Ab Initio Simulation of Charge Transfer at the Semiconductor Quantum Dot/TiO₂ Interface in Quantum Dot‐Sensitized Solar Cells

Abstract

to most dyes, due to their quantum-confi ned nature, semiconductor quantum dots (QDs), such as CdE (E = S, Se) or lower bandgap PbE, possessing size tunable broad absorption from visible (e.g., CdE) to near-infrared (e.g., PbE) range, narrow symmetric emission, and resistance to photobleaching under low-oxygen environment, [ 3 ] are ideally suited for light harvesting and mimicking photosynthesis.

Improving the Efficiency of Quantum Dot Sensitized Solar Cells beyond 15% via Secondary Deposition

“…The distinguished optoelectronic properties of quantum dot (QD) light absorbers, such as high stability toward light, heat, and moisture, high extinction coefficient, multiple exciton generation (MEG) possibility, solution processability, etc., make QD-sensitized solar cells (QDSCs) a promising low-cost third-generation photovoltaic cell with a theoretical power conversion efficiency (PCE) of up to 44%. − Benefiting from the exploitation of near-infrared absorption QD sensitizers and the high-efficiency photogenerated electron extraction, − the development of high-performance counter electrodes, especially Ti-mesh-supported mesoporous carbon counter electrodes (MC/Ti CE), − as well as an interface engineering strategy using an energetic barrier layer to suppress undesired charge recombinations, − the certified PCE record of QDSCs has been improved from less than 5% to over 14% in the past decade. , However, this value is still far less than the corresponding theoretical value (44%), as well as those for other emerging solar cells. , One of the main reasons for the intermediate photovoltaic performance of QDSCs is due to the low QD loading (i.e., low QD coverage) on the photoanode surface and the concomitant insufficient light-harvesting capacity as well as charge recombination loss in the cell device. ,− A high QD loading on photoanode surface is a prerequisite for high photocurrent and high photovoltaic performance in QDSCs. − This is because high QD loading can reduce the necessary thickness of a sensitized photoanode to capture all incident solar photons. A thin photoanode means a short transportation path for photogenerated electrons through the photoanode to the current collector plate and consequently a reduced possibility for undesirable charge recombination. − Furthermore, a high QD loading corresponds to a decrease in the proportion of the uncovered TiO 2 surface directly exposed to the electrolyte and less possibility of photogenerated electrons captured by the redox couple in the electrolyte, thereby benefiting the photovoltaic performance, the especially photovoltage and fill factor, of a cell device. − …”

First-Principles Study of the Electron–Hole Recombination Rate at the Interface of the CdSe Quantum Dot and TiO2 Substrate

“…Most studies have focused on electron injection because the initial charge separation in the photoanode is the primary event that leads to the conversion of solar energy into electric current. The electron injection rate ( k inj ) can be modulated by tuning the conduction band offsets between the oxide semiconductor and QDs. − Experimentally, the electron injection rate from CdSe QDs to TiO 2 measured with the transient absorption spectrum was much faster for smaller-sized QDs, k inj , changing from 10 11 to 10 7 s –1 when the size increased from 2.4 to 7.5 nm in diameter . The k inj variations are interpreted with the nonadiabatic Marcus theory, in which a wider conduction band offset between the QD and TiO 2 leads to a larger driving force for charge transfer in small QDs. − Theoretically, Tafen and Prezhdo found that the small-sized Cd 6 Se 6 QD has the fastest electron injection among their simulated CdSe–TiO 2 systems.…”

“…The electron injection rate ( k inj ) can be modulated by tuning the conduction band offsets between the oxide semiconductor and QDs. − Experimentally, the electron injection rate from CdSe QDs to TiO 2 measured with the transient absorption spectrum was much faster for smaller-sized QDs, k inj , changing from 10 11 to 10 7 s –1 when the size increased from 2.4 to 7.5 nm in diameter . The k inj variations are interpreted with the nonadiabatic Marcus theory, in which a wider conduction band offset between the QD and TiO 2 leads to a larger driving force for charge transfer in small QDs. − Theoretically, Tafen and Prezhdo found that the small-sized Cd 6 Se 6 QD has the fastest electron injection among their simulated CdSe–TiO 2 systems. Xin et al demonstrated that the conduction band offsets between CdSe (PbSe) QDs and the TiO 2 substrate increase with the decreasing QD size, which facilitates the charge transfer from QD donors to TiO 2 acceptors.…”

Artificial Intelligence in Materials Modeling and Design

“…Applicable to high-dimensional data; Good performance on feature classification Need for a large number of samples; Ambiguous physical meaning [48] In the past few decades, with rapid advances in computing power, material science and numerical modeling, many researchers have focused on the development of new techniques to accurately and efficiently predict the properties of several important engineering materials [49][50][51][52][53]. For instance, quantum mechanics and density functional theory (DFT) can be used to study and predict the properties of isolated molecules, bulk solids, and materials interface [54][55][56].…”