Volume 33 Issue 5
Nov.  2020
Turn off MathJax
Article Contents

Zhen Chi, Hui-hui Chen, Zhuo Chen, Hai-long Chen. Unveiling Defect-Mediated Carrier Dynamics in Few-Layer MoS2 Prepared by Ion Exchange Method via Ultrafast Vis-NIR-MIR Spectroscopy†[J]. Chinese Journal of Chemical Physics , 2020, 33(5): 547-553. doi: 10.1063/1674-0068/cjcp2007123
Citation: Zhen Chi, Hui-hui Chen, Zhuo Chen, Hai-long Chen. Unveiling Defect-Mediated Carrier Dynamics in Few-Layer MoS2 Prepared by Ion Exchange Method via Ultrafast Vis-NIR-MIR Spectroscopy[J]. Chinese Journal of Chemical Physics , 2020, 33(5): 547-553. doi: 10.1063/1674-0068/cjcp2007123

Unveiling Defect-Mediated Carrier Dynamics in Few-Layer MoS2 Prepared by Ion Exchange Method via Ultrafast Vis-NIR-MIR Spectroscopy

doi: 10.1063/1674-0068/cjcp2007123
More Information
  • Corresponding author: Hai-long Chen, E-mail:hlchen@iphy.ac.cn
  • Part of the special issue for "the Chinese Chemical Society's 16th National Chemical Dynamics Symposium".
  • Received Date: 2020-07-11
  • Accepted Date: 2020-09-02
  • Publish Date: 2020-10-27
  • Part of the special issue for "the Chinese Chemical Society's 16th National Chemical Dynamics Symposium".
  • 加载中
  • [1] Q. H. Wang, K. Kalantar-Zadeh, A. Kis, J. N. Coleman, and M. S. Strano, Nat. Nanotechnol. 7, 699 (2012). doi:  10.1038/nnano.2012.193
    [2] X. Xu, W. Yao, D. Xiao, and T. F. Heinz, Nat. Phys. 10, 343 (2014). doi:  10.1038/nphys2942
    [3] K. F. Mak, C. Lee, J. Hone, J. Shan, and T. F. Heinz, Phys. Rev. Lett. 105, 136805 (2010). doi:  10.1103/PhysRevLett.105.136805
    [4] A. M. Van Der Zande, P. Y. Huang, D. A. Chenet, T. C. Berkelbach, Y. You, G. H. Lee, T. F. Heinz, D. R. Reichman, D. A. Muller, and J. C. Hone, Nat. Mater. 12, 554 (2013). doi:  10.1038/nmat3633
    [5] K. F. Mak, K. He, J. Shan, and T. F. Heinz, Nat. Nanotechnol. 7, 494 (2012). doi:  10.1038/nnano.2012.96
    [6] T. Cao, G. Wang, W. Han, H. Ye, C. Zhu, J. Shi, Q. Niu, P. Tan, E. Wang, and B. Liu, Nat. Commun. 3, 887 (2012). doi:  10.1038/ncomms1882
    [7] M. Chhowalla, H. S. Shin, G. Eda, L. J. Li, K. P. Loh, and H. Zhang, Nat. Chem. 5, 263 (2013). doi:  10.1038/nchem.1589
    [8] L. Britnell, R. Ribeiro, A. Eckmann, R. Jalil, B. Belle, A. Mishchenko, Y. J. Kim, R. Gorbachev, T. Georgiou, and S. Morozov, Science 340, 1311 (2013). doi:  10.1126/science.1235547
    [9] J. Shi, Y. Li, Z. Zhang, W. Feng, Q. Wang, S. Ren, J. Zhang, W. Du, X. Wu, X. Sui, Y. Mi, R. Wang, Y. Sun, L. Zhang, X. Qiu, J. Lu, C. Shen, Y. Zhang, Q. Zhang, and X. Liu, ACS Photonics 6, 3082 (2019). doi:  10.1021/acsphotonics.9b00855
    [10] Z. Zeng, Z. Yin, X. Huang, H. Li, Q. He, G. Lu, F. Boey, and H. Zhang, Angew. Chem. Int. Ed. 50, 11093 (2011). doi:  10.1002/anie.201106004
