Minlin Jiang

Enhancing the performance of planar organo-lead halide perovskite solar cells by using a mixed halide source

Cl incorporation has been reported to be able to significantly increase the diffusion length of carriers in organo-lead halide perovskite thin films. However, due to the low solubility of PbCl2, Cl incorporation has been rarely reported in a two-step process, which is capable of producing uniform organo-lead halide perovskite thin films. In this letter, we report a novel growth method that combines the two-step process and Cl incorporation. The Cl-incorporated organo-lead halide perovskite solar cell made from a two-step process demonstrated a power conversion efficiency of 10.5%, which is 27% higher than that without Cl incorporation. Further investigation was performed to explore the fundamentals of the enhancement. Kelvin probe force microscopy (KPFM) measurement revealed a larger band bending at grain boundaries with Cl incorporation, which brings the Fermi level of the bulk perovskite thin film closer to the center of the bandgap. As a result, a p–i–n type of junction is formed in the devices with Cl incorporation, which facilitates the charge carrier collection. Additional carrier lifetime measurements indicate that the electron lifetime in the perovskite thin film with Cl incorporation is longer, indicating reduced recombination in the devices with Cl incorporation.

Observation of lower defect density in CH3NH3Pb(I,Cl)3 solar cells by admittance spectroscopy

The introduction of Cl into CH3NH3PbI3 precursors is reported to enhance the performance of CH3NH3PbI3 solar cell, which is attributed to the significantly increased diffusion lengths of carriers in CH3NH3Pb(I,Cl)3 solar cell. It has been assumed but never experimentally approved that the defect density in CH3NH3Pb(I,Cl)3 solar cell should be reduced according to the higher carrier lifetime observed from photoluminescence (PL) measurement. We have fabricated CH3NH3Pb(I,Cl)3 solar cell by adding a small amount of Cl source into CH3NH3PbI3 precursor. The performance of CH3NH3Pb(I,Cl)3 solar cell is significantly improved from 15.39% to 18.60%. Results from scanning electron microscopy and X-ray diffraction indicate that the morphologies and crystal structures of CH3NH3PbI3 and CH3NH3Pb(I,Cl)3 thin films remain unchanged. Open circuit voltage decay and admittance spectroscopy characterization jointly approve that Cl plays an extremely important role in suppressing the formation of defects in perovskite solar cells.

The application of Al2TiO5 at the TiO2/perovskite interface to decrease carrier losses in solar cells

Herein we describe the mechanism of crystallization of Al3+ with amorphous TiO2 on a TiO2 film, and found that the resultant Al2TiO5 was able to improve the average efficiency of planar perovskite solar cells (PSCs) from 15.4 to 17.2%. The electron lifetime in these Al-PSCs was about 3 times longer than that in unmodified PSCs. The O 1s X-ray photoelectron spectrum of the TiO2/CH3NH3PbI3 interface indicated that the oxygen vacancy or defect peak decreased by more than 31% on modifying the interface with Al2TiO5. The stability of the PSCs was improved substantially: Al-PSCs had still 88% of their initial value after 1440 hours, while the efficiency of the reference PSCs decreased to 68% of their initial value.

Effects of Se vapor annealing on water-based solution-processed Cu2ZnSn(S,Se)4 thin-film solar cells

Abstract. Among various deposition techniques to deposit Cu2ZnSn(S,Se)4 (CZTSSe) thin films, solution-based processes have attracted considerable attention because of their potentially low cost. Most of the reported solution-based methods are based on nonaqueous solvents, such as hydrazine, and organic solvents. We report the deposition of CZTSSe thin films and fabrication of CZTSSe solar cells by a water-based, solution-processed method followed by Se vapor annealing. The effects of Se vapor feeding time on the properties of CZTSSe thin films and the performance of CZTSSe solar cells are investigated. The ratio of Se/(Se+S) can be tuned by changing the Se vapor feeding time. Our results indicate that extending the Se vapor feeding time increases the band gap, slightly increases the lattice constant, and significantly improves the morphologies of the CZTSSe thin films. A remarkable enhancement in the performance was observed from the CZTSSe solar cells annealed with a longer Se vapor feeding time compared with those without Se feeding.

