Vacuum distillation purification method of high-purity chalcogenide semiconductor material

Abstract

This invention relates to the field of high-purity chalcogenide semiconductor material preparation technology, and discloses a vacuum distillation purification method for high-purity chalcogenide semiconductor materials. The method includes: setting and maintaining a constant temperature in the component control zone to establish a preset chalcogenide vapor pressure; introducing an inert gas background pressure into the vacuum system; monitoring the heating power in the condensation zone to generate a power change signal characterizing the transport rate; and, while maintaining a constant temperature in the component control zone, adjusting the inert gas background pressure based on the difference between the power change signal and a preset target power change. This invention orthogonally decouples the establishment of a thermodynamic benchmark from the control of the kinetic rate, solving the technical problem of mutual constraint between the two, and achieving independent control of the kinetic process without interfering with a constant chemical background.

Description

Technical Field

[0001] This invention relates to a vacuum distillation purification method for high-purity chalcogenide semiconductor materials, belonging to the field of high-purity chalcogenide semiconductor material preparation technology. Background Technology

[0002] Currently, in the preparation of high-purity chalcogenide semiconductor materials, such as cadmium telluride or gallium arsenide, purification via vacuum distillation or transport is the core process for obtaining high-purity raw materials. Those skilled in the art recognize that utilizing the vapor pressure of one of the material's components, particularly the chalcogenide components, as a thermodynamic control parameter allows for the chemical selective separation of the target material from specific impurities. This method provides an effective process path for controlling the stoichiometry of the product. However, the purification process based on component vapor pressure exhibits technical conflicts related to process control objectives when applied to industrial production. The purification process pursues thermodynamic goals of final chemical purity and stoichiometry, and kinetic goals of controlling the material transport rate and deposition morphology. Too slow a transport rate is not economical for industrial production, while too fast a transport rate leads to excessive supersaturation in the condensation zone, resulting in poor product crystal quality and even inducing physical entrainment, affecting the final purification effect.

[0003] Faced with this complex situation where thermodynamic and kinetic objectives are intertwined, those skilled in the art have tried various process paths, but often they have lost sight of the bigger picture. For example, Chinese invention patent CN120646778A discloses a method for purifying high-purity tellurium, which uses a multi-stage separation process combining wet and pyrometallurgical methods. Impurities are gradually removed through chemical dissolution, multiple recrystallizations, and subsequent stepped vacuum distillation. The purification task is broken down into multiple physical and chemical steps, each targeting specific impurities. This multi-step, long-process method is not only cumbersome to operate and prone to introducing secondary pollution, but the final vacuum distillation step is a traditional separation method based on high vacuum and temperature difference. It fails to solve the technical contradiction of how to independently control the chemical background and transport rate of a compound in situ within a single vacuum system.

[0004] Therefore, the technical problem to be solved by this invention is how to provide a purification method that can separate the establishment of the thermodynamic background from the regulation of the kinetic process from the control architecture, so as to achieve separate control of chemical purity and physical quality. Summary of the Invention

[0005] To address the problems mentioned in the background art, the technical solution of the present invention is as follows: A vacuum distillation purification method for high-purity chalcogenide semiconductor materials, wherein the chalcogenide semiconductor materials are composed of metal components and chalcogenide components, the method is carried out in a vacuum system including a source region, a condensation region, and a component control region, wherein the temperature of the condensation region is lower than the temperature of the source region, and the method includes the following steps:

[0006] Step 101: Set the crude material to be purified in the source area;

[0007] Step 102: Set pure chalcogenide components in the component control area;

[0008] Step 103: Heat the source region to the first temperature and set the condensation region to the second temperature;

[0009] Step 104: Set and maintain the temperature of the component control zone at a constant first set temperature to establish a preset chalcogenide vapor pressure. The chalcogenide vapor pressure asymmetrically inhibits the decomposition and volatilization of impurity compounds with stronger chemical bond energy in the source region. The combination of chalcogenide vapor pressure and second temperature is used to regulate the stoichiometry of the deposited products.