    [11] B. Radisavljevic, A. Radenovic, J. Brivio, I. V. Giacometti, and A. Kis, Nat. Nanotechnol. 6, 147 (2011). doi:  10.1038/nnano.2010.279
    [12] O. Lopez-Sanchez, D. Lembke, M. Kayci, A. Radenovic, and A. Kis, Nat. Nanotechnol. 8, 497 (2013). doi:  10.1038/nnano.2013.100
    [13] B. W. Baugher, H. O. Churchill, Y. Yang, and P. Jarillo-Herrero, Nat. Nanotechnol. 9, 262 (2014). doi:  10.1038/nnano.2014.25
    [14] H. Wang, L. Yu, Y. H. Lee, Y. Shi, A. Hsu, M. L. Chin, L. J. Li, M. Dubey, J. Kong, and T. Palacios, Nano Lett. 12, 4674 (2012). doi:  10.1021/nl302015v
    [15] B. Radisavljevic and A. Kis, Nat. Mater 12, 815 (2013). doi:  10.1038/nmat3687
    [16] K. F. Mak, K. He, C. Lee, G. H. Lee, J. Hone, T. F. Heinz, and J. Shan, Nat. Mater. 12, 207 (2013). doi:  10.1038/nmat3505
    [17] C. Lui, A. Frenzel, D. Pilon, Y. H. Lee, X. Ling, G. Akselrod, J. Kong, and N. Gedik, Phys. Rev. Lett. 113, 166801 (2014). doi:  10.1103/PhysRevLett.113.166801
    [18] E. J. Sie, A. J. Frenzel, Y. H. Lee, J. Kong, and N. Gedik, Phys. Rev. B 92, 125417 (2015). doi:  10.1103/PhysRevB.92.125417
    [19] X. Duan, C. Wang, A. Pan, R. Yu, and X. Duan, Chem. Soc. Rev. 44, 8859 (2015). doi:  10.1039/C5CS00507H
    [20] H. Chen, Z. Chen, B. Ge, Z. Chi, H. Chen, H. Wu, C. Cao, and X. Duan, Chem. Mater. 29, 10019 (2017). doi:  10.1021/acs.chemmater.7b03523
    [21] D. Voiry, M. Salehi, R. Silva, T. Fujita, M. Chen, T. Asefa, V. B. Shenoy, G. Eda, and M. Chhowalla, Nano Lett. 13, 6222 (2013). doi:  10.1021/nl403661s
    [22] J. Hong, Z. Hu, M. Probert, K. Li, D. Lv, X. Yang, L. Gu, N. Mao, Q. Feng, L. Xie, J. Zhang, D. Wu, Z. Zhang, C. Jin, W. Ji, X. Zhang, J. Yuan, and Z. Zhang, Nat. Commun. 6, 6293 (2015). doi:  10.1038/ncomms7293
    [23] H. Qiu, T. Xu, Z. Wang, W. Ren, H. Nan, Z. Ni, Q. Chen, S. Yuan, F. Miao, F. Song, G. Long, Y. Shi, L. Sun, J. Wang, and X. Wang, Nat. Commun. 4, 2642 (2013). doi:  10.1038/ncomms3642
    [24] Z. Lin, B. R. Carvalho, E. Kahn, R. Lv, R. Rao, H. Terrones, M. A. Pimenta, and M. Terrones, 2D Materials 3, 022002 (2016). doi:  10.1088/2053-1583/3/2/022002
    [25] G. Ye, Y. Gong, J. Lin, B. Li, Y. He, S. T. Pantelides, W. Zhou, R. Vajtai, and P. M. Ajayan, Nano Lett. 16, 1097 (2016). doi:  10.1021/acs.nanolett.5b04331
    [26] J. Xie, H. Zhang, S. Li, R. Wang, X. Sun, M. Zhou, J. Zhou, X. W. D. Lou, and Y. Xie, Adv. Mater. 25, 5807 (2013). doi:  10.1002/adma.201302685
    [27] H. Shi, R. Yan, S. Bertolazzi, J. Brivio, B. Gao, A. Kis, D. Jena, H. G. Xing, and L. Huang, ACS Nano 7, 1072 (2013). doi:  10.1021/nn303973r
    [28] P. D. Cunningham, K. M. McCreary, A. T. Hanbicki, M. Currie, B. T. Jonker, and L. M. Hayden, J. Phys. Chem. C 120, 5819 (2016). doi:  10.1021/acs.jpcc.6b00647
    [29] M. Palummo, M. Bernardi, and J. C. Grossman, Nano Lett. 15, 2794 (2015). doi:  10.1021/nl503799t
