Open Access
Add to BibSonomy
Abstract
Titanium trisulfide (TiS3) is an important representative of nanomaterials, which are regarded as significant candidates for nanoscale optoelectronics. Due to the promising high intrinsic charge-carrier mobility, it has been proposed as a high performance visible and near-infrared-resistive photodetector with excellent stability. However, the conventional diagnostics of such materials require specific sample preparation. Here, we present the effective utility of the optical pump–terahertz probe technique for the additional photodetector characterization, which is deprived of these shortcomings. In particular, we non-invasively determined the lifetime of charge carriers of 80 ps at a pump fluence F = 1.6 μJcm−2. The responsivity, which is defined in our research, was R = Δσ/F = 84 × 104 m2/ΩJ. The quantum efficiency Q.E. = Ne/Nph = 3 × 104. These characteristics confirm that TiS3 nanoribbons are a highly desirable material for optoelectronic devices.
© 2024 Optica Publishing Group under the terms of the Optica Open Access Publishing Agreement
1. Introduction
The transition metal trichalcogenides (TMT) are a very rich family of materials, which includes many layered structures with intriguing properties. Being intensively studied because of the increased scientific attention titanium trisulfide (TiS$_{3}$) reveals strong anisotropy, high carrier mobility and pronounced photo-response [1–4]. These remarkable features make TiS$_{3}$ a promising candidate for electronic and optoelectronic applications. In particular, in our previous paper, we have shown that TiS$_{3}$ nanoribbons might be considered as an auspicious material for photodetection [5].
TiS$_{3}$ nanoribbons perfectly fit for the optoelectronic applications as it is an n-doped semiconductor with a bandgap energy of approximately 1.1 eV, which can be fine-tuned by geometrical structure [6]. Such a narrow band gap allows TiS$_{3}$ nanoribbons efficiently absorb light, which is advantageous for the potential applications. It is accompanied by the relatively high electrical mobility of 25–80 cm$^2$V$^{-1}$s$^{-1}$ obtained for a single nanoribbon [3]. Moreover, as a quasi one-dimensional material, it possesses giant anisotropy of the electronic coefficients resulting from the backscattering of charge carriers [7]. This property might improve the nanotechnological device.
Recently, it was shown that the photodetectors based on TiS$_{3}$ nanoribbons demonstrate a quick response (with the response-to-recovery time of approximately 0.093 s) at a variety of temperatures [5]. This photodector demonstrated high photocurrent (6.95 $\times$ 10$^{-5}$ A) which is accompanied by high photoresponsivity (1.624 $\times$ 10$^{8}$ AW$^{-1}$), Q.E. (3.3 $\times$ 10$^{8}$ AW$^{-5}$), and detectivity (4.328 $\times$ 10$^{15}$ Jones) under 617 nm continuous light excitation at a low intensity of about 1.0 $\times$ 10$^{-5}$ Wcm$^{-2}$.
The fatigue test of these devices is still a challenging task due to the accurate positioning of nanomaterials on electrodes produced by complex high-precision lithography methods. In addition, the traditional dc-measurements do not allow to track the charge carrier dynamics, which plays a crucial role in the understanding of the charge separation process and transport efficiency. From this point of view, optical pump-terahertz probe (OPTP) spectroscopy is a powerful method, which allows to overcome these obstacles. It is a non-contact nondestructive electrical probe technique, which capable of measuring the transient photoconductivity of numerous nanoscale semiconductors [8–11].
In this work, we used OPTP to study the photoconductivity under 800 nm fs-duration laser excitation. It allows us to complete the performance characterisation of TiS$_{3}$ nanoribbons photodetectors. The responsivity, which is determined as $R=\Delta \sigma /\Delta F$, where $\sigma$ is the photoconductivity and $F$ is the fluence, has been found to be 84$\times$10$^{4}$ m$^{2}$/$\Omega J$. The quantum efficiency (Q.E.) calculated as $N_{e}/N_{ph}$, where $N_{e}$ is the number of carriers involved in conductivity and $N_{ph}$ the number of incident photons, was 3$\times$10$^{4}$. Finally, we obtained the charge carrier lifetime of 80 ps at 1.6 $\mu$Jcm$^{-2}$ pump fluence. These characteristics together with previous photodetector measurements make TiS$_{3}$ nanoribbons highly desirable for high-gain photodetectors.