[0010] Step 105: Introduce an inert gas background pressure into the vacuum system;

[0011] Step 106: Monitor the heating power required to maintain the condensation zone at a constant second temperature;

[0012] Step 107: Based on the change in heating power caused by the exothermic deposition reaction of the volatile components of the chalcogenide semiconductor material in the condensation zone, a power change signal characterizing the transport rate of the chalcogenide semiconductor material is generated.

[0013] Step 108: Based on the difference between the power change signal and the preset target power change, while keeping the first set temperature, the first temperature and the second temperature constant, the background pressure of the inert gas is adjusted accordingly.

[0014] Preferably, step 106 further includes monitoring the second heating power required to maintain the source region at a constant first temperature; step 107 specifically involves: generating a leading power change signal based on the change in the second heating power caused by the endothermic decomposition of the coarse material; generating a lagging power change signal based on the change in heating power; and the feedback adjustment in step 108 is performed by integrating the leading power change signal and the lagging power change signal.

[0015] Preferably, the method further includes: periodically applying a preset thermal disturbance to a first temperature; monitoring the dynamic response of the second heating power caused by the thermal disturbance; identifying the actual value of the chalcogenide vapor pressure based on the difference between the dynamic response and the standard response model; and calibrating the first set temperature based on the deviation between the actual value of the chalcogenide vapor pressure and the preset chalcogenide vapor pressure.

[0016] Preferably, the method further includes: continuously monitoring the natural fluctuations of the first temperature and the corresponding fluctuations of the second heating power; and calculating the dynamic gain between the natural fluctuations and the corresponding fluctuations through cross-correlation analysis. : ,in, This represents the cross-power spectral density between the corresponding wave and the natural wave. The self-power spectral density of natural fluctuations. Inverse Fourier transform; based on dynamic gain By comparing with the standard gain model, the actual value of the chalcogenide vapor pressure is passively identified; based on the deviation between the actual value of the chalcogenide vapor pressure and the preset chalcogenide vapor pressure, the first set temperature is calibrated.

[0017] Preferably, the method further includes: monitoring the actual deposition rate of the material in the condensation zone; identifying the actual value of the vapor pressure of the chalcogen components, the first temperature, and the second temperature, and calculating the theoretical deposition rate of the material in a forward direction; when the deviation between the actual deposition rate of the material and the theoretical deposition rate of the material exceeds a preset model confidence threshold, stopping the calibration of the first set temperature based on the actual value of the vapor pressure of the chalcogen components.

[0018] Preferably, the method further includes: monitoring the actual deposition rate of the material in the condensation zone; identifying the actual value of the vapor pressure of the chalcogen components, the first temperature, and the second temperature, and calculating the theoretical deposition rate of the material in a forward direction; when the deviation between the actual deposition rate of the material and the theoretical deposition rate of the material exceeds a preset model confidence threshold, stopping the calibration of the first set temperature based on the actual value of the vapor pressure of the chalcogen components.

[0019] Preferably, the chalcogenide semiconductor material is cadmium telluride, the chalcogenide component is tellurium, and the impurity compound is zinc telluride.

[0020] Preferably, the chalcogenide semiconductor material is gallium arsenide, the chalcogenide component is arsenic, and the impurity compound is indium arsenide.

[0021] Preferably, the range of the first set temperature is 550°C. Up to 650 .

[0022] Preferably, the background pressure of the inert gas is adjustable within the range of 1 Pa to 500 Pa.

[0023] Compared with the prior art, the beneficial effects of the present invention are:

[0024] 1. By using the chalcogenide components of the material to be purified as the gas-phase control medium in the purification process, the vapor pressure of the components can be precisely controlled. At the same time, it has an asymmetric inhibitory effect on the decomposition and volatilization of target products and impurities of compounds with different chemical stability from a thermodynamic perspective. The vapor pressure of the components also defines the chemical boundary conditions when the material is deposited in the condensation zone. The two core process objectives of chemical purification selectivity and product stoichiometry are both controlled by the same process parameter, namely the vapor pressure of the chalcogenide components, thus avoiding the risk of secondary contamination caused by the introduction of external chemical reagents.