    [30] Y. Li, J. Shi, H. Chen, Y. Mi, W. Du, X. Sui, C. Jiang, W. Liu, H. Xu, and X. Liu, Laser Photo. Rev. 13, 1800270 (2019). doi:  10.1002/lpor.201800270
    [31] H. Wang, C. Zhang, and F. Rana, Nano Lett. 15, 339 (2014).
    [32] K. Chen, R. Ghosh, X. Meng, A. Roy, J. S. Kim, F. He, S. C. Mason, X. Xu, J. F. Lin, D. Akinwande, S. K. Banerjee, and Y. Wang, NPJ 2D Mater. Appl. 1, 15 (2017). doi:  10.1038/s41699-017-0019-1
    [33] H. Wang, C. Zhang, and F. Rana, Nano Lett. 15, 8204 (2015). doi:  10.1021/acs.nanolett.5b03708
    [34] H. Chen, X. Wen, J. Zhang, T. Wu, Y. Gong, X. Zhang, J. Yuan, C. Yi, J. Lou, P. M. Ajayan, W. Zhuang, G. Zhang, and J. Zheng, Nat. Commun. 7, 12512 (2016). doi:  10.1038/ncomms12512
    [35] V. Nicolosi, M. Chhowalla, M. G. Kanatzidis, M. S. Strano, and J. N. Coleman, Science 340, 1226419 (2013). doi:  10.1126/science.1226419
    [36] P. Mao, Z. Wang, W. Dang, and Y. Weng, Rev. Sci. Instrum. 86, 123113 (2015). doi:  10.1063/1.4938187
    [37] X. He, G. Zhu, J. Yang, H. Chang, Q. Meng, H. Zhao, X. Zhou, S. Yue, Z. Wang, J. Shi, L. Gu, D. Yan, and Y. Weng, Sci. Rep. 5, 17076 (2015). doi:  10.1038/srep17076
    [38] T. Cheiwchanchamnangij and W. R. Lambrecht, Phys. Rev. B 85, 205302 (2012). doi:  10.1103/PhysRevB.85.205302
    [39] S. Sim, J. Park, J. G. Song, C. In, Y. S. Lee, H. Kim, and H. Choi, Phys. Rev. B 88, 075434 (2013). doi:  10.1103/PhysRevB.88.075434
    [40] B. Chakraborty, H. Matte, A. Sood, and C. Rao, J. Raman Spectrosc. 44, 92 (2013). doi:  10.1002/jrs.4147
    [41] C. Lee, H. Yan, L. E. Brus, T. F. Heinz, J. Hone, and S. Ryu, ACS Nano 4, 2695 (2010). doi:  10.1021/nn1003937
    [42] Z. Nie, R. Long, L. Sun, C. C. Huang, J. Zhang, Q. Xiong, D. W. Hewak, Z. Shen, O. V. Prezhdo, and Z. H. Loh, ACS Nano 8, 10931 (2014). doi:  10.1021/nn504760x
    [43] Z. Chi, H. Chen, Z. Chen, Q. Zhao, H. Chen, and Y. X. Weng, ACS Nano 12, 8961 (2018). doi:  10.1021/acsnano.8b02354
    [44] B. Peng, G. Yu, X. Liu, B. Liu, X. Liang, L. Bi, L. Deng, T. C. Sum, and K. P. Loh, 2D Materials 3, 025020 (2016). doi:  10.1088/2053-1583/3/2/025020
    [45] Z. Chi, H. Chen, Q. Zhao, and Y. X. Weng, Nanotechnology 31, 235712 (2020). doi:  10.1088/1361-6528/ab79ad
    [46] Z. Chi, H. Chen, Q. Zhao, and Y. X. Weng, J. Chem. Phys. 151, 114704 (2019). doi:  10.1063/1.5115467
    [47] M. M. Ugeda, A. J. Bradley, S. F. Shi, F. H. da Jornada, Y. Zhang, D. Y. Qiu, W. Ruan, S. K. Mo, Z. Hussain, Z. X. Shen, F. Wang, S. G. Louie, and M. F. Crommie, Nat. Mater. 13, 1091 (2014). doi:  10.1038/nmat4061
    [48] W. Zhou, X. Zou, S. Najmaei, Z. Liu, Y. Shi, J. Kong, J. Lou, P. M. Ajayan, B. I. Yakobson, and J. C. Idrobo, Nano Lett. 13, 2615 (2013). doi:  10.1021/nl4007479
    [49] C. P. Lu, G. Li, J. Mao, L. M. Wang, and E. Y. Andrei, Nano Lett. 14, 4628 (2014). doi:  10.1021/nl501659n
  • 加载中
通讯作者: 陈斌, bchen63@163.com
  • 1. 

