Modulation of opto-electronic properties of InSe thin layers via phase transformation

Abstract

Phase engineering of two-dimensional (2D) materials offers unique opportunities for acquiring novel opto-electronic properties and allows for the searching of outstanding candidates for applications in opto-electronic detectors, sensors, catalysis, or phase-change memory devices. Here, we report the phase-transformation from β-InSe to γ-In₂Se₃, exploiting the thermal annealing route to trigger the process starting at 200 °C that expands the family of phase-change materials. The presence of γ-In₂Se₃ is solidly confirmed by the characteristic peaks in X-ray diffraction (XRD) and energy dispersive X-ray (EDX), and is quite stable at ambient condition, thus facilitating substantial application in phase-change memory devices. A Raman shift in the A′₁ mode from 225 cm⁻¹ to 230 cm⁻¹ further illustrates the phase transformation. Besides the photoluminescence (PL) peak of β-InSe, the ∼2 eV PL peak, ascribed to γ-In₂Se₃, is observed in the annealed nanosheet. The increased PL band gap of β-InSe as a function of annealing temperature during phase transformation was possibly affected by the suppressed interlayer coupling, as well as the planar quantum confinement of photo-excited carriers by the external surfaces of the sheets. The photodetector performance with respect to photocurrent, mobility, detectivity, responsivity, and external quantum efficiency was subsequently evaluated after thermal annealing, showing deteriorated optical performance. The present work proved that thermal annealing could induce the successful phase transformation, and adjusted the opto-electronic properties in some extent, providing useful information for processing 2D materials based nano-devices.

---

Introduction

Two-dimensional (2D) materials have attracted much attention mainly owing to their exotic physical properties, which are strikingly different from their bulk counterparts.^1–5 Graphene, the most famous member of the 2D material family, has extraordinary physical properties and showed promising potential in various applications.^3,6 The difficulties of creating semiconducting graphene, and the lack of its native band gap, have led to extend studies focusing on other 2D semi-conducting materials. For example, transition metal dichalcogenides (TMDs) layered semiconductors^7,8 can be prepared in a single or few-layers structure by mechanical or liquid exfoliation,^9,10 due to the weak interlayer interactions, or other techniques such as chemical vapor deposition.^11,12 TMDs down to a mono-layer or few-layers show extremely fascinating optical properties compared to the bulk, and have been widely studied in terms of opto-electronic devices, showing promising applications in valleytronics and spintronics.¹³⁻¹⁵ Bulk Mo and W dichalcogenides (MoS₂, MoSe₂, WS₂, and WSe₂) possess an indirect band gap. Strikingly, single layers have a direct band gap and become strongly photoluminescent.^16,17 Such layered TMDs structures have paved the way for the fabrication of highly efficient field-effect transistors^10,18 as well as high performance photodetectors.^19,20

The phase-transition of 2D materials has received tremendous attention in recent years because of its promising application in opto-electronic devices, catalysis, and especially in phase-change memories.²¹ Phase conversion in TMDs can be realized by chemical routes at room temperature, pressure, or thermal annealing.²² For instance, the 2H phase of MoS₂ is semiconducting, but the low electrical conductivity makes it not suitable as a good electrode. In contrast, the 1T phase of MoS₂ is metallic, which can be prepared by alkali doping via intercalation or substitutional doping with elements with more valence electrons, shows a much improved electrical conductivity. The 1T phase of MoS₂ shows great promise in the construction of logic devices, not only due to the structurally coherent interface between the 2H and 1T phases of MoS₂, but also for the ohmic contact built using the metallic 1T phase of MoS₂ as a source and drain electrodes.²³ First-principle calculations predicted that hexagonal (H) and trigonal (T) crystal structures of VS₂ are both thermodynamically stable, and the dimension greatly affects the relative stability and related magnetism.²⁴ Recent work found that the polymorphic In₂Se₃ has multiple crystalline phases, such as α, β, and γ phases, which can be prepared by thermal treatment at various temperatures. The temperature of the crystalline–crystalline (α → β) phase transformation of In₂Se₃ was as high as 130 K with a layer thickness of 4 nm. The β-phase In₂Se₃ showed an electrical resistivity about 1–2 orders of magnitude smaller than the α-phase, providing a promising candidate for multi-level phase-change memories in a single material system.²⁵

