1.

Introduction

Recent advances in two-dimensional (2-D) semiconductors, particularly monolayer transition-metal dichalcogenides (TMDs), have attracted rapidly growing scientific interests for next-generation electrical and optoelectronic device applications.^1–11 A single layer of TMDs exhibits a totally new phenomenon, which can lead to 2-D nanoelectronic and nanophotonic applications owing to its unique physical, electrical, and optical properties.^12–15 Chemical vaporized deposition method was used to synthesize monolayer TMDs^16–19 mainly; however, magnetron sputtering technique can be used to produce large thin layered TMDs films or even monolayer TMDs.²⁰ Simply, a pulsed laser deposition (PLD) method was used.^21,22 However, the size of most large stable monolayer TMDs is of several hundred micrometers scales.^1,2 It is much easier to have a wafer size thin layer ${\mathrm{WS}}_2$ film (tens of nm thickness or over), especially in optoelectronic device applications; it is necessary to investigate the optical natures of thin layered TMDs film.

In this study, a new photoelectric material is reported, which is a few layer ${\mathrm{WS}}_2$ film on an n-doped SiC substrate. The 15-nm-thick ${\mathrm{WS}}_2/\mathrm{SiC}$ film and 150-nm-thick ${\mathrm{WS}}_2/\mathrm{SiC}$ film were superimposed by ${\mathrm{WS}}_2-{\mathrm{WS}}_2$ face to face together to form a film stack, and both n-doped SiC substrates on each side was used as electric electrode. For this ${\mathrm{WS}}_2/\mathrm{SiC}$ film stack, new optical properties about this material are discovered as follows: photovoltaic effect is related to a cutoff wavelength of the incident monochromatic light or laser, and the power of laser, as well as the polarization of the laser electric field. It is found that the maximum open circuit voltage output is 6.3 V with the 40 mW of a 532-nm linearly polarized laser at the present experiment.

The second new phenomenon concerning ${\mathrm{WS}}_2/\mathrm{SiC}$ material is the wavelength of the laser beam passing through these film stacks was blueshifted, when an external electric field was applied perpendicularly. It was well known that the laser frequency does not change any more when it passes through most of the transparent medium. However, in this paper, a tunable blueshift of laser wavelength by an electrically driven thin layered ${\mathrm{WS}}_2$ film was reported for the first time, and a tunable blueshift by changing the driven DC voltage is also demonstrated.

To explore the mechanism of blueshift of the laser wavelength when the laser is passing through the external field driven film stack, the inverse Compton scattering model of photon by both electron and hole was used to explain this photon energy gain. Both theoretical and experimental data matches well.

The article is organized as follows. In Sec. 2, we describe the photovoltaic effect of a ${\mathrm{WS}}_2/\mathrm{SiC}$ film stack. In Sec. 3, we study the effect of transverse electric field on the wavelength blueshift of the film experimentally. It is demonstrated that the blueshift can be modified smoothly by changing the strength of applied electric field. A theoretical explanation based on the inverse Compton scattering about the blueshift changes is investigated. In Sec. 4, we briefly describe design, formation, and characterization of the ${\mathrm{WS}}_2/\mathrm{SiC}$ film. Finally, we summarize our findings and give our discussions in Sec. 5.

2.

Photovoltaic Effects in WS₂/SiC Film

The photovoltaic effect experiment setup was shown as Fig. 1. The thickness of 15 nm ${\mathrm{WS}}_2/\mathrm{SiC}$ sample and 150 nm thickness of ${\mathrm{WS}}_2/\mathrm{SiC}$ sample were overlapped (with ${\mathrm{WS}}_2$ surface to ${\mathrm{WS}}_2$ surface) face-to-face tightly to form a ${\mathrm{WS}}_2/\mathrm{SiC}$ film stack. An excitation laser beam passes through the surface of the 15-nm ${\mathrm{WS}}_2$ film first and then through 150-nm ${\mathrm{WS}}_2$ film perpendicularly. The beam splitter (transparency 50% @ 532 nm) was placed in front of the ${\mathrm{WS}}_2/\mathrm{SiC}$ film stack and two optical power meters (HIOKI 3664 together with the HIOKI 9742 detector) were used to monitor the reflected and transparent laser powers. A HIOKI 3238 Digital Hitester was used to measure the open circuit photovoltaic voltage of stacked ${\mathrm{WS}}_2/\mathrm{SiC}$ films under laser irradiation.

