Solution Processed Polymer–Inorganic Semiconductor Solar Cells Employing Sb₂S₃ as a Light Harvesting and Electron Transporting Material

Abstract

Hybrid poly-3-hexylthiophene (P3HT)–Sb2 S3 nanocomposite films are formed by means of an in-situ low-temperature thermal decomposition of a solution-processable antimony xanthate precursor in a polymer film. Transient optical spectroscopy and photovoltaic device studies suggest that charge separation and current generation in Sb2S3:P3HT based devices results mainly from Sb2S3 light absorption and subsequent hole-transfer from the inorganic semiconductor to the organic hole transporting material. Hybrid solar cells based upon organic–inorganic semiconductor heterojunctions are currently the subject of significant interest as they incorporate the attractive properties of both organic and inorganic materials, including the ability to tune both the electronic and structural properties over a wide range using solution-based fabrication methods.1-7 A configuration of particular promise is the hybrid inorganic nanocrystal–polymer bulk heterojunction solar cell. A typical device consists of a photoactive layer composed of a blend of inorganic nanoparticles and a semiconducting polymer, which is sandwiched between two charge-collecting electrodes. The operation of such a system is based upon a photoinduced charge separation reaction at the inorganic–organic semiconductor heterojunction, followed by charge carrier transport and collection at the device electrodes. To date, a variety of inorganic semiconductors have been used in solution processed polymer solar cells including metal oxides, sulfides and selenides. Metal chalcogenide nanocrystals are especially attractive for use in photovoltaic device applications as they offer the potential to extend the light harvesting capability of the device into the near infrared region of the solar spectrum. For example, impressive solar-light to electrical power conversion efficiencies have been recently reported for photovoltaic devices based upon CdSe:PCPDTBT (> 3%)5, 8 and CdS:P3HT nanocomposite films (4.1%).6 Key challenges to the design of high-performance hybrid solar cells are (i) the development of new fabrication routes for hybrid thin films that enable the achievement of high yields of charge separation whilst maintaining good electrical connectivity between the inorganic nanocrystals in the photoactive layer and (ii) the development of alternative inorganic electron acceptors that exhibit light harvesting properties superior to the typically used cadmium-based materials. To address challenge (i), we have recently reported a new approach to the fabrication of hybrid metal sulfide–polymer solar cell photoactive layers, which is based upon the in-situ thermal decomposition of a single source metal xanthate precursor in a polymer film.9 The use of metal xanthate (or metal o-alkyl dithiocarbonate) precursors for the in-situ growth of metal sulfide nanocrystals in polymer films is of particular interest due to their high solubility, low decomposition temperature and the volatility of the side products generated upon thermal decomposition. As such, we have implemented this design strategy in the fabrication of CdS:P3HT and CuInS2:polymer nanocomposite films and demonstrated efficient charge photogeneration at the donor–acceptor heterojunction.9-11 Furthermore, the integration of such photoactive layers into solar cell architectures yielded impressive power conversion efficiencies approaching 3% under AM1.5 simulated solar illumination.11 In this paper, we extend our previous work and report on a bulk heterojunction hybrid solar cell composed of antimony sulfide (Sb2S3) nanocrystals and a semiconducting polymer poly-3-hexylthiophene (P3HT). Sb2S3 is an attractive material for use in photovoltaic devices due to its narrow band-gap (1.7 eV) and high absorption coefficient (∼1.8 ×105 cm−1 at 450 nm).12 Consequently, Sb2S3 has been recently employed as a light-harvesting material in extremely thin absorber (ETA) and solid-state nanocrystal-sensitized solar cells.13-19 Herein, we present the first example of a solution processed polymer/Sb2S3 blend solar cell, in which Sb2S3 acts as both a light absorber and an electron-transporting material. Specifically, we report the fabrication of an Sb2S3:P3HT film utilizing an antimony triethyldithiocarbonate (antimony ethyl xanthate) precursor complex that decomposes into the metal sulfide at relatively low temperatures (∼160 °C). A combination of steady state and time-resolved optical spectroscopy, transmission electron microscopy and Raman techniques are used to characterize the nanomorphology and photo-induced interfacial charge transfer in the Sb2S3:P3HT nanocomposite films. Transient absorption spectroscopy measurements provide evidence for charge separation at the Sb2S3:polymer