Among dozens of reported TMDs, bulk InSe has a direct band gap (∼1.26 eV at room temperature²⁶), showing interesting optical properties,^27,28 that can be changed under different pressure states.^29,30 Furthermore, it has a range of applications for Li-batteries³¹ as well as photovoltaic devices with an efficiency of 11%.³² Very recently, layered indium selenide (InSe) has experienced renewed interests, due to its optical, electronic, and mechanical properties with a tendency towards applications in memory devices, opto-electronic sensors, and thermoelectric tools.^33,34 Furthermore, layered InSe has a crossover transition from the direct to indirect band gap when the layer thickness is reduced to ≤6 nm, and its narrower band gap, with smaller exciton reduced mass, leads to stronger quantum confinement more than any other semiconductor belonging to the group IIIA–VIA materials.³⁵ These behaviors show the opportunity for controlling the band gap, providing tunable nano-devices in a single material system. A very recent report displayed that photodetectors fabricated from few-layered InSe on an SiO₂/Si substrate had a photo-response of 34.7 mA W⁻¹ at 532 nm.³⁶ Briefly, InSe showed various optical properties that can be controlled by pressure,³⁷ heat treatments,³⁸ and a capping layer.³⁵ However, the effect of thermal annealing on the optoelectronic properties of InSe in low dimension is seldom reported up to date. Taken together, we tend to apply thermal annealing as an alternative process for modulating the optical or electronical properties of InSe nanosheets.

In this work, we have studied the post-thermal annealing effects on the phase transformation of 2D InSe nanosheets under different temperatures. Single-crystalline 2D InSe nanosheets prepared by mechanical exfoliation and transferred onto a SiO₂/Si substrate. These nanosheets were annealed under argon gas flow under different temperatures (200 °C, 300 °C, and 400 °C) for 30 minutes. Based on the results of XRD, EDX, Raman, and PL spectra, the thermal annealing process causes a new phase generation of γ-In₂Se₃ from β-InSe. Moreover, the PL band gap of InSe nanosheets increases as a function of thermal annealing temperature. The effect of the phase transformation on the opto-electric properties, in terms of a photodetector device, was also carried out and gave a tentative explanation for the deteriorated device performance due to the phase transformation.

Experimental

Materials and methods

The bulk InSe crystals were prepared by chemical vapor deposition (CVD). Selenium powder (>99.99%, Aladdin Co.) and indium (>99.99%, Aladdin Co.) were placed in two quartz boats, respectively. Then, the boats were placed in a horizontal furnace with a fused silica tube and separated by a distance of 20 cm between them. The closed system was purged under 100 sccm Ar gas for 30 min. Then boats with In and Se powders were heated to 973 K and 673 K, respectively, for 2 h under 10 sccm Ar gas. After that, the system was cooled down to room temperature naturally. InSe nanosheets were prepared by mechanical exfoliation using adhesive tape from the bulk single crystal and transferred onto a silicon wafer with an oxide layer of 300 nm. Subsequently, some of these sheets remained as-exfoliated, and the other sheets were annealed at 200 °C, 300 °C, and 400 °C under 10 sccm Ar gas, for 30 minutes. After the annealing process, the furnace was cooled down to room temperature by natural cooling, the sheets were taken out from the furnace, and directly transferred to the Raman chamber for optical measurements. For device fabrication, InSe nanosheets were deposited on SiO₂/Si by mechanical exfoliation. The electrodes were Cr/Au (Cr is 5 nm thick and Au is 35 nm thick) deposited by thermal evaporation using a shadow mask. After opto-electronic measurements of the as-exfoliated sheet, the device was annealed at 200 °C for 30 min with 100 sccm (Ar/H₂ = 9 : 1) to investigate the effect of phase transformation on the opto-electronic properties.

Characterization of InSe nanosheets. The structure of the grown InSe was characterized by X-ray diffraction (XRD, Diffractometer-6000 with Cu Kα radiation, λ = 0.1542 nm). The optical absorption and transmittance spectra were measured using an UV-vis-NIR spectrometer (Hitachi U-3410). The elemental composition ratio was determined using a scanning electron microscope (SEM, Hitachi S-4300 with an accelerating voltage of 20 kV attached with an energy dispersive X-ray spectroscopy (EDS)). The thicknesses of InSe nanosheets were determined by atomic force microscopy (AFM, Nanoscope IIIa Vecco). Raman, as well as photoluminescence (PL), spectra of InSe nanosheets were achieved by Raman microscopy (LabRAM XploRA, incident power of 1 mW, pumping wavelength of 532 nm). The electrical properties of the photodetectors, based on the as-exfoliated and annealed InSe nanosheets, were determined via a semiconductor characterization system (Keithley 4200 SCS) with a Lakeshore probe station. Monochromatic 490–850 nm light was used via an optical filter of 500 W Xenon lamp as the light source. The fixed intensity was determined with a power and energy meter (Model 372, Scienteck).

