Ferroelectric polymer tuned two dimensional layered MoTe₂ photodetector

Abstract

Two dimensional material based photodetectors have attracted wide attention in recent years. In this work, a few-layer MoTe₂ based phototransistor with a ferroelectric polymer P(VDF-TrFE) topgate is fabricated. The remanent polarization of the ferroelectrics could deplete the channel effectively to decrease the dark current of the device by more than one magnitude. As a result, the MoTe₂ phototransistor has an appreciable photoresponse for visible light and near infrared. The device has a broad photoresponse range (0.6–1.5 μm), the responsivity and detectivity reach 16.4 mA W⁻¹ and 1.94 × 10⁸ Jones for 1060 nm light. The device works without an external gate voltage, which makes for higher reliability and lower power dissipation for practical application.

---

MoTe₂, a transition metal dichalcogenide (TMDC) semiconductor, possesses the typical remarkable properties of TMDC two dimensional (2D) semiconductors, which makes it promising in the development of next-generation optoelectronics.^1–6 In particular, MoTe₂ has a bandgap of ∼1.0 eV,^7–9 which is significant for designing and fabricating near infrared photodetectors with high performance. In fact, many investigations have been conducted on the electronic properties of MoTe₂,^3,10–13 but photodetectors have rarely been explored.¹⁴ Yin et al. reported a high responsivity in a MoTe₂ photodetector under an illumination of a 473 nm laser.¹⁵ In their work, since a very high backgate voltage is utilized, a fairly dark current results. Meanwhile, the high backgate voltage raises the risk of fatal leakages, which will degrade the performance and reliability of the devices.^16–18 To our knowledge, so far there is no report of MoTe₂ has been carried out for the near infrared photodetector. To develop ultra-sensitive photodetectors, suppressing the dark current is critical. In recent years, ferroelectrics (P(VDF-TrFE), PZT and LiNbO₃) have been used as dielectric layer to tune nano electronic devices.^19–23 In this work, we developed an organic ferroelectric polymer P(VDF-TrFE) gated MoTe₂ phototransistor. For the existence of remanent polarization, the dark current can be suppressed at a fairly low lever when no gate voltage is applied and the photocurrent changes very little because the photocurrent is mainly controlled by the photo-generated carriers.²⁴ The P(VDF-TrFE) polymer acts as a floating gate, so the device can work at ZERO gate voltage, leading to low power dissipation. And low leakage current increases the reliability of the device. In addition, the MoTe₂ phototransistor exhibits broad photoresponse range (covers from visible light to near infrared), and high performance for 1060 nm light.

The 2H–MoTe₂ crystal consists of layers of a trigonal prismatic Te–Mo–Te structure,²⁵ as presented in Fig. 1(a). The MoTe₂ nanoflakes were obtained by mechanical exfoliation of a semiconducting 2H–MoTe₂ bulk crystal (purchased from HQ graphene). Then the nanoflakes were transferred to a heavily doped silicon substrate covered with a 285 nm-thick silicon oxide. Fig. 1(b) shows the optical image of as-exfoliated MoTe₂ nanoflakes. Flakes suitable for electrical characterization were identified by an optical microscope and the corresponding thickness was accurately determined by atomic force microcopy (AFM). The thickness is about 3.4 nm as shown Fig. 1(c) along the white line. The Raman spectra were conducted by using the Lab Ram HR800 from HORIBA with a 532 nm excitation laser. Three Raman active peaks have been observed, A_1g mode at ∼173 cm⁻¹, E¹_2g mode at ∼235.3 cm⁻¹, and B¹_2g mode at ∼290.4 cm⁻¹, as shown in Fig. 1(d). The Raman modes are in good agreement with the results in bulk MoTe₂ crystal, which suggests good crystal qualities of the few-layer MoTe₂ nanoflakes.²⁶

Fig. 1 (a) Molecule structure of layered 2H–MoTe₂. (b) Typical optical microscopic image of as-exfoliated MoTe₂ nanoflakes. The scale bar represents 20 μm. (c) The thickness data of the MoTe₂ nanoflakes fabricated into a phototransistor along the white line in the inset image. The inset shows the AFM 3D image of the MoTe₂ phototransistor before P(VDF-TrFE) is covered on it. (d) Raman spectra of MoTe₂ nanoflakes given in (b). (e) Optical microscopic image of the fabricated topgate FET based on few-layer MoTe₂. The scale bar stands for 20 μm. (f) Schematic structure of P(VDF-TrFE) topgate FET based on few-layer MoTe₂ with a laser illuminated on it.