Fig. 1

Principle diagram of photovoltaic effect experiment setup.

Several laser wavelengths (532, 650, 661, 780, 808, 850, 980, and 1064 nm) were utilized to investigate this photovoltaic effect. Each laser was tuned to an output power of 5 mW to incident on the ${\mathrm{WS}}_2/\mathrm{SiC}$ films stack, and the correspondent open circuit photovoltaic voltage ($V_{\mathrm{oc}}$) and short-cut circuit current ($I_{\mathrm{sc}}$) were measured as shown in Table 1.

Table 1

Photovoltaic effect of WS2/SiC film stack measurement.

It is found that (1) no incident laser, no $V_{\mathrm{oc}}$ can be tested. (2) There must exist a cutoff wavelength ${\lambda }_{\mathrm{c}}$ ($>661\mathrm{nm}$ for this film) less than which there is a photovoltaic phenomenon, otherwise, no photovoltaic effect can be observed. (3) This photovoltaic effect is much related to the polarization of the incident light. No polarization, no photovoltaic effect (refer to Sec. 5.2).

No laser no $V_{\mathrm{oc}}$ means that there are no junctions between the ${\mathrm{WS}}_2$ film and the SiC substrate or between ${\mathrm{WS}}_2$ films, and the main cause of this photovoltaic effect is supposed that a photon is absorbed to produce positive and negative charges. Therefore, ${\lambda }_{\mathrm{c}}$ must be much related to the band gap of the ${\mathrm{WS}}_2$ film. ${\lambda }_{\mathrm{c}}=1240/Eg$, where $Eg$ is the band gap. Those lasers with the wavelength less than ${\lambda }_{\mathrm{c}}$ will produce charges and thus photovoltaic phenomena happen, whereas, lasers with the wavelength greater than ${\lambda }_{\mathrm{c}}$ will not produce any charges and thus no photovoltaic phenomena can be observed.

As we used SiC substrate as the electrode, the short cut circuit current $I_{\mathrm{sc}}$ is much limited by the resistance of SiC substrate.

In order to investigate influences of thickness of ${\mathrm{WS}}_2$ film on photovoltaic effect of ${\mathrm{WS}}_2/\mathrm{SiC}$ stack, three stacks are combined by using $10+15\mathrm{nm}$, $5+150\mathrm{nm}$, and $15+150\mathrm{nm}$ configurations. Each configuration stack is tested and the photovoltaic output voltage is regarded as criteria for choosing the best configuration.

The tested results are shown in Fig. 2. It was noted that the $15+150\mathrm{nm}$ configuration stack presents a better performance in photovoltaic output voltage measurement.

Fig. 2

Photovoltaic effects of ${\mathrm{WS}}_2/\mathrm{SiC}$ stacks with different ${\mathrm{WS}}_2$ film thickness configurations. The red line is for $15+150\mathrm{nm}$ thickness case; the green line is for $5+150\mathrm{nm}$ thickness case; the blue line is for $10+15\mathrm{nm}$ case.

The maximum photovoltaic output voltage measurement was conducted by applying a commercial 532-nm laser (YRL-532-100) to a ${\mathrm{WS}}_2/\mathrm{SiC}$ film stack. The output power of the laser can be tuned from 0 to 50 mW continuously. The relation between $V_{\mathrm{oc}}$ and the laser output power is shown in Fig. 3. $V_{\mathrm{oc}}$ varies linearly with the increase of laser power, and it reaches to a maximum of 6.3 V at 40 mW.

Fig. 3

Photovoltaic open circuit voltage versus laser power (@532 nm).

3.

Wavelength Blueshift and the Inverse Compton Scattering in WS₂/SiC Film

3.1.