    沈阳化工大学材料科学与工程学院 沈阳 110142

  1. 本站搜索
  2. 百度学术搜索
  3. 万方数据库搜索
  4. CNKI搜索

Figures(11)  / Tables(1)

Article Metrics

Article views(53) PDF downloads(8) Cited by()

Proportional views
Related

Unveiling Defect-Mediated Carrier Dynamics in Few-Layer MoS2 Prepared by Ion Exchange Method via Ultrafast Vis-NIR-MIR Spectroscopy

doi: 10.1063/1674-0068/cjcp2007123
Part of the special issue for "the Chinese Chemical Society's 16th National Chemical Dynamics Symposium".
Zhen Chi, Hui-hui Chen, Zhuo Chen, Hai-long Chen. Unveiling Defect-Mediated Carrier Dynamics in Few-Layer MoS2 Prepared by Ion Exchange Method via Ultrafast Vis-NIR-MIR Spectroscopy†[J]. Chinese Journal of Chemical Physics , 2020, 33(5): 547-553. doi: 10.1063/1674-0068/cjcp2007123
Citation: Zhen Chi, Hui-hui Chen, Zhuo Chen, Hai-long Chen. Unveiling Defect-Mediated Carrier Dynamics in Few-Layer MoS2 Prepared by Ion Exchange Method via Ultrafast Vis-NIR-MIR Spectroscopy[J]. Chinese Journal of Chemical Physics , 2020, 33(5): 547-553. doi: 10.1063/1674-0068/cjcp2007123
  • Two dimensional (2D) transition metal dichalcogenides (TMDs), such as MoS2, MoSe2, WSe2, and WS2, have recently gained extensive research interests, acting as a new type of atomically thin materials [1-4]. This class of materials exhibits fascinating physical properties including layer-dependent energy band gap, strong Coulombic and light-matter interactions, distinctive spin-valley coupling, etc. [3, 5-9]. These extraordinary properties make TMDs suitable for many applications in electronics and photonics, such as field-effect transistors, integrated logic circuits, photodetectors and phototransistors [10-14]. Up to now, many studies have largely focused on the 2D TMDs prepared by mechanical exfoliation [15, 16] or chemical vapor deposition (CVD) methods [17, 18]. Although some of the highest-quality layered TMDs can be formed via mechanical exfoliation, this method is unsuitable for practical technologies due to its limited scalability for production. CVD can also result in high quantities of TMDs, however, it is expensive and is generally restricted by the substrate [19]. In contrast, using ion exchange method to prepare layered TMDs in suitable solvents is a low-cost and precisely controllable way [20]. Especially, the nanosheets produced in solution phase by this strategy can be directly used for the application of photocatalysis and also enables an easy transformation into films or coatings [21].

    As previously reported, many kinds of defects exist in layered MoS2 grown by CVD and ion exchange methods, such as chalcogen atom vacancies, antisite defects, impurities, interstitials, and dislocations [20, 22-24]. These defects commonly play a significant role in the photocatalytic processes and other photoinduced carrier dynamics. For example, more active sites introduced via the formation of defects within the layered MoS2 can lead to a significant improvement of the hydrogen evolution activity [25, 26]. Recently, a number of ultrafast experiments have been reported that nonradiative relaxation pathways dominate the carrier dynamics in layered MoS2 samples due to the fast trapping by mid-gap defect states [27-29], resulting in extremely low quantum yields of current 2D TMDs based light detectors and emitters [11-13, 30]. However, the mid-gap defects serving as carrier traps in these materials still remains poorly understood [31]. Meanwhile, rarely spectroscopic signal has been designated as a direct defect trapping signature in ultrafast spectroscopy measurements [32], which increases difficulties in unveiling defect-mediated carrier dynamics in 2D TMDs.

    In this work, we utilize ultrafast transient absorption (TA) spectroscopy to study the defect-assisted carrier dynamics in few-layer MoS2 prepared by ion exchange method. Compared with previously reported optical pump-probe studies of MoS2, the probe wavelength used in our experiments can be extended from visible to near-infrared (NIR) to mid-infrared (MIR). It has been demonstrated that the NIR spectrum is mainly contributed by the intraband absorption from both free carriers and excitons [31] as well as the bleaching of optically active mid-gap defect states [33], while the MIR spectrum is predominantly affected by the free carriers and weakly bound carriers with binding energies well below the detecting energy [34]. As a result, the dynamics of various kinds of bound and unbound carriers in few-layer MoS2 can be unambiguously investigated. Our experimental results indicate that the midgap defect states are widely distributed in few-layer MoS2 and greatly affect the dynamics of photoinduced carriers. Especially, the MIR TA spectrum reveals some shallow defect states occupied by electrons which are located at less than 0.24 eV below the conduction band minimum. These defect states can act as effective carrier trap centers to assist the nonradiative recombination process of photo-induced carriers on the picosecond time scale.

  • Layered SnS2 was first prepared by liquid exfoliation [35]. Next, few-layer SnS2 nanosheets (0.1 mmol), oleylamine (10 mmol), and 5 mL of 1-octadecene were loaded into a 50 mL three-necked flask. Then, the mixture was degassed under a nitrogen atmosphere at temperature of 120 ℃. In the following, MoCl5 (2 mmol), oleic acid (3 mL) and trioctylphosphine (3 mL) were injected together into the precursor mixture. The reaction was proceeded at 320 ℃ for 5 h, and then the mixture was cooled down to room temperature. Finally, the products were precipitated by adding ethanol and collected by centrifugation. More detailed process was described in Ref.[20]. The sample of MoS2 film was prepared by evaporating its solution on a CaF2 window. All experiments in this work were performed at room temperature.

