A deep level transient spectroscopy system, its testing method and application

Abstract

The present invention discloses a deep level transient spectroscopy (DLTS) test system. This test system is built based on the impedance analyzer MFIA developed by Zurich Instrument. This system omits the pulse generator part in the traditional DLTS test system and uses the built-in pulse generation module of the MFIA to generate the reverse bias voltage and the forward fill voltage in the DLTS test. During the DLTS test, the parasitic capacitance existing in the fixture and cables will affect the measurement accuracy. Therefore, the parasitic capacitance in the sample fixture and cables is compensated to improve the test accuracy. During the DLTS test, it is necessary to perform a temperature scan on the sample. By adjusting the PID algorithm to control the Cryocon22C temperature controller, certain improvements are made to the temperature control stability, thereby improving the accuracy of the temperature scan. Compared with the complete and expensive DLTS systems on the market, the DLTS system built by the present invention not only meets the test requirements of laboratory samples, but also greatly reduces the system cost and has good engineering applicability.

Description

Technical Field

[0001] The present invention belongs to the technical field of defect information detection of solar thin-film batteries, and relates to a deep level transient spectroscopy (DLTS) system, an improvement method thereof, a testing method, and an application. Background Art

[0002] Solar cells, also known as photovoltaic cells, are currently a major research focus both at home and abroad. However, impurities are inevitably introduced during the manufacturing process of solar cells, or various defects are generated due to uncontrollable factors such as stress between materials. Defects will affect the efficiency and durability of the battery. Therefore, the defect analysis of solar cells has always been one of the core points in the research process.

[0003] Currently, the mainstream detection methods for solar cell defects are admittance spectroscopy (AS) and deep level transient spectroscopy (DLTS). Among them, deep level transient spectroscopy is one of the most effective means for detecting defects at present. Most of the deep level transient spectroscopy systems on the market are mainly the DLS-83D of Semilab Company or the PhysTech FT series, which are expensive and have few purchase channels. Summary of the Invention

[0004] In order to solve the deficiencies of the existing technology, the object of the present invention is to propose a deep level transient spectroscopy testing system based on an MFIA impedance analyzer.

[0005] The testing system of the present invention can simultaneously generate pulse signals and has high-speed data acquisition capabilities based on the MFIA impedance analyzer. The MFIA is used as a pulse excitation source and a capacitance information acquisition part, which is equivalent to realizing the functions of a pulse generator and a capacitance meter with one MFIA at the same time.

[0006] The impedance analyzer of the present invention has a built-in pulse generation module for replacing the components of a traditional pulse generator. The built-in pulse generation module provides a wider frequency, higher pulses, lower noise, and better data acquisition capabilities. These functions provide more flexible testing conditions and more accurate data acquisition. The impact on the test results is that the noise is smaller, and different test frequencies and test pulse amplitudes can be selected according to the characteristics of different samples.

[0007] This system is applicable to the detection of defects in various types of solar thin-film batteries. By performing a temperature scan on the solar thin-film battery, the DLTS energy spectrum curve of the solar cell can be quantitatively obtained, and defect information such as defect type, defect activation energy, defect concentration, and capture cross-section can be identified through mathematical method processing and analysis.

[0008] Compared with the existing deep level transient spectroscopy (DLTS) test system, in the system of the present invention, the impedance analyzer is responsible for both pulse generation and signal acquisition. Although it can perform the same tasks as the traditional Boonton 7200, it cannot change the pulse frequency. Meanwhile, the noise of the capacitance signal is larger than that of the MFIA, resulting in poorer accuracy of the capacitance information. For the cryogenic system, we adopt a helium circulation refrigeration system, which is different from the liquid nitrogen cryogenic systems used in most traditional DLTS test systems.

[0009] The present invention provides a deep level transient spectroscopy test system for defect detection of solar cells. The test system includes: a vacuum pump, a compressor, a sample chamber, a temperature controller, an impedance analyzer, and a computer.