Results and discussion

Fig. 1a displays the top view of a hexagonal structure of a single layered InSe and the lattice constant along the a-axis or b-axis is 0.40 nm. Fig. 1b and S1† show the AFM images of InSe nanosheets on the SiO₂/Si and the corresponding height profiles, indicating thickness of ∼6 nm, ∼9 nm, ∼11 nm, ∼13 nm, ∼15 nm, ∼18 nm, ∼21 nm, and ∼24 nm, respectively. The new phase generation of γ-In₂Se₃ was studied by XRD, Raman spectroscopy, and PL spectra. The XRD technique is a useful method to give information about the occurrence of a secondary phase.³⁹ Raman and PL analyses are also the main tools to characterize secondary phases or phase transformation in the crystal structure.^25,40 XRD of as-exfoliated and annealed β-InSe are shown in Fig. 2a. Pure β-phase InSe without any impurities such as In₄Se₃, In₆Se₇, or Se, was confirmed by a blue trace in Fig. 2a, compared with the standard XRD curve of β-InSe (JCPDS card no. 4-1431), showing a hexagonal unit cell and lattice parameters of a = b = 4.005 Å and c = 16.64 Å.⁴¹ The peak with strong diffraction at 2θ = 21.68° corresponds to the diffraction from the (004) planes, and the other peaks at 2θ = 10.9°, 32.4°, 43.59°, 50.78°, and 67.5° are the diffractions on the (002), (006), (008), (114), and (0012) planes, respectively. However, thermally annealed β-InSe at various temperatures (200–400 °C) showed some new peaks in Fig. 2a at 2θ = 27.5° corresponding to (006),⁴² 2θ = 29.27° of (201), 2θ = 24.96° of (110),³⁹ and 2θ = 30.3° of (202).⁴³ All these new peaks are indexed to a hexagonal γ-In₂Se₃ (JCPDS card no. 89-0658). These observations of the characteristic peaks of β-InSe and γ-In₂Se₃ after thermal annealing are in agreement with the literature that stated the coexistence of InSe phases.^44–46 This transformation can be attributed to the diffusion of In into the Se lattice, causing the formation γ-In₂Se₃, as reported in previous studies.^43,47

Fig. 1 (a) Top-view of the hexagonal structure of a single layer of InSe where the lattice constant along a-axis or b-axis is 0.40 nm. Side view of the lattice structure. The distance between the two nearest layers is 0.84 nm. (b and d) AFM images of InSe nanosheets on SiO₂/Si. (c and e) The height profiles of the corresponding dotted lines of A and B, respectively.

Fig. 2 (a) XRD curves of the as-exfoliated β-InSe and samples annealed at 400 °C, 300 °C, and 200 °C. Standard curves of β-InSe and γ-In₂Se₃ are shown in bars. (b) XRD intensity ratio of I_{(0 0 6)} of γ-In₂Se₃ to I_{(0 0 6)} of β-InSe.

The intensity ratio of characteristic peaks between different phases is a semi-empirical XRD analysis that is useful to determine the facet's growth⁴⁸ or the degree of phase transformation and crystallization.^49,50 The intensity ratio of one typical peak (I_{γ(0 0 6)}) of γ-In₂Se₃ to the other peak intensities of β-InSe as a function of annealing temperature are plotted in Fig. 2b and S2.† The intensity ratio of I_{γ(0 0 6)} of γ-In₂Se₃ to the I_{β(0 0 6)} of β-InSe with annealing temperature indicate the percentage of γ-In₂Se₃ phase increases with annealing temperature.

The structure and the existence of the new phase transformation of annealed InSe were further checked by Raman spectroscopy. The layered β-InSe belongs to a D⁴_6h symmetry and has six Raman modes.⁵¹ The corresponding modes are shown in Fig. 3a. The Raman modes of the as-exfoliated β-InSe, annealed at 200 °C, 300 °C, and 400 °C, are shown in Fig. 3b, c, and S3,† respectively. These modes are classified as E′′ at 41 cm⁻¹, A′₁ at 117 cm⁻¹, E′′ at 176 cm⁻¹, E′ at 202 cm⁻¹, A′′₂ at 209 cm⁻¹ and A′₁ at 225 cm⁻¹, respectively. It is worth noting that the mode A′₁ at 225 cm⁻¹ in the as-exfoliated sheets shifted to 230 cm⁻¹ for the annealed sheets, which belongs to γ-In₂Se₃.^52,53 Such a result solidly indicated the existence of the γ-In₂Se₃ phase. Additionally, it is noted that the peak positions of all modes in both as-exfoliated and annealed InSe nanosheets were independent of the thickness, agreeing well with previous experimental and theoretical studies.^35,54 Considerable decreases in their intensities were also observed as the sheet thickness decreased, as well as the increase of annealing temperature in Fig. S3c.†⁵⁵

Fig. 3 (a) Schemes of the six Raman modes of InSe. Raman spectra of (b) as-exfoliated sheets and (c) annealed sheets at 200 °C as a function of thickness. The thickness from bottom to top in both b and c are about 6, 9, 11, 13, 15, 18, 21, and 24 nm.