heterojunction. In particular, we show that charge separation and current generation in the device results primarily from Sb2S3 light absorption followed by hole-transfer from the inorganic semiconductor to the organic hole transporting material. We discuss the implications of our findings for the design of hybrid inorganic–organic solar cells. Figure 1 shows the chemical structure of the antimony ethyl xanthate (Sb(S2COEt)3) complex employed in this study. Sb2S3:P3HT films were prepared by spin coating a blend composed of varying volume ratios of polymer (25 mg/mL in chlorobenzene) and precursor (400 mg/mL in chlorobenzene) solutions. The resultant films were thermally annealed at 160 °C to decompose the precursor and to generate the Sb2S3 nanocrystals in the polymer film. Figure 1a shows the steady-state UV-Vis absorption characteristics for films comprising P3HT and Sb(S2COEt)3 (50:50 volume ratio) before and after thermal annealing at 160 °C. The un-annealed film exhibits the characteristic absorption of the P3HT (Figure 1a, blue curve). However, upon thermal annealing at 160 °C, the appearance of a broad absorption feature (400–750 nm) is observed, which is consistent with the presence of Sb2S3 (Figure 1a, black curve). This broad feature is also observed in the absorption spectrum of thermally annealed Sb(S2COEt)3:polystyrene film, which is used here as a control sample (Figure 1a, red curve). Once again, this feature is not seen in the un-annealed film (Figure 1a, green curve), confirming that the Sb2S3 semiconductor is only generated upon thermal annealing at 160 °C. (a) Absorption spectra of thin films comprising P3HT and Sb(S2COEt)3 before (blue curve) and after (black curve) annealing at 160 °C. Also shown are the equivalent data for thin films comprising polystyrene (PS) and Sb(S2COEt)3 (b) Raman spectrum of a 60:40 (Sb2S3:P3HT) film and (c) top-down transmission electron microscopy (TEM) image for a 60:40 (Sb2S3:P3HT) film. Further evidence for the presence of Sb2S3 in the polymer film was obtained by Raman spectroscopy measurements. Typical Raman data for a hybrid film annealed at 160 °C is shown in Figure 1b. The Raman spectrum shown in Figure 1b exhibits peaks at ∼240, ∼280 and ∼310 cm−1 which are characteristic of crystalline stibnite,20, 21 thus confirming the presence of Sb2S3 in the P3HT film. To investigate the nanomorphology of the Sb2S3:P3HT film, transmission electron microscopy studies were performed. Figure 1c shows bright field TEM image of a Sb2S3:P3HT (60:40) film annealed at 160 °C. The image exhibits a distinct phase separation pattern with dark regions of Sb2S3 (∼100–200 nm sized domains) and lighter regions of P3HT. We now consider the charge separation and recombination reactions in the Sb2S3:P3HT nanocomposite film. For this purpose, microsecond–millisecond transient absorption spectroscopy was used. Transient absorption spectroscopy is a pump-probe technique that can be used to identify photogenerated charge separated states and monitor their decay dynamics.22-25 Full experimental details of our transient absorption spectrometer have been described previously and are provided in the Supplementary Information to this manuscript.26 Briefly, transient absorption data were obtained using an optical pump-probe set-up employing nitrogen-pumped dye-laser (600 ps pulse duration) as an excitation source and a tungsten lamp as a probe source. Figure 2a shows a transient absorption spectrum obtained 10μs after pulsed excitation (567 nm, ∼21 μJ/cm2) of a Sb2S3:P3HT film. The transient absorption data presented in Figure 2a exhibits a broad absorption maximum centered at ∼ 950 nm, which is typical of P3HT+ polarons formed as a result of photoinduced interfacial charge separation.27-29 The kinetics of the charge recombination reaction between the photogenerated electrons and holes were determined by monitoring the decay of the P3HT+ polaron band at 950 nm. Typical decay dynamics as a function of increasing Sb2S3 volume fraction (within the hybrid film) are shown in Figure 2b. In all cases, the transients exhibit micro to millisecond power-law (change in optical density ΔOD ∝ t−α) behavior with an exponent α ≈ 0.4–0.5 (see Figure 1, Supporting Information). This is consistent with the presence of thermal trap states, as have been reported in polymer/PCBM,30-32 polymer/CdS,9 and perylene bisimide-arylamine based block copolymers,33 which limit the diffusion of charges and thus the rate of recombination, extending the lifetime of separated charges to the microsecond timescale. Furthermore, no transient absorption features are observed when exciting pristine P3HT and Sb2S3:polystyrene films (see Figure 2, Supporting Information). As such, the amplitude of the transient absorption signal (mΔOD) shown in Figure 2b, which is proportional to the concentration of P3HT+, is indicative of the yield