[0010] The vacuum pump, the compressor, the sample chamber, the temperature controller, and the impedance analyzer are all connected to the sample chamber, and the computer is connected to the impedance analyzer.

[0011] There is no refrigerator in this system. The low temperature is achieved by compressing low-temperature helium gas through the compressor. The sample chamber base is connected to the compressor through two pipes. The vacuum pump is connected to the sample chamber through one pipe. The MFIA impedance analyzer is connected to the sample chamber base through four BNC cables. The temperature controller is connected to the temperature sensor inside the sample chamber. The MFIA is connected to the PC through USB.

[0012] The computer is installed with software for controlling the impedance analyzer, which is used to trigger the test, record the test data, and save the test data on the PC side.

[0013] The impedance analyzer is built-in with a pulse generation module for the data acquisition module. The MFIA impedance analyzer can collect transient capacitance signals at a maximum rate of 807KSa / S and perform mathematical processing of averaging the collected transient capacitance signals multiple times.

[0014] The sample chamber includes a probe station, a resistance temperature sensor, and a heater. The probe station is used to fix the solar thin film battery to be tested. The resistance temperature sensor is used to measure the temperature inside the sample chamber. The heater is located inside the probe station and heats the sample through the copper block wrapping the heater.

[0015] The vacuum pump is used to pump out the air inside the sample chamber to create a vacuum heat insulation environment for the purpose of accurately controlling the temperature.

[0016] The compressor is used to compress low-pressure helium gas into high-pressure helium gas. The sample chamber base is connected to the compressor through two pipes. Pipe 1 releases the high-pressure helium gas as low-pressure helium gas in the sample chamber, and the temperature drops rapidly during this process. Then the low-pressure helium gas returns to the compressor through Pipe 2 to form a refrigeration cycle.

[0017] The temperature controller is used to control the 50-ohm cylindrical heater inside the probe station. If the temperature in the sample chamber is lower than the set value, the heater under the probe station is turned on, and then the sample is heated through the copper block that wraps the heater. If the temperature is higher than the set value, the heater is controlled to stop working, and only the refrigeration work of the compressor is carried out.

[0018] The refrigerator is a refrigeration system composed of an 8200 compressor / M22 cold head / helium gas pipeline, rather than a single component.

[0019] The present invention also proposes an improvement method for the above deep level transient spectroscopy test system, including the following two aspects:

[0020] Aspect 1: Any cable or fixture used to connect the device under test (DUT) to impedance measurement will inevitably introduce measurement deviation (accuracy relative to the true value). Eliminating the residual impedance of the fixture and the influence of the cable is an important part of high-precision measurement. In our system, the DUT refers to a solar cell, and the MFIA that constitutes the equivalent circuit of the test fixture, as Figure 2 shown. Z M represents the measured value between the current leads (HCUR, LCUR) and the voltage leads (HPOT, LPOT). Zs represents the residual impedance when short-circuited (DUT impedance is zero), and Yo represents the residual conductance when open-circuited (DUT impedance is infinite). Due to these influences, the true impedance value Zx of the DUT can be expressed as follows:

[0021]

[0022] Using the built-in compensation module of the MFIA, calculate Z S and Y O after short-circuiting and disconnecting the probes respectively, and the compensation information will be stored in the compensation file. Just load the compensation file into the MFIA before each test. Figure 3 represents the sweep frequency curves with and without compensation.