The chemical compositions of the 2D InSe nanosheets were characterized by EDS, as shown in Fig. S4,† showing the presence of In and Se in addition to Si and O peaks from the substrate, which confirms the high pure nature of the prepared InSe nanosheets. The EDS results summarized in Table S1† show that the atomic percentage of In and Se are around 52.4% and 47.6%, respectively, for the as-exfoliated InSe. However, the ratio of In/Se decreased from 1.1 to 0.83 with the increase in annealing temperature. As is known, the indium re-evaporated from the crystalline sheet surfaces due to the loss of volatile selenium during the annealing process.⁵⁶ Such an atomic ratio change also indicated the phase transformation through the annealing process, as observed in XRD and Raman spectra. Additionally, the β → γ transformation is further supported by the PL spectra of the annealed sheets. Fig. 4a displays that the as-exfoliated sheets have only one PL peak, confirming the single-phase of the β-InSe. In contrast, the annealing process causes the dominant higher PL emissions at ∼2 eV in Fig. 4b, S5a and S5b.† The higher PL emissions at ∼2 eV correspond to the γ-In₂Se₃ phase caused by the thermal annealing process, consistent with previous reports.^39,47,57,58 Taken together, based on the analysis of XRD, EDX, Raman modes, and PL spectra, it is safe to conclude that the occurrence of the phase transformation, or mixed phases of β-InSe and γ-In₂Se₃ via the thermal annealing process, indeed occurred.

Fig. 4 PL spectra of (a) as-exfoliated sheets and (b) annealed sheets at 200 °C. The thickness from bottom to top in both (a) and (c) are about 6, 9, 11, 13, 15, 18, 21, and 24 nm, respectively. (c, d, and e) PL band gap, intensity, and FWHM of the as-exfoliated and annealed β-InSe nanosheets as a function of thickness, respectively.

In order to disclose the influence of the phase transformation caused by the annealing process on the optical or electronic properties of InSe, we first evaluated the PL band gap of β-InSe nanosheets in detail. The decrease in the thickness and the increase in the annealing temperature caused the photon energy of β-InSe to shift to higher energies and the FWHM of the PL peak of β-InSe become larger. Such PL shifting trends indicated that the phase transformation enhances the progressive development of higher energy components at the consecutive thin parts of the annealed nanosheets. Fig. 4c displays the PL band gap of β-InSe nanosheets as a function of the thickness from ∼6 to ∼24 nm for both the as-exfoliated and annealed samples. The PL band gap of InSe nanosheets in all thickness increases with the increase of thermal annealing temperature. For example, PL shifted ∼40 MeV for annealed sheets in ∼6 nm at 400 °C in comparison to the as-exfoliated sheet. This phenomenon is consistent with the planar quantum confinement of photo-excited carriers by the external surfaces of the sheets. The quantum confinement effect for annealed sheets is more prominent than that of the as-exfoliated sheets.^33,59 The decreased intensity and widening of the FWHM of the PL emission as a function of annealing temperature were also observed in Fig. 4d, agreeing well with previous work.^35,59,60 Meanwhile, the decreased PL signal can be also observed as the thickness decreased, which is attributed to the enhancement of non-radiative surface carrier recombination processes with decreasing thickness.⁶¹ The widened FWHM can be attributed to the surface roughness, or increase in the binding energies, due to strong quantum confinement.^35,61

Fig. 5a clarifies the influence of phase transformation temperature on the increase in the PL band gap of β-InSe nanosheets. Fig. 5b shows the assumed sketch of band structures for the as-exfoliated and annealed sheets. The top of the valence band is constituted mainly by the p_z orbital of the Se atom. Below the valence band, there are p_x,y orbitals from the Se atom, ascribed to the interband transition E′.²⁸ The top conduction band is shaped by the overlap of the p_z orbital of Se with a small part of the Se p_x,y orbitals. Such an overlap phenomenon will lead to conduction bands behaving as s-like orbitals, and thus, results in intense interlayer interactions and finally causes energy level splitting by a narrow band gap in the bulk InSe and GaSe.^28,62 During thermal annealing, this suppressed coupling was responsible for the reduction of the photocurrent response and the increase of the band gap, mainly due to the shifting of both p_z and p_x,y orbitals and the enlarged electronic band structure.³⁶ Thus, the increase in PL band gap for annealed sheets, more than the exfoliated sheets as in Fig. 5a, as well as the decrease in photocurrent as in Fig. S7† after annealing process, confirmed the enhancement of the suppressed interlayer coupling and related orbitals shifting in Fig. 5b. Fig. S5c† shows the decreases in peak intensity of γ-In₂Se₃ for each thickness with increasing annealing temperature, attributed to the enhancement of the non-radiative recombination effect in γ-In₂Se₃. The intensity ratio of the γ-In₂Se₃ peak (I_γ) to the β-InSe peak (I_β) was defined as . It is found that this ratio for each thickness decreased as the annealing temperature increased in Fig. S5d.† The decreased intensity ratio indicated that the non-radiative process in recombination centers of the secondary phase γ-In₂Se₃ was strengthened more than that in the main phase β-InSe.

