Abstract

Interfacial carrier transfer kinetics is critical to the efficiency and stability of perovskite solar cells. Herein, we measure the regeneration rate constant, absorption cross-section, reduction rate constant, and conductivity of hole transport layered perovskites using scanning electrochemical microscopy (SECM). The SECM feedback revealed that the regeneration rate constant, absorption cross-section, and reduction rate constant of the nickel oxide (NiO) layer perovskite layer are higher than those of the poly (3,4-ethyenedioxythiophene)-poly (styrenesulfonate) layered perovskite. Also, at a specific flux density (), the value of the regeneration rate constant (keff) in both blue and red illuminations for the NiO/CH3NH3PbI3 film is significantly higher than in both PEDOT: PSS/CH3NH3PbI3 and FTO/CH3NH3PbI3 films. The difference in keff between layered and nonlayered perovskite conforms to the impact of the hole conducting layer on the charge transfer kinetics. According to the findings, SECM is a powerful approach for screening an appropriate hole transport layer for stable perovskite solar cells.

1. Introduction

Recent high-performanceorganic-inorganic perovskite solar cells have sparked a lot of interest in the photovoltaic field [1, 2]. They have been regarded as the most promising candidates for being at the forefront of next-generationthin-film solar cells because of their distinct advantages, such as their excellent photovoltaic properties, simple fabrication process, and low cost [2, 3]. The quality of perovskite films plays a key role in fabricating stable and efficient PSCs [4]. The fast growth in efficiency of perovskite solar cells [3, 4] and the limitations of PSCs for outdoor applications have driven the scientific community to do much more research on the absorption characteristics, interface layer, and optoelectrical properties of the perovskite material [5, 6]. In this regard, organometallic perovskite is an excellent light harvester due to unique material features such as a broad absorption spectrum range with a high extinction coefficient, bipolar diffusion, and a long carrier diffusion length. When light strikes the CH3NH3PbI3 active layer, the excitation of carrier electron results in the formation of electron-hole pairs with a small exciton binding energy [712]. Recent research reports on hole transporting materials (HTMs) have revealed their significant effects on the processes of hole extraction, transportation, and blocking recombination processes [13]. In perovskite solar cells, the hole transporting layer (HTL) plays significant roles such as avoiding unwanted electron transfer, extracting the holes in the active layer, splitting the perovskite layer from the anode, minimizing degradation, improving the device’s stability, and enhancing the open-circuit voltage (Voc) [2, 1316]. Among different hole transporting materials, inorganic NiO is stable, of low cost, earth abundant, nontoxic, has high optical transmittance, sufficient conductivity, and excellent charge extraction ability [1723]; and PEDOT: PSS offers promising organic candidates for perovskite solar cells due to their high optical transmittance, low cost, outstanding stability, high mechanical flexibility, adjustable conductivity (10−2 to 103 S/cm), suppression interfacial recombination, and accelerate hole transfer [2, 24, 25]. They have been attributed to back contact as an electron blocking layer, with higher conduction band energy in CH3NH3PbI3 versus vacuum [24, 25]. According to a recent report, photo-induced absorption spectra and photoluminescence measurements have shown that NiO can enhance the stability [2628], and PEDOT: PSS can act as a good hole transport layer but affect the stability due to moisture absorption [29, 30]. For high performance, stability, and low-cost PSCs, it is crucial to investigate the charge transfer reactions of NiO/CH3NH3PbI3 and PEDOT: PSS/MAPbI3 layers after photo-excitation and the conductivity of CH3NH3PbI3 deposited on both NiO and PEDOT: PSS hole transporting layers. SECM has arisen as an innovative approach for a variety of study disciplines, including biology [31], corrosion [32], energy [33, 34], electron transfer processes [33], and interfacial kinetics [3543].

In this paper, we present SECM feedback mode studies on the effect of hole conducting layers (NiO and PEDOT: PSS) on the regeneration kinetics of the hole conductor/CH3NH3PbI3 interface. With illumination, kinetic parameters such as the regeneration rate constant (keff), reduction rate (kred), and absorption cross-section () were investigated. Furthermore, the Hall effect measurement model was used to characterize the conductivity measurements.

