A dynamic detection system for charge transfer process of semiconductor heterojunction interface

Abstract

This invention relates to a dynamic detection system for the charge transfer process at the interface of a semiconductor heterojunction, belonging to the fields of semiconductor characterization and transient spectroscopy. It includes an induced pump optical path, a detection signal optical path, a biased white light source, a circuit synchronization control module, and a sample stage, enabling real-time monitoring of the transient absorption dynamics curves of optoelectronic devices on the scale of hundreds of nanoseconds to milliseconds. Its advantages lie in utilizing a delayed pulse generator with a laser clock as the core to precisely control the timing of each optoelectronic hardware component, ensuring extremely high temporal resolution; the introduction of a filter wheel effectively reduces the overall intensity of the detection light, minimizing additional interference from the detection light on the interface charge transfer process; and the inclusion of an independent biased white light source to simulate real sunlight irradiation enables dynamic characterization of optoelectronic devices in practical application environments, providing a novel and efficient analytical method for optimizing photovoltaic device performance and photocatalytic system design.

Description

Technical Field

[0001] This invention belongs to the field of semiconductor characterization and transient spectroscopy technology, specifically relating to a dynamic detection system for the charge transfer process at the interface of a semiconductor heterojunction. Background Technology

[0002] Against the backdrop of global energy structure transformation and rapid development of semiconductor technology, charge transfer processes at semiconductor heterojunction interfaces have become a core physical mechanism determining the performance of optoelectronic devices, photocatalytic systems, and novel high-energy physics conversion devices. Whether in third-generation thin-film solar cells (such as dye-sensitized solar cells and perovskite solar cells), efficient semiconductor photocatalytic hydrogen production systems, or artificial photosynthesis simulation devices, the injection, separation, migration, and subsequent recombination and regeneration kinetics of photoexcited charge carriers at the heterojunction interface are directly related to the device's photoelectric conversion efficiency, quantum yield, and long-term operational stability. For researchers, obtaining time-resolved signals of these microscopic kinetic processes on the nanosecond (ns) to millisecond (ms) scale is a crucial prerequisite for revealing internal energy loss paths, optimizing material energy level matching, and improving electrode interface construction processes. Therefore, developing a high-precision, in-situ monitoring system for dynamic testing of interface charge transfer is not only an urgent need for basic scientific research to delve deeper into the subject, but also a vital link in promoting independent innovation in my country's semiconductor characterization technology and strengthening its research support capabilities.

[0003] Currently, the most common and effective technique for characterizing the charge dynamics of semiconductor interfaces is time-resolved transient absorption spectroscopy based on the principle of "flash photolysis." This technique typically employs a "pump-induction" logic, using a high-energy pulsed laser as an induction source to instantaneously excite the sample, causing it to transition from the ground state to a non-equilibrium excited state. Subsequently, a steady-state white light beam is used as the probe light to penetrate the sample, and the decay of the concentration of excited-state species or free carriers over time is tracked by monitoring the transient evolution curve of the transmitted light intensity. However, although this technique is relatively mature at the theoretical physics level, existing testing equipment and characterization schemes still have many significant systemic defects in practical engineering implementation and in-depth scientific research applications, making it difficult to meet the growing demand for high-performance characterization.

[0004] First, existing commercial high-precision transient absorption spectroscopy testing systems (represented by Edinburgh Instruments and Nippon Spectroradiometer) are extremely complex in structure, typically integrating expensive nanosecond lasers, high-sensitivity dynamo detectors, and sophisticated optical monochromator systems. This results in prohibitively high purchase costs, with the investment in a single unit often reaching millions of RMB, and subsequent operating environment requirements are extremely stringent, leading to substantial maintenance and upkeep costs. For the vast majority of small and medium-sized research institutions, local university laboratories, and R&D departments of startups in China, this constitutes a very high financial barrier and entry threshold, severely limiting the adoption rate of interfacial charge transfer, a core characterization technology, in basic research and industrial improvement.

[0005] Secondly, existing technologies have significant gaps in simulating the in-situ operating conditions of semiconductor devices. In real-world applications, photovoltaic devices or photocatalytic systems are not located in dark rooms but operate under continuous solar irradiation. Under these conditions, constant background illumination generates a high density of steady-state carriers. These carriers occupy defect state energy levels within the semiconductor, significantly altering the Fermi level distribution at the interface and the electric field strength of the space charge layer, thereby greatly modulating the recombination lifetime and charge transfer rate of transient carriers. However, most commercially available characterization equipment is designed to operate in complete darkness. The kinetic parameters obtained from this "dark-state testing" are essentially the material's response near equilibrium, often deviating significantly from the dynamic behavior of the device under real illumination. For example, in the study of dye-sensitized solar cells, without a bias light simulation of the operating environment, the test results will be completely unable to assess the true impact of carrier concentration under real light intensity on the rate of electrolyte reduction of dye cations (dye regeneration). This disconnect between laboratory static measurements and actual service performance makes it difficult to effectively translate a large amount of scientific research data into engineering improvement suggestions, becoming a hidden bottleneck restricting the development of high-performance devices.