  • Absorption spectroscopy was performed using Hitachi U-3900 UV-Vis spectrophotometer. Raman measurement was performed using a confocal micro-Raman spectrometer (WITec Alpha 300R). The excitation laser with a wavelength of 532 nm was focused by a ×100 objective lens onto the sample surface. The scattered signals were collected with the same lens. Atomic force microscopy (AFM) was carried out with a NT-MDT Solver P47H-PRO (Russia NT-MDT Corporation) in the tapping mode.

  • The ultrafast spectroscopy was performed with a femtosecond amplifier laser system (Spitfire Ace, Spectra Physics) that generated laser pulses with time duration of ~35 fs, central wavelength of 800 nm, and repetition rate of 1 kHz. The output was split into three paths. The first path was directed into an optical parametric amplifier (TOPAS, Spectra Physics) to generate tunable excitation pulses. The second path with weaker energy was focused onto a sapphire or an yttrium aluminum garnet plate to produce the white-light supercontinuum as visible or near-infrared probe pulses. A motorized delay stage was used to control the time delay between the pump beam and the probe beam, both of which were focused onto the sample. After frequency resolved by a spectrograph, the excitation-induced transmission change of the probe light was collected by a home-built 46-channel synchronous digital lock-in amplifier [36], with optional Si photodiodes and InGaAs sensors for visible and near-infrared detection, respectively. For mid-infrared measurement, the above probe light was replaced by the third path, which was directed to generate ultra-broadband super-continuum pulses that covered almost the whole mid-IR region by focusing 800 nm fundament light and 400 nm second harmonic simultaneously on air [37]. The spectrum of mid-infrared probe light was detected by a liquid-nitrogen-cooled mercury-cadmium-telluride array detector after frequency resolved by a spectrograph (iHR 320, HORIBA JobinYvon).

  • FIG. 1(a) shows the absorption spectrum of few-layer MoS2 nanosheets on calcium fluoride window. The absorption peaks located at 676 nm (1.83 eV) and 625 nm (1.98 eV) correspond to A- and B-exciton resonances as labeled. The A- and B-excitons are associated with direct transitions at the K point of the Brillouin zone between the spin-split valance band maxima and the conduction band minimum, respectively [3, 38]. An additional broad absorption peak at around 465 nm (2.67 eV), named C-exciton, is ascribed to transitions from deeper within the valence band to the conduction band [39]. Raman spectrum as displayed in FIG. 1(b) shows two prominent features which are assigned to E$ ^1_{ \rm{2g}} $ (381.6 cm-1) and A$ _{1 \rm{g}} $ (406.9 cm-1) modes, respectively. The peak distance between the two modes (25.3 cm-1) confirms the dominance of 7-8 layers in the measured MoS2 nanosheets [40]. The number of layers was further verified by AFM, as illustrated in FIG. 1(c). The AFM profile shows that the height of MoS2 nanosheets is primarily distributed around 6 nm, which is in good agreement with the layer number estimated by Raman spectroscopy [41].

    Figure 1.  (a) Absorption spectrum of MoS2 nanosheets on calcium fluoride window with A-, B-, and C-exciton features labeled. (b) Raman spectrum of MoS2 nanosheets showing two prominent modes as labeled. Open circles are the experimental points, and solid lines are the Gauss fits. (c) AFM image of MoS2 nanosheets; the inset is the height profile of the AFM image for three nanosheets.