2. Experimental Methods

2.1. Chemicals

Terpineol, ethyl cellulose, polycarbonate solvent, and NiO nanoparticles were purchased from Aldrich. Organic redox T_, PEDOT: PSS (Figure 1(a)), and CH3NH3PbI3 were synthesized using the methods described in the literature [3841]. Figure 1(a) shows the chemical structure of PDOT: PSS, Figure 1(b) shows the energy level diagrams of perovskite solar cells, Figure 1(c) shows the chemical structure of persivate, and Figures 1(d) and 1(e) show the SECM top view of PEDOT: PSS/CH3NH3PbI3 and NiO/CH3NH3PbI3, with mean particle diameters of about 200 nm for NiO nanoparticles, respectively.

2.2. Film Preparation

In this study, NiO nanoparticles (with a particle size of ≅20 nm) were ball milled in ethanol with a few drops of acetic acid. To obtain a fine dispersion, the aforementioned colloidal solution, ethyl cellulose (Aldrich), and terpinol were sonicated and mixed alternatively. A paste was made by evaporating ethanol from the mixture on a rotary evaporator. FTO glass Nippon sheets (15 Ω square−1) were dip coated with nickel acetate ethanol solution (0.05 M) and then dried at 80°C. The photocathode films were screen-printed with NiO paste and dried for 5 min at 125°C. The NiO electrodes were sintered at 450°C for 30 min with a ramping time of 30 min from room temperature to 450°C and then posttreated at 550°C for 15 min with a ramping time of 10 min from 450°C to 550°C [4245].

2.3. SECM Measurements

SECM experiments were carried out on the CHI 920C electrochemical workstation (CH Instruments, Shanghai). A Pt wire counter electrode and an Ag/Ag+ reference electrode were held in a handmade Teflon cell (with a volume of 2 mL). The sensitized electrodes FTO/CH3NH3PbI3, FTO/NiO/CH3NH3PbI3, and FTO/PEDOT: PSS/CH3NH3PbI3 were positioned at the bottom and short-circuited to the electrolyte by a Pt wire [46, 47]. The LEDs were placed close to the cell from the back and focused on film electrodes by an objective lens with a photo-illuminated spot area of 0.0785 cm2. A 25 μm diameter Pt wire (Good Fellow, Cambridge, UK) was sealed into a 5 cm long glass capillary prepared with a vertical pull-pin instrument (PC-10, Japan). The ultramicroelectrode (UME) was polished by a grinding instrument (EG-400, Japan) and micropolishing cloth with 1.0, 0.3, and 0.05 mm of alumina powder. The ultramicro electrode (UME) was then conically sharpened to an RG of 10, where RG is the ratio of the glass sheath and Pt disc diameters. All experiments were carried out at room temperature. The approach curves are given in the form of normalized UME current IT vs normalized distance (L), the position zmax at which the UME makes mechanical contact with the sensitized sample, and the distance (do) of the active electrode area to the sample at zmax. L is obtained from the vertical position z, which increases with approach toward the sample. As a result, the possible distance between the substrate and the tip is 200 μm ± 10 and has a potential of ET = 0.7 V [44, 4648].

2.4. Hall Effect Measurement

The Ecopia Hall Effect measurement model number HMS-5500 was used to measure conductivity. A glass substrate was cleaned for 15 min each using detergent, deionized water, acetone, and IPA. The cleaned substrate was air dried and cured for 40 min using a UV ozone cleaner. The CH3NH3PbI3 solution was spin coated at 3000 rpm for 50 seconds on the NiO/CH3NH3PbI3 paste and PEDOT: PSS/CH3NH3PbI3 coated glass substrate on top of the cleaned glass substrate. A conductivity measurement was performed in the dark for both samples. The sample that was initially stored under illumination was placed in the dark measurement chamber 10 min before the measurement [4951].