[0023] Aspect 2: In the DLTS test, after obtaining the C-T curve by performing a temperature scan on the sample, further mathematical processing is required to obtain the Arrhenius straight line, thereby obtaining relevant defect information. Therefore, the influence of temperature on the test results is undoubtedly a problem to be solved, which is an inevitable requirement for improving measurement accuracy. Temperature setting is not as immediate as voltage, pulse, etc. It requires a certain time delay to reach the target temperature value, and the temperature will oscillate up and down to a certain extent. Therefore, we hope that the temperature in the sample chamber can reach the preset value and remain stable in the shortest possible time. The Cryocon22C temperature controller uses a proportional-integral-derivative (PID) algorithm for temperature control. Algorithm 1 illustrates the basic process of temperature setting in the DLTS system. It takes a temperature setting list {S T1 ,S T2 ,…,S Ti} with a step size of 5K and a series of feedback values for each target temperature i as inputs. There is a certain deviation between the preset temperature and the actual temperature received by the sensor, which is generally divided into three parts: current deviation, historical deviation, and recent deviation. The current deviation is considered to be the difference between the kth expected value and the sensor acquisition value. The historical deviation is considered to be the sum of the current deviations, that is . Positive and negative values respectively represent that most of the time in the past period did not reach or exceeded the standard. The recent deviation is regarded as the difference between the two most recent current deviations, and its value reflects the trend of the deviation increasing or decreasing. K p ,K i ,K d represent the coefficients of the proportional (current), integral (historical), and derivative (recent) parts. The sum of these three deviation parts is the output of Algorithm 1. To make the temperature of the system more stable, the settings of K p ,K i ,K d play an almost decisive role. Figure 4 shows the temperature offset under different series settings of K p ,K i ,K d . According to experience, the order of parameter adjustment is first proportional, then integral, and finally derivative, and the values increase. The proportional coefficient in Cryocon 22C is usually fixed at 20. Under fixed conditions, increasing and from 5 to 10 will effectively reduce temperature fluctuations. From Figure 4 it can be seen that when K p / K i / K d = 20 / 10 / 10, the temperature fluctuation is very small, almost less than 0.1K, which meets the requirements of temperature system stability.

[0024]

[0025]

[0026] The present invention also provides a method for testing a solar cell using the above deep level transient spectroscopy test system, and the method includes the following steps:

[0027] Step 1: Fix the solar cell to be tested on the probe stage, close the probe stage and evacuate the chamber to vacuum, turn on the compressor, and set the desired temperature through the temperature controller;

[0028] Step 2: Start the impedance analyzer, and load the instrument configuration file and compensation file into the software on the computer that controls the impedance analyzer;

[0029] Step 3: After the temperature in the chamber reaches the set temperature of the temperature controller, keep the temperature for a period of time, increase the temperature in set step sizes, and scan the sample at each temperature point until the temperature reaches the set final temperature, and collect data on the computer;

[0030] Step 4: Process the data collected on the computer to obtain whether the solar cell to be tested has defects and the corresponding defect information.

[0031] The beneficial effects of the present invention include: The deep level transient spectroscopy test system described in the present invention can be used to detect deep level defects of various solar cells. The test system in the present invention uses the pulse generation module built in the MFIA to generate a bias voltage and apply it to the sample to be tested, replacing the pulse generator in the traditional DLTS system. While accurately detecting information, it greatly reduces the cost of the system and simplifies the system structure.

[0032] System cost reduction: The MFIA impedance analyzer costs about 100,000 yuan, and the low temperature system costs about 30 yuan (including compressor / vacuum pump / sample chamber). Compared with the DLTS test system sold as a complete set on the market with a selling price of about 900,000 - 1,000,000 yuan, it not only realizes the test function, but also greatly saves the system cost;

[0033] System structure simplification: The system can directly adjust the pulse amplitude on the control panel of the MFIA, omitting the pulse generator. Description of the Drawings

[0034] Figure 1 It is the system structure diagram of the deep level transient spectroscopy detection system of the present invention.

[0035] Figure 2 It is a schematic diagram of the equivalent circuit model of the fixture (probe stage) and cable parasitic capacitance. Among them, Z MRepresents the measured value between the current leads (HCUR, LCUR) and the voltage leads (HPOT, LPOT). Zs represents the residual impedance at short circuit (DUT impedance is zero), Yo represents the residual conductance at open circuit (DUT impedance is infinite), and Zx represents the value of the DUT under test.