The electrodes of the device were made by standard electron-beam lithography, followed by thermal evaporation of a Cr/Au film (15/45 nm thick). The devices were then annealed at 200 °C in vacuum with 100 sccm Argon atmosphere for 2 hours to remove photoresist residue and to decrease contact resistance. After the backgate FET device was fabricated and annealed as described in our previous work, the P(VDF-TrFE) (70 : 30 mol%) solution (dissolved in the diethyl carbonate with 2.5% wt) was spin coated on top of the MoTe₂. Then the P(VDF-TrFE) film was annealed at 135 °C in an oven for four hours. Finally, 10 nm-thick aluminum was deposited on the top of the P(VDF-TrFE) film by electron beam evaporation and patterned by photolithography as the top-gate semitransparent electrodes. The optical image of the topgate FET device is shown in Fig. 1(e). Fig. 1(f) shows the schematic structure of the device with a laser illuminated on it. The electric and optoelectronic measurement were carried out using an Agilent B2902A semiconductor parameter analyzer under probe station in ambient air at room temperature. Between measurements, the samples are stored in a vacuum drying oven.

At first, the ferroelectric properties of P(VDF-TrFE) copolymer films have been characterized. The typical hysteresis loop of 300 nm P(VDF-TrFE) capacitor is obtained as shown in Fig. 2(a). The coercive voltage is about 23 V and the remanent polarization value P_r is 7 μC cm⁻², which indicates a good quality of the P(VDF-TrFE) film. The transfer characteristics of the MoTe₂ FET with a P(VDF-TrFE) topgate were investigated at room temperature when source–drain voltage (V_ds) at 1 V, and the results are shown in Fig. 2(b). When the topgate voltage swept from −35 V to 35 V, high on/off ratio (10⁵) is observed. It is noteworthy that a large nonvolatile hysteresis window of 40 V is present resulted from the polarization of the ferroelectric layer, which is different from a volatile hysteresis of the ordinary backgate transfer characteristics.²⁷ As a negative topgate voltage (V_tg) larger than the coercive voltage is applied to the gate, the P(VDF-TrFE) would be the polarization up (P_up) state, and holes are accumulated in the MoTe₂ channel. It is found that the accumulation state can keep for a very long time even if the V_tg is removed, which is due to a local electric field acting on the MoTe₂ channel aroused from the remanent polarization of P(VDF-TrFE). The negative bound charge at the interface would induce a high electrostatic field, which leads to the accumulation of holes in the MoTe₂ channel. The process is shown in Fig. 2(b), corresponds to the section V_tg when it changes from −35 V to 0 V. Similarly, when a positive topgate voltage larger than the coercive voltage is applied, the P(VDF-TrFE) would be in the polarization down (P_down) state. The MoTe₂ channel remains depleted in this state even after the V_tg is reduced to ZERO, due to the remanent polarization induced opposite polarity electrostatic field. The process corresponds to the section V_tg when it changes from 35 V to 0 V in Fig. 2(b). For the ferroelectric characteristics of P(VDF-TrFE), the gate capacitance was not a constant under different topgate voltage.²⁸ Fig. 2(c) depicts the relationships between the capacitance and the applied voltage. So we calculate the field effect mobility by the following equations,²⁹

where L and W are the length and width of the channel, C_tg represents gate capacitance. The maximum μ value of the device was calculated to be 68 cm² V⁻¹ s⁻¹. The relationship of mobility and topgate voltage is shown in ESI.† The mobility and on/off ratio are higher than the results reported before.³