[0025] 2. The intermediate capture temperature zone set between the source zone and the condensation zone is a synergistic and enhanced function of regulating the chalcogenide vapor. The background pressure of the chalcogenide components established in the main scheme and the specific intermediate temperature window allow the chalcogenide vapor to transform into a chemical precipitant when it flows through the intermediate capture zone after acting as a decomposition inhibitor in the source zone. This functional reuse in the mechanism enables this method to handle stable impurities with stronger chemical bond energy in a continuous process flow, suppressing them in the source zone and precipitating volatile metallic impurities in the intermediate capture zone, thus expanding the limitations of single physical distillation or chemical transport methods in terms of the breadth of impurity treatment.

[0026] 3. The process thoroughly separates the execution of two control objectives with different properties: thermodynamic background and kinetic rate. By maintaining a constant temperature in the component control zone, the thermodynamic benchmarks upon which chemical purification depends—namely, chemical purity and stoichiometry—are protected. An inert gas background pressure, which is purely physical and does not participate in the chemical reaction, is introduced as an independent kinetic regulator. When it is necessary to adjust the material transport rate, the method adjusts the pressure of this inert gas to physically regulate the diffusion resistance of gaseous components. Without interfering with the constant chemical background, independent and decoupled control of the kinetic process is achieved, enabling the synergistic protection of the two mutually constraining objectives of chemical purity and physical quality of the product in traditional processes. Attached Figure Description

[0027] Figure 1 This is a schematic diagram of the decoupling control logic of the purification method of the present invention;

[0028] Figure 2 This is a comparison diagram of the synergistic purification effect between the sample group and the control group of the present invention;

[0029] Figure 3 This is a schematic diagram showing the complete operation state switching of the purification process of the present invention. Detailed Implementation

[0030] To make the objectives, technical solutions, and advantages of this invention clearer, the technical solutions of this invention will be described in detail below. Obviously, the described embodiments are only some embodiments of this invention, not all embodiments. All other embodiments obtained by those skilled in the art based on the embodiments of this invention without creative effort are within the scope of protection of this invention.

[0031] This invention discloses a vacuum distillation purification method for high-purity chalcogenide semiconductor materials. The method is implemented in a vacuum system comprising a source region, a condensation region, and a component control region. The chalcogenide semiconductor material is composed of metallic and chalcogenide components. The temperature of the condensation region is lower than that of the source region, thus separating the establishment of a thermodynamic baseline from the control of the kinetic rate. A preset chalcogenide vapor pressure is established by maintaining a constant temperature in the component control region to stabilize the chemical purification selectivity and product stoichiometry. An inert gas background pressure is introduced as an independent kinetic regulator, which adjusts the kinetic rate based on a signal characterizing the material transport rate. The pressure of the inert gas is adjusted to independently control the kinetic process without disturbing the constant chemical background. In a specific embodiment, the chalcogenide semiconductor material is cadmium telluride, the chalcogenide component is tellurium, and the impurity compound may be zinc telluride with stronger chemical bond energy. The crude cadmium telluride to be purified is placed in the source region; pure chalcogenide components, in this case, pure tellurium, are placed in the component control region; the vacuum system is evacuated and then heated; the source region is heated to a first temperature, and the condensation region is set to a second temperature; the temperature of the component control region is set and maintained at a constant first set temperature, which can be in the range of 550°C. Up to 650 A preset constant vapor pressure of chalcogenide components is established in the vacuum system. This vapor pressure is used to asymmetrically suppress the decomposition and volatilization of impurity compounds with stronger chemical bond energies in the source region, taking zinc telluride as an example. The combination of chalcogenide vapor pressure and a second temperature is used to regulate the stoichiometry of the deposited products. An inert gas background pressure is introduced into the vacuum system; this inert gas can be argon or nitrogen, etc., and the pressure adjustment range can be 1 Pa to 500 Pa. To control the material transport rate, the heating power required to maintain a constant second temperature in the condensation zone is monitored. The heating power is due to the exothermic reaction of the volatile components of the chalcogenide semiconductor material in the condensation zone. The changes generated by the deposition reaction produce a power change signal characterizing the transport rate of the chalcogenide semiconductor material. When the material transport rate increases, the heat released during deposition increases, and the external heating power required to maintain the constant temperature of the condensation zone decreases accordingly. Based on the difference between the power change signal and the preset target power change, the background pressure of the inert gas is adjusted in feedback while keeping the first set temperature, the first temperature, and the second temperature constant. If the transport rate is too fast, the background pressure of the inert gas is increased to increase the diffusion resistance and reduce the rate. This method is also applicable to the purification of gallium arsenide, where the chalcogenide component is arsenic and the impurity compound is indium arsenide.