[0036] Figure 3 Is the C-f curve of the same device before and after the compensator.

[0037] Figure 4 Is a schematic diagram of temperature fluctuations at different PID ratios.

[0038] Figure 5a Is Sb 2 S 3 Temperature scan C-T diagram of the thin film battery from 180K - 480K.

[0039] Figure 5b Is Sb 2 S 3 Capacitance difference change diagram of the thin film battery under different time rate windows.

[0040] Figure 5c Is Sb 2 S 3 Arrhenius straight line diagram of the thin film battery. Detailed implementation mode

[0041] Combined with the following specific embodiments and drawings, the invention will be further described in detail. The processes, conditions, experimental methods, etc. for implementing the present invention, except for the specifically mentioned content below, are all common knowledge and well-known common sense in the art, and the present invention has no particularly restricted content.

[0042] The present invention discloses a deep level transient spectroscopy test system. This test system is built based on the impedance analyzer MFIA developed by Zurich Instrument. This system omits the pulse generator part in the traditional deep level transient spectroscopy test system and uses the built-in pulse generation module of MFIA to generate the reverse bias voltage and forward fill voltage in DLTS testing. During the testing of deep level transient spectroscopy, the parasitic capacitance existing in the fixture and cable will affect the measurement accuracy. Therefore, the parasitic capacitance in the sample fixture and cable is compensated to improve the testing accuracy. During the testing of deep level transient spectroscopy, it is necessary to perform a temperature scan on the sample. By adjusting the PID algorithm to control the Cryocon22C temperature controller, certain improvements are made to the temperature control stability, and the accuracy of the temperature scan is improved. Compared with the complete and expensive DLTS systems on the market, the DLTS system built by the present invention not only meets the testing requirements of laboratory samples, but also greatly reduces the system cost and has good engineering applicability.

[0043] The present invention provides a deep level transient spectroscopy test system for defect detection of solar cells. The test system includes the following devices:

[0044] Device 1: The temperature controller can be a Cryocon22C temperature controller. During the test of deep level transient spectroscopy, temperature scanning of the sample is required. The American Cryocon22C temperature controller is used to control the 50-ohm cylindrical heater inside the probe station. The heater temperature is displayed by the thermometer DT670. The sample temperature is accurately and quickly fed back to the temperature inside the sample chamber through the resistance temperature sensor of the probe station and compared with the set value of the temperature controller. If the temperature inside the sample chamber is lower than the set value, the heater under the sample stage (i.e., inside the probe station) is turned on, and then the sample is heated through the copper block wrapping the heater. If it is lower than the set value, the heater is controlled to start working. If it is higher than the set value, the heater is controlled to stop working, and only the refrigeration work of the compressor is carried out. A temperature control range of 200 mK - 1500 K can be achieved, and flexible programming is possible.

[0045] Device 2: The cryocooler can be a CCS-150 helium cryogenic system from Janis.Ltd. This is a closed-loop helium cryogenic system. The working principle is that first, the compressor compresses low-pressure helium into high-pressure helium, and then it is sent to the sample stage base of the cryogenic system through a pipeline. While releasing it as low-pressure helium, the sample in the chamber is cooled. The low-pressure helium then returns to the compressor to continue pressurization, repeating the above actions to form multiple cycles. This cryogenic system can create a low-temperature environment as low as 10 K.

[0046] Device 3: An impedance analyzer. Specifically, the MFIA high-speed impedance analyzer from Zurich Instrument can be used. This is an advanced and user-friendly impedance analyzer. It can use the built-in pulse generation module to replace the pulse generator in the traditional DLTS system. At the same time, it can provide a wider frequency setting (1 Hz - 5 MHz), a higher pulse selection (-10 V - 10 V), a lower noise impact, and better data acquisition capabilities.