Fig. 2 (a) The ferroelectric hysteresis loop 300 nm P(VDF-TrFE) film capacitor. It is measured by using Sawyer–Tower circuit at 1 Hz applied voltage frequency. (b) The transfer curves of MoTe₂ phototransistor with P(VDF-TrFE) ferroelectric polymer gate at room temperature when V_ds = 1 V. The dark curve corresponds to the linear coordinate (left axis); the navy curve corresponds to the logarithmic coordinates (right axis). (c) Capacitance–voltage (C–V) relationship obtained from Au/P(VDF-TrFE)/Al structure, where P(VDF-TrFE) is 300 nm thick. (d) The backgate transfer curves of MoTe₂ phototransistors when the P(VDF-TrFE) at P_up state (red curve) and P_down state (dark curve), V_ds = 1 V. (e) The output characteristics (V_bg = 0 V) with three states of ferroelectric layer. The three states are fresh state (ferroelectric layer without polarization), P_up state and P_down state. (f) The photoresponse for 830 nm light of the MoTe₂ phototransistor at P_up state and P_down state when V_ds = 1 V and no external gate voltage applied.

The backgate transfer characteristics of the device were obtained when the topgate P(VDF-TrFE) under different polarization state, as shown in Fig. 2(d). The backgate transfer curves show different hysteresis with the ferroelectric topgate. The I_ds of the P_up state is much larger than that of the P_down state, which suggests the large difference in the density of charge carriers in the channel. For the existence of three states of the P(VDF-TrFE) topgate: fresh state without polarization, P_up state and P_down state, the charge carrier density of the MoTe₂ channel would present three different states as is shown in Fig. 2(e). Fig. 2(e) shows the logarithmic output characteristics of the device at three different states with ZERO gate voltage. Comparing to the channel current at the fresh state, the P_up state could enhance the channel current significantly and the P_down state can suppress the channel current effectively. So it is a good way to decrease the dark current of a photodetector by the ferroelectric gate. Then we measured the photoresponse of the phototransistor when the topgate at different polarization states with an 830 nm laser periodically illuminated on the sample at a frequency of 0.5 Hz, as shown in Fig. 2(f). From Fig. 2(f), we find the dark current of the device at P_down state is about two magnitudes lower than that at P_up state, but the on/off ratio (I_illum/I_dark) at P_down state is much larger than that of P_up state. The lower dark current and higher signal noise ratio ensure a better performance of the photodetector.

Fig. 3(a) and (c) show the schematic structure of the P(VDF-TrFE) topgate MoTe₂ phototransistor, and the arrows show the direction of polarization. Fig. 3(a) depicts the P_up state, the remanent polarization of P(VDF-TrFE) makes the channel accumulated with holes for the negative bound charge facing MoTe₂ channel. The band diagram of the P_up states is shown in Fig. 3(b). Due to the accumulation of charge carriers, the channel current at this state is enhanced than the fresh state (not poled yet). The P_down state is shown in Fig. 3(c) and (d). At this state the positive bound charge is facing the MoTe₂ channel leading to the depletion of holes in the channel, so the current is suppressed at a low level than the fresh state.

Fig. 3 (a) Schematic structure of “P_up state” of the MoTe₂ based topgate FET. The dark red arrows in the P(VDF-TrFE) layer represent the polarization. (b) The corresponding band diagram in the MoTe₂ channel when topgate is at P_up state. It shows the holes accumulated in the channel. (c) Schematic structure of “P_down state” of the MoTe₂ based topgate FET. (d) The corresponding band diagram in the MoTe₂ channel when topgate is at P_down state. It shows the holes depleted in the channel.