3. Results and Discussion

Previous studies [42, 51] described the SECM feedback mode’s principle for investigating heterogeneous responses under illumination. The UME current (IT) varies with distance (L) and T regeneration rate at the sample. Under blue illumination, the photo-electrochemical effects at the Pt UME probe approach curves were recorded for the FTO/CH3NH3PbI3, NiO/CH3NH3PbI3, and PEDOT: PSS/CH3NH3PbI3 films (as shown in Figure 2(a)). Furthermore, we used Van Der Pauw four-point probes in all of the reliability tests [50, 52]. Figure 2(b) shows the temperature-dependent conductivity, which decreases with time and temperature. Because the temperature-dependent conductivity curves showed a change in slope from 7 S (W−1 cm−1) to 4 S (W−1 cm−1) with the crossponding temperature rising from 280 K to 340 K at all times from 10 min to 3 h as presented; moreover, at low temperatures, the thermal conductivity obeys the power law of that the temperature-dependent conductivity is subjugated by acoustic phonon scattering [53]. The PODET : PSS/CH3NH3PbI3 indicated a fast reduction in conductivity across the entire temperature range in the first half hour, but in the dark, there is a continuing reduction in conductivity, especially at high temperatures. Figure 2(c) depicts the conductivity of NiO/CH3NH3PbI3 as a function of temperature, while Figure 2(d) shows the conductivity of PEDOT : PSS/CH3NH3PbI3 as a function of temperature. As a result, the conductivity and activation energy of PEDOT : PSS/CH3NH3PbI3 were much higher than those of NiO/CH3NH3PbI3. This could be due to the high amount of free electrons in the PEDOT : PSS/CH3NH3PbI3 film, as well as the highly conductive PEDOT chains acting as a hole transport path [54, 55]. Moreover, this could be because of PEDOT : PSS, which has a dominant conductivity over the conductivity of NiO [53, 56], whereas NiO has a low conductivity [52, 54, 57]. Furthermore, the conductivity of NiO and PEDOT : PSS was reported to be 74 S/cm and 3026 S/cm, respectively [56]. This shows the conductivity of PEDOT: PSS is more than ten times higher than the conductivity of NiO [56].

The Arrhenius model, represented by equation (1), was commonly employed for conductivity as a function of temperature [49]:where is the conductivity of the film at temperature T (K), is the preexponential factor and represents the limiting conductivity at infinite temperature, Ea (eV) is the activation energy that results from the variation of the conducting level, Fermi energy, and Boltzmann’s constant (k = 8.616×10−5 eVK−1). Furthermore, the activation energy that governs conductivity in the ionic conduction process is about 1 eV [48]. The activation energy Ea was calculated by the natural logarithm of the Arrhenius equation and the slope of the plots indicating the variation of activation with temperature in equation (2):

The activation energies of NiO/CH3NH3PbI3 and PDOT: PSS/CH3NH3PbI3 were calculated to be 20.26 meV and 235.2 meV, respectively. Furthermore, the conductivity of NiO/CH3NH3PbI3 and PDOT: PSS/CH3NH3PbI3 was reported to be 74 S/cm and 870 S/cm, respectively.

The photons excite the charge carriers CH3NH3PbI3, and the electrons and hole diffusion in the conduction band and in the valence band can be drifted by the potential difference. The holes transfer from the hole conductor to CH3NH3PbI3, resulting in electron injection into the layers from CH3NH3PbI3. There is an indicator reaction between CH3NH3PbI3 as light absorbers and the redox species, which emerged on the UME. A previous study elucidated photo-electrochemical reactions on sensitized NiO electrodes under illumination at the NiO electrode/electrolyte interface. The electrochemical reactivity of the substrate determines the electrochemical reaction of tip current [58, 59]. The excited holes in NiO/CH3NH3PbI3 and PEDOT: PSS/CH3NH3PbI3 generate electron-hole pairs at the perovskite.

