Introduction

A system that generates a high photocurrent in response to light may be used as a photosensor. The light penetration distance is very short; thus, the photoresponse properties are dominated by the carriers near the material surface. A material with a relatively high surface carrier dominance can be expected to perform as a relatively efficient photodetector. To optimize the photoresponse, various types of nanostructured materials, with high surface-to-volume ratios and high levels of photoresponse, were investigated1,2,3,4,5,6. Recently, two-dimensional materials such as graphene7,8, graphene-based heterostructures1,2,3,4, and two-dimensional transitional metal dichalcogenides (TMDs) have attracted noteworthy attention9,10,11,12,13,14,15,16. These two-dimensional materials demonstrate excellent photoelectrical performance because they have high surface-to-volume ratios and abundant surface carriers.

Three-dimensional topological insulators are promising materials because they offer insulating bulk states and a gapless conducting surface state. These insulators have a surface state that is topologically protected by a time reversal symmetry, which is induced by a strong spin-orbit interaction. This remarkable surface state has garnered intensive theoretical and experimental attention and had been a recent research topic17,18. The linear dispersions in the surface state and the extremely high carrier mobility levels make these insulators promising candidates for optical electrical devices19,20. The photoelectrical characteristics of the Bi-based topological insulators have been investigated and have revealed promising responses21,22. It is reported that the Bi2Te3 topological insulator based heterostructures1,23,24 and PLD-grown Bi films25 reveal ultrahigh responsivity in wide wave range. Recently, it was reported that Sb2Te3 thin films offer higher photoelectrical responses than that in Bi-based topological insulators26.

In this paper, we report on the photocurrent produced by a 532-nm wavelength in a Sb2SeTe2 topological insulator. The experimental results reveal extremely high performance; specifically, the responsivity and the photoconductive gain reached 2293 AW−1 and 5344 at a bias of 1 V. These observations are orders of magnitude higher than most reported values in other topological insulators and two-dimensional TMDs, which suggests that Sb2SeTe2 nanoflakes have great potential for future optoelectronic device applications.

Experimental method

Single crystals of Sb2SeTe2 were grown by a homemade resistance-heated floating zone furnace (RHFZ). The starting raw materials of Sb2SeTe2 were mixed according to the stoichiometric ratio. At first, the stoichiometric mixtures of high purity elements Sb (99.995%), Se (99.995%) and Te (99.995%) were melted at temperatures of 700 ~800 °C for 20 h, and then slowly cooled to room temperature in an evacuated quartz glass tube. The resulting material was used as a feeding rod for the following RHFZ experiment. After growth, the crystals were then furnace cooled to room temperature. The as-grown crystals were cleaved along the basal plane, producing a silvery shining mirror-like surface, and then prepared for the further experiments. The Raman27, EDS and XPS28 spectrum support that the crystal is Sb2SeTe2.

The Sb2SeTe2 nanoflakes were obtained by exfoliating bulk crystals using dicing tape and were then dispersed on the insulating SiO2 (300 nm)/n-Si templates with pre-patterned Ti/Au circuits. Two platinum (Pt) metal contacts were subsequently deposited on the selected Sb2SeTe2 nanoflakes using focused-ion beam (FIB) technique (shown in the right-bottom inset of Fig. 1). The thickness of a nanoflake is determined by the atomic force microscopy; here, the nanoflake was 181-nm thick, 708-nm long, and 1667-nm wide. The current-voltage characteristic reveals a linear dependence that indicates the ohmic contacts in the sample; the conductivity was approximately 33.7 S/cm. The left-top inset within Fig. 1 shows the X-ray diffraction of the Sb2SeTe2; the sharp peaks indicate that the Sb2SeTe2 crystal has high crystallinity. Our previous works show that the physical parameters extracted from XPS, Raman spectrum, ARPES and the quantum SdH oscillation are consistent. That supports the Sb2SeTe2 crystal reveals high quality and uniformity. Figure 2 presents the schematic of the Sb2SeTe2 nanoflake device, illustrating the photoelectrical measurement setup and the light that illuminated it. The wavelength of the illuminating light was 532 nm.