Low dark current is essential for a high detectivity photodetector. It is expected that MoTe₂ photodetector operates with a rather low dark current at a P_down state. The output and transfer characteristics were measured under different powers of incident light (λ = 1060 nm) and the results are shown in Fig. 4(a) and (b). The photocurrent increased significantly with increasing power of incident laser at a fixed bias voltage V_ds. The dark current can be suppressed at a relative low level attributes to the depletion of intrinsic carriers in the channel, while the photocurrent changes very little because the photocurrent is mainly controlled by the photo-generated carriers. According to the Fig. 4(b), the channel current increases with the power of light and the photo current minimums shift to the negative polarity. It suggests the photogating effect plays a role in the photocurrent generation. The photogating effect, which is related to the trapped states, could lead to a photoconductive gain.³⁰ The sublinear relationship of photocurrent I_ph (I_ph = I_illum − I_dark) and the power of incident laser is shown in Fig. 4(c), when V_ds = 0.2 V and λ = 1060 nm. The relationship can be expressed as I_ph ∝ P^α, where P is the power of incident light, α is a constant. According to the power function fitting method, we can obtain α = 0.63. The nonlinear relationship also reflects the trap states take part in the photocurrent generation process.^31,32 The current responsivity (R) is defined as the photocurrent generated per unit power of incident light. It is expressed as R = I_ph/P,^33,34 where P is the light intensity irradiated on MoTe₂. So it can be derived that the photo responsivity R ∝ P^(α−1). For α < 1, the photo responsivity decreases with increasing power of the incident light. The responsivity under illumination of 1060 nm laser is calculated to be 16.3 mA W⁻¹ when the power of incident light is 10 μW. The detectivity (D*) which considered both responsivity and dark current is a much more proper assessment of a photodetector. The detectivity is given by D* = RA^1/2/(2eI_dark)^1/2, where R is the responsivity, A is the area of the detector, e is the unit charge, and I_dark is the dark current.³⁵ And it is calculated to be 1.94 × 10⁸ Jones for 1060 nm light when incident light is 10 μW. With a ferroelectric topgate the dark current can be suppressed at a low level but the photocurrent changes a little because the photocurrent is mainly controlled by the photo-generated carriers. Previous research in ferroelectric gated MoS₂ and CdS nanowire photodetectors also showed the same results.^19,24

Fig. 4 (a) The output characteristics under different power of incident light (λ = 1060 nm) at P_down state without gate voltage. (b) The transfer curves under different incident light powers (λ = 1060 nm) when V_ds = 1 V. (c) Photocurrent as a function of power of incident light (λ = 1060 nm). (d) Photoresponse of MoTe₂ topgate FET as a function of light wavelengths from 600 to 1650 nm at V_ds = 1 V, with incident light power of 1.5 mW. (e) Time resolved response of photocurrent and fitting curves by exponent function, the rise time constant is about 1.4 ms. (f) Time resolved response of photocurrent and fitting curves by exponent function, the fall time constant is about 1.3 ms.

MoTe₂ has a band gap of about 1.0 eV, so it is expected to be good photodetector materials for visible light and near infrared. A monochromator was used for wavelength dependent measurements of the photocurrent. The photoresponse of a MoTe₂ based topgate phototransistor as a function of light wavelength from 600 to 1650 nm at V_ds = 1 V is shown in Fig. 4(d) (the incident light power is 1.5 mW). It has largest photoresponse for about 900 nm light and keeps weak photoresponse until 1500 nm light which is attributed to the absorption of band tails states and charge traps. The broad range photoresponse can be also attributed to the bandgap changes under ultrahigh electrostatic field induced by the remanent polarization of P(VDF-TrFE), as proved in MoS₂ phototransistors before.¹⁹ These results illustrate that the topgate MoTe₂ phototransistor is a good photodetector for visible light and near infrared.

Another important factor for a photodetector is the response time. As can be seen from Fig. 4(e) and (f), it shows the rise and fall procedure when the light turned on and off for a 1060 nm laser. The experimental data were collected by a high speed Tektronix MDO3014 oscilloscope. We obtain the response time by fitting the curves with the exponential function, I(t) = I_dark + A exp((t − t₀)/τ), where I_dark is the dark current, A is the scaling constant, τ is a time constant, and t₀ is the time when the illumination starts or ends. The rise and fall times are 1.4 ms and 1.3 ms, respectively. It demonstrates that the ferroelectric topgate MoTe₂ phototransistor has a fairly fast photoresponse among 2D materials photodetectors.^36–38

In summary, few-layered MoTe₂ transistors have been fabricated for photodetector. The MoTe₂ device with a ferroelectric topgate exhibits interesting photodetection capabilities: broad spectral range detection, stable and fast photoresponse, and low dark current for visible light and near infrared. The device can work efficiently without an external gate voltage, which makes for higher reliability and lower power dissipation for practical application. It suggests that ferroelectric polymer gate could be an effective method to tune the 2D materials based phototransistors to achieve high performance. There are many fruitful and fundamental physics issues in ferroelectric/photoelectric 2D material hybrid systems to be explored in future studies.