The SECM feedback mode on redox mediator reaction can be expressed with equation (3):

When illuminated, perovskite absorbs light and forms electron-hole pairs, which can promote the production of exciton. The injection of electrons into electron-transporting materials (ETM) and the injection of holes into hole transporting materials (HTM) are multicharge separation reactions. This excitation of CH3NH3PbI3 leads to electron-hole pairs (equation (4)) which results in hole injection (as shown in Figure 3) into the valence band (VB) of NiO and PEDOT: PSS (equations (5) and (7) as shown in Figure 3). This represents the charge transport kinetics at the interface of a hole conductor and a perovskite-electrolyte-UME [54, 60]. The charge separation reaction is based on the kinetic competition of exaction, which results in photoluminescence (equations (8) and (9)), as well as back charge transfer at the HTM surface (Figure 3) (equation (9)) [52, 61]:

In order to investigate the effect of the hole conducting layer on the regeneration kinetics, we used SECM under illumination with a blue LED with varying light intensity. Figure 4(a) shows the normalized approach curves of UME to FTO/NiO/CH3NH3PbI3 under blue illumination (curves #1–7). The normalized approach curves of UME towards FTO/NiO/CH3NH3PbI3 are documented. There is no diffusion reaction in the dark as demonstrated by curve #0, implying that layered perovskites act as an insulator. For FTO/NiO/CH3NH3PbI3 film, Keff increased from 5.94 × 10−3 to 16.9 × 10−3 cms−1 as the photon flux density increased from 2.02 × 10−9 to 22.4 × 10−9molcm−2s−1. In addition, the keff of the FTO/PEDOT: PSS/CH3NH3PbI3 film increased from 3.99 × 10−3 to 13.4 × 10−3 cms−1. The keff increased from 1.28 × 10−3 to 4.12 × 10−3 cms−1 as the photon flux density increased from 2.02 × 10−9 to 22.4 × 10−9 molcm−2s−1 for FTO/CH3NH3PbI3 (Table 1).

The increase in keff with rising flux intensity indicates that the regeneration process can only be driven by the absorber’s light absorption. As a result, the FTO/NiO/CH3NH3PbI3 film has a higher keff than the FTO/PEDOT: PSS/CH3NH3PbI3 film and the FTO/CH3NH3PbI3 film. This could be due to the fact that the NiO thickness allowed for greater perovskite loading than the PEDOT: PSS sample and bar FTO. The loading concentration of CH3NH3PbI3 on NiO sample (ΓD, NiO) was significantly higher than ΓD, PDOT: PSS and ΓD, FTO), with an estimated value of (ΓD, NiO 1.304 ) × (ΓD, PDOT: PSS), and (ΓD, NiO 2.54 ) × (ΓD, FTO) [46, 62]. Furthermore, the absorption cross-section of FTO/NiO/CH3NH3PbI3 is 6.87 × 106 mol−1 cm2 under blue light, while the absorption cross-section of FTO/PEDOT: PSS/CH3NH3PbI3 is 4.15 × 105 mol−1 cm2 under blue illumination. Similarly, with red illumination, the absorption cross-section of FTO/NiO/CH3NH3PbI3 was significantly higher than of FTO/PEDOT:PSS/CH3NH3PbI3, and , FTO/CH3NH3PbI3. This pronounced variation in absorption cross-section confirms the influence of absorber concentration on regeneration kinetics. Figure 4(b) shows the UME approach curves to PEDOT: PSS/CH3NH3PbI3 under red illumination (curves #1–7). The UME current increased significantly when the back side of the hole conductor’s layered CH3NH3PbI3 film was illuminated. With red illumination, the photon flux density increases from 2.01 × 10−9 to 14.7 × 10−9 molcm−2s−1. The keff of NiO/CH3NH3PbI3 increases from 1.83 × 10−3 to 5.42 × 10−3 cms−1 in the organic electrolyte of T, while the keff of PEDOT: PSS/CH3NH3PbI3 increases from 0.82 × 10−3 to 2.43 × 10−3 cms−1 (Table 2).