Fig. 5 (a) PL band gap of β-InSe as a function of annealing temperature. The thickness from top to bottom as indicated by the dotted arrow are about 6, 9, 11, 13, 15, 18, 21, and 24 nm. (b) Electronic band energy changes of β-InSe in p_z orbitals during phase transformation under thermal annealing. The InSe band gap is well defined between p_z-like orbitals.

Moreover, Fig. S6† shows the absorption and transmission spectra of the as-exfoliated InSe and annealed samples at 200 °C and 400 °C for 30 minutes. There is an obvious peak in the range from 850 to 1100 nm in the case of the as-exfoliated InSe, ascribing to the band gap of β-InSe. After thermal annealing at 200 and 400 °C, the position of this peak still remains. The exact central position of this peak is difficult to determine, but the peak showed a slight blue shift compared to that of the as-exfoliated InSe, as indicated by the dotted line in Fig. S6a.† Such a trend is consistent with the increase of the PL band gap for the annealed β-InSe at higher temperature. Meanwhile, a new peak appeared in the range from 400 to 600 nm, which can be ascribed to the new phase generation of γ-In₂Se₃. This peak also agrees well with the PL band gap of γ-In₂Se₃, which is ∼2 eV. The transmission spectra is provided in Fig. 6b. One obvious valley corresponding to a relatively lower transmission is observed in the range from 850 to 1100 nm. In comparison, two valleys corresponding to the relatively lower transmission in annealed InSe are observed in the range from 850 to 1100 nm and from 400 to 600 nm. As is known, the largest phonon loss will happen when the incident phonon energy is close to the intrinsic band gap of the tested semiconductors. Therefore, the two absorptions and transmissions peaks can be ascribed to the partial phase transformation from β-In₂Se₃ to γ-In₂Se₃ during the thermal annealing process, consistent with the results of the XRD and PL spectra.

Fig. 6 (a) Transfer curve of the as-exfoliated and annealed InSe nanosheets based FET. The inset is a schematic drawing of a FET or photodetector device. (b and c) Drain–source characteristics of the as-exfoliated and annealed device at 200 °C that are measured at V_g = 0 V with various wavelength of light under light intensity of 0.29 mW cm². (d, e, and f) Responsivity, detectivity, and EQE as a function of incident wavelength, respectively.

The phase transformation will affect the optical or electrical properties of InSe. We finally evaluated the effect after the introduction of γ-In₂Se₃, in terms of photodetector, of the as-exfoliated and the annealed InSe nanosheets at 200 °C. The inset of Fig. S7a† shows the optical image of the device with a thickness of ∼24 nm, length (L) of ∼20 nm, and width (W) of 25 nm. Fig. 6a displays that both photodetectors exhibit typical n-type FET properties. The inset in Fig. 6a designates the schematic structure of the device. The current on/off ratio is calculated by assuming the ratio of maximum to minimum of the drain current I_ds of the curve of V_g vs. I_ds. The current on/off ratio is 5.51 × 10⁴ for the as-exfoliated device and 7.42 × 10³ for the annealed device (Table S2†). The carrier mobility can be obtained using the following equation:

μ = [L/(W × (ε₀ε_r/d × V_ds)) × dI_ds/dV_g]

where L = 20 mm, W = 25 mm, ε₀ = 8.854 × 10¹² is the vacuum permittivity, ε_r is 3.9 for SiO₂, and d (285 nm) is the thickness of SiO₂. The calculated mobilities found to be 10.32 cm² v⁻¹ s⁻¹ for the as-exfoliated device and 2.37 cm² v⁻¹ s⁻¹ for the annealed device. To measure the effect of the phase transformation on other parameters such as photocurrent, photo-responsivity, detectivity, and external quantum efficiency (EQE), a monochromatic light illumination was focused vertically with a fixed intensity of 0.29 mW cm⁻² under illumination wavelengths ranging from 490 to 850 nm, as seen in Fig. 6b and c. The photocurrents (I_p = I_illumination − I_dark) for the annealed device decreased compared to the device consisting of the as-exfoliated nanosheets (Fig. S7b†). This decrease in the photocurrent is consistent with the observed effect of phase transformation on the PL band gap. It is assumed that the photocurrent is related to the absorption in the range of the wavelength, i.e., I_p = α × d × η, where α is the absorption, d is the thickness of InSe nanosheet, and η is the quantum efficiency. The band gap behavior of the semiconductor can be expressed as follows:

