct implications for their use in electromechanical and optomechanical devices. We apply this ability towards making tunable range pressure sensors that are also combined with the light harvesting properties of these films resulting in light powered tunable pressure sensors. The concept is based on modulating the interface between the ZnO nanosheets and the PS–MAPbI₃ films by an applied pressure (Fig. 2f). The response of a 1 wt% PS–MAPbI₃ film under 0.1 sun illumination to applied pressure after poling at 5 V μm⁻¹ is shown in Fig. 3b, where a direct correlation between J_sc and the applied pressure is observed. The derivative of current density and pressure shown in Fig. 3c further illustrates that the current accurately tracks the changes in applied pressure both in magnitude and rate. The interface of the sheets with the perovskite film can be taken as being similar to a flat surface indenter interacting with a plain film. The increase in the interaction area with load then follows a linear relationship, similar to the observed behaviour.²² Further the observed increase in current can be attributed to the increase in contact area (and not piezoelectric or triboelectric effect) as with time under a constant load the current response is maintained (Fig. S7†). The sudden variations in the pressure rates (dP/dt) during the loading steps are the result of the feedback loop of the motor, as confirmed by the similar loading curves of pressure on using just glass slides (Fig. S8†). The response from the sensor is correlated with the magnitude of the applied pressure as seen in cycling at different pressure loads in Fig. 3d. Further, the sensor is highly stable as there is no loss in response over more than 200 rapid loading cycles (Fig. 3e). The FESEM images of the ZnO sheets after testing (Fig. S9†) show that their morphology is not altered, further attesting to the stability of the device. Control experiments show that the effect is due to the ZnO sheets, and these were done by observing current modulation in just the perovskite layer and ZnO layer (Fig. S10†) in the same set-up. No modulations were observed in current through the perovskite layer with load while the current through the ZnO sheets followed identical modulation behaviour to the device.

Fig. 3 Mechanical properties and pressure-sensing capabilities of the monolithic 1 wt% PS–MAPbI₃ pressure sensor. (a) Elastic modulus for plain MAPbI₃ and PS–MAPbI₃ films with varying amount of PS shows the stiffness of the films. The standard error is based on sampling over 8–10 different indentation spots. (b) J_sc response is in step with the dynamic and static pressure modulation. (c) Derivative of the J_sc response tracks that of the applied pressure with accuracy. (d) J_sc cycles with various applied pressure stimuli show a consistent response. (e) Cyclic stability and durability tests of the pressure sensor under a repetitive high-pressure loading of 333 kPa. The response from the pressure sensor was monitored after 5 V μm⁻¹ poling for 5 min under 0.1 sun illumination in air.