of charge separation at the Sb2S3:P3HT heterojunction. It is apparent from Figure 2b that upon increasing the Sb2S3 volume fraction in the Sb2S3:P3HT film, the yields of P3HT+ polarons are seen to increase. Figure 2b (inset) shows the plot of mΔOD determined at 1μs versus Sb2S3 fraction. It is clear from the data presented in Figure 2b that the highest yield of charge generation is observed in a film comprising 60:40 volume ratio of Sb2S3 to P3HT. We note that similar signal amplitudes of P3HT+ (magnitude of mΔOD at 950 nm) have been observed for other metal chalcogenide:P3HT systems, including CdS:P3HT films.9, 10, 34 This indicates that the Sb2S3:P3HT films reported here exhibit high yields of charge photogeneration. Transient absorption spectroscopy of Sb2S3:P3HT films. (a) Transient absorption spectrum of Sb2S3:P3HT (60:40) film recorded at a time interval of 10 μs following laser excitation at 567 nm, pump intensity (∼21 μJ/cm2). (b) Transient kinetics (following photoexcitation at 567 nm), obtained using a probe wavelength of 950 nm so as to monitor the recombination kinetics of P3HT+ polarons in the Sb2S3:P3HT films as a function of increasing Sb2S3 fraction. (b) inset shows a plot of the transient absorption signal at 1 μs versus Sb2S3:P3HT volume fraction. The data presented in Figure 2 indicate that P3HT+ polarons are generated upon photoexcitation the Sb2S3:P3HT film. Such P3HT+ polarons may be generated in two ways: (i) electron transfer from the photoexcited state of the polymer to the Sb2S3 conduction band and/or (ii) photoinduced hole transfer from the Sb2S3 to the P3HT polymer. To better understand the charge generation process in the Sb2S3:P3HT film, transient absorption spectroscopy studies were performed as a function of laser excitation wavelength. In the present study, the excitation pulses were tuned to excite different regions of the Sb2S3:P3HT absorption spectrum from 400–650 nm. We further note that the amplitudes of the resulting transient absorption signals were corrected for the number of photons incident upon the film at the excitation wavelength. In all cases, excitation of the Sb2S3:P3HT sample resulted in the appearance of P3HT+ polarons. Figure 3a shows the amplitude of the signal (mΔOD) measured at 1 μs as a function of excitation wavelength. It is evident from the data presented in Figure 3a that the P3HT+ polaron yield is strongly dependent on the excitation wavelength. Comparing the absorption spectra of Sb2S3:P3HT and Sb2S3:polystyrene films with the transient absorption data reveals that the P3HT+ polaron generation is considerably lower when the polymer component of the hybrid film is excited. Furthermore, it is apparent that the transient absorption data follows the light harvesting profile of the Sb2S3 semiconductor; this observation indicating that the electron transfer from the polymer to the metal chalcogenide is relatively inefficient. We therefore conclude that the most likely mechanism for the generation of P3HT+ polarons in the Sb2S3:P3HT film is a photoinduced hole-transfer from Sb2S3 to the polymer. (a) Transient absorption signals (ΔOD) (red squares) at a time delay of 10 μs presented as function of excitation wavelength. These transient absorption data monitor the P3HT+ polaron band at 950nm. The external quantum efficiency of the device is plotted as green circles. Also shown is the light harvesting profile of the Sb2S3:P3HT film (grey dash), Sb2S3 film (black curve). UV spectra are normalized to an arbitrary scale to highlight their shape. (b) Current–voltage characteristics of photovoltaic devices based upon a Sb2S3:P3HT (60:40) active layer. Devices were prepared in the “inverted” ITO/TiO2/CdS/Sb2S3:P3HT/PEDOT:PSS/Ag configuration. Further details of device fabrication can be found in the Experimental Section of the manuscript. We now turn our attention to photovoltaic devices based upon the Sb2S3:P3HT layers. Photovoltaic devices were fabricated using an inverted architecture as illustrated in Figure 3. Figure 3a shows the external quantum efficiency (EQE) of the device measured under monochromatic excitation (Figure 3a, green squares). It is clear from the data presented in Figure 3a that the EQE spectrum shows a close correlation with mΔOD excitation spectrum (and thus polaron yield) obtained from the transient absorption measurements. Both of these spectra follow the absorbance profile of the Sb2S3 component of the blend. Moreover, a comparison of the photocurrent generation and the transient absorption data suggests that the optical excitation of the P3HT polymer leads to less efficient charge generation and thus lower photocurrent. These observations confirm that charge separation and therefore current generation in the device results primarily from a hole transfer reaction, as discussed above. Typical current density–voltage traces for