[0047] Device 4: A computer. LabOne is installed on the computer. LabOne is a PC-side operation interface based on controlling the MFIA impedance analyzer. It can trigger the start of the test and record the test data, and manually save the test data on the PC side.

[0048] Device 5: The vacuum pump used is an EDWARD EXT75DX turbo molecular drag pump without oil, with a pumping speed of up to 61 s-1. It is required to provide oil-free and vibration-free operation to ensure the purity of the sample chamber and the test stability while evacuating the sample chamber.

[0049] Equipment Six: Sample Chamber; the cold head model in the sample chamber is M-22, the helium compressor model is 8200, equipped with a standard vacuum cover and a heat radiation shield, with a built-in temperature sensor and a heater, and the probe station is a gold-plated oxygen-free high-conductivity copper (OFHC) probe station;

[0050] In the deep level transient spectroscopy test system of the present invention, the temperature setting cannot reach stability instantaneously like voltage, pulse, etc. It takes a certain time delay to reach the target temperature value, so the temperature will oscillate up and down to a certain extent. Equipment One needs to adjust the temperature control stability by modifying the PID algorithm so that the temperature of the vacuum cold well (sample chamber) can be stabilized in the shortest possible time. The proportionality coefficient of Cryocon 22C is usually fixed at 20. With Kp fixed, increasing Ki and Kd from 5 to 10 can effectively reduce the temperature fluctuation. As can be seen from Figure 4 it, when Kp / Ki / Kd = 20 / 10 / 10, the temperature fluctuation is the smallest, almost less than 0.1K, which well meets the requirements for the stability of the temperature system. Kp / Ki / Kd represents the coefficients of proportional (current), integral (history), and derivative (recent). The sum of these three deviation parts is the output of the PID algorithm, and the ratio of Kp / Ki / Kd will directly affect the magnitude of the temperature drift.

[0051] In the deep level transient spectroscopy test system of the present invention, for Equipment Two, it is necessary to perform capacitance compensation and debug the MFIA test parameters, and compensation files and configuration files (XML documents) can be generated respectively.

[0052] Among them, capacitance compensation is to eliminate the parasitic capacitance existing in the cables and fixtures in the system; in the DLTS test, the stability and accuracy of the capacitance signal will be affected by the parasitic capacitance existing in the cables and fixtures in the system. Therefore, the Compention module in LabOne is used to compensate the parasitic capacitance existing in the cables and fixtures in the system and generate a compensation file recording the compensation information. In each test, only the compensation file needs to be loaded into LabOne for testing.

[0053] After the measurement, it is necessary to perform mathematical processing on the obtained C-T curve to calculate the defect information.

[0054] Using the deep level transient spectroscopy detection system of the present invention to detect a solar thin film battery includes the following steps:

[0055] Step One: Fix the solar thin film battery on the probe station, close the probe station and evacuate the chamber to vacuum, turn on the compressor and set the desired temperature of 180K on the Cryocon22C low-temperature temperature controller;

[0056] Step 2: Start the MFIA and open the LabOne operation interface on the PC side, and load the instrument configuration file and compensation file into LabOne;

[0057] Step 3: The temperature controller shows 180K. After maintaining for a period of time, the sample is heated and scanned in steps of 5K up to 480K (which can be flexibly set according to experimental requirements), and data is collected on the PC side;

[0058] Step 4: Mathematically process the data collected on the PC side to obtain defect information.

[0059] In the method for identifying the defect type of the solar cell described in the present invention, in Step 1, the desired temperature is adjusted according to the test requirements. At the same time, in order to prevent the probe from scratching the thin-film battery due to the jitter of the compressor and vacuum pump over a long period of time as much as possible, a smooth and appropriately sized nickel metal sheet is added above the battery point of the thin-film battery to be measured, that is, the probe is connected to the thin-film battery point through the nickel metal sheet.

[0060] Embodiment

[0061] This test case is described using a antimony sulfide (Sb 2 S 3 ) solar thin-film battery prepared by gas transport deposition method.