Varying the PS content in the PS–MAPbI₃ films will directly affect their functioning as a pressure sensor due to the change in their mechanical modulus. This is confirmed by observing the response of PS–MAPbI₃ films with varying amounts of PS in the precursor solution (Fig. 4a). Two effects are observed: first, as the PS content increases the dynamic range for pressure sensing increases, and second, however, the sensitivity does not follow a monotonic trend. Plain MAPbI₃ films are limited to an ∼100 kPa pressure range before saturation in J_sc is observed. Introducing PS increases the dynamic range of the device progressively to more than 400 kPa with 7 wt% PS–MAPbI₃ films. The maximum sensitivity in response is, however, observed for the 1 wt% PS–MAPbI₃ films, as seen in Fig. 4b. At a maximum poling field of 5 V μm⁻¹, the 1 wt% PS–MAPbI₃ devices attain a high sensitivity of 19.77 kPa⁻¹ (with a linear response up to 333 kPa), which is 30 times more than the maximum sensitivity possible with the plain MAPbI₃ films (0.64 kPa⁻¹) (Fig. 4c). We observe that the sensitivity of the 1 wt% PS–MAPbI₃ device increases with the strength of the poling electric field (Fig. 4d). This is consistent with the increased internal polarization which will lead to more effective charge separation, thus improving both the charge collection efficiency by the ZnO films and the sensitivity to pressure modulations. However, the sensitivity of the plain MAPbI₃ device increases up to 3 V μm⁻¹ poling field; subsequently at higher poling fields (>3 V μm⁻¹) the performance decreases due to the segregation of the ions which results in the formation of PbI₂ and hence the observed polarization effect decreases. Further, the 1 wt% PS–MAPbI₃ device can sense pressure as low as 4 Pa (50 μL water droplet) as shown in Fig. S11.† A softer perovskite film due to the incorporated polymer is better able to dissipate the mechanical energy and hence extend the linear operating pressure range of these devices, which is similar to the observation with indentation of softer films.²³ The sensitivity is however dependent on both the poling effects (resulting in generation of J_sc) and the mechanical properties. Hence, better sensitivity is observed with the 1 wt% PS–MAPbI₃ devices. In general, the incorporation of the polymer increases the sensitivity and linearity range of the PS–MAPbI₃ films compared to plain MAPbI₃ films. This allows us to tune the operating range, sensitivity, and linear range of these pressure sensors based on the polymer content. Further, the 1 wt% PS–MAPbI₃ device once poled at 5 V μm⁻¹ for 5 min can be easily operable for more than 120 h without a power source and after that the device can be repoled to recover the performance (Fig. 4e), whereas the plain MAPbI₃ and 3 wt% and 7 wt% PS–MAPbI₃ devices are operable only for 48 h, 72 h, and 48 h, respectively (see Fig. S12†). It can be clearly observed that compared with the previous literature, the monolithic 1 wt% PS–MAPbI₃ pressure sensor exhibits one of the best combinations of high sensitivity with a linear response over a broad dynamic pressure range, as well as the device can be self-powered (Fig. 4f and Table S1†).^{13,14,24–44} Further this is achieved in a simple device structure with the ability to sense both constant static stimuli and also dynamic stimuli, which is a challenge in many architectures.

Fig. 4 Pressure sensitivity, linearity, and self-powered operation of the devices. (a) Variation of J_sc as a function of applied pressure for plain MAPbI₃ and 1 wt%, 3 wt%, and 7 wt% PS–MAPbI₃ devices after 1 V μm⁻¹ poling for 5 min in air under 0.1 sun illumination. (b) Relative current of the devices (after 1 V μm⁻¹ poling) in response to the applied pressure shows a linear response over a broad dynamic range. (c) The sensitivity and pressure range of the 1 wt% PS–MAPbI₃ device (after 5 V μm⁻¹ poling) is more than that of the plain MAPbI₃ device. (d) The sensitivity of the 1 wt% PS–MAPbI₃ device increases with the poling field strength. (e) J_sc response is maintained over 120 h to load cycles of 333 kPa for the 1 wt% PS–MAPbI₃ device after initial poling at 5 V μm⁻¹ for 5 min. (f) Comparison of the sensitivity, linear sensing response, and dynamic pressure range of this work with the previously reported pressure sensors.

3. Conclusion

Polystyrene chains having specific π–cation interactions with the perovskite matrix are incorporated into perovskite films and this modulates their mechanical and electrical properties. As a result, we show that the PS–MAPbI₃ films have highly reduced ionic currents with greater stability that allows them to be poled at higher electric fields. The poling effects result in a stable and continuous power generation of more than 1.1 W m⁻² under light illumination, comparable to the best reported energy harvesting densities with triboelectric generators, which are intermittent in nature. The incorporation of PS also makes the films softer which combined with the stable energy generation leads to light powered tactile sensors that have a wide dynamic range and linear response and high sensitivity in a single structure. The ability to develop such organic–inorganic films with tunable properties using semiconducting perovskites has wide ranging implications for their application in energy harvesting and detection, e.g., light powered sensors, quantum electronics and electro-optical devices, due to their greater stability and higher polarization effects.

Conflicts of interest

There are no conflicts to declare.

Acknowledgements

This work was supported by the University of Waterloo, Canada Foundation for Innovation, Early Researcher Award from Ministry of Research and Innovation and Science, Ontario and NSERC Canada.