[0032] A preferred embodiment integrates leading and lagging signals. This scheme further includes monitoring the second heating power required to maintain a constant first temperature in the source region; generating a leading power change signal based on the change in the second heating power caused by the endothermic decomposition of the coarse material; generating a lagging power change signal based on the change in the heating power in the condensation zone; feedback regulation is performed by integrating the leading and lagging power change signals to achieve more precise regulation of the inert gas background pressure; for calibrating the chalcogenide vapor pressure, one method includes: applying a preset thermal disturbance to the first temperature; monitoring the dynamic response of the second heating power caused by the thermal disturbance; identifying the actual value of the chalcogenide vapor pressure based on the difference between the dynamic response and the standard response model; calibrating the first set temperature based on the deviation between the actual value of the chalcogenide vapor pressure and the preset chalcogenide vapor pressure; another calibration method includes: continuously monitoring the natural fluctuations of the first temperature and the corresponding fluctuations of the second heating power; calculating the dynamic gain between the natural fluctuations and the corresponding fluctuations through cross-correlation analysis. : ,in, This represents the cross-power spectral density between the corresponding wave and the natural wave. The self-power spectral density of natural fluctuations. Inverse Fourier transform; based on dynamic gain The method involves comparing the actual vapor pressure of the chalcogenide components with that of the standard gain model to identify the actual vapor pressure of the chalcogenide components; calibrating the first set temperature based on the deviation between the actual vapor pressure of the chalcogenide components and the preset vapor pressure of the chalcogenide components; and, to verify the accuracy of the above calibration, further comprising: monitoring the actual deposition rate of the material in the condensation zone; calculating the theoretical deposition rate of the material forward based on the identified actual vapor pressure of the chalcogenide components, the first temperature, and the second temperature; and stopping the calibration of the first set temperature based on the actual vapor pressure of the chalcogenide components when the deviation between the actual deposition rate of the material and the theoretical deposition rate of the material exceeds the preset model confidence threshold.

[0033] Example 1: In the industrial preparation scenario of vacuum distillation purification of high-purity cadmium telluride (CdTe) semiconductor materials, the crude material to be purified contains a certain amount of zinc telluride (ZnTe) impurities with stronger chemical bond energy, as well as volatile impurities. The process challenge in this scenario lies in the initial heating stage of the purification process. The source material is heated unevenly, and volatile impurities and some cadmium telluride crude material will undergo sudden and uncontrolled flash evaporation, generating a large-flux of instantaneous mass transport. If traditional methods of fixing the high vacuum or adjusting the source temperature are used, this instantaneous kinetic impact will cause a surge in local supersaturation in the condensation zone, inducing disordered crystal nucleus formation, damaging the physical quality of the product, and volatile impurities will also be physically entrained, leading to a decrease in chemical purity. Cadmium telluride crude material is placed in the source region, and pure tellurium is placed in the component control region. The temperature of the component control zone is set and maintained at a constant first set temperature, establishing a preset chalcogenide vapor pressure to asymmetrically suppress the decomposition and volatilization of zinc telluride impurities. An initial inert gas background pressure of 5 Pa is introduced into the vacuum system. When the source region is heated to the first temperature, flash evaporation occurs, and the mass transport rate surges instantaneously. The system detects a sharp increase in heat release due to mass deposition in the condensation zone, causing a sudden drop in the heating power required to maintain it at a constant second temperature, generating a power change signal far exceeding the preset target power change. While maintaining a constant first set temperature, the system adjusts the inert gas background pressure based on the difference between the power change signal and the preset target power change, increasing it from 5 Pa to 150 Pa in a short time. The increased inert gas background pressure increases the vapor phase composition, including... , The diffusion resistance of volatile impurities is reduced, and the material transport rate is suppressed and falls back to the preset target level. During this adjustment process, the first set temperature is kept constant to ensure the stability of the thermodynamic and chemical background. After a smooth transition to the steady-state purification stage, the control system stabilizes the inert gas background pressure at a new operating point based on the power change signal, so that the purified product, namely high-purity cadmium telluride crystals, meets the process requirements in terms of chemical purity and physical quality.