Figure 1: Left-top inset shows the XRD of the Sb2SeTe2.
figure 1

The sharp peaks indicate the high crystallinity of the Sb2SeTe2 crystal. The right-bottom inset shows a SEM picture of the Sb2SeTe2 nanoflake. The red line is the scale bar of SEM. The blue curve is the AFM thickness profile and the sample thickness is 181 nm. Two Pt contacts were deposited on the nanoflake to measure the photocurrent. The linear current-voltage curve indicates the ohmic contact between the Pt electrodes and the Sb2SeTe2 nanoflake.

Figure 2: Schematic of the Sb2SeTe2 nanoflake device illustrating the photoelectrical measurement setup and the light that illuminated it.
figure 2

The wavelength of the light was 532 nm.

Results and Discussion

The inset of the Fig. 3 shows that the measured current of our Sb2SeTe2 nanoflake under light illumination with light power that ranges from 1 to 50 mW that is corresponding to the power intensity of 40 to 2000 Wm−2. It clearly indicates that the current increases with increasing light power. The overall response time is approximately 10 s; which is shorter than the reported values of Sb2Te3 films26, but longer then the values of Bi-based topological insulators21,22. Here the photocurrent is presented as a function of the power intensity at bias of 0.1 V (Fig. 3). For quantitative analysis, the relationship between the photocurrent and the light intensity can be fitted to the simple power law relation, Ip = APθ, where the A is a constant for the wavelength of the illuminating light, P is the power intensity of the light that illuminates the device, and θ is a constant related the photosensitivity of the device. As Fig. 3 reveals, the experimental data agrees with the power law relation and the fitting result gives a θ of 0.85.

Figure 3: Left-top inset depicts the photocurrents produced by illumination at different power levels.
figure 3

The measured photocurrent is a function of light power intensity, and can be accurately described by a simple power law relation.

To quantitatively determine the performance of the Sb2SeTe2 nanoflake under illumination, responsivity, R, and the photoconductive gain, G, are calculated through the following equations;

where IP, P, S, h, c, e, η and λ are the photocurrent, the light intensity, the effective area, Planck’s constant, the velocity of light, the charge of an electron, the quantum efficiency (for convenience, we assume η = 1) and the wavelength, respectively. The G for a wavelength is proportional to the R at the same wavelength. Figure 4 depicts R and G as functions of the light intensity at a constant bias of 0.1 V, and reveals that the R and G decrease as the power intensity increases. Specifically, the R and G are 276 AW−1 and 643 at a power intensity of 120 Wm−2.

Figure 4: Responsivity and photoconductive gain as functions of the light power intensity at a wavelength of 532 nm.
figure 4

Both responsivity and photoconductive gain increase as the light power intensity decreases.

The photocurrent is strongly related to the applied bias. To comprehensively investigate the intrinsic optoelectronic characteristics in the Sb2SeTe2 nanoflake, an experiment on bias-dependent photocurrents was performed. As shown in the inset of the Fig. 5, the photocurrent was linearly related to the applied bias; specifically, the observed photocurrent was approximately 0.8 μA at 1 V and a light power intensity of 280 Wm−2. This linear bias-dependent increment of the photocurrent can be attributed to the increment drift velocity and the reduced carrier transit time caused by applied bias. Expressed as , T is the carrier transit time, l is the device length, μ is the carrier mobility, and Vsd is the applied bias. This indicates a system with higher carrier mobility; and, a higher bias might decrease the carrier transit time, and produce a higher photocurrent. Figure 5 also indicates that the evaluated R and G are functions of bias, to which both linearly relate. At 1 V, The R and G reach 2293 AW−1 and 5344, respectively.