2. Methods and sample fabrication
TiS$_{3}$ nanoribbons were synthesized by a solid-gas reaction between titanium powder (2 g, good fellow, 99.5 % purity) and sulfur gas produced by heating of sulfur powder (6 g, Merck, 99.9 % purity) [12]. This reaction was carried out in a vacuum-sealed ampule to remove impurities, such as atmospheric oxygen and water, and prevent oxidation of titanium during heating. Prior to this reaction, the first Ti disks were etched in an acidic mixture of hydrofluoric acid and HNO$_{3}$ (4:30 wt %), which cleans the surface of Ti disks and removes any oxide impurities. Ampules were then heated at a rate of 250 $^{\circ }$C per h and kept at 500 $^{\circ }$C for 24 h (temperature and time of sulfuration have been previously optimized) in a horizontal furnace [4]. After approximately 20 h of growth process, the ampule is cooled in ambient condition; TiS$_3$ nanoribbons were then obtained from the ampule.
The obtained powder was dispersed in isopropanol for 10 mins using tip-sonication. The solution was spray-coated on z-cut quartz substrate. The structural, morphological as well as optical properties of the fabricated samples were examined by transmission electron microscopy (TEM), scanning electron microscopy (SEM), Raman spectroscopy and UV–Vis spectroscopy techniques.
The arrangement properties and size distributions of TiS$_{3}$ were studied by the high-resolution TEM (HR-TEM) system (JEOL JEM2100) operating at 200 kV with a Gatan Multiscan CCD in imaging modes. The structural morphology of the deposited NT films was visualized via scanning electron microscopy (SEM, JSM7001F, JEOL, Tokyo, Japan) working in the secondary electron imaging mode with a Schottky emitter, an acceleration voltage of 10 kV and a working distance of about 4.7 mm.
Absorbance spectra were obtained using a Cary UV/vis/NIR spectrophotometer (250–800 nm wavelength) working in a transmission mode and calculated as $A=-\log _{10}(T)$, where $T$ is the transmission. Raman and PL spectra were obtained using a confocal microspectrometer (Labram, Jobin-Yvon Horiba) with a 532 nm laser wavelength (with a spectral resolution of 0.5 cm$^{-1}$).
The photoconductivity measurement was performed using the in-house OPTP setup, the key element of which is an amplified Ti:Sapphire laser (Spectra Physics Spitfire) (Fig. 1). In our OPTP, terahertz pulsed radiation was produced by phase-matched optical rectification of femtosecond laser pulses in lithium niobate crystal [13]. Terahertz radiation was detected via electro-optic sampling in a <110> ZnTe crystal [14]. The sample was excited by 40 fs, 1 kHz repetition rate fundamental laser pulses at 800 nm. This OPTP system is quite similar to those detailed in Refs [8,10].
Fig. 1. The scheme for OPTP. The femtosecond beam is divided by a set of beamsplitters into three paths: the THz generation path (via tilted pulse front excitation scheme based on LiNbO$_{3}$), the gate path (for detection with electro-optic sampling in a <110> ZnTe crystal) and pump excitation path (at fundamental harmonic of Ti:Sa laser).
3. Preliminary microscopy and optical characterization
The low and high magnification SEM images revealed the TiS$_{3}$ "belt-like" structures, which form the percolated network (Fig. 2(a-c)). TEM images (Fig. 2(d)) confirm that the formation of TiS$_{3}$ nanoribbons has been obtained with lengths of several tens of micrometers and an average width of approximately 80 nm with high quality crystallinity.
Fig. 2. (a)-(c) High to low magnification SEM images of TiS$_{3}$ nanoribbons. (d) Bright field TEM microimage of one individual nanoribbon. (e) Averaged Raman spectra of the TiS$_{3}$ nanoribbons measured using excitation wavelength of 532 nm. (f) Optical absorption spectra of TiS$_{3}$ nanoribbons.
Under 532 nm excitation TiS$_{3}$ nanoribbons show the Raman modes at 172 cm$^{-1}$, 300 cm$^{-1}$, 369 cm$^{-1}$ and 558 cm$^{-1}$, which correspond to the I-A$_{g}$ rigid, II-A$_{g}$ internal, III-A$_{g}$ internal and A$_{g}$ S–S types of vibration (see Fig. 2(e)). The mode at 175 cm$^{-1}$ is a signature of out-of-phase oscillations of the TiS$_{3}$ chains along b-direction. The peaks at 300 cm$^{-1}$ and 369 cm$^{-1}$ arise from internal vibrations of individual layers of nanoribbons. Finally, the peak at 558 cm$^{-1}$ is an in-plane out-of-phase vibration of S–S molecules as reported in [6]. It is known that the difference between I-peak and III-peak positions determine the number of layers in the TiS$_{3}$ nanostructures [15]. We obtained on average a 197 cm$^{-1}$ frequency difference between the peak modes, which corresponds to a 6-layer structure.