[0006] From a deeper physical mechanism perspective, the lack of this "in-situ simulation" can lead to misjudgments of the interface recombination mechanism. The recombination rate on semiconductor surfaces typically exhibits strong nonlinear characteristics, limited by the degree of surface state filling. Under real-world service conditions, continuous bias light irradiation produces a "shielding effect" or "filling effect," causing certain recombination channels to be closed or activated. If the test system lacks the ability to superimpose bias light sources, it only captures a simplified physical picture under "cold equilibrium," neglecting the most critical competitive reaction dynamics in the photoelectric conversion process. Therefore, how to apply precisely controllable simulated background light while simultaneously performing transient detection is a key technical challenge for achieving high-quality dynamic characterization.

[0007] Furthermore, the contradiction of "observation interference" with the probe light is another major obstacle restricting the accuracy of testing. In transient absorption experiments, in order to obtain a sufficient signal-to-noise ratio within an extremely short time (nanosecond level), the system often needs to significantly increase the output intensity of the steady-state probe light. However, for semiconductor heterojunctions with broad spectral responses or narrow bandgap nanomaterials, the high-intensity probe white light itself contains enough energy to induce secondary excitation of charge carriers or generate local thermal effects in the micro-regions of the sample, thereby causing severe secondary perturbations to the evolution process of charge carriers that were originally in the laser-excited state. This phenomenon results in the observed lifetime data not being an inherent property of the sample, but rather a distorted result due to interference from the observation method. Existing equipment lacks flexible modulation methods to balance "signal clarity" and "probe interference," especially when it comes to long-lifetime (millisecond-level) signal acquisition, where the effects of photochemical degradation or thermal effects are particularly prominent.

[0008] Furthermore, the timing integration and high-precision scheduling of multiple devices and dimensions are also bottlenecks in engineering implementation. Interface charge transfer signals are typically extremely weak, and their dynamic evolution spans multiple orders of magnitude from ultrafast injection (hundreds of nanoseconds) to slow diffusion and recombination (milliseconds). To achieve complete capture of the entire process, the system must achieve high coordination at the nanosecond level with a nanosecond laser (as a pump source), a filter wheel (as a modulator), a monochromator (as a spectrum selection unit), a bias light source, and a high-speed digital oscilloscope (as an acquisition terminal). In currently common home-built systems, due to the lack of a unified clock reference among the hardware components, there is often significant trigger jitter and uncontrollable system latency, resulting in signal loss or severe distortion in the acquired curves near t=0. This timing instability directly leads to inaccurate fitting of the dynamics of the ultrafast interface process.

[0009] Finally, data post-processing and multi-field coupling analysis are also weak points in existing technologies. Interface charge processes are often accompanied by spectral overlap and complex non-exponential decay behavior, requiring comprehensive analysis combining various data such as device JV characteristics and electrochemical impedance spectroscopy. Existing single-function devices lack overall control software and physical model fitting algorithms, forcing researchers to manually switch between multiple software platforms to process data. This is not only inefficient but also makes it difficult to establish a precise mapping relationship between "material microdynamics and device macroscopic performance".

[0010] In summary, to address a series of core pain points in existing characterization equipment, such as extremely high cost, lack of service condition simulation, significant interference from probe light, insufficient system synchronization accuracy, and fragmented data analysis, we have developed a dynamic testing device for semiconductor heterojunction dynamics based on the flash photolysis principle, integrating a bias light simulation system, a controllable light intensity modulation unit, and high-precision hardware delay scheduling logic. This device has profound technical significance and value for enhancing my country's original innovation capabilities and engineering R&D level in the fields of photovoltaics, photoelectric detection, and photochemistry. Summary of the Invention

[0011] This invention provides a dynamic detection system for the charge transfer process at the semiconductor heterojunction interface, which solves many technical problems of existing semiconductor heterojunction interface dynamic detection systems, such as complex structure, extremely high cost, inability to effectively simulate real-world illumination conditions, and the tendency of high-intensity detection light to generate severe secondary interference.

[0012] The technical solution adopted in this invention includes an induced pump optical path, a detection signal optical path, a biased white light source, a circuit synchronization control module, and a sample stage. In the induced pump optical path, a nanosecond laser serves as the induction source to excite the semiconductor heterojunction sample placed on the sample stage, inducing an interface charge transfer process. In the detection signal optical path, the detection light emitted by the steady-state white light source passes through a filter wheel and then through the semiconductor sample. The light is then spectrally selected by a monochromator and coupled to a photomultiplier tube. An independent biased white light source applies continuous bias light irradiation to the semiconductor heterojunction sample to simulate a solar background. The high-precision delay pulse generator in the circuit synchronization control module controls the dynamic detection time of the interface charge transfer in the semiconductor heterojunction sample. A digital oscilloscope measures and records the dynamic curve of the interface charge transfer process, and the data is transmitted in real-time to a computer for data storage and post-processing.

[0013] The induced pump optical path of the present invention includes a nanosecond laser, a reflector, and a convex lens. The high-energy pulsed laser emitted by the nanosecond laser is reflected by the reflector and focused by a convex lens, so that it irradiates the surface of the semiconductor heterojunction sample on the sample stage, instantly generating a non-equilibrium carrier distribution.

[0014] The focal length of the convex lens described in this invention is 5cm.