Notes and references

1. N. Reznikov, M. Bilton, L. Lari, M. M. Stevens and R. Kröger, Science, 2018, 360, eaao2189.
2. G. Falini, S. Albeck, S. Weiner and L. Addadi, Science, 1996, 271, 67–69.
3. B. Wang, T. N. Sullivan, A. Pissarenko, A. Zaheri, H. D. Espinosa and M. A. Meyers, Adv. Mater., 2019, 31, 1804574.
4. J. S. Peng and Q. F. Cheng, Adv. Mater., 2017, 29, 1702959.
5. V. K. Thakur and R. K. Gupta, Chem. Rev., 2016, 116, 4260–4317.
6. W. Liu, S. W. Lee, D. C. Lin, F. F. Shi, S. Wang, A. D. Sendek and Y. Cui, Nat. Energy, 2017, 2, 17035.
7. Y. Y. Kim, J. D. Carloni, B. Demarchi, D. Sparks, D. G. Reid, M. E. Kunitake, C. C. Tang, M. J. Duer, C. L. Freeman, B. Pokroy, K. Penkman, J. H. Harding, L. A. Estroff, S. P. Baker and F. C. Meldrum, Nat. Mater., 2016, 15, 903–910.
8. F. Bouville, E. Maire, S. Meille, B. Van de Moortèle, A. J. Stevenson and S. Deville, Nat. Mater., 2017, 16, 1271.
9. M. Chen, Y. Zhang, L. Xing, Y. Liao, Y. Qiu, S. Yang and W. Li, Adv. Mater., 2017, 29, 1607015.
10. J. Kang, D. Son, G. J. N. Wang, Y. Liu, J. Lopez, Y. Kim, J. Y. Oh, T. Katsumata, J. Mun, Y. Lee, L. Jin, J. B.-H. Tok and Z. Bao, Adv. Mater., 2018, 30, 1706846.
11. D. Son, J. Kang, O. Vardoulis, Y. Kim, N. Matsuhisa, J. Y. Oh, J. W. To, J. Mun, T. Katsumata, Y. Liu, A. F. McGuire, M. Krason, F. M. Lopez, J. Ham, U. Kraft, Y. Lee, Y. Yun, J. B.-H. Tok and Z. Bao, Nat. Nanotechnol., 2018, 13, 1057–1065.
12. R. Saraf, L. Pu and V. Maheshwari, Adv. Mater., 2018, 30, 1705778.
13. L. Lin, Y. Xie, S. Wang, W. Wu, S. Niu, X. Wen and Z. L. Wang, ACS Nano, 2013, 7, 8266–8274.
14. X. Pu, M. Liu, X. Chen, J. Sun, C. Du, Y. Zhang, J. Zhai, W. Hu and Z. L. Wang, Sci. Adv., 2017, 3, e1700015.
15. Y. Jie, X. T. Jia, J. D. Zou, Y. D. Chen, N. Wang, Z. L. Wang and X. Cao, Adv. Energy Mater., 2018, 8, 1703133.
16. J. Chen, Y. Huang, N. Zhang, H. Zou, R. Liu, C. Tao, X. Fan and Z. L. Wang, Nat. Energy, 2016, 1, 16138.
17. H. Zou, Y. Zhang, L. Guo, P. Wang, X. He, G. Dai, H. Zheng, C. Chen, A. C. Wang, C. Xu and Z. L. Wang, Nat. Commun., 2019, 10, 1427.
18. J. Chen, S. K. Oh, H. Zou, S. Shervin, W. Wang, S. Pouladi, Y. Zi, Z. L. Wang and J. H. Ryou, ACS Appl. Mater. Interfaces, 2018, 10, 12839–12846.
19. R. Saraf and V. Maheshwari, ACS Appl. Energy Mater., 2019, 2, 2214–2222.
20. E. Edri, S. Kirmayer, A. Henning, S. Mukhopadhyay, K. Gartsman, Y. Rosenwaks, G. Hodes and D. Cahen, Nano Lett., 2014, 14, 1000–1004.
21. J. Feng, APL Mater., 2014, 2, 081801.
22. W. C. Oliver and G. M. Pharr, J. Mater. Res., 1992, 7, 1564–1583.
23. D. Tranchida, S. Piccarolo and M. Soliman, Macromolecules, 2006, 39, 4547–4556.
