**2.3 WS2 electron transport layer**

Tungsten disulfide (WS2) share common basic properties of TMD with other systems, such as high-mobility properties, unique optoelectronic properties, large exciton-binding energy, and good physical and chemical stability as well as ambipolar properties [11]. In addition, WS2 has an energy level that is suitable for the perovskite layer of three types of cations (**Figure 9**) and can be easily synthesized by the ultrasonic spray pyrolysis method. WS2 also has high stability as well as having fast interface charge transfer properties [32]. Among the available 2D TMD, the energy band structure of WS2 is a much better match with the common perovskite of MAPbI3 (**Figure 10**). Furthermore, it also has a relatively larger bandgap if compared with the other system in this class of materials, promising facile excitonic separation during the photovoltaic process and producing better power conversion efficiency.

Recently, we have realized the PSC device utilizing the WS2 layer as ETL and evaluated how the number of layers of WS2 influences the carrier dynamic in the device [5]. We prepared the WS2 atomic layer via ultrasonic spray pyrolysis. **Figure 11** shows a schematic diagram of the 2D atomic layer preparation. A modified commercially available ultrasonic spray system (Daiso, Japan) was used. A homemade solution container was placed on the top of the ultrasonic membrane of the system (**Figure 11**). Ultrasmall solution precursor mist can be produced from the process and fall on the ITO substrate surface that is positioned approximately 5 cm below the membrane. The temperature of the substrate was set at 350°C.

#### **Figure 9.**

*Energy levels of dichalcogenide transition metal materials (TMDs) as ETLs and MAPbI3 as perovskite layers in perovskite solar cells.*

**Figure 10.** *Energy level diagram for n-i-p perovskite solar cells using WS2 ETL.*

#### **Figure 11.**

*Schematic diagram of ultrasonic spray pyrolysis for the preparation of TMD ETL.*

The typical morphology of the WS2 atomic layer on the ITO substrate is shown in **Figure 12A**. The WS2 nanosheet's morphology resembles a circular structure that is produced from the precursors' mist that emerged from the ultrasonic spray membrane. Confocal Raman imaging further indicated the existence of a very thin layer of structure from the circular structure as shown in **Figure 12B**. Raman analysis then confirmed the phase crystallinity of the WS2 (**Figure 12C**). As the figure reveals, there are two sharp peaks obtained from the Raman spectrum that is centered at 348.9 cm−1

*Two-Dimensional Transition Metal Dichalcogenide as Electron Transport Layer of Perovskite… DOI: http://dx.doi.org/10.5772/intechopen.103854*

#### **Figure 12.**

*The morphology, phase crystallinity, chemical state properties of WS2 nanosheet. (A) FESEM image of WS2 nanosheet on the ITO substrate. (B-C) Raman imaging and spectrum of WS2 were obtained using 532 nm laser excitation. The inset in (C) shows the corresponding main vibration mode of Raman. (D) XRD spectrum for WS2 nanosheet showing 2H phase. (E-F) Low and high-resolution TEM image of WS2 nanosheet. (G) SAED pattern of WS2 nanosheet showing at least three stacking WS2 nanosheets. (H-I) High-resolution scan of XPS at W and S binding energy of WS2 nanosheet. (Reprinted from [5]. © 2020 Wiley-VCH GmbH).*

and 412.3 cm−1, which are associated with the in-plane (E2g) and the out-of-plane (A1g) vibration modes of the lattice (see inset in **Figure 12C**) [33–39]. According to the value of the separation between these two peaks, the thickness of the atomic layer is estimated to be in the range of 10 L. The X-ray diffraction analysis further confirmed the phase crystallinity of the WS2 layer (**Figure 12D**) [40–42]. The high-resolution transmission electron microscopy (HRTEM) and selected area electron diffraction (SAED) analysis results (**Figure 12F** and **G**) show that the sample is single crystalline. However, the presence of SAED composed of a triple spot is related to the stacking of the WS2 atomic layer during the transfer to the lacey grid for HRTEM analysis. The XPS analysis then further confirmed the Raman and XRD analysis results on the phase crystallinity of the sample of which it belongs to WS2 (**Figure 12H**-**I**).

PSCs device was fabricated using the WS2 atomic layer as ETL and investigated how the thickness of the WS2 ETL influenced the photovoltaic process. The structure of the PSC device is ITO/WS2 nanosheets/Perovskite/Spiro-OMeTAD/Au. Perovskite used was triple cations system of Cs0.05[MA0.13FA0.87]0.95Pb (I0.87Br0.13)3 [43].

It was found that the thickness, represented by the number of layers, of the WS2 atomic layer ETL, strongly influences the power conversion efficiency of the PSC device (**Figure 13**). The results show that the PCE performance improves with the increase of thickness from 4 L to the optimum thickness of 7 L (WS30 sample in the figure). The optimized WS2 ETL thickness can produce a PSC device with PCE as high as 18.21% with *J*sc, *V*oc, and FF as high as 22.24 mA cm−2, 1.12 V, and 0.731,

#### **Figure 13.**

*The photovoltaic performance of PSC using different thicknesses of WS2 ETL. (A) J-V curves for the champion PSC device. (B–E) The statistic plot for PCE, Voc, Jsc, and FF, respectively. (F) EQE and integrated current density (J) of the corresponding device. (Reprinted from [5]. © 2020 Wiley-VCH GmbH).*

respectively. The average performance was 17.84%, 22.33 mA cm−2, 1.10 V, 0.731 for PCE, *J*sc, *V*oc, and FF, respectively. However, due to an increase in the energetic disorder when using the WS2 ETL, the device performance then declined when the thickness of the ETL increased above 7 L. From **Figure 13**, we can also see that the values of *V*oc and fill factor (FF) are impressively high, which is higher than 1.1 V for *V*oc and approximately 74% for FF. This reflects that the photogenerated carrier dynamic in the device is high and the photogenerated carrier is effectively extracted to the external circuit to produce photocurrent [44]. This is verified by the high-external quantum efficiency (EQE) of the device as shown in **Figure 13F**.

To understand the extent effect of the WS2 atomic layer as ETL in the PSC device, the device performance was compared with the reference PSC utilizing well-known SnO2 ETL. In the typical process, the performance of SnO2-based PSC shows lower performance than the WS2 atomic layer–based device (**Figure 14**). Steady-state and transient photoluminescence analysis revealed that the interfacial charge transfer from the perovskite to ETL is high in the WS2 atomic layer [45], the result of

*Two-Dimensional Transition Metal Dichalcogenide as Electron Transport Layer of Perovskite… DOI: http://dx.doi.org/10.5772/intechopen.103854*

#### **Figure 14.**

*The comparison of the photovoltaic parameter between WS2 (7 L, WS30 sample) and SnO2-based PSC device. (A) J-V curves for the champion device. (B–E) The comparison of PCE, Voc, Jsc, and FF for the two devices, respectively. (F–H) Steady-state PL, time-resolved PL spectra (TRPL), and electrochemical impedance spectra for WS2 and SnO2-based PSC devices, respectively. (Reprinted from [5]. © 2020 Wiley-VCH GmbH).*

optimized coupling due to ultra-flat surface morphology offered by the WS2 atomic layer. This phenomenon is further confirmed by the electrochemical impedance spectroscopy analysis result where it is obtained that the interface charge transfer resistance is lower in the WS2-based PSC device than the SnO2-based device. Thus, it can be remarked that the WS2 atomic layer enables highly active interfacial charge transfer for a high-performance PSC device.