Experiment Setup of the Blueshift of Laser Wavelength Measurement

Figure 4 shows the principle diagram of blueshift of laser wavelength measurement setup. A self made n-doped (doping density of $7.37\times {10}^{19}{\mathrm{cm}}^{-3}$, tested with PHYS TECH RH-2035) 4H-SiC²³ substrate (2 in. in diameter, 4 deg tilt cutting) with a thickness of ($300\mu \mathrm{m}$) was used, it was coated of 15-nm thickness thin layer ${\mathrm{WS}}_2$ film with a ${\mathrm{WS}}_2$ ceramic bulk material (99.9% purity) as a target by using the PLD technique. The substrate deposited with ${\mathrm{WS}}_2$ was cut into a size of $10\times 10{\mathrm{mm}}^2$ square. An adjustable DC power supply (KXN-305D) with a voltage output range from 0 to 30 V was used as an external electric supply. N-doped SiC substrate was used as a transparent electrode to which the external DC supply voltage was applied. The DC voltage applied between the two SiC substrates was measured by a microammeter (HIOKI 3238 Digital Hitester). The commercial 532-nm laser that utilizes a second harmonic generation technique was used as a monochromatic light source. The laser is linearly polarized. The optical output powers were set to be 1.5 mW. The laser beam direction is perpendicular to the surface of the thin layer ${\mathrm{WS}}_2$ film. Two optical spectrometers (Ocean 2000 plus from Ocean Optics with a spectra resolution of 0.2 nm) were used to monitor the laser wavelength changes. The wavelength of the laser in front of and behind the ${\mathrm{WS}}_2/\mathrm{SiC}$ film stack can be measured simultaneously. The wavelength shift of the laser is defined as the difference of the spectrum peak in front of the ${\mathrm{WS}}_2$ film and the correspondent spectrum peak behind the ${\mathrm{WS}}_2$ film.

Fig. 4

Principle diagram of blueshift of laser wavelength measurement.

It was found that, for 532-nm laser, the wavelength is shifted linearly from 532.99 to 531.66 nm when the external DC electrical supply is tuned from 0 to 30 V. A 1.33-nm blueshift of laser wavelength was observed.

Figure 5 showed the blueshift of the laser wavelength changes for 532-nm laser. The relation between the blueshift of the wavelength and the driven DC voltage was linear.

Fig. 5

The blueshift of 532-nm laser wavelength versus external electric DC voltage.

3.2.

Theory

The inverse Compton scattering effect was used to explain our experiment results. The average energy change of the photon per Compton scattering is²⁴

Eq. (1)

$$⟨\frac{{\mathrm{\Delta }}_i}{{\omega }_0}⟩=\frac{4K{T}_i-\hslash {\omega }_0}{m_{\mathrm{eff}}^i{C}^2},i=e,h,$$

where $[{\mathrm{\Delta }}_i,i=e,h]$ is the frequency shift of an incident photon with a photon energy of by an electron or a hole, respectively. $C$ is the light speed in vacuum, $m_{\mathrm{eff}}^i$, $[i=e,h]$ is the electron (hole) effective mass in ${\mathrm{WS}}_2$ film, $T_i[i=e,h]$ is the accelerated electron (hole) temperature, and $K$ is the Boltzmann constant. Here, $0.5K{T}_i$ was the energy of an electron (hole) accelerated by an external electric field $E$, or $0.5K{T}_i=e{U}_0$, where $U_0$ was the voltage applied to the ${\mathrm{WS}}_2$ film. So Eq. (1) was modified as

Eq. (2)

$$⟨\frac{\mathrm{\Delta }}{{\omega }_0}⟩=\frac{8e{U}_0-\hslash {\omega }_0}{m_{\mathrm{eff}}{C}^2},$$

where $\mathrm{\Delta }={\mathrm{\Delta }}_e+{\mathrm{\Delta }}_h$ was the total frequency shift contributed by both electron and hole. And $1/m_{\text{eff}}=1/m_{\mathrm{eff}}^e+1/m_{\mathrm{eff}}^h$ was the reduced effective mass of the electron and hole pair. Take $m_{\mathrm{eff}}^e=0.39m_0$ and $m_{\mathrm{eff}}^h=0.4m_0$, here $m_0$ was the static mass of an electron.²⁵ This results in a reduced mass $m_{\text{eff}}=0.197m_0$.