α = (E − E_g)^1/2

where E is the incident photon energy and E_g is the band gap of the semiconductor. Consequently,

I_p ∝ (E − E_g)^1/2

Thus, we can conclude that the decrease in photocurrent is consistent with the increase in the band gap. The R, D* and EQE are vital parameters to estimate the performance of a photodetector.⁶³ R is defined as the photocurrent produced per unit power of the incident light upon the effective area of a photodetector as calculated by:

R = I_P/PS

where P is the incident light intensity and S (500 μm²) is the effective area. The shot noise caused by the dark current is the main subscriber to the total noise and in this event, the detectivity can be estimated from:

D* = RA^1/2/(2eI_d)^1/2

where A is the photodetector channel area, e is the electron charge, and I_d is the dark current. EQE is defined as the number of electrons per incident photon and calculated from:

EQE = hcR/eλ,

where h is Planck's constant, and λ is the wavelength. Fig. 6d to f shows the responsivity, detectivity, and EQE as a function of wavelength. It is found that all these parameters decreased in the annealed device compared to those of the as-exfoliated device in Table S2.† The introduction of the new phase into the pure β-InSe will somehow destroy the crystalline structure, and thus, lead to a non-ideal carrier mobility transfer path due to more scattering centers at grain boundaries, which is highly responsible for the decreased photocurrent in the thermally annealed device. In most previous work, researchers widely applied the thermal annealing method around 200 °C to enhance the contact between the metal electrodes and semiconductor. However, on the basis of the present work, one should take great caution to process the annealing treatment when taking into account the possible phase transformation of the semiconductor. As confirmed in the present work, the phase transformation from β-InSe to γ-In₂Se₃ provided opportunities to manipulate the band gap that resulted in deteriorated photodetector performance.

Conclusions

In summary, our results indicate that β-InSe nanosheets undergo partial phase transformation under thermal annealing, and the generation of γ-In₂Se₃ resulted in variations of band gap and electronic properties. The phase transformation process effectively occurred above the thermal annealing temperature of 200 °C, and the characteristic features of γ-In₂Se₃ can be confirmed by XRD, EDX, Raman, and PL spectra. The PL band gap of the InSe nanosheets increases as a function of thermal annealing temperature from 200 °C to 400 °C, attributed to the suppressed interlayer coupling as well as the planar quantum confinement of the photo-excited carriers by the external surfaces of the sheets. The effects of the phase transformation on the opto-electronic properties in terms of the photodetector device were also carried out, and the deteriorated device performance was ascribable to the phase transformation during thermal annealing. This work provides useful clues to modulate the electronic orbital structure of the InSe thin layers via phase transformation for opto-electronic applications, providing useful information for processing 2D materials-based nano-devices.

Acknowledgements

This study is supported by the National key Basic Research Program of China (973 Program) under Grant No. 2013CB632900, the National Natural Science Foundation of China (NSFC, 61390502, 21373068), Project supported by the Foundation for Innovative Research Groups of the National Natural Science Foundation of China (Grant No. 51521003) and by Self-Planned Task (No. SKLRS201607B) of State Key Laboratory of Robotics and System (HIT).