[0015] The detection signal optical path of the present invention includes a steady-state white light source, a filter wheel, a second convex lens, a third convex lens, a photomultiplier tube, and a monochromator. The steady-state white light source emits continuous broadband detection light. This beam first passes through the filter wheel to modulate its intensity. The modulated detection light is then focused by the second convex lens onto the stimulated region of the sample stage. After penetrating the semiconductor heterojunction sample, it carries an absorption signal indicating the change in carrier concentration. The third convex lens then collects the transmitted light and guides it to the monochromator. The monochromator extracts monochromatic light according to a preset wavelength and couples it to the photomultiplier tube. The photomultiplier tube converts the time evolution of the light intensity into a current signal, which is finally input to a digital oscilloscope for recording and displayed on a computer as a dynamic curve of absorbance decay over time.

[0016] The steady-state white light source described in this invention utilizes xenon gas to emit light.

[0017] The filter wheel of this invention has 12 standard slots, in which 6 neutral density filters with different transmittances are installed, with OD values ​​of 0.3, 0.5, 1.0, 1.5, 2.0, and 3.0, respectively, to precisely modulate the radiation intensity of the probe light to avoid secondary excitation interference to the sample; at the same time, 6 narrowband interference filters with center wavelengths of 400nm, 450nm, 515nm, 532nm, 600nm, and 700nm are also installed, to screen the probe signal of a specific wavelength band according to experimental requirements, to ensure the purity of the signal and prevent stray light interference.

[0018] The focal length of the second convex lens and the focal length of the third convex lens of the present invention are both 5 cm.

[0019] The biased white light source described in this invention uses an adjustable intensity white LED, whose spectral range and light intensity have been calibrated to simulate the lighting environment of a solar cell under real-world operating conditions.

[0020] The circuit synchronization control module of this invention includes a digital oscilloscope, a computer, and a high-precision delay pulse generator. The high-precision delay pulse generator serves as the trigger core of the entire system. Its output port is connected to the external trigger interface of the nanosecond laser via a BNC shielded cable. Simultaneously, another synchronization output port of the high-precision delay pulse generator is connected to the external trigger channel of the steady-state white light source via a signal line. The signal input channel of the digital oscilloscope is connected to the electrical signal output port of the photomultiplier tube, converting the captured transient light intensity changes into voltage signals that are transmitted to the digital oscilloscope for waveform recording. The digital oscilloscope is connected to the computer via a USB or GPIB data cable. At the same time, the computer is connected to the monochromator via a control interface to precisely adjust the rotation angle of the internal grating of the monochromator, thereby realizing the automatic selection and switching of the detection wavelength.

[0021] The high-precision delay pulse generator described in this invention is a Stanford DG645.

[0022] The advantage of this invention lies in its novel structure. A semiconductor nanosecond laser serves as the induced pump source, emitting high-energy nanosecond pulsed laser light to instantaneously excite the semiconductor heterojunction sample in the test area, thereby inducing a non-equilibrium interface charge transfer process. To capture this transient evolution process in real time, the system uses a steady-state white light source to emit a broadband probe beam that penetrates the sample. The probe light transmitted through the sample then enters a high-resolution monochromator for spectral separation. The monochromator extracts monochromatic transmitted light of a specific wavelength and couples it to a high-sensitivity photomultiplier tube. The photomultiplier tube is responsible for converting the captured weak transient photoelectric signal into a current signal, which is then input to a high-bandwidth digital oscilloscope. Finally, the oscilloscope measures and records the dynamic evolution curve of the transient photoelectric absorption signal over time in real time. Details are as follows:

[0023] (I) This invention adds an independently biased white light source with adjustable intensity around the sample cell to apply continuous background light irradiation to the sample. This breaks the limitation that traditional transient absorption tests must be performed in a dark room, and can realistically simulate the actual service environment of dye-sensitized solar cells or semiconductor photocatalytic systems under sunlight irradiation. This allows the system to accurately diagnose key dynamic processes such as dye regeneration, carrier migration, and interfacial charge recombination under device service conditions. This invention successfully bridges the gap between "dark-state testing" and "service conditions," realizing dynamic diagnosis of semiconductor devices under real light conditions. The obtained parameters such as interfacial charge transfer lifetime and recombination rate directly correspond to the actual operating performance of the device, and have extremely high scientific research guidance value.

[0024] (ii) This invention utilizes a high-precision delayed pulse generator to coordinate the overall timing. Without relying on expensive ultrafast detection equipment, it still achieves extremely high hardware synchronization accuracy and can seamlessly cover the entire time window signal capture from ultrafast charge injection to slow electrolyte regeneration.

[0025] In terms of timing control and environmental simulation in this invention, the core control logic involves the introduction of a high-precision delayed pulse generator. This generator uses the internal emission clock pulse of the nanosecond laser as the global time reference for the entire system, and sends synchronous trigger pulses with nanosecond-level precision to the steady-state white light detection unit. Through this precise timing allocation mechanism centered on laser triggering, the system can stably control the relative time offset between the laser pulse and the data sampling point, synchronously triggering the laser pumping and signal acquisition windows. This ensures that pumping excitation and signal acquisition are tightly synchronized over an extremely wide timescale from hundreds of nanoseconds to several milliseconds, effectively eliminating timing jitter during multi-device joint debugging, and enabling continuous and stable data acquisition of transient absorption signals generated by interface charge transfer.