Figure 5 shows the relationship between and keff for FTO/NiO/CH3NH3PbI3, FTO/PEDOT: PSS/CH3NH3PbI3, and FTO/CH3NH3PbI3 films (a) under blue illumination with a maximum wavelength of 494 nm and (b) red illumination with a maximum wavelength of 647 nm in 1 mM T electrolyte. It is noted that under blue illumination, the keff of NiO/CH3NH3PbI3 is slightly higher than the keff of PEDOT: PSS/CH3NH3PbI3, but under red illumination, the keff of NiO/CH3NH3PbI3 is two times higher than the keff of PEDOT: PSS/CH3NH3PbI3 [36, 50]. This implies that the significant impact layer on the registration kinetics rate was represented by loading concentration of CH3NH3PbI3 on the film. Therefore, Γ𝐷 of the FTO/NiO/CH3NH3PbI3 is significantly higher than ΓD of the FTO/PEDOT: PSS/CH3NH3PbI3 and FTO/CH3NH3PbI3 [46, 62]. These results agree with the light absorbance of both NiO/CH3NH3PbI3 film and PEDOT: PSS/CH3NH3PbI3 film, revealing the maximum photon absorption in the range of 350 nm to 800 nm in the visible region. The red illumination (with the highest wavelength of 647 nm) correlates with the strong spectral absorption range of both samples. Thus, the reduction rate constant kred for the regeneration of the CH3NH3PbI3 by the reduced electrolyte after hole injection into the hole transport layers is obtained by fitting the experimental keff values as a function of according to equation (4) [42, 51]:where [T] is the concentration of the reduced electrolyte, Γ𝐷 is the sample thickness per concentration of the absorber on the film, is the incident photon flux, and is the excitation cross-section.

Therefore, the kinetic reduction rate constant kred and absorption cross-section of NiO/CH3NH3PbI3 and FTO/CH3NH3PbI3 films were evaluated to be kred = 9.98 × 106 mol cm3s1,  = 6.87 × 106 mol−1 cm2 in blue illumination, and  = 4.89 × 106 mol−1 cm2 in red illumination (Table 3). The regeneration rate constants of PEDOT: PSS/CH3NH3PbI3 were evaluated as kred = 3.15 × 106 mol cm3s−1,  = 4.56 × 106 mol−1 cm2 in blue LED illumination, and  = 2.45 × 106 mol−1 cm2 in red illumination. The kred of the NiO/CH3NH3PbI3 film is higher than both PODT: PSS/CH3NH3PbI3 and FTO/CH3NH3PbI3 in blue illumination. The excitation cross-section of the NiO/CH3NH3PbI3 film was significantly bigger than that of PEDOT: PSS/CH3NH3PbI3 in both red and blue illumination (Table 3). As a result, at a specific flux density , the value of keff in both blue and red illuminations for the NiO/CH3NH3PbI3 film is significantly higher than in both PEDOT: PSS/CH3NH3PbI3 and FTO/CH3NH3PbI3 film. This clear difference in keff of layered and nonlayered perovskite conforms to the impact of the hole conducting layer on the charge transfer kinetics.

The SECM measurement for approach curves in the feedback mode is a steady-state measurement for valence band energy (EVB) perovskite, photoexcited perovskite at the HOMO, and photo-induced hole injection h+. The absorption cross-section from the effective rate constant at low light intensities and the hole injection process were obtained by equation (5) [46, 63]:

Figure 6(a) shows the plot of vs of NiO/CH3NH3PbI3 (right ordinates) under blue illumination that for increasing from 2.02 × 10−9 to 22.4 × 10−9 molcm−2s−1, its corresponding absorption cross-section decreased from 2.94 × 10−6 to 0.75 × 10−6 mol cm−2 for NiO/CH3NH3PbI3 (Table 1). Figure 6(b) shows the plot of vs in red illumination PODT: PSS/CH3NH3PbI3 (right ordinates). As increased from 2.02 × 10−9 to 14.7 × 10−9 mol cm−2s−1, its corresponding absorption cross-section deceased from 2.01 × 10−6 to 0.5 mol cm−2 (Table 2). This implies that photon flux density and absorption cross-section are inversely related. Upon light illumination, hole-electron pairs are generated in CH3NH3PbI3 followed by hole transfer to NiO or PODT: PSS. The free energy for hole transfer can be calculated by equation (2)) for hole conductors. Furthermore, under illumination, the quasi-Fermi energies of holes in the solar cell are related to the band gap, Eg [64]. At the HTM-layered interface, the quasi-Fermi energy of holes at the interface increases. Hence, there is balance between the driving energy for hole extraction, ΔE, and the open-circuit voltage (voc).