[0062] First, fix the antimony sulfide solar thin-film battery on the probe stage, close the chamber and evacuate it; then turn on the compressor and adjust the temperature controller to 180K. Finally, turn on the MFIA and the LabOne operation interface on the PC side, and load the configuration file and compensation file into it. After the temperature drops to 180K, stay for a period of time, and then start the heating test in steps of 5K, measure up to 480K to end the test and save the data.

[0063] In DLTS analysis, it is necessary to mathematically process the DLTS energy spectrum to obtain relevant defect information; three different time intervals are defined as time windows, and the two ends of each time window are marked with t1 and t2 respectively. When the temperature is scanned, the capacitance difference between times t1 and t2 will change. At the same time, this capacitance difference (ΔC) will form a relationship curve with temperature. For the maximum (or minimum) difference, a derivative of the signal should be equal to zero. Therefore, the emission rate at this maximum peak height can be calculated through the expression of the time constant, as shown below:

[0064]

[0065] Selecting different rate windows will result in corresponding changes in the emission rate. Each emission rate corresponds to a certain temperature at which the DLTS peak appears. Then, an Arrhenius plot is drawn using these points. The method to obtain the Arrhenius plot is to fit a straight line through the formula, as shown below:

[0066]

[0067] Among them, e n (T) is the carrier emission rate, γ is a constant, k is the Boltzmann constant, T is the Kelvin temperature, σ n is the capture cross-sectional area, and Ea is the defect activation energy. From the defect density can be calculated, where N D is the effective acceptor state density; C 0 refers to the steady-state capacitance of the sample before applying a pulse to the sample;

[0068] Through the above mathematical processing of the data, Figure 5a as shown is for Sb 2 S 3 the temperature scan energy spectrum of the thin film battery. It can be seen that in the high-temperature region, the capacitance shows a process of gradually decreasing and then gradually returning to the steady state, and in the low-temperature region, the capacitance shows a process of first decreasing, then increasing, and finally gradually returning to the steady state. The former is the DLTS energy spectrum characteristic of minority carrier defects, and the latter is the DLTS energy spectrum characteristic of majority carrier defects; Figure 5b is the schematic diagram of the change in the capacitance difference under different time rate windows. A peak and a trough are observed, that is, a minority carrier defect and a majority carrier defect are detected; Figure 5c is the fitted Arrhenius straight line. Thus, the activation energy of the electron trap center is obtained as 0.61 eV, the trap center capture area is 2.17E-20 cm -2 and the electron trap center concentration is 5.75E14 cm -3 ; the activation energy of the hole trap center is 0.43 eV, the trap center capture area is 2.65E-21 cm -2 , and the hole trap center concentration is 2.54E15 cm -3 .

[0069] Figure 5a : DLTS energy spectrum curves of the sample at different temperatures;

[0070] illustrates different changing trends of the observed transient capacitance;

[0071] Figure 5b : Fitted curves of taking different time windows for the DLTS energy spectrum;

[0072] illustrates that both majority and minority carrier defects (i.e., peaks and troughs) can be observed under different time windows;

[0073] Figure 5c : Arrhenius curve fitting diagram;

[0074] According to the peaks and valleys observed under different time windows, the temperature corresponding to the extreme points is taken, and the temperature and the time window value are substituted into Equation to fit and obtain defect information such as the defect activation energy.

[0075] The protection scope of the present invention is not limited to the above embodiments. Without departing from the spirit and scope of the inventive concept, the changes and advantages that can be conceived by those skilled in the art are included in the present invention, and the appended claims are taken as the protection scope.