[0026] (III) This invention incorporates a filter wheel in the detection optical path to modulate the intensity and wavelength of the detection light output from the steady-state white light source. By inserting filters of specific wavelengths or transmittance into the optical path in real time, the radiation intensity of the detection light is precisely modulated. While maintaining the signal-to-noise ratio of the monochromatic detection light, the overall radiation intensity of the white light is significantly reduced, minimizing observational interferences such as secondary excitation and local thermal effects caused by the detection light on the sample. The introduction of the filter wheel cleverly solves the problem of secondary interference from high-frequency steady-state white light on weak charge transfer signals, thereby effectively suppressing thermal effect errors or additional induced interference caused by high-intensity detection light on the interface charge transfer process, improving the authenticity and accuracy of the test results, and ensuring the purity and authenticity of the kinetic curve from a physical source.

[0027] In summary, the overall architecture of this invention is based on the physical detection principle of "flash photolysis". It can not only achieve high-precision detection with a wide time window through precise global timing control, but also realistically simulate the actual working environment of photovoltaic devices. It has excellent scalability and in-situ detection capability. It uses nanosecond laser pulses to induce interface charge transfer and combines steady-state detection light and time-resolved detection technology to detect the dynamic signal of interface charge transfer process in semiconductor devices or photocatalytic systems, and accurately monitor and analyze the lifetime of interface charge process.

[0028] This invention provides a powerful tool for understanding the working mechanism of photovoltaic devices and optimizing photocatalytic systems, and offers an innovative, reliable, and cost-effective physical characterization method for performance diagnosis and mechanism analysis of high-efficiency semiconductor devices. It possesses diagnostic capabilities for key kinetic processes within high-performance dye-sensitized solar cells and photocatalytic systems. By probing signals at different time scales, it can distinguish and measure the "dye regeneration" process between dye cations and electrolytes (nanosecond to microsecond scale), and the "charge recombination" process between semiconductor interfaces and electrolyte materials (microsecond to millisecond scale). By analyzing the rate or lifetime of these interfacial charge transfer channels, it provides data support for optimizing device material ratios and improving manufacturing processes, providing a novel high-precision characterization system for the development of semiconductor optoelectronic devices and energy conversion systems. Attached Figure Description

[0029] Figure 1 This is a schematic diagram of the present invention. Detailed Implementation

[0030] See Figure 1The system includes an induced pump optical path, a detection signal optical path, a biased white light source, a circuit synchronization control module, and a sample stage 7. In the induced pump optical path, a nanosecond laser 1 serves as the induction source to excite the semiconductor heterojunction sample placed on the sample stage 7, generating an interface charge transfer process. In the detection signal optical path, the detection light emitted by the steady-state white light source 4 passes through the semiconductor sample after passing through the filter wheel 5. The monochromator 10 performs spectrum selection and couples it to the photomultiplier tube 9, converting the instantaneous spectral information and intensity evolution law contained in the transmitted detection light into an electrical signal. The independent biased white light source 13 applies continuous bias light irradiation to the semiconductor heterojunction sample to simulate the solar background, enabling the photovoltaic device to enter a real service working state, thereby realizing real-time detection of the interface charge dynamics under device service conditions. The high-precision delay pulse generator 14 of the circuit synchronization control module controls the dynamic detection time of the interface charge transfer of the semiconductor heterojunction sample. The dynamic curve of the interface charge transfer process is measured and recorded by a digital oscilloscope 11 and transmitted in real time to a computer 12 for data storage and post-processing.

[0031] The induced pump optical path includes a nanosecond laser 1, a reflector 2, and a convex lens 3. The high-energy pulsed laser emitted by the nanosecond laser 1 is reflected by the reflector 2 and focused by a convex lens 3, so that it irradiates the surface of the semiconductor heterojunction sample on the sample stage 7, instantly generating a non-equilibrium carrier distribution.

[0032] The focal length of the convex lens 3 is 5cm.

[0033] The nanosecond laser is: Spectrophysic, PRIMOSCAN / ULD-240.

[0034] The detection signal optical path includes a steady-state white light source 4, a filter wheel 5, a second convex lens 6, a third convex lens 8, a photomultiplier tube 9, and a monochromator 10. The steady-state white light source 4 emits continuous broadband detection light. This beam first passes through the filter wheel 5 to modulate its intensity. By switching neutral density filters with different transmittances, the intensity of the detection light is modulated to a lower level to avoid additional thermal effects or secondary excitation of the sample by high-intensity white light. The modulated detection light is focused by the second convex lens 6 onto the stimulated region of the sample stage 7. After penetrating the semiconductor heterojunction sample, it carries an absorption signal indicating the change in carrier concentration. The third convex lens 8 then collects the transmitted light and guides it to the monochromator 10. The monochromator 10 extracts monochromatic light according to a preset wavelength and couples it to the photomultiplier tube 9. The photomultiplier tube 9 converts the time evolution of light intensity into a current signal, which is finally input to a digital oscilloscope 11 for recording and displayed on a computer 12 as a dynamic curve of absorbance decay over time.