    To reveal the defect-mediated carrier dynamics in the few-layer MoS2 nanosheets, the visible and NIR pump-probe measurements with resonant excitation of the A-exciton (676 nm) were firstly carried out. FIG. 2 (a) and (b) show the pseudocolor representation of waiting time dependent TA spectra in the visible and NIR regions, respectively. FIG. 2(c) displays the TA spectra at 0.2 ps time delay extracted from FIG. 2(a) (blue line) and FIG. 2(b) (red line). In the visible TA spectrum, the highly structured features with a series of alternating narrow positive and negative bands are observed, which is in accord with a previous report for few-layer MoS2 nanosheets [27]. The negative bands with peaks located at around 680 and 630 nm are assigned to the photobleaching of A- and B-excitonic transitions, respectively. The positive bands symmetrically situated on both sides of the photobleaching bands are attributed to the excited state absorption caused by the carrier-induced broadening of the excitonic transitions [42]. As reported in our previous work for few-layer MoS2 [43], the exciton resonance energy would be redshifted and the line width would be broadened after ultrafast excitation, leading to the current excited state absorption feature. In contrast, the NIR TA spectrum shows a relatively weak but clearly observable broadband bleach signal, which can extend to the spectral region with the wavelength much larger than 1500 nm. We have confirmed that the NIR bleach signal measured for the few-layer MoS2 nanosheets is predominantly contributed by the optical transitions from midgap defect states to the conduction bands [43]. Therefore, considering the photon energy of 1500 nm ($ \sim $0.82 eV) probe light is much smaller than the band gap of few-layer MoS2 ($ \sim $1.35 eV) [3], the broadband bleach signal observed in the NIR TA spectrum indicates that the mid-gap defect states are widely distributed in the measured few-layer MoS2 nanosheets. FIG. 2(d) and FIG. S1 (see supplementary materials) present the ultrafast dynamics detected at 676 nm (A-exciton, blue line) and 1026 nm (defect states, red line) for the few-layer MoS2 nanosheets. By fitting the dynamics curve detected at 676 nm with a multiexponential decay function, two fast decays with time constants of 1.1$ \pm $0.1 and 5.1$ \pm $0.3 ps followed by a slower process of 780$ \pm $60 ps are obtained, while the dynamics at 1026 nm exhibits only two decay components with time constants of 2.2$ \pm $0.8 ps and $ \sim $6$ \pm $2 ps, respectively. The observed fast decay components in both visible and NIR spectra indicate the mid-gap defect states play a significant role in the photoinduced carrier dynamics. According to previous reports [31, 43], the decay component of 1-2 ps can be ascribed to the fast trapping of carriers by defect states [27, 31]. Shi et al. reported that the nonradiative relaxation pathway rather than the radiative relaxation pathway dominate the dynamics in the monolayer and few-layer MoS2 [27]. In addition, the lifetime of radiative recombination of the photoinduced carriers in MoS2 is at least tens of picoseconds [44, 45]. Therefore, the decay component of 5-6 ps should be assigned to the nonradiative recombination of trapped carriers rather than the radiative recombination process. The remaining slower component observed at 676 nm that can last hundreds of picoseconds is strongly dependent on the pump fluence (see FIG. S2 in supplementary materials), and should be attributed to the thermalization of MoS2 lattice. The detailed discussion can be seen in our previous work [43]. In addition, the similar results were also obtained by nonresonant excitation at 400 nm, as displayed in FIG. S3 and S4 (see supplementary materials). It is worth noting that the signal amplitude in the visible region for the 400 nm excitation is much larger than that for the 676 nm excitation under similar pump fluence, while the contrary result was observed in the NIR region (see FIG. S3 in supplementary materials). This is because 676 nm excitation leads to a relatively large proportion of transitions from defect states to the conduction band compared with that for 400 nm excitation.

    Figure 2.  Waiting time ($ t_ \rm{w} $) dependent TA spectra in (a) visible and (b) NIR regions for the few-layer MoS2 nanosheets after photoexcitation at 676 nm with the pump fluence of 89 μJ/cm2. (c) TA spectra at 0.2 ps time delay extracted from FIG. 2(a) (blue line) and FIG. 2(b) (red line). (d) Extracted temporal evolution of excitation-induced transmission change detected at 676 nm and 1026 nm. Dots are data, and curves are multi-exponential fitting with the consideration of instrument response function (~150 fs).

    To further clarify the mechanism of defect-mediated nonradiative recombination process, we performed pump fluence dependent measurements with NIR probe pulses. FIG. 3(a) shows TA traces at 1026 nm recorded under different pump fluences after photoexcitation at 676 nm. The peak intensity of the signal extracted from FIG. 3(a) is proportional to the pump fluence (FIG. 3(b), yellow balls), indicating that the measurement was performed in the regime where the optical transitions are not saturated. Besides, all collected dynamics shown in FIG. 3(a) exhibit biexponential decay feature with the two decay time constants nearly independent of the pump fluence (FIG. 3(b)). This phenomenon is different from those reported in previous studies for the monolayer and few-layer MoS2, where a significant pump fluence dependence of the carrier dynamics was observed, and the Auger carrier capture model was thus proposed to explain the data [31, 33]. Here, our experimental results suggest that the defect-mediated nonradiative recombination (Shockley-Read-Hall recombination) mechanism serves as a dominated role in our sample, as illustrated in FIG. 4(a). The detailed discussion can be seen in our previous studies [43, 46]. The same phenomenon upon excitation at 400 nm was also clearly observed (see FIG. S5 and Table S1 in supplementary materials).