Theoretical and experimental study of dielectrophoretic force controlled nanowires assembly

Because of their appealing characteristics, quasi one-dimensional nanostructures like semiconducting nanowires have drawn much attention in the past decade. They can be utilized in a variety of applications such as field-effect transistors, chemical sensors, and photodectors. These devices are expected to deliver unprecedented performance because of the quantum confinement and the large surface-to-volume ratio. To fabricate these practical devices, nanowires need to be manipulated towards pre-defined electrodes in a controllable fashion. Enlightened by Pohls pioneering contribution in dielectrophoresis (DEP), many researchers have utilized an electric field to align the nanowires. However, assembly of nanowires by DEP are studied mainly through experimental observations so far, very little theoretical studies have been reported, which could provide useful guidance in real application. In this paper, we aim to theoretically analyze the assembly of nanowires by DEP under real conditions. In a non-uniform electric field, the nanowire is polarized and can be treated as an effective dipole, thereby receiving dielectrophoretic (DEP) force and torque, while the motion of the nanowire is hampered by the hydrodynamic drag force and drag torque, which increases the difficulty and complexity in theoretical analysis on nanowire trajectory. In our previous work, a model involving different forces and torques is constructed to simulate the nanowire trajectory. However, the Brownian motion of nanowires, which could be significant, is ignored. In this paper, a more comprehensive model considering DEP force, DEP torque, drag force, drag torque, and Brownian motion is constructed to predict the nanowire trajectory by simulation. Depending on the attaching point of the nanowire on the electrode and the length of the nanowire, the nanowire is predicted to either align with the direction of the electric field or bridge the electrodes ultimately. We also fabricated a pair of microelectrodes and implemented experiments to dielectrophoretically assemble nanowires. The experimental results agree with the simulation results. Such agreement validated our theoretical analysis.

The characterization of defects states and charge injection barriers in perovskite solar cells

The power conversion efficiency of perovskite solar cells has exceeded 22% within a few years. To further optimize their performance, the focus should be on minimizing the defects and barriers in the devices. In this work, Kelvin probe force microscopy (KPFM) together with admittance spectroscopy (AS) is utilized in the characterization of charge injection barriers and defects states in perovskite solar cells. Our measurements indicate that perovskite has a defect level of 0.25 eV above its valance band. In addition, charge injection barriers are present in the interface between electron transporting layer (ETL) and perovskite layer. To improve the performance, passivation technique is needed for the removal of defects and charge injection barriers which could act as recombination centers in perovskite solar cells.

Fabrication, calibration, and recovery of chemical nanosensor array for ammonia detection

The low selectivity of nanosensors is one of their major obstacles for their wide deployment. To enhance the selectivity, nanosensor array made from zinc oxide (ZnO) nanowires and carbon nanotubes (CNTs) was assembled through dielectrophoresis (DEP). The fabricated nanosensor array was used to detect ammonia (NH3) in a well-controlled environment at room temperature. Because of their opposite material types, ZnO nanowire based sensor behaved oppositely to CNT based sensor. In this study, it is also demonstrated that DC biases can quickly recover both sensing elements. After collecting sensing signals from two transducers under different NH3 concentrations, the concentration of NH3 can be estimated through regression methods. It is shown that quadratic model with the lasso performs well on the collected data.

Reconstruction of Kelvin probe force microscopy image with experimentally calibrated point spread function

A Kelvin probe force microscopy (KPFM) image is sometimes difficult to interpret because it is a blurred representation of the true surface potential (SP) distribution of the materials under test. The reason for the blurring is that KPFM relies on the detection of electrostatic force, which is a long-range force compared to other surface forces. Usually, KPFM imaging model is described as the convolution of the true SP distribution of the sample with an intrinsic point spread function (PSF) of the measurement system. To restore the true SP signals from the blurred ones, the intrinsic PSF of the system is needed. In this work, we present a way to experimentally calibrate the PSF of the KPFM system. Taking the actual probe shape and experimental parameters into consideration, this calibration method leads to a more accurate PSF than the ones obtained from simulations. Moreover, a nonlinear reconstruction algorithm based on total variation (TV) regularization is applied to KPFM measurement to reverse the blurring caused by PSF during KPFM imaging process; as a result, noises are reduced and the fidelity of SP signals is improved.

Study of annealing induced nanoscale morphology change in organic solar cells with machine learning

Nanoscale morphology of the active layer in organic solar cells (OSCs) plays a significant role in affecting the overall performance of OSCs. In this work, by using the domain size to describe the morphology change with different annealing temperatures, the influence of annealing to nanoscale morphology in OSCs is intensively studied. Also, machine learning is utilized in investigating the relation between domain size and the annealing temperature. The realization of mapping the domain sizes to their annealing temperatures is significant in bridging the simulation work of OSCs to the actual fabrication process, indicating that the optimized domain sizes obtained through simulation could now be converted to actual annealing temperature in device fabrication.