[0035] The steady-state white light source 4 uses xenon gas to emit light and employs Energetiq, EQ-99X-NM-S-NA.

[0036] The monochromator model is: Zhuoli Hanguang, Omni-λ3027i.

[0037] The filter wheel 5 has 12 standard slots, in which 6 neutral density filters with different transmittances are installed, with OD values ​​of 0.3, 0.5, 1.0, 1.5, 2.0, and 3.0, respectively, to precisely modulate the radiation intensity of the probe light to avoid secondary excitation interference to the sample; at the same time, 6 narrowband interference filters with center wavelengths of 400nm, 450nm, 515nm, 532nm, 600nm, and 700nm are also installed, to screen the probe signal of a specific wavelength band according to experimental requirements, to ensure the purity of the signal and prevent stray light interference.

[0038] The filter wheel modulates the intensity of the probe light to a lower level by inserting specific filters into the optical path, so as to avoid the high-intensity white light from generating additional thermal effects or secondary excitation on the sample. While ensuring that the intensity of a single probe wavelength meets the signal-to-noise ratio requirements, it significantly reduces the overall radiation intensity of the probe white light, thereby effectively suppressing the thermal effect error or additional induced interference generated by the high-intensity probe light on the interface charge transfer process, and improving the authenticity and accuracy of the test results.

[0039] The focal length of the second convex lens 6 is 5cm, and the focal length of the third convex lens 8 is 5cm.

[0040] The biased white light source 13 employs an adjustable-intensity white LED, whose spectral range and light intensity have been calibrated to simulate the lighting environment of a solar cell under real-world operating conditions. The steady-state background light provided by the biased light source can effectively fill the trapped states inside the semiconductor and induce a shift in the Fermi level.

[0041] The circuit synchronization control module includes a digital oscilloscope 11, a computer 12, and a high-precision delay pulse generator 14. The high-precision delay pulse generator 14 serves as the trigger core of the entire system. Its output port is connected to the external trigger interface of the nanosecond laser 1 via a BNC shielded cable to control the emission time of the pulsed laser. This point is defined as the zero point of physical evolution, t=0. Meanwhile, another synchronization output port of the high-precision delay pulse generator 14 is connected to the external trigger channel of the steady-state white light source 4 via a signal line. By manually or by preset adjustment of the relative delay Δt between the two signals output by the high-precision delay pulse generator 14, the dynamic detection time of charge transfer at the interface of the semiconductor heterojunction sample can be precisely controlled. The signal input channel of the digital oscilloscope 11 is connected to the electrical signal output port of the photomultiplier tube 9, which converts the captured transient light intensity change into a voltage signal and transmits it to the digital oscilloscope 11 for waveform recording. The digital oscilloscope 11 is connected to the computer 12 via a USB or GPIB data cable to ensure that the acquired dynamic curve can be transmitted to the computer 12 in real time for data storage and post-processing. At the same time, the computer 12 is connected to the monochromator 10 via a control interface to precisely adjust the rotation angle of the internal grating of the monochromator, thereby realizing the automatic selection and switching of the detection wavelength. In the entire circuit logic, the high-precision delay pulse generator 14 is independently responsible for the hardware-level high-precision timing phase-locking, while the computer 12 is mainly responsible for the acquisition of data at the data terminal and spectral analysis.

[0042] The system employs highly synchronized timing control logic, using the clock signal inside the semiconductor nanosecond laser as the core reference. It utilizes a digital delay pulse generator to allocate and regulate the timing of various hardware devices, such as the semiconductor laser and the steady-state white light source. Through this precise timing scheduling, the system can synchronously trigger the laser pump and signal acquisition window, ensuring continuous and stable data acquisition of transient absorption signals generated by interface charge transfer within a timescale of hundreds of nanoseconds to several milliseconds.

[0043] The system integrates specially developed control software and a user interface. The software is responsible for remotely tuning and setting the operating parameters of various subsystems such as nanosecond lasers, monochromators, and digital oscilloscopes. It is also used for real-time data recording, graphical display, and physical model fitting. By combining the energy conversion efficiency and IV characteristic curves of solar cells, the software can help researchers fully reveal the key charge transfer bottlenecks that restrict the performance of semiconductor devices.

[0044] The digital oscilloscope model is: Tektronix, DPO70404C.

[0045] The high-precision delay pulse generator 14 is a Stanford DG645.

[0046] Working principle of the invention:

[0047] (1) The operating principle of this invention is based on the "pump-probe" transient absorption technology and in-situ service condition simulation logic. Its core process is as follows: a nanosecond pulsed laser is used as an induction source to instantaneously excite the semiconductor heterojunction sample, causing it to generate non-equilibrium carriers and triggering interface charge transfer; at the same time, continuous white light modulated by intensity is used as a sensing medium (probe) to penetrate the sample. According to the Lambert-Beer law, the transient fluctuation of the probe light intensity reflects the evolution law of the carrier concentration in real time. A hardware-level time-locked phase between laser excitation and signal acquisition is established through a high-precision delayed pulse generator to achieve time-resolved detection of weak absorption signals. Biased white light radiation is introduced to simulate the real illumination service environment. By pre-filling the trap state and modulating the Fermi level, it is ensured that the measured dynamic parameters can truly reflect the charge transfer and recombination mechanism of the device under load. After photoelectric conversion and multiple accumulation and averaging, the charge transfer lifetime and rate constant reflecting the interface performance can be quantitatively obtained through mathematical fitting.