Figure 5: Left-top inset shows the linear relation of a photocurrent to the applied bias at a wavelength of 532 nm and a power intensity of 280 W/m2.
figure 5

The main figure shows the responsivity and the photoconductive gain as functions of the applied bias at a wavelength of 532 nm and the power intensity of 280 W/m2.

To qualitatively identify the optical performance of the Sb2SeTe2 nanoflake, the reported values were collected. Table 1 presents a list of the reported R and G values for topological insulators and two-dimensional TMDs, and clearly reveals that the R and G values for our Sb2SeTe2 are orders of magnitude higher than most the reported values in topological insulators and two-dimensional TMDs under similar conditions. That suggests that the Sb2SeTe2 has the potential to deliver extremely high-performance photocurrent-related applications.

Table 1 List of the reported responsivity and gain values of photocurrents in topological insulators and two-dimensional transition metal dichalcogenides.

Aside from the high quality of the crystalline sample and the large surface-to-volume ratio, several possible causes might lead to this extremely high photoresponse. First, the photoresponse is extremely sensitive to the condition of sample surface. In addition to the reduction of the effective response area, surface defects and oxidation reduce carrier mobility and life time. One prior study reported that adsorbed molecules on a surface reduce the carrier’s life time; thus, the photoresponse of a material in a vacuum is higher than the photoresponse of the same material in the air29. Our previous work revealed that the surface state carrier transport characteristics in our Sb2SeTe2 topological insulator can tolerant surface oxidation and molecules adsorbed on the sample’s surface; such molecules might come from unavoidable pollution during the fabrication process or from sample transference28. Therefore, less effective defective materials might impair the surface electron transport properties of our Sb2SeTe2 sheet, and the proposed nanoflake might be very effectively by comparison. Second, in addition to the artificial and extrinsic factors, R and G values are directly related to carrier mobility. The reported R and G in MoS2 and WSe2 flakes were positively related to the field-effect carrier mobility30. The surface state carrier mobility of our Sb2SeTe2 topological insulator was approximately 55.5 cm2V−1s−1 at room temperature31; that is one order larger than the previously reported value (4 cm2V−1s) for a single-layer MoS2 flake30. Third, it is noteworthy that previous works have revealed that graphene-based heterostructure greatly enhances photoresponse because electron have high mobility in graphene and two-dimensional TMDs demonstrate enhanced adsorption ratios. The R and G values of our Sb2SeTe2 are orders of magnitude higher than most reported values in topological insulators and two-dimensional TMDs, and are only lower than the reported values in the nanowires5,6,32. and graphene-MoS2 hybrid structure1. Theoretical calculation shows that the surface state Dirac point lies at the energy gap of the bulk state in Sb2SeTe2, and our previous work supported that the Fermi level is below the Dirac point. This energy band structure is similar to the graphene-MoS2 hybrid structure and might lead to the observed high photoresponse.

Detectivity, that is an important figure-of-merit in evaluating the ability of a photodetector to detect weak signal, is another important indices used to characterize the performance of photodetectors33. The specific detectivity (D*) is calculated through the relation:

where R, S, q, and Id are the responsivity, effective area of light illumination, electronic charge, and dark current. By using the experimental data, the detectivity is determined to be 4.5 × 108 Jones.

Conclusion

A photocurrent experiment was performed in a Sb2SeTe2 topological insulator nanoflake at a wavelength of 532 nm. It exhibited extremely high performance; the responsivity and the photoconductive gain were 2293 AW−1 and 5344 at 1 V, respectively. This high photoresponse was orders of magnitude higher than most reported values in topological insulators and two-dimensional TMDs. This finding suggests that the Sb2SeTe2 nanoflake has remarkable potential for future optoelectronic device applications.

Additional Information

How to cite this article: Huang, S.-M. et al. Extremely high-performance visible light photodetector in the Sb2SeTe2 nanoflake. Sci. Rep. 7, 45413; doi: 10.1038/srep45413 (2017).

Publisher's note: Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.