Author contributions

J. W., and W. H. conceived and supervised the research. H. H. and X. W. fabricated the devices. H. H., P. W., and X. W. performed the measurements. J. W., W. H., and H. H. wrote the paper. All authors discussed the results and revised the manuscript.

Conflict of interest

The authors declare no competing financial interests.

Acknowledgements

This work was partially supported by the Major State Basic Research Development Program (Grant No. 2014CB921600), Ten Thousand Talents Program for Young Talents, Natural Science Foundation of China (Grant No. 11374320, 11322441, and 61574152), CAS Interdisciplinary Innovation Team, and Fund of Shanghai Science and Technology Foundation (Grant No. 14JC1406400).

References

1. H. Yuan, X. Wang, B. Lian, H. Zhang, X. Fang, B. Shen, G. Xu, Y. Xu, S. C. Zhang, H. Y. Hwang and Y. Cui, Nat. Nanotechnol., 2014, 9, 851.
2. Q. H. Wang, K. Kalantar-Zadeh, A. Kis, J. N. Coleman and M. S. Strano, Nat. Nanotechnol., 2012, 7, 699.
3. Y. F. Lin, Y. Xu, S. T. Wang, S. L. Li, M. Yamamoto, A. Aparecido-Ferreira, W. Li, H. Sun, S. Nakaharai, W. B. Jian, K. Ueno and K. Tsukagoshi, Adv. Mater., 2014, 26, 3263.
4. G. Fiori, F. Bonaccorso, G. Iannaccone, T. Palacios, D. Neumaier, A. Seabaugh, S. K. Banerjee and L. Colombo, Nat. Nanotechnol., 2014, 9, 768.
5. F. Schwierz, J. Pezoldt and R. Granzner, Nanoscale, 2015, 7, 8261.
6. J. Kwon, Y. K. Hong, G. Han, I. Omkaram, W. Choi, S. Kim and Y. Yoon, Adv. Mater., 2015, 27, 2224.
7. S. Fathipour, N. Ma, W. Hwang, V. Protasenko, S. Vishwanath, H. Xing, H. Xu, D. Jena, J. Appenzeller and A. Seabaugh, Appl. Phys. Lett., 2014, 105, 192101.
8. Y. Ma, Y. Dai, M. Guo, C. Niu, J. Lu and B. Huang, Phys. Chem. Chem. Phys., 2011, 13, 15546.
9. I. G. Lezama, A. Arora, A. Ubaldini, C. Barreteau, E. Giannini, M. Potemski and A. F. Morpurgo, Nano Lett., 2015, 15, 2336.
10. S. Fathipour, N. Ma, W. S. Hwang, V. Protasenko, S. Vishwanath, H. G. Xing, H. Xu, D. Jena, J. Appenzeller and A. Seabaugh, Appl. Phys. Lett., 2014, 105, 192101.
11. C. Ruppert, O. B. Aslan and T. F. Heinz, Nano Lett., 2014, 14, 6231.
12. I. G. Lezama, A. Ubaldini, M. Longobardi, E. Giannini, C. Renner, A. B. Kuzmenko and A. F. Morpurgo, 2D Mater., 2014, 1, 021002.
13. S. Nakaharai, M. Yamamoto, K. Ueno, Y.-F. Lin, S.-L. Li and K. Tsukagoshi, ACS Nano, 2015, 9, 5976.
14. G. Cunningham, D. Hanlon, N. McEvoy, G. S. Duesberg and J. N. Coleman, Nanoscale, 2015, 7, 198.
15. L. Yin, X. Zhan, K. Xu, F. Wang, Z. Wang, Y. Huang, Q. Wang, C. Jiang and J. He, Appl. Phys. Lett., 2016, 108, 043503.
16. A. M. Ionescu and H. Riel, Nature, 2011, 479, 329.