[0048] This invention can accurately capture the charge transfer lifetime, recombination rate, and dynamic response of the interface barrier at the semiconductor heterojunction interface under real illumination in a simulated service environment. This design, which combines in-situ bias light simulation with high-precision pump-probe technology, significantly improves the correlation between characterization results and actual device performance, providing rigorous data support for the development of high-performance optoelectronic materials.

[0049] (2) Analyze the obtained dynamic signals:

[0050] The dynamic signal analysis process obtained by this invention is the core link in revealing the charge transfer mechanism at the semiconductor heterojunction interface. The original signal is captured in real time by a digital oscilloscope 11 and transmitted to a computer 12. Essentially, it is the voltage waveform V(t) that detects the evolution of the transmitted light intensity over time after the light passes through the sample.

[0051] In computer 12, the original voltage signal V(t) is converted into a transient absorbance change. The process involves a complex computational flow combining physical law transformation and mathematical noise reduction. First, the system uses a digital oscilloscope 11 to capture a voltage signal before the laser pulse reaches the sample, and the computer 12 defines its average value as the steady-state reference voltage V0. This value represents the background voltage generated at the photodetector end after the probed white light penetrates the sample in the material's unexcited equilibrium state. When the nanosecond laser 1 triggers the excitation pulse, charge transfer at the interface causes a transient change in the sample's absorption of the probed light. The photomultiplier tube 9 then outputs a transient voltage waveform V(t) that dynamically evolves over time. The calculation program then performs the conversion according to the difference form of the Lambert-Beer law. Specifically, the transient transmittance T(t) is defined as the ratio of the voltage at the excitation moment to the reference voltage, i.e., V(t) / V0. Since absorbance and transmittance have a negative logarithmic relationship, the change in transient absorbance... It can be done through the formula The calculation yields the following result. In this formula, when the detected voltage V(t) is lower than the reference voltage V0, A positive value physically corresponds to the photo-induced absorption process, representing enhanced absorption of the probe light by excited-state species or trapped-state carriers; conversely, if the calculated value is negative... A negative value indicates ground state bleaching.

[0052] In the signal processing stage, in order to obtain highly reliable dynamic parameters, the computer 12 performs cumulative averaging on the waveforms acquired under multiple trigger cycles. This step is particularly critical for detection in in-situ service environments, because the introduction of the biased white light source 13 inevitably increases the photon shot noise of the system. By averaging multiple times (typically 500 to 2000 excitations), the signal-to-noise ratio can be significantly improved, thereby clearly separating weak signals on the microsecond or even nanosecond scale from the baseline noise.

[0053] The obtained absorbance evolution curve This includes information on multiple competing paths for interfacial charges. For semiconductor heterojunction systems, due to the inhomogeneity of energy level distribution within the material and the complex contributions of trapped states, the charge recombination or injection process often does not follow a simple single-exponential decay law, but rather exhibits multi-component characteristics. Therefore, the analysis process requires the use of a multi-exponential function fitting model:

[0054]

[0055] in, This typically corresponds to ultrafast interface charge injection or in-band hot carrier relaxation processes, while This corresponds to a slower interface charge recombination or hole diffusion process. The amplitude obtained through fitting... and time constant By performing integral analysis, the average carrier lifetime can be calculated. This allows for a quantitative assessment of charge transfer efficiency. .

[0056] Among them, charge transfer efficiency ( The core of the evaluation lies in the quantitative comparison of the rates of competing physical processes at the interface. In this system, this evaluation process first relies on the amplitude obtained by fitting the transient absorbance curve. With time constant Perform an integral analysis. Physically, the charge transfer efficiency depends on the charge transfer rate (k... CT ) at the total decay rate (k total The proportion of the total charge carrier lifetime. Computer 12 calculates the average carrier lifetime. This reflects the outcome of the competition. Typically, average lifespan can be calculated using a weighted integral formula: In practical evaluations, efficiency information is often extracted by comparing the difference in average lifetime between states with and without charge acceptors. This is achieved using the formula... It can accurately calculate the percentage reduction in the total carrier lifetime caused by the opening of the interface injection path, which numerically corresponds to the percentage of charge that successfully undergoes interface transfer.

[0057] In addition, the amplitude A of each fitted component i The proportion is also a key criterion for evaluating efficiency. This is because of the fast-decaying components. Typically, for an effective interface charge injection process, the magnitude of the pre-factor A1 directly reflects the initial carrier concentration crossing the interface energy barrier; while for the slowly decaying component... The baseline offset represents the amount of charge that failed to transfer, recombinated at the interface, or was trapped. By calculating the contribution weight of A1 in the total amplitude and combining it with the trap filling effect induced by the in-situ biased light source 13, the system can quantitatively determine what proportion of photoexcited carriers avoid the recombination loss path under real service conditions, thus providing quantitative efficiency diagnostic data for optimizing the interface construction process of semiconductor heterojunctions.