Figure 2(f) shows the absorption spectra of TiS$_{3}$ nanoribbons. Broadband absorption peaks appeared due to the superposition of several interband transitions overlapping with the pronounced background. In further photoconductivity measurement, we used 800 nm excitation, which is located above the interband absorption lines of TiS$_{3}$ nanoribbons.
4. OPTP measurements
We used OPTP spectroscopy to determine the main photovoltaic characteristics of TiS$_{3}$ nanoribbons for their implementation as photodetectors.
The photoconductivity transient was determined by monitoring photoconductivity at the peak position of the transmitted terahertz pulse as a function of pump-probe delay time. The dynamics show the fast almost resolution-limited rise followed by mono-exponential decay evident from the log-scale $y$-axis (Fig. 3(a)). It originates from the recombination of the electrons in the conduction band with holes in the valence band. The absence of the fast ($\leqslant$10 ps) exponential decay indicates the low number of defects involved in the recombination dynamics. We determined the lifetime of photoexcited charge carriers. It varies from 50 to 80 ps with the increase of pump fluence from 0.1 to 0.56 $\mu$Jcm$^{-2}$ with no obvious saturation in this fluence range (Fig. 3(b)).
Fig. 3. (a) The photoconductivity transients of TiS$_{3}$ nanoribbons for different excitation fluences. (b) The lifetime extracted from the monoexponentional fit (see text) vs excitation fluence. (c) Real (filled) and imaginary (empty) parts of photoconductivity spectra at about 3 ps after photoexcitation. The lines are fits to the data in the frames of the localized surface plasmon model as described in the main text.
To obtain the number of charges and mobility, we measured frequency-resolved photoconductivity at different fluences and pump-probe delay times. In all experiments, the photoconductivity shows a peak in the real part and a consequent zero-crossing of the imaginary part, which is a signature of localized surface plasmons (LSP) in nanoribbons (Fig. 3(c)). The photoconductivity of TiS$_{3}$ nanoribbons is well-fitted within the plasmon model [8,9]:
The electron mobility can be estimated as
The resonance frequency of the LSP depends on the geometrical factor $g$ and the charge-carrier concentration:
$\epsilon _{0}$ is the vacuum permittivity, $\epsilon _{w}$ is the relative permittivity of the material.In Figs. 3(c), 4(a-c) theoretical solid lines fit the measured data (points) at different pump fluences and pump-probe delays. The momentum scattering rate for the highest fluence $\gamma$=1.76 THz allowed us to estimate the mobility $\mu =2500$ cm$^{2}$V$^{-1}$s$^{-1}$ using the effective masses $m_\mathrm {eff}=0.369 m_\mathrm {e}$ reported for 6-layer structures [16]).
Fig. 4. Real (filled points) and imaginary (empty points) parts of the photoconductivity spectra at $F=2.2\;\mu$Jcm$^{-2}$ at 2.66, 9.33 and 22.7 ps after photoexcitation. The lines are fits to the data in the frame of the LSP model as described in the main text.
5. Photodetection performance
Commonly, photodetectors are characterized by several key performance parameters such as Q.E., responsivity $R$, E$_{ON}$ to E$_{OFF}$ signal ratio, the photoconductive gain [17]. The Q.E. is usually defined as Q.E.=$N_{e}/N_{ph}$, where $N_{e}$ is the number of carriers involved in the conduction, $N_{ph}$ the number of the incident photons. Using $N_{e}$, which was obtained by the fit of the photoconductivity with Eq. (1) and assuming that the sample has the absorbance of 0.5 OD, we estimated $N_{ph}$ and $N_{e}$ at different fluences at 3 ps after photoexcitation (Fig. 5). With the increase of pump power, and therefore, fluence, the Q.E. tends to linearly increase with the maximum value of 3.2$\times$10$^{4}$.
Fig. 5. The quantum efficiency (a), responsivity (b) and ON/OFF ratio (c), the characters defined in the main text, as a function of pump fluence.