17. Y. Ouyang and J. Guo, Appl. Phys. Lett., 2006, 89, 183122.
18. A. Szabo, S. J. Koester and M. Luisier, Device Research Conference (DRC), CA, 2014.
19. X. Wang, P. Wang, J. Wang, W. Hu, X. Zhou, N. Guo, H. Huang, S. Sun, H. Shen, T. Lin, M. Tang, L. Liao, A. Jiang, J. Sun, X. Meng, X. Chen, W. Lu and J. Chu, Adv. Mater., 2015, 27, 6575.
20. Y. Zheng, G.-X. Ni, C.-T. Toh, C.-Y. Tan, K. Yao and B. Özyilmaz, Phys. Rev. Lett., 2010, 105, 166602.
21. R. C. G. Naber, C. Tanase, P. W. M. Blom, G. H. Gelinck, A. W. Marsman, F. J. Touwslager, S. Setayesh and D. M. de Leeuw, Nat. Mater., 2005, 4, 243.
22. C. Baeumer, S. P. Rogers, R. Xu, L. W. Martin and M. Shim, Nano Lett., 2013, 13, 1693.
23. Y. T. Lee, H. Kwon, J. S. Kim, H.-H. Kim, Y. J. Lee, J. A. Lim, Y.-W. Song, Y. Yi, W.-K. Choi, D. K. Hwang and S. Im, ACS Nano, 2015, 9, 10394.
24. D. Zheng, J. Wang, W. Hu, L. Liao, H. Fang, N. Guo, P. Wang, F. Gong, X. Wang, Z. Fan, X. Wu, X. Meng, X. Chen and W. Lu, Nano Lett., 2016, 16, 2548.
25. D. Puotinen and R. E. Newnham, Acta Crystallogr., 1961, 14, 691.
26. M. Yamamoto, S. T. Wang, M. Ni, Y.-F. Lin, S.-L. Li, S. Aikawa, W.-B. Jian, K. Ueno, K. Wakabayashi and K. Tsukagoshi, ACS Nano, 2014, 8, 3895.
27. A. Lipatov, P. Sharma, A. Gruverman and A. Sinitskii, ACS Nano, 2015, 9, 8089.
28. C. J. Brennan, Integr. Ferroelectr., 1992, 2, 73.
29. K. Lee, H. Y. Kim, M. Lotya, J. N. Coleman, G. T. Kim and G. S. Duesberg, Adv. Mater., 2011, 23, 4178.
30. B. E. A. Saleh and M. C. Teich, in Fundamentals of Photonics, John Wiley & Sons, Inc., 2001, p. 937.
31. H. Kind, H. Yan, B. Messer, M. Law and P. Yang, Adv. Mater., 2002, 14, 158.
32. N. Guo, W. Hu, L. Liao, S. Yip, J. C. Ho, J. Miao, Z. Zhang, J. Zou, T. Jiang and S. Wu, Adv. Mater., 2014, 26, 8203.
33. G. Konstantatos and E. H. Sargent, Nat. Nanotechnol., 2010, 5, 391.
34. Y. Guo, G. Yu and Y. Liu, Adv. Mater., 2010, 22, 4427.
35. Y. Hu, J. Zhou, P. H. Yeh, Z. Li, T. Y. Wei and Z. L. Wang, Adv. Mater., 2010, 22, 3327.
36. H. Huang, P. Wang, Y. Gao, X. Wang, T. Lin, J. Wang, L. Liao, J. Sun, X. Meng, Z. Huang, X. Chen and J. Chu, Appl. Phys. Lett., 2015, 107, 143112.
37. P. Hu, Z. Wen, L. Wang, P. Tan and K. Xiao, ACS Nano, 2012, 6, 5988.
38. M. Buscema, J. O. Island, D. J. Groenendijk, S. I. Blanter, G. A. Steele, H. S. van der Zant and A. Castellanos-Gomez, Chem. Soc. Rev., 2015, 44, 3691.

---

Footnote

† Electronic supplementary information (ESI) available: The gate–source current (I_gs) of the P(VDF-TrFE) topgate MoTe₂ phototransistor and the gate voltage dependent field effect mobility of P(VDF-TrFE) topgate MoTe₂ phototransistor. See DOI: 10.1039/c6ra18238k

---