photovoltaic devices based upon Sb2S3:P3HT layers are shown Figure 3b. Despite the apparent lack of photocurrent generation from the P3HT, the overall power conversion efficiency of the device was measured under simulated AM1.5G sunlight to be 1.29% (FF = 0.46, Jsc = 3.85 mAcm−2, Voc = 0.73 V). We note that improvement in the device performance can be expected through optimization of the nanoscale morphology and the use of selective electrode contacts and interlayers. The next question that arises relates to the reason for superior charge generation yield upon photoexcitation of the Sb2S3 component of the hybrid film, as seen in Figure 3a. At present however, we can only speculate on the origin of these observations. Optical excitation of the Sb2S3:P3HT nanocomposite film is likely to result in the formation of electron-hole pairs in both P3HT and Sb2S3 components. The dissociation of tightly bound P3HT excitons into free charges requires the morphology of the hybrid film to be structured on the scale of the exciton diffusion length in the polymer (∼10 nm). This ensures that the excitons are generated in close proximity to a donor-acceptor interface, at which the coulombic attraction can be overcome and the charges separated. On the basis of the TEM data presented in Figure 1c, it is likely that only a small fraction of the P3HT excitons are able to reach the Sb2S3:P3HT heterojunction and undergo dissociation to free charges, in agreement with the data presented in Figure 3a. Our observations are consistent with those reported for Sb2S3:P3HT-based semiconductor-sensitised solar cells (SSSC). In these systems, excitons generated in the polymer hole transport phase cannot be separated, which causes a depression in IPCE in the spectral region where the polymer absorbs.18, 35 Additionally, we note that the yield of charge separation at the donor/acceptor heterojunction is expected to be sensitive to specific interactions between the organic and inorganic components,35, 36 which may not be conducive to electron transfer in our system. On the other hand, the photogenerated electrons and holes within the Sb2S3 nanocrystals would be easier to spatially separate into free charges owing to the higher dielectric constant (relative to P3HT) and the crystalline nature of the inorganic semiconductor. In the presence a good hole-extractor such as P3HT, a hole-transfer reaction could occur from the Sb2S3 to the polymer generating P3HT+ polarons leading to long-lived charge separation, as observed here. This is in agreement with our recent work on Sb2S3- based SSSCs, in which we show that hole transfer from Sb2S3 to an organic hole transport material can occur even when electron injection to TiO2 does not, so highlighting the crucial role of hole transfer in the charge generation process.37 This picture would help explain the current observations. More detailed studies addressing the charge generation process in Sb2S3:polymer films are currently underway and will be reported in due course. In summary, we have reported the fabrication and characterization of Sb2S3:P3HT photoactive layers and their application in photovoltaic devices. The hybrid nanocomposite films were fabricated by employing a method based on the in-situ thermal decomposition of an antimony ethyl-xanthate precursor in a P3HT polymer film. A combination of optical spectroscopy, Raman and transmission electron microscopy techniques has been used to confirm the growth of Sb2S3 nanocrystals in the polymer film. Transient absorption spectroscopy and photovoltaic device measurements indicate that charge separation and photocurrent generation in Sb2S3:P3HT devices results from Sb2S3 light absorption followed by hole-transfer from the Sb2S3 to P3HT. More generally, the present findings (particularly our observation of photo-induced hole transfer) should foster new strategies for the optimization of hybrid nanocomposite solar cells as well as the development of novel device architectures. Supporting Information is available from the Wiley Online Library or from the author. We thank the Engineering and Physical Sciences Research Council (EPSRC) for financial support via the Excitonic Supergen (EP/G031088/1) and UK-India (EP/H040218/2) programs. S.A.H thanks the Royal Society for a University Research Fellowship (RS-URF). We thank Dr. Simon King and Mr. Andrew MacLachlan for assistance with the transmission electron microscopy measurements. As a service to our authors and readers, this journal provides supporting information supplied by the authors. Such materials are peer reviewed and may be re-organized for online delivery, but are not copy-edited or typeset. Technical support issues arising from supporting information (other than missing files) should be addressed to the authors. Please note: The publisher is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.