[0058] Under in-situ service conditions, the analysis program focuses on comparing the kinetic differences under different intensities of bias light irradiation. As the intensity of the bias white light source 13 increases, the trapped states inside the sample are gradually pre-filled, which is reflected in the signal curve as a decrease in the proportion of slow components and an increase in the recombination rate. In the computer 12, the rate constant k=1 / under different light intensities is fitted. Fitting the rate constant under different light intensities is a multidimensional data processing procedure based on nonlinear least squares. First, for each specific bias light intensity, the system acquires a set of transient absorbance curves A(t) after cumulative averaging. The fitting program uses an iterative algorithm to match the experimental curves with a multi-exponential decay model, the mathematical expression of which is typically set as:

[0059]

[0060] In this model, k is the rate constant that needs to be extracted. i (corresponding to the reciprocal of lifespan 1 / ) A i y0 represents the weight magnitude of the corresponding path, reflecting the offset of the long-lived species signal or system baseline drift. For semiconductor heterojunction systems, double exponential fitting is typically used to analyze the rate constants k1 and k2 for fast processes (such as interface injection) and slow processes (such as recombination or trapping), respectively.

[0061] In the actual fitting process, the calculation program processes the curve sequences acquired under different bias light intensities in parallel or serially. As the intensity of the bias white light source 13 increases stepwise, the Fermi level position inside the sample shifts, leading to changes in carrier concentration and trap filling state. This is reflected in the fitting results as the rate constant k increases significantly with increasing light intensity, while the proportion of the slow component A2 usually decreases gradually due to the pre-filling of trap states. The fitting software automatically records the optimal fitting parameter k for each light intensity point (corresponding to a specific quasi-Fermi level position). To ensure the physical reliability of the fitting, certain system parameters (such as the instrument response function IRF) are usually fixed during calculation, and only the rate constant and amplitude are used as free variables for iteration until the sum of squared residuals reaches its minimum value. Finally, the computer 12 summarizes the fitting results under different light intensities, extracts the functional relationship between the rate constant k and the light intensity (or charge density), and constructs a mapping spectrum reflecting the intrinsic connection between interface energy level evolution and recombination dynamics.

[0062] Subsequently, by combining the changes in interface charge density, a mapping relationship between Fermi level shift and recombination dynamics can be constructed. Constructing this mapping relationship is a deep analysis process that uses quantitative changes in external physical conditions (bias light intensity) to detect the coupling relationship between the internal energy level states and charge lifetime of a semiconductor.

[0063] (3) The specific construction steps are as follows:

[0064] 1) Multigradient bias photoexcitation: By adjusting the output intensity of the bias light source 13, the sample is made to be in different steady-state quasi-Fermi level states. As the light intensity increases, the photoinduced carrier concentration increases, the trapped states inside the semiconductor are gradually filled, and the quasi-Fermi level shifts accordingly.

[0065] 2) Rate constant extraction: Under each specific bias light intensity, the transient absorbance curve is acquired using the system, and the corresponding decay time constant is extracted through multi-exponential fitting. Then, the recombination rate constant k under this state can be calculated.

[0066] 3) Charge density / energy level correlation: Combining the changes in interfacial charge density or open-circuit voltage data of the sample under different light intensities, computer 12 converts the light intensity parameter into a specific Fermi level shift. ).

[0067] 4) Coordinate mapping plotting: Plot the Fermi level shift (or corresponding charge density) on the x-axis and the recombination rate constant k (usually the logarithm ln k) on the y-axis.

[0068] 5) Physical mechanism discrimination:

[0069] Slope analysis: The order of the recombination process can be determined by the slope of the curve in the mapping graph. For example, a linear relationship usually corresponds to one-molecule recombination (trap-assisted recombination), while a nonlinear relationship may indicate that two-molecule recombination (band-band recombination) is dominant.

[0070] Loss path identification: Based on the degree of change of k with the Fermi level, the analysis program can determine whether the main charge loss comes from bulk recombination of the material or defect trapping at the interface, thus providing accurate data for energy level optimization.

[0071] Ultimately, the mapping generated by fitting these data points using computer 12 can intuitively demonstrate the quantitative logical chain between "ambient light intensity - internal band position - charge recombination rate".

[0072] This deep data mining not only provides the timescale of charge residence but also identifies whether the key loss path limiting device efficiency is due to bulk recombination or interface trapping, thus providing precise quantitative basis for energy level optimization and surface passivation processes in semiconductor heterojunctions. Finally, all analysis results, fitting curves, and extracted physical parameters are integrated and output by computer 12 to form a complete interface charge dynamics diagnostic report.

[0073] In summary, this invention constructs a complete dynamic characterization platform for charge transfer at semiconductor heterojunction interfaces. It not only addresses the technical limitations of traditional testing equipment in simulating real-world illumination conditions from a physical architecture perspective, enabling dynamic probing to penetrate into the actual operational state of devices, but also achieves high-fidelity capture of weak transient absorption signals spanning multiple time scales (nanoseconds to milliseconds) through hardware-level precise timing alignment and multi-dimensional signal gain control. Through integrated computer processing and nonlinear dynamic fitting, this invention can quantitatively reveal the injection efficiency, recombination lifetime, and loss mechanism of interface charge under real loads. Its high system integration, excellent anti-interference capabilities, and accurate reproduction of actual operating conditions provide crucial scientific evidence and technical support for the study of the microscopic mechanisms and macroscopic process optimization of high-performance photovoltaic devices, photocatalytic materials, and related energy conversion systems.