24. C. Yeom, K. Chen, D. Kiriya, Z. Yu, G. Cho and A. Javey, Adv. Mater., 2015, 27, 1561–1566.
25. Y. Joo, J. Yoon, J. Ha, T. Kim, S. Lee, B. Lee, C. Pang and Y. Hong, Adv. Electron. Mater., 2017, 3, 1600455.
26. S. Jang, E. Jee, D. Choi, W. Kim, J. S. Kim, V. Amoli, T. Sung, D. Choi, D. H. Kim and J.-Y. Kwon, ACS Appl. Mater. Interfaces, 2018, 10, 31472–31479.
27. S. H. Shin, S. Ji, S. Choi, K. H. Pyo, B. Wan An, J. Park, J. Kim, J. Y. Kim, K. S. Lee, S. Y. Kwon, J. Heo, B. G. Park and J. U. Park, Nat. Commun., 2017, 8, 14950.
28. G. Schwartz, B. C.-K. Tee, J. Mei, A. L. Appleton, D. H. Kim, H. Wang and Z. Bao, Nat. Commun., 2013, 4, 1859.
29. Q. Sun, D. H. Kim, S. S. Park, N. Y. Lee, Y. Zhang, J. H. Lee, K. Cho and J. H. Cho, Adv. Mater., 2014, 26, 4735–4740.
30. S. Gong, W. Schwalb, Y. Wang, Y. Chen, Y. Tang, J. Si, B. Shirinzadeh and W. Cheng, Nat. Commun., 2014, 5, 3132.
31. H. B. Yao, J. Ge, C. F. Wang, X. Wang, W. Hu, Z. J. Zheng, Y. Ni and S. H. Yu, Adv. Mater., 2013, 25, 6692–6698.
32. X. W. Wang, Y. Gu, Z. P. Xiong, Z. Cui and T. Zhang, Adv. Mater., 2014, 26, 1309.
33. C. Pang, G.-Y. Lee, T.-i. Kim, S. M. Kim, H. N. Kim, S.-H. Ahn and K.-Y. Suh, Nat. Mater., 2012, 11, 795–801.
34. L. Pan, A. Chortos, G. Yu, Y. Wang, S. Isaacson, R. Allen, Y. Shi, R. Dauskardt and Z. Bao, Nat. Commun., 2014, 5, 3002.
35. Y. Pang, K. Zhang, Z. Yang, S. Jiang, Z. Ju, Y. Li, X. Wang, D. Wang, M. Jian, Y. Zhang, R. Liang, H. Tian, Y. Yang and T.-L. Ren, ACS Nano, 2018, 12, 2346–2354.
36. H. Tian, Y. Shu, X. F. Wang, M. A. Mohammad, Z. Bie, Q. Y. Xie, C. Li, W. T. Mi, Y. Yang and T. L. Ren, Sci. Rep., 2015, 5, 8603.
37. J. Park, Y. Lee, J. Hong, M. Ha, Y. -D. Jung, H. Lim, S. Y. Kim and H. Ko, ACS Nano, 2014, 8, 4689–4697.
38. Y. Lee, J. Park, S. Cho, Y. E. Shin, H. Lee, J. Kim, J. Myoung, S. Cho, S. Kang, C. Baig and H. Ko, ACS Nano, 2018, 12, 4045–4054.
39. S. C. B. Mannsfeld, B. C. K. Tee, R. M. Stoltenberg, C. V. H. H. Chen, S. Barman, B. V. O. Muir, A. N. Sokolov, C. Reese and Z. Bao, Nat. Mater., 2010, 9, 859–864.
40. Q. Hua, J. Sun, H. Liu, R. Bao, R. Yu, J. Zhai, C. Pan and Z. L. Wang, Nat. Commun., 2018, 9, 244.
41. C. G. Núñez, W. T. Navaraj, E. O. Polat and R. Dahiya, Adv. Funct. Mater., 2017, 27, 1606287.
42. D. H. Ho, Q. Sun, S. Y. Kim, J. T. Han, D. H. Kim and J. H. Cho, Adv. Mater., 2016, 28, 2601–2608.
43. D. Kwon, T. I. Lee, J. Shim, S. Ryu, M. S. Kim, S. Kim, T. S. Kim and I. Park, ACS Appl. Mater. Interfaces, 2016, 8, 16922–16931.
44. D. J. Lipomi, M. Vosgueritchian, B. C. Tee, S. L. Hellstrom, J. A. Lee, C. H. Fox and Z. Bao, Nat. Nanotechnol., 2011, 6, 788–792.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: 10.1039/c9ta03982a

---