The calculated photon energy gain and measured wavelength blueshift were shown in Fig. 6. Good coincidence between experiment and the theory was found.

Fig. 6

532-nm photon blueshift versus applied voltage.

4.

Design and Characterization of a Thin Layered ${\mathrm{WS}}_2/\mathrm{SiC}$ Film

Thin layer ${\mathrm{WS}}_2$ film has been formed by the PLD method. The thickness of ${\mathrm{WS}}_2$ film, which was formed on the n-doped SiC substrate of 2 in. in diameter and thickness of $400\mu \mathrm{m}$, is about 15 and 150 nm, respectively, in our experiment. The ${\mathrm{WS}}_2$ film was cut in into a size of $10\times 10{\mathrm{mm}}^2$. Raman scattering data, atomic force microscopy data of the thin layer film, and STEM were used to characterize the material, which was revealed that the thin layer ${\mathrm{WS}}_2$ film on the n-doped SiC substrate is a fine ${\mathrm{WS}}_2$ film,²⁶ the details of the film formation and the correspondent characterization data can be found in SPIE-PLD/TFPA 2019 (Qingdao, China, May 19 to 22, 2019).

5.

Conclusions and Discussions

5.2.

Discussions

Several points of view were addressed as follows:

• (1) The photovoltaic effect reported in this study is quite different from the traditional solar cell or photodetectors in which there is a p-n junction. There is no junction in this ${\mathrm{WS}}_2/\mathrm{SiC}$ film stack. No laser irradiation on the film, no photovoltaic voltage detected. This confirms our conclusion that no junction exists in ${\mathrm{WS}}_2/\mathrm{SiC}$ film stack.

• (2) A hypothesis is assumed that the polarized laser electric field plays an impotent role in this new kind of photovoltaic effect, i.e., laser photons initiate the electrons and holes, and then laser electric field separates the positive and negative charges, which results in an open circuit voltage output formation. To verify this hypothesis, an ordinary white light source—a halogen lamp with a 10-W output light (conducted by a coupled optical fiber bundle) was used to illuminate the ${\mathrm{WS}}_2/\mathrm{SiC}$ film stack, the measured voltage by HIOKI 3238 Digital Hitester is almost zero. However, when a polarizer was placed in front of the ${\mathrm{WS}}_2/\mathrm{SiC}$ film stack, a voltage of about 100 mV was detected.

• (3) Two optical spectrometers were utilized to observe the changes of the laser spectrum peak in front of the ${\mathrm{WS}}_2$ film and behind the ${\mathrm{WS}}_2$ film. This arrangement permits that the reflected and transmitted laser beams be measured simultaneously. The wavelength blueshift is defined by the difference of the peak positions as shown in Fig. 7.

Fig. 7

Behaviors of blue-shift of a 532-nm laser reflected and transmitted with the change of DC voltage applied on ${\mathrm{WS}}_2/\mathrm{SiC}$ film stack. (a)–(e) The applied voltage of 4, 11,16,22, and 30 V, respectively.

In Fig. 7, the red curve is the laser spectrum in front of the ${\mathrm{WS}}_2/\mathrm{SiC}$ film stack, and the green one is the laser spectrum behind the ${\mathrm{WS}}_2/\mathrm{SiC}$ film stack. Figures 7(a)–7(e) clearly show that the red peaks keep unchanged no matter how the applied voltage changes, but the green peak changes along with the applied voltage.

• (4) The inverse Compton scattering model can explain the blueshift behavior. From Eq. (2), it may conclude that in order to increase the photon energy gain, a higher voltage may play an important role. But the maximum voltage is limited because of a breakdown of the film. Another way to increase this blueshift is to find a small reduced effective mass of electron–hole pair, this will be our next effort.²⁷

• (5) Our discoveries in this study reveal a clue that an external tunable blueshift material is of great importance in the optoelectronic device application field.