    Figure 3.  (a) Temporal evolution of excitation-induced transmission change of few-layer MoS2 detected at 1026 nm, under the excitation of 676 nm with different pump fluences. Circles are data, and curves are bi-exponential fitting. (b) The excitation-induced peak amplitude (orange balls, left vertical axis) and the fitted time constants (red and green balls, right vertical axis) as the function of incident pump fluence

    Figure 4.  (a) A schematic illustration of photoinduced carrier relaxation processes in few-layer MoS2. The blue solid lines and green dotted line indicate the mid-gap states and Fermi level ($ E_ \rm{F} $), respectively. The yellow and red vertical arrows indicate the initial excitation with 676 nm and 1150 nm pump pulses, respectively. The red dashed vertical arrows indicate the NIR and MIR probe pulses. The wavy gray arrows illustrate different relaxation processes: (1) trap of carriers by defect states and (2) nonradiative recombination of trapped carries via defect-mediated processes. (b) and (c) Temporal evolution of excitation-induced absorption change of MoS2 detected at 3500 nm, under the excitation of (b) 676 nm with a pump fluence of 100 μJ/cm$ ^2 $, and (c) 1150 nm with a pump fluence of 1.2 mJ/cm$ ^2 $. Circles are data and curves are exponential fitting. (d) Normalized temporal evolution of excitation-induced absorption change of MoS2 detected at different wavelengths, under the excitation of 676 nm with a pump fluence of 100 μJ/cm$ ^2 $. Circles are data and curves are global fitting results. (e) MIR TA spectra at four typical time delays. The TA spectrum at 0.16 ps is fitted in the form of $ \lambda^2 $ (gray dashed line). The distortion of the spectrum close to 4260 nm is caused by the light absorption of CO2 in the atmosphere. (f) Decay associated difference spectrum corresponding to the 10.2 ps slow decay component, which was obtained from the global fitting analysis for the MIR TA data. The orange arrow indicates the position of the spectrum converting from negative to positive.

    In the following, we tuned the probe wavelength to the MIR region to further explore the properties of the mid-gap defect states distributed in MoS2. Since the photon energy of MIR pulse is well below the exciton binding energy of layered MoS2 [47], the MIR detection pulse is only sensitive to free charge carriers and weakly bound electron/hole pairs, and barely affected by the tightly bound carriers such as excitons [34]. FIG. 4(b) displays the temporal evolution of excitation-induced absorption change detected at 3511 nm for the MoS2 under the excitation of 676 nm. It shows that a positive signal reaches its maximum immediately after photoexcitation, and then a positive to negative crossover of the signal at around 2 ps is clearly observed. In general, the collected MIR TA signal is caused by a superposition of intra- and inter-band transitions, which contributes to the differential optical conductivity with opposite signs. Here, the negative signal ($ \Delta $OD) can be attributed to the photobleaching caused by the interband electron transitions. Since the probe photon energy ($ \sim $0.35 eV) is much lower than the band gap of MoS2, the interband electron transition should be from the defect states distributed below the conduction band to the conduction band, which is similar to that observed in the NIR TA spectrum. In contrast, the positive signal is predominantly due to the absorption of free charge carriers as well as weakly bound electron/hole pairs that is generated by the photoexcitation (see FIG. 4(a)). After 2 ps, most of these carriers will form tightly bound carriers, such as excitons and trapped carriers. As a result, the negative photobleaching signal dominates the excitation-induced absorption change at a relatively longer time delay, leading to a positive to negative signal conversion.

    To confirm this explanation, we further tuned the pump wavelength to 1150 nm, which could generate a relatively large proportion of transitions from defect states to the conduction band as discussed above (see FIG. 4(a)). As shown in FIG. 4(c), the negative signal rapidly appears within only hundreds of femtoseconds, and becomes more remarkable compared with that after 676 nm-excitation (FIG. 4(b)). It further indicates that some shallow defect states occupied by electrons distribute below the conduction band since the detecting photon energy is only about 0.35 eV. By exponential fitting for the dynamics in FIG. 4(c), a $ \sim $8.7 ps recovery time for the bleaching signal can be obtained, which is in agreement with the time scale of defect-assisted nonradiative recombination process observed by visible and NIR probe (FIG. 2(d) and 3(a)).

    The above results also suggest that the photobleaching signal can be used to uncover the occupied states of the electrons distributed below the conduction band. For this purpose, we collected the TA data ranging from 3000 nm to 7700 nm under the excitation of 676 nm. FIG. 4(d) displays the measured dynamics at several probe wavelengths, with all the datasets normalized by their maximum values. For the dynamics detected at shorter wavelength (e.g., 3128 nm), a rapid conversion of the signal from positive to negative values can be clearly observed, which has been discussed above. In contrast, no obvious negative bleach signal can be detected as the probe wavelength tuned to be larger (e.g., 7498 nm).