Energy level alignment between the photoactive hole conducting layer and its adjacent interface is one of the criteria in designing an efficient solar cell device. According to the energy band diagram shown in Figure 7, the HOMO levels of PEDOT: PSS (−5.0 eV) and of NiO (−5.2 eV) are very close to those of CH3NH3PbI3, which makes it suitable for SECM analysis [51]. When a photon activates CH3NH3PbI3, it generates free electrons and holes [51]. Under illumination, the excitation of the CH3NH3PbI3 layer results in the creation of electron-hole pairs with small exciton-binding energy [7]. The ambipolar carrier transport properties as well as the elongated carrier lifetime of CH3NH3PbI3 allow the direct transport of hole charge carriers to the charge collective contacts. The CH3NH3PbI3/HTM junction shows a restoration layer because of the hole transfer from the electron-accepting contact to the CH3NH3PbI3 layer. On illumination, the CH3NH3PbI3 perovskite should absorb the light, and as a result, holes are transferred from the CH3NH3PbI3 perovskite to the HTM and electrons are transported to the electrolyte as shown in Figure 7 [8, 9]. In this work, we investigate the effect of layers on the kinetics of charge regeneration at the illuminated HTM/CH3NH3PbI3/electrolyte interface with SCEM.

The energetic driving force for hole transfer can be approximated by equation (12):where EHOMO of HTMs is relative to the CH3NH3PbI valence band energy, EVB (−5.4 eV) [58]. The resulting energetic driving for hole transfer across the interface of PEDOT: PSS/CH3NH3PbI3 and NiO/CH3NH3PbI3 estimated as and . This implies that the valance band energy of the CH3NH3PbI3 excited state lies below the NiO and PEDOT: PSS layers . Furthermore, the hole regeneration at the interface must be allowed, imposing that the valance band energy of the HTM can be more negative than that of CH3NH3PbI3 [41, 65].

4. Conclusion

To summarize, we investigated the effect of the HTM layer on hole transfer kinetics at the HTM/CH3NH3PbI3 interface using Ecopia Hall Effect measurement and the SECM feedback approach. The SECM study’s findings include variations in the effective rate constant keff, the reduction rate constant kred, and the absorption cross-section (hv) for FTO/NiO/CH3NH3PbI3, FTO/PEDOT: PSS/CH3NH3PbI3, and FTO/CH3NH3PbI3. To determine absorption rates, the conductivity of HTM-layered NiO/CH3NH3PbI3 and PEDOT: PSS/CH3NH3PbI3 was also investigated. The approach curves attributable to the kinetics of hole regeneration at the interface are affected by incident light wavelength, photon flux density, and layer optical properties. The dominant conductivity of PEDOT: PSS/CH3NH3PbI3 was found to be significantly higher than that of NiO/CH3NH3PbI3. Furthermore, the energetic driving force for hole transfer was found to be and . This demonstrates that the HOMO level of PEDOT: PSS is very close to that of CH3NH3PbI3, making it an excellent layer for fast hole regeneration interfaces. This allows for simple testing of highly efficient photon absorbers with a wide range of absorption. Thus, SECM is a novel method for screening an appropriate hole transport layer for stable perovskite solar cells.

Data Availability

The data used in this study are included in the article. If further information be required, this is available from the authors upon request.

Conflicts of Interest

The authors declare that they have no conflicts of interest.

Acknowledgments

The authors are grateful for the Ph.D. fellowship provided by the Chinese Scholarship Council (CSC). They would also like to thank Prof. Dr. Mingkui Wang for his contributions. They acknowledge that the experiments were performed on a CHI920C electrochemical workstation (CH Instruments, Shanghai) provided by the Chinese Scholarship Council (CSC) for Ph.D. fellowship.