Notes and references

1. K. Novoselov, A. K. Geim, S. Morozov, D. Jiang, M. Katsnelson, I. Grigorieva, S. Dubonos and A. Firsov, Nature, 2005, 438, 197–200.
2. S. Z. Butler, S. M. Hollen, L. Cao, Y. Cui, J. A. Gupta, H. R. Gutierrez, T. F. Heinz, S. S. Hong, J. Huang and A. F. Ismach, ACS Nano, 2013, 7, 2898–2926.
3. A. K. Geim and K. S. Novoselov, Nat. Mater., 2007, 6, 183–191.
4. M. Xu, T. Liang, M. Shi and H. Chen, Chem. Rev., 2013, 113, 3766–3798.
5. Y. Guo, K. Xu, C. Wu, J. Zhao and Y. Xie, Chem. Soc. Rev., 2015, 44, 637–646.
6. X. Li, M. Rui, J. Song, Z. Shen and H. Zeng, Adv. Funct. Mater., 2015, 25, 4929–4947.
7. Q. H. Wang, K. Kalantar-Zadeh, A. Kis, J. N. Coleman and M. S. Strano, Nat. Nanotechnol., 2012, 7, 699–712.
8. A. Geim and I. Grigorieva, Nature, 2013, 499, 419–425.
9. A. Ayari, E. Cobas, O. Ogundadegbe and M. S. Fuhrer, J. Appl. Phys., 2007, 101, 14507.
10. Y. Zhang, J. Ye, Y. Matsuhashi and Y. Iwasa, Nano Lett., 2012, 12, 1136–1140.
11. W. S. Hwang, M. Remskar, R. Yan, V. Protasenko, K. Tahy, S. D. Chae, P. Zhao, A. Konar, H. G. Xing and A. Seabaugh, Appl. Phys. Lett., 2012, 101, 013107.
12. K.-K. Liu, W. Zhang, Y.-H. Lee, Y.-C. Lin, M.-T. Chang, C.-Y. Su, C.-S. Chang, H. Li, Y. Shi and H. Zhang, Nano Lett., 2012, 12, 1538–1544.
13. K. F. Mak, K. He, J. Shan and T. F. Heinz, Nat. Nanotechnol., 2012, 7, 494–498.
14. H. Zeng, J. Dai, W. Yao, D. Xiao and X. Cui, Nat. Nanotechnol., 2012, 7, 490–493.
15. T. Cao, G. Wang, W. Han, H. Ye, C. Zhu, J. Shi, Q. Niu, P. Tan, E. Wang and B. Liu, Nat. Commun., 2012, 3, 887.
16. K. F. Mak, C. Lee, J. Hone, J. Shan and T. F. Heinz, Phys. Rev. Lett., 2010, 105, 136805.
17. S. Tongay, J. Zhou, C. Ataca, K. Lo, T. S. Matthews, J. Li, J. C. Grossman and J. Wu, Nano Lett., 2012, 12, 5576–5580.
18. D. Braga, I. Gutiérrez Lezama, H. Berger and A. F. Morpurgo, Nano Lett., 2012, 12, 5218–5223.
19. O. Lopez-Sanchez, D. Lembke, M. Kayci, A. Radenovic and A. Kis, Nat. Nanotechnol., 2013, 8, 497–501.
20. B. W. Baugher, H. O. Churchill, Y. Yang and P. Jarillo-Herrero, Nat. Nanotechnol., 2014, 9, 262–267.
21. M. Chhowalla, D. Voiry, J. Yang, H. S. Shin and K. P. Loh, MRS Bull., 2015, 40, 585–591.
22. Y.-C. Lin, D. O. Dumcenco, Y.-S. Huang and K. Suenaga, Nat. Nanotechnol., 2014, 9, 391–396.
23. R. Kappera, D. Voiry, S. E. Yalcin, W. Jen, M. Acerce, S. Torrel, B. Branch, S. Lei, W. Chen and S. Najmaei, APL Mater., 2014, 2, 092516.
24. H. Zhang, L.-M. Liu and W.-M. Lau, J. Mater. Chem. A, 2013, 1, 10821–10828.
25. X. Tao and Y. Gu, Nano Lett., 2013, 13, 3501–3505.
26. J. Camassel, P. Merle, H. Mathieu and A. Chevy, Phys. Rev. B: Condens. Matter Mater. Phys., 1978, 17, 4718.
27. E. Bringuier, A. Bourdon, N. Piccioli and A. Chevy, Phys. Rev. B: Condens. Matter Mater. Phys., 1994, 49, 16971.
28. A. Segura, J. Bouvier, M. Andrés, F. Manjón and V. Munoz, Phys. Rev. B: Condens. Matter Mater. Phys., 1997, 56, 4075.
29. F. Manjón, D. Errandonea, A. Segura, V. Munoz, G. Tobías, P. Ordejón and E. Canadell, Phys. Rev. B: Condens. Matter Mater. Phys., 2001, 63, 125330.
30. D. Errandonea, A. Segura, F. Manjón, A. Chevy, E. Machado, G. Tobias, P. Ordejón and E. Canadell, Phys. Rev. B: Condens. Matter Mater. Phys., 2005, 71, 125206.
31. C. Julien and M. Balkanski, Appl. Surf. Sci., 1991, 48, 1–11.
32. J. Martínez-Pastor, A. Segura, J. Valdes and A. Chevy, J. Appl. Phys., 1987, 62, 1477–1483.
33. M. Camara, A. Mauger and I. Devos, Phys. Rev. B: Condens. Matter Mater. Phys., 2002, 65, 125206.
34. D. Olguín, A. Rubio-Ponce and A. Cantarero, Eur. Phys. J. B, 2013, 86, 1–9.