[0034] Example 2: This example demonstrates the method of the present invention in a vacuum distillation purification system. The system includes a source zone, a condensation zone, and a component control zone, each with independently controllable temperature and a temperature control accuracy of ±0.5°C. To establish an independent execution space for the chemical precipitation mechanism at the physical hardware level, the purification system incorporates an intermediate capture zone at the physical flow channel transition between the source and condensation zones. The internal quartz channel diameter of this zone maintains strict hydrodynamic continuity with the evaporation liquid surface diameter of the source zone. It is externally equipped with independently controlled closed-loop heating and insulation components. The control unit can precisely set and maintain its operating temperature within a specific intermediate capture temperature window that is higher than the second temperature of the condensation zone and lower than the first temperature of the source zone. During the purification process, when the gaseous fluid mixture containing trace amounts of metallic impurities flows through this intermediate capture zone driven by the system's macroscopic pressure difference, it is constrained by the localized supersaturated thermodynamic state caused by the intermediate cooling in this zone, and by the chemical potential barrier established by the preset vapor pressure of the chalcogen components. The weaker chemical bonds within this zone further limit its impact. Volatile metallic impurity particles are forced to encounter excess chalcogenide vapors within this narrow temperature range, triggering an irreversible reverse chemical reaction. This preferentially forms a stable solid-phase compound deposition film on the inner wall of the capture tube, thus being physically trapped in situ. The system is equipped with an inert gas (Ar) pressure control unit, including a mass flow controller and a high-precision capacitive thin-film gauge, with a pressure setting and monitoring range of 0.1 Pa to 1000 Pa. The power monitoring module monitors the heating power required to maintain a constant temperature in the condensation zone, with a monitoring resolution of 0.1 W. The experimental material used is crude cadmium telluride (CdTe), initially analyzed by glow discharge mass spectrometry (GDMS) to contain 150 ppmw of zinc telluride (ZnTe) impurities. The component control zone uses a 6N purity tellurium source. A set of constant process temperature parameters was set for the experiment: the first temperature of the source region was set at 850°C. The second temperature of the condensation zone is set to 750. The preset target power change, which characterizes the target transport rate, is set to -5.0W. It is determined by subtracting the exothermic enthalpy change corresponding to the preferred growth rate from the baseline heating power of the condensation zone when no material is deposited. The experimental setup is shown in Table 1, which includes the sample group of the present invention using the method of the present invention, control groups 1 to 3 for comparing synergistic effects, and out-of-range control groups 4 and 5 for verifying the boundary of the inert gas background pressure (1 Pa to 500 Pa).

[0035] Table 1: Comparative Experiment Table of the Impact of Different Control Strategies on Purification Efficiency

[0036]

[0037] The analysis of the experimental results in Table 1 is as follows: Control group 1 used high vacuum (<0.1 Pa) without activating the component control zone, resulting in low ZnTe impurity removal rate (145 ppmw). Simultaneously, the transport rate was too high (power change -15.2 W) and fluctuated significantly (±4.5 W), leading to high supersaturation in the condensation zone and a loose powder as the product physical quality. Control group 2, under a constant temperature of 600°C... The vapor pressure of the chalcogenide components was established at the first set temperature, still using a high vacuum (<0.1 Pa). Under these conditions, ZnTe impurities were suppressed (4.3 ppmw), and chemical purity was improved; however, the transport rate was too fast (-13.8 W) and fluctuated (±4.2 W), resulting in polycrystalline and dendritic morphologies in the product. In control group 3, the component control zone was not activated, but inert gas background pressure feedback was used for adjustment; the transport rate stabilized at the target value of -5.1 W (fluctuation ±0.2 W), and the product exhibited dense, large-grained physical quality; however, the chemical purity was low (142 ppmw). The sample group of this invention used a constant 600... The first setting temperature and inert gas background pressure feedback regulation are enabled; the system transport rate is stabilized at -5.0W (fluctuation ±0.1W), and the product has both high chemical purity (4.1ppmwZnTe) and high physical quality (dense large grains). The inert gas background pressure of control group 4 is set at 0.5Pa (below the lower limit of 1Pa), the transport rate is -9.7W, and the physical quality of the product decreases to polycrystalline; the pressure of control group 5 is set at 600Pa (above the upper limit of 500Pa), the transport rate drops to -0.8W, and the purification efficiency is low; by controlling the inert gas background pressure within the range of 1Pa to 500Pa, the process conditions of product physical quality and purification rate are synergistically guaranteed.