    We further plot the MIR TA spectra at four typical time delays, as shown in FIG. 4(e). At 0.16 ps after photoexcitation, the TA spectrum can be well fitted by a function of $ C $$ \cdot $$ \lambda^2 $ ($ C $ is a constant coefficient, and $ \lambda $ denotes the probe wavelength), which can be assigned as the absorption of free carriers generated by the pump light [37]. As demonstrated above, most of these carriers will form tightly bound carriers after several picoseconds. As a consequence, the negative photobleaching feature appears at longer time delays (FIG. 4(e)). We analyze the entire spectral and temporal pump-probe dataset with multiexponential global fitting analysis, and three decay time constants of 0.7 ps, 3.1 ps and 10.2 ps are therefore obtained (see FIG. S6 in supplementary materials), which are related to the formation of excitons [43], trap of carriers by defects and nonradiative recombination of trapped carriers, respectively. Especially, we depict the decay associated difference spectrum that is related to the 10.2 ps slow decay component in FIG. 4(f). It is clearly shown that the photobleaching signal can be observed in a wide range of spectral region, and disappears as the probe wavelength increasing to larger than 5100 nm ($ \sim $0.24 eV). Here, the transition point of around 0.24 eV implies the lower limit of the shallow defect states occupied by electrons (FIG. 4(a)).

    Zhou et al. have reported that a S2 column substituting a Mo atom (S2Mo) can induce deep and shallow levels [48]. In addition, both monosulfur vacancy (V$ _ \rm{S} $) and disulfur vacancy (V$ _{ \rm{S}_2} $) induce two levels about 0.6 eV below conduction band minimum [48]. Besides, Lu et al. also reported that the Fermi level was at $ \sim $0.35 eV below the conduction band for MoS2 [49]. Our experimental results indicate that some shallow defect states occupied by electrons distribute below the conduction band ($ < $0.24 eV), which is consistent with n-doping property of MoS2 [23]. These defect states will serve as effective carrier trap centers to assist the nonradiative recombination of photo-induced carriers in few-layer MoS2.

  • In summary, we carried out an ultrafast Vis-NIR-MIR spectroscopy to investigate the defect-mediated carrier dynamics in few-layer MoS2 prepared by ion exchange method. We have demonstrated that the mid-gap defect states are widely distributed in few-layer MoS2 nanosheets and play an essential role in the nonradiative recombination process of photoinduced carriers. From the MIR TA spectrum, we further revealed some shallow defect states occupied by electrons which are located at less than 0.24 eV below the conduction band minimum. Our results provide a comprehensive understanding of the defect-related carrier dynamics in few-layer MoS2 and will be helpful in evaluating the performance of 2D TMDC-based optoelectronic devices and photocatalytic materials in the future.

    Supplementary materials: Visible and NIR TA dynamics of few-layer MoS2 nanosheets excited at 676 nm and 400 nm, pump-fluence dependent TA dynamics of few-layer MoS2 detected at 676 nm under the excitation of 676 nm and 400 nm, TA spectra at 0.6 ps time delay for the few-layer MoS2 after photoexcitation at 400 nm, fitted time constants for the measured dynamics, and results of global fitting analysis for the MIR TA data are available.

  • This work was supported by the National Key Research and Development Program of China (No.2018YFA0208700), the National Natural Science Foundation of China (No.21603270 and No.21773302), and the Strategic Priority Research Program of Chinese Academy of Sciences (No.XDB30000000).

  • Figure S1.  Temporal evolution of excitation-induced transmission change for the few-layer MoS2 nanosheets detected at 676 nm and 1026 nm, under the excitation of 676 nm (89μJ/cm2).

    Figure S2.  Normalized temporal evolution of excitation-induced transmission change for the few-layer MoS2 detected at 676 nm, under the excitation of 676 nm with pump fluences of 38 and 89 μJ/cm2. The slow decay component that can last hundreds of picoseconds is strongly dependent on the pump fluence, which can be attributed to the thermalization of MoS2 lattice.

    Figure S3.  (a) Transient absorption (TA) spectra at 0.6 ps time delay for the few-layer MoS2 after photoexcitation at 400 nm. The pump fluences are 110 μJ/cm2 for the visible probe and 50 μJ/cm2 for the NIR probe. (b) Temporal evolution of excitation-induced transmission change detected at 676 nm and 1026 nm under the excitation of 400 nm. Dots are data, and curves are multi-exponential fitting with the consideration of instrument response function (~150 fs).

    Figure S4.  Temporal evolution of excitation-induced transmission change for the few-layer MoS2 detected at 676 nm, under the excitation of 400 nm with different pump fluences.

    Figure S5.  (a) Temporal evolution of excitation-induced transmission change for the few-layer MoS2 detected at 1026 nm, under the excitation of 400 nm with different pump fluences. Circles are data, and curves are multi-exponential fitting with the consideration of instrument response function (~150 fs).

    Table S1.  Fitted time constants for decay curves (-ΔT/T) in Figure S5.

    Figure S6.  Results of global fitting analysis for the MIR TA data with three exponential decay functions showing (a) three exponential decay dynamics with time constants 0.7 ps, 3.1 ps and 10.2 ps, respectively, and (b) three corresponding decay associated difference spectra (DADS). The data was collected under the excitation of 676 nm with a pump fluence of 100 μJ/cm2.

Reference (49)

Catalog

    /

    DownLoad:  Full-Size Img  PowerPoint
    Return
    Return