The responsivity $R$ is another key parameter, which is used to characterise a photodetector performance. Responsivity is conventionally determined as the ratio of photocurrents to the illuminated light power. Therefore, it determines the current change under certain light power illumination. Alternatively, in our experiment we determined responsivity as $R$ = $\Delta \sigma _{ph}/\Delta F$. In the measured fluence range, the photoconductivity scales almost linearly. Therefore, we determined the responsivity of 84$\times$10$^{4}$ m$^{2}$/$\Omega$J. To compare, we extracted the required data for carbon nanotubes from our another research [10], and responsivity appeared to be approximately 4 times lower than for TiS$_{3}$ nanoribbons
For the development of a photodetector, it is also important to determine the ON/OFF signal ratio before and after the light illumination of the device. In the traditional photodetector characterization techniques, it is determined as a ratio of the currents before and after excitation. In a similar way, we expressed the ON/OFF ratio as a ratio between the maximum of the THz pulse signal before the illumination with the femtosecond laser irradiation and after it. We observed typical for the traditional semiconductors reduction of the transmitted pulse, which corresponds to positive photoconductivity. The ON/OFF ratio changes from 0.99 to 0.94 in the measured fluence range.
Finally, the photoconductive gain ($G$) of a device is the important parameter, which determines the ability of photogenerated carries to circulate in the channel. The gain can be defined as the ratio of the light-generated charge carrier lifetime to the drift transit time, $G$ = $\tau _{lt}/\tau _{tt}$. The increase of the charge carrier lifetime is evident from Fig. 3(b). The photoconductive gain therefore can be estimated using the measured in the current experiment the charge carrier lifetime and the drift transit time, which can be determined in the conventional photoconductivity experiments.
6. Discussion
As a benchmark for our technique, we have measured the subpicosecond photo-generated carrier dynamics and frequency-dependent photoconductivity for TiS$_{3}$ nanoribbons. This material has already been known as a candidate for photodetection devices. OPTP allows to determine several important parameters of charge carriers: lifetime, mobility, photoconductivity, etc.; and identify the factors that can limit these parameters without the need for device fabrication. All these parameters can be restricted by the material geometries, grain boundaries, defects, etc., which can be revealed and engineered using OPTP as a diagnostic tool and microscopy tools for visualization. It is clear that the comparison of OPTP with DC measurements may give a valuable insight into the material characterization due to the several parameters, such as the exceptional time-resolution and local sampling. From some point of view, it can give more conclusive information about the photodetector performance. In addition, it can be accompanied by pressure changes, temperature effects, saturation effects, doping effects and information on certain resonant absorption, laser power and polarization. Moreover, many important characteristics can be revealed by femtosecond-picosecond-nanosecond multi-timescale excitation experiments.
Moving forward, the insights gained using OPTP spectroscopy for the photodetectors characterization can be adopted for different optoelectronic device characterisation. In particular, TiS$_{3}$ nanoribbons can be used in a wide range of optical-to-electrical transducers based on the extracted parameters. It is worthy noting that potentially it is important to compare OPTP and DC measurements and reveal the similarities and differences. In particular, OPTP shows rather localized time-dependent response from free charges, which is different from time-averaged long-distance responses in DC measurements. This difference must be taken into consideration if the direct comparison of the parameters is required.
Overall, the OPTP spectroscopy has proven to be a significant tool in understanding the photodetection properties of the emerging materials. The insights provided by this technique are unobtainable with other conventional methodologies. Besides, it is a non-contact method, and the absence of any specific sample preparation makes the OPTP setup essentially useful for the photodetector characterization.
7. Conclusions
In this paper we demonstrated that the optical pump - THz probe technique is a facile and powerful approach for the investigation of the photodetectors performance. This research can be regarded as an experimental simulation of the real conditions, which might be useful for the studies under continuous illumination of sunlight. We calculated Q.E., the responsivity, $E_{ON}/E_{OFF}$ for TiS$_{3}$ nanoribbons for the first time. Overall, we propose that the OPTP technique may form an important complementary approach to other conventional techniques, which are used to investigate the performance of photodetectors.
Funding
Ministry of Science and Higher Education of the Russian Federation (075-15-2020-790).
Acknowledgements
This work is supported by the Ministry of Science and Higher Education of the Russian Federation, Project ${\%}$ 075-15-2020-790. TEM analyses were performed using the equipment of the Center of Shared Research Facilities (MIPT).