[0038] Example 3: This example combines Figures 1 to 3 The vacuum distillation purification method for high-purity chalcogenide semiconductor materials is described, such as... Figure 1 As shown, a constant temperature is maintained in the pure chalcogenide component control zone, and a preset chalcogenide component vapor pressure is established as a thermodynamic reference. Simultaneously, the crude material to be purified enters two monitoring channels. One channel monitors the second heating power in the source zone to sense the decomposition endothermic effect and generate an advanced power change signal. The other channel monitors the heating power in the condensation zone to sense the deposition exothermic effect and generate a lagging power change signal. These two signals, after merging the advanced and lagging signals, characterize the overall material transport rate and are compared with a preset target. The comparison result is used to adjust the inert gas background pressure. This adjustment, together with the introduced inert gas background pressure kinetic adjustment actuator, independently controls the kinetic process to obtain semiconductor materials with high purity and high physical quality.

[0039] like Figure 2As shown, the left Y-axis represents the ZnTe impurity content in ppmw, and the right Y-axis represents the power change in W. Data shows that: control group 1 had a ZnTe impurity content of 145 ppmw and a power change of 15.2 W; control group 2 had a ZnTe impurity content of 4.3 ppmw and a power change of 13.8 W; control group 3 had a ZnTe impurity content of 142 ppmw and a power change of 5.1 W; and the sample group of this invention had a ZnTe impurity content of 4.1 ppmw and a power change of 5.0 W, achieving both low impurity content and stable power change. Figure 3 As shown, the process begins with system initialization, material setup, heating, and introduction of initial inert gas. After reaching the set temperature, it enters a stable purification state. In this state, a constant thermodynamic baseline is maintained, and the kinetic rate is continuously monitored. The process includes three branch paths: First, when a rate deviation is detected, it enters a rate adjustment state to adjust the background pressure of the inert gas. Once the rate recovers to the target value, it returns to stable purification. Second, when periodic calibration is triggered, it enters an online calibration state to identify and calibrate the vapor pressure of the components. After calibration, it returns to stable purification. Third, when the source material is depleted, it enters a final processing state to smoothly shut down the process by incorporating the lead signal. Finally, the process ends.