Claims

1. A deep level transient spectroscopy system, characterized in that, it includes: a vacuum pump, a compressor, a sample chamber, a temperature controller, an impedance analyzer, and a computer; wherein, the vacuum pump, the compressor, the sample chamber, the temperature controller, and the impedance analyzer are all connected to the sample chamber, and the computer is connected to the impedance analyzer; the base of the sample chamber and the compressor are connected by two pipes; the vacuum pump is connected to the sample chamber through one pipe; the impedance analyzer is an MFIA impedance analyzer and is connected to the base of the sample chamber through four BNC cables; the MFIA impedance analyzer is connected to the PC end through USB; the temperature controller is connected to the temperature sensor inside the sample chamber; the computer is installed with software for controlling the impedance analyzer, which is used to trigger tests and record test data, and save the test data on the PC end; the impedance analyzer is built-in with a pulse generation module for the data acquisition module.

2. The deep level transient spectroscopy system according to claim 1, characterized in that, the MFIA impedance analyzer acquires transient capacitance signals at a rate of 807 KSa / S and performs mathematical processing of averaging the acquired transient capacitance signals multiple times; the sample chamber includes a probe station, a resistance temperature sensor, and a heater; the probe station is used to fix the solar thin film battery to be tested; the resistance temperature sensor is used to measure the temperature inside the sample chamber; the heater is located inside the probe station and heats the sample through the copper block wrapping the heater; the vacuum pump is used to pump out the air inside the sample chamber to create a vacuum heat insulation environment for the purpose of accurately controlling the temperature.

3. The deep level transient spectroscopy system according to claim 1, characterized in that, the compressor is used to compress low-pressure helium gas into high-pressure helium gas. The base of the sample chamber and the compressor are connected by two pipes. The high-pressure helium gas is released as low-pressure helium gas in the sample chamber through pipe one, and the temperature drops rapidly during this process. Then the low-pressure helium gas returns to the compressor through pipe two to form a refrigeration cycle; the temperature controller is used to control the 50-ohm cylindrical heater inside the probe station. If the temperature inside the sample chamber is lower than the set value, the heater below the probe station is turned on, and then the sample is heated through the copper block wrapping the heater; if it is higher than the set value, the heater is controlled to stop working, and only the refrigeration work of the compressor is carried out.

4. A test method for the deep level transient spectroscopy system according to any one of claims 1-3, characterized in that, it includes the following steps: Step one: Connect various instruments used in the system; Step two: Debug the test parameters of the MFIA impedance analyzer and compensate and eliminate the parasitic capacitance and resistance in the probe station and cables; Step three: By adjusting the PID algorithm, make the temperature fluctuation in the sample chamber of the Cryocon22C temperature controller tend to be stable; Step four: When performing a temperature scan on the sample, first lower the temperature to <100K, and then increase the temperature in steps of 5K for testing after stabilization; Step five: Measure and save the data, perform mathematical processing on the data, and obtain the sample defect information.

5. The method according to claim 4, characterized in that, In Step 1, the system uses the pulse generation module built in the MFIA impedance analyzer to generate pulses. The built-in DC pulses are connected to the auxiliary output 1 of the MFIA to the auxiliary input 1 through a BNC cable.

6. The method according to claim 4, wherein, in Step 2, to eliminate the parasitic capacitance existing in the probe station and the cable, and to compensate for the parasitic capacitance existing in the probe station and the cable, the Short / Open mode in the MFIA compensation module is adopted.

7. The method according to claim 4, wherein, in Step 2, the compensation method adopts the Short / Open mode in the LabOne compensation module; the anode and cathode of the probe are short-circuited and disconnected respectively, and then the compensation calculation is carried out by using the compensation module in LabOne.

8. The method according to claim 4, wherein, in Step 2, the experimental parameter settings of the MFIA impedance analyzer are saved as a file in XML format by the LabOne control panel; the compensation data is saved as a file in CAL format, and the saved document can be directly loaded into the LabOne control panel.

9. The method according to claim 4, wherein, in Step 2, when the temperature controller Kp / Ki / Kd is 20 / 10 / 10.

10. An application of a deep level transient spectroscopy system according to any one of claims 1-3 in the detection of thin film solar cells.

Images