35. G. W. Mudd, S. A. Svatek, T. Ren, A. Patanè, O. Makarovsky, L. Eaves, P. H. Beton, Z. D. Kovalyuk, G. V. Lashkarev and Z. R. Kudrynskyi, Adv. Mater., 2013, 25, 5714–5718.
36. S. Lei, L. Ge, S. Najmaei, A. George, R. Kappera, J. Lou, M. Chhowalla, H. Yamaguchi, G. Gupta and R. Vajtai, ACS Nano, 2014, 8, 1263–1272.
37. C. Ulrich, M. Mroginski, A. Goñi, A. Cantarero, U. Schwarz, V. Muñoz and K. Syassen, Phys. Status Solidi B, 1996, 198, 121–127.
38. S. Shigetomi and T. Ikari, Jpn. J. Appl. Phys., 2000, 39, 1184.
39. R. Sreekumar, T. Sajeesh, T. Abe, Y. Kashiwaba, C. Sudha Kartha and K. Vijayakumar, Phys. Status Solidi B, 2013, 250, 95–102.
40. S. Marsillac, A. Combot-Marie, J. Bernede and A. Conan, Thin Solid Films, 1996, 288, 14–20.
41. I.-H. Choi and P. Y. Yu, J. Appl. Phys., 2003, 93, 4673–4677.
42. X. Tan, J. Zhou and Q. Yang, CrystEngComm, 2011, 13, 2792–2798.
43. R. Sreekumar, R. Jayakrishnan, C. S. Kartha, K. Vijayakumar, S. Khan and D. Avasthi, J. Appl. Phys., 2008, 103, 023709.
44. Y. Igasaki and T. Fujiwara, J. Cryst. Growth, 1996, 158, 268–275.
45. A. Ates, M. Kundakci, A. Astam and M. Yildirim, Phys. E, 2008, 40, 2709–2713.
46. C. Viswanathan, G. Rusu, S. Gopal, D. Mangalaraj and S. K. Narayandass, J. Optoelectron. Adv. Mater., 2005, 7, 705–711.
47. R. Sreekumar, R. Jayakrishnan, C. S. Kartha, K. Vijayakumar, Y. Kashibawa and T. Abe, Sol. Energy Mater. Sol. Cells, 2006, 90, 2908–2917.
48. P. Dong, Y. Wang, H. Li, H. Li, X. Ma and L. Han, J. Mater. Chem. A, 2013, 1, 4651–4656.
49. K. Keong, W. Sha and S. Malinov, J. Alloys Compd., 2002, 334, 192–199.
50. B.-W. Park, B. Philippe, T. R. Gustafsson, K. R. Sveinbjörnsson, A. Hagfeldt, E. M. Johansson and G. Boschloo, Chem. Mater., 2014, 26, 4466–4471.
51. C. Carlone, S. Jandl and H. Shanks, Phys. Status Solidi B, 1981, 103, 123–130.
52. C.-H. Ho, Y.-C. Chen and C.-C. Pan, J. Appl. Phys., 2014, 115, 033501.
53. C.-H. Ho, Sci. Rep., 2014, 4, 4764.
54. R. Schwarcz, M. Kanehisa, M. Jouanne, J. Morhange and M. Eddrief, J. Phys.: Condens. Matter, 2002, 14, 967.
55. J. F. Sánchez-Royo, G. Muñoz-Matutano, M. Brotons-Gisbert, J. P. Martínez-Pastor, A. Segura, A. Cantarero, R. Mata, J. Canet-Ferrer, G. Tobias and E. Canadell, Nano Res., 2014, 7, 1556–1568.
56. M. Parlak and C. Ercelebi, Thin Solid Films, 1998, 322, 334–339.
57. C. Rincón, S. Wasim, G. Marın, R. Márquez, L. Nieves, G. S. Pérez and E. Medina, J. Appl. Phys., 2001, 90, 4423–4428.
58. J. I. Pankove, Optical processes in semiconductors, Courier Corporation, 2012.
59. S. Yang and D. F. Kelley, J. Phys. Chem. B, 2005, 109, 12701–12709.
60. S. Yang and D. F. Kelley, J. Phys. Chem. B, 2006, 110, 13430–13435.
61. G. Mudd, A. Patanè, Z. Kudrynskyi, M. Fay, O. Makarovsky, L. Eaves, Z. Kovalyuk, V. Zólyomi and V. Falko, Appl. Phys. Lett., 2014, 105, 221909.
62. S. Lei, L. Ge, Z. Liu, S. Najmaei, G. Shi, G. You, J. Lou, R. Vajtai and P. M. Ajayan, Nano Lett., 2013, 13, 2777–2781.
63. P. Hu, Z. Wen, L. Wang, P. Tan and K. Xiao, ACS Nano, 2012, 6, 5988–5994.

---

Footnote

† Electronic supplementary information (ESI) available: AFM images and corresponding height profiles; intensity ratios of I_{(0 0 6)} of γ-In₂Se₃ to the other peaks of β-InSe; Raman modes of annealed sheets; intensity of A′₁ mode; EDS spectrum; In/Se ratio; PL spectra and peak intensity of γ-In₂Se₃; absorption and transmission spectra; intensity ratio of I_γ/I_β; optical image of InSe photodetector; photocurrents of as-exfoliated and annealed devices as a function of wavelength; photodetector parameters of as-exfoliated and annealed sheet. See DOI: 10.1039/c6ra13543a

---