Claims

1. A dynamic probing system for semiconductor heterojunction interface charge transfer processes, characterized by: The system includes an induced pump optical path, a probe signal optical path, a biased white light source, a circuit synchronization control module, and a sample stage. In the induced pump optical path, a nanosecond laser serves as the induction source to excite the semiconductor heterojunction sample placed on the sample stage, inducing an interface charge transfer process. In the probe signal optical path, the probe light emitted by the steady-state white light source passes through a filter wheel and then through the semiconductor sample. The light is then spectrally selected by a monochromator and coupled to a photomultiplier tube. An independent biased white light source applies continuous bias light irradiation to the semiconductor heterojunction sample to simulate a solar background. The high-precision delay pulse generator of the circuit synchronization control module controls the dynamic detection time of the interface charge transfer of the semiconductor heterojunction sample. The dynamic curve of the interface charge transfer process is measured and recorded by a digital oscilloscope and transmitted in real time to a computer for data storage and post-processing.

2. The dynamic detection system of a semiconductor heterojunction interface charge transfer process according to claim 1, wherein: The induced pump optical path includes a nanosecond laser, a reflector, and a convex lens. The high-energy pulsed laser emitted by the nanosecond laser is reflected by the reflector and focused by a convex lens, so that it irradiates the surface of the semiconductor heterojunction sample on the sample stage, instantly generating a non-equilibrium carrier distribution.

3. The dynamic detection system for the charge transfer process at the interface of a semiconductor heterojunction according to claim 2, characterized in that: The focal length of the first convex lens is 5cm.

4. The dynamic detection system for charge transfer process at the interface of a semiconductor heterojunction according to claim 1, characterized in that: The detection signal optical path includes a steady-state white light source, a filter wheel, a second convex lens, a third convex lens, a photomultiplier tube, and a monochromator. The steady-state white light source emits continuous broadband detection light. This beam first passes through the filter wheel to modulate its intensity. The modulated detection light is then focused by the second convex lens onto the stimulated region of the sample stage. After penetrating the semiconductor heterojunction sample, it carries an absorption signal indicating the change in carrier concentration. The third convex lens then collects the transmitted light and guides it to the monochromator. The monochromator extracts monochromatic light according to a preset wavelength and couples it to the photomultiplier tube. The photomultiplier tube converts the time evolution of light intensity into a current signal, which is finally input to a digital oscilloscope for recording and displayed on a computer as a dynamic curve of absorbance decay over time.

5. The dynamic detection system for charge transfer process at the interface of a semiconductor heterojunction according to claim 4, characterized in that: The steady-state white light source uses xenon gas to emit light.

6. The dynamic detection system of a semiconductor heterojunction interface charge transfer process according to claim 4, wherein: The filter wheel has 12 standard slots, in which 6 neutral density filters with different transmittances are installed, with OD values ​​of 0.3, 0.5, 1.0, 1.5, 2.0, and 3.0, respectively. These filters are used to precisely modulate the radiation intensity of the probe light to avoid secondary excitation interference to the sample. At the same time, 6 narrowband interference filters with center wavelengths of 400nm, 450nm, 515nm, 532nm, 600nm, and 700nm are also installed. These filters are used to select probe signals of specific wavelengths according to experimental requirements, ensuring signal purity and preventing stray light interference.

7. The dynamic detection system of a semiconductor heterojunction interface charge transfer process according to claim 4, wherein: The focal length of the second convex lens is 5cm, and the focal length of the third convex lens is 5cm.

8. The dynamic detection system of a semiconductor heterojunction interface charge transfer process according to claim 1, wherein: The biased white light source uses an adjustable intensity white LED, whose spectral range and light intensity have been calibrated to simulate the lighting environment of a solar cell under real operating conditions.

9. The dynamic detection system of a semiconductor heterojunction interface charge transfer process according to claim 1, wherein: The circuit synchronization control module includes a digital oscilloscope, a computer, and a high-precision delay pulse generator. The high-precision delay pulse generator serves as the trigger core of the entire system. Its output port is connected to the external trigger interface of the nanosecond laser via a BNC shielded cable. Simultaneously, another synchronization output port of the high-precision delay pulse generator is connected to the external trigger channel of the steady-state white light source via a signal line. The signal input channel of the digital oscilloscope is connected to the electrical signal output port of the photomultiplier tube, converting the captured transient light intensity changes into voltage signals that are transmitted to the digital oscilloscope for waveform recording. The digital oscilloscope is connected to the computer via a USB or GPIB data cable. At the same time, the computer is connected to the monochromator via a control interface to precisely adjust the rotation angle of the internal grating of the monochromator, thereby realizing the automatic selection and switching of the detection wavelength.

10. The dynamic probing system of a semiconductor heterojunction interface charge transfer process according to claim 9, wherein: The high-precision delay pulse generator is a Stanford DG645.

Images