Disclosures
The authors declare no conflicts of interest.
Data availability
Data underlying the results presented in this paper are not publicly available at this time but may be obtained from the authors upon reasonable request.
References
1. J. Dai and X. C. Zeng, “Titanium trisulfide monolayer: theoretical prediction of a new direct-gap semiconductor with high and anisotropic carrier mobility,” Angew. Chem. Int. Ed. 54(26), 7572–7576 (2015). [CrossRef]
2. K. Wu, E. Torun, and H. Sahin, “Unusual lattice vibration characteristics in whiskers of the pseudo-one-dimensional titanium trisulfide TiS3,” Nat. Commun. 7(1), 12952 (2016). [CrossRef]
3. J. O. Island, R. Biele, and M. Barawi, “Titanium trisulfide (TiS3 ): a 2d semiconductor with quasi-1d optical and electronic properties,” Sci. Rep. 6(1), 22214 (2016). [CrossRef]
4. M. Talib, R. Tabassum, and Abid, “Improvements in the performance of a visible–NIR photodetector using horizontally aligned TiS3,” ACS Omega 4(4), 6180–6191 (2019). [CrossRef]
5. M. Talib, N. Tripathi, and S. Manzoor, “TiS3 nanoribbons: a novel material for ultra-sensitive photodetection across extreme temperature ranges,” Sensors 23(10), 4948 (2023). [CrossRef]
6. N. Tripathi, V. Pavelyev, and P. Sharma, “Review of titanium trisulfide (TiS3 ): A novel material for next generation electronic and optical devices,” Mater. Sci. Semicond. Process. 127, 105699 (2021). [CrossRef]
7. S. Liu, W. Xiao, and M. Zhong, “Highly polarization sensitive photodetectors based on quasi-1d titanium trisulfide (TiS3),” Nanotechnology 29(18), 184002 (2018). [CrossRef]
8. J. Lloyd-Hughes and T.-I. Jeon, “A review of the terahertz conductivity of bulk and nano-materials,” J. Infrared, Millimeter, Terahertz Waves 33(9), 871–925 (2012). [CrossRef]
9. H. J. Joyce, J. L. Boland, and C. L. Davies, “A review of the electrical properties of semiconductor nanowires: insights gained from terahertz conductivity spectroscopy,” Semicond. Sci. Technol. 31(10), 103003 (2016). [CrossRef]
10. M. I. Paukov, V. V. Starchenko, and D. V. Krasnikov, “Ultrafast optomechanical terahertz modulators based on stretchable carbon nanotube thin films,” Ultrafast Sci. 3, 0021 (2023). [CrossRef]
11. M. G. Burdanova, A. P. Tsapenko, and M. V. Kharlamova, “A review of the terahertz conductivity and photoconductivity of carbon nanotubes and heteronanotubes,” Adv. Opt. Mater. 9(24), 2101042 (2021). [CrossRef]
12. J. Singh, P. Sharma, and N. Tripathi, “Synthesis of highly sensitive nanomaterial for ultra-fast photocatalytic activity: a detailed study on photocatalytic capabilities of rod-shaped TiS3 nanostructures,” Catal. Commun. 162, 106381 (2022). [CrossRef]
13. Y.-S. Lee, N. Amer, and W. C. Hurlbut, “Terahertz pulse shaping via optical rectification in poled lithium niobate,” Appl. Phys. Lett. 82(2), 170–172 (2003). [CrossRef]
14. B. Pradarutti, G. Matthäus, and C. Brückner, “Electrooptical sampling of ultra-short THz pulses by fs-laser pulses at 1060 nm,” Appl. Phys. B 85(1), 59–62 (2006). [CrossRef]
15. A. Lipatov, M. J. Loes, and H. Lu, “Quasi-1d TiS3 nanoribbons: mechanical exfoliation and thickness-dependent Raman spectroscopy,” ACS Nano 12(12), 12713–12720 (2018). [CrossRef]
16. J. Kang, H. Sahin, and H. D. Ozaydin, “TiS3 nanoribbons: width-independent band gap and strain-tunable electronic properties,” Phys. Rev. B 92(7), 075413 (2015). [CrossRef]
17. M. Malik, M. A. Iqbal, J. R. Choi, et al., “2d materials for efficient photodetection: overview, mechanisms, performance and UV-IR range applications,” Front. Chem. 10, 1 (2022). [CrossRef]