[0040] Example 4: During the tenth batch of continuous operation of the vacuum distillation system used for purifying cadmium telluride (CdTe), a slow drift in the system's thermal parameters caused the temperature reading in the component control zone to remain constant at the first set temperature of 600°C during step 104. The established preset vapor pressure of the chalcogenide components has deviated from its initial calibration value, affecting the asymmetric suppression effect on zinc telluride (ZnTe) impurities. The system then initiates an active online calibration procedure. While maintaining the condensation zone (second temperature) and the component control zone (first set temperature) constant, the system periodically applies a preset thermal perturbation to the first temperature of the source zone, with an amplitude of +2. The system monitors a 300-second step signal and the dynamic response curve of the second heating power used to maintain a constant temperature in the source region due to the thermal disturbance. A built-in standard response model, based on a first-order transfer function model established by the system's heat capacity and the rate of heat absorption during the decomposition of coarse material in the source region, describes the standard dynamic response characteristics of the second heating power to the first temperature step under a preset chalcogenide vapor pressure. Based on the difference between the measured dynamic response and the standard response model, the system identifies the actual value of the current chalcogenide vapor pressure by fitting the transfer function gain using the least squares method. In this extraction and identification calculation step, the system's data processing unit first performs bandpass filtering and steady-state periodic integration on the acquired power dynamic response signal. This directly removes and eliminates the low-frequency phase shift error caused by the inherent heat capacity inertia of the quartz crucible and the sensible heat hysteresis of the undecomposed remaining coarse material from the physical signal stream, thereby accurately extracting the heat absorption purely from the decomposition reaction of the purified material. The net increment of the steady-state amplitude triggered by the drastic thermal fluctuation is then divided by the fixed amplitude of the first temperature thermal disturbance step, given by the system, to obtain the pure actual dynamic response gain coefficient after eliminating external thermal inertial coupling interference. Since the decomposition and volatilization endothermic rate of the crude material in the distillation system is strictly suppressed by the equilibrium thermodynamic potential established by the ambient sulfide vapor pressure, this purified dynamic gain coefficient has a corresponding monotonically nonlinear decaying mapping relationship with its current actual sulfide vapor pressure. The system algorithm substitutes the fitted output gain coefficient into the pre-calibrated system internal mapping function library using different standard pressures for reverse indexing, thereby obtaining the actual sulfide vapor pressure data with clear physical meaning after removing stray parameters from the equipment. It is identified that the actual vapor pressure is approximately 4.5% lower than the preset value. Based on this deviation, the control target value of the first set temperature is adjusted from 600... Automatic calibration to 602.8 This restores the actual vapor pressure of the chalcogen components to the preset level.

[0041] As a parallel passive calibration method, the natural fluctuation of the first temperature is monitored, here within ±0.3. The random fluctuations of the second heating power are compared with the corresponding fluctuations of the second heating power; the dynamic gain is calculated in real time through cross-correlation analysis. The calculation is as follows: ,in, This represents the cross-power spectral density between the corresponding wave and the natural wave. The self-power spectral density of natural fluctuations. This is the inverse Fourier transform; it should be noted that, when calculating the dynamic gain... In the implementation method, the listed formulas middle Let be the frequency domain transfer function of the system. The result is obtained by inverse Fourier transform. The system's time-domain impulse response; dynamic gain. Specifically, through the pass function At zero frequency The value at that location, i.e. To determine the value The steady-state response gain of the second heating power to the natural fluctuations of the first temperature is objectively reflected by the standard gain model. The system will establish a real-time correspondence between the vapor pressures of the chalcogen components and the vapor pressures of the chalcogen components. The numerical values ​​are compared with the pre-calibrated standard gain model stored in the system, i.e. By comparing the correlation curves with the vapor pressures of chalcogenide components, the actual values ​​of the vapor pressures of chalcogenide components are identified, and the first set temperature is calibrated. During the execution of the spectral density mapping analysis formula by the underlying digital signal processor, in order to effectively filter out the strong and irregular low-frequency random thermal noise fluctuations caused by the violent boiling of the solid-liquid multiphase flow in the industrial-grade source region, the system deliberately uses the intersection product of the cross-power spectrum and the self-power spectrum of the two inverse Fourier time-domain transform equations to dynamically construct an adaptive broadband correlation time-domain weight mask with strong attenuation truncation characteristics. The processor then... By utilizing the steep impulse response of the autocorrelation function of natural temperature fluctuations at the origin of time as the integration trigger condition, a mask is applied to the cross-correlation characteristic containing original thermal noise within a strictly defined, extremely short finite time window, and a weighted truncated definite integral operation is performed. This forcibly isolates the time-domain diffusion of long-tailed low-frequency noise from the mathematical integration mechanism level, achieving unbiased anti-interference extraction of the DC component of the signal system, i.e., the zero-frequency response characteristic. Thus, even under conditions with severe thermal disturbances, it can still provide accurate gain equivalent to the pure zero-frequency absolute steady-state thermal response. The system is supported by quantitative results; to prevent model mismatch from causing incorrect calibration, the system also performs a verification step; the system independently monitors the actual deposition rate of the material in the condensation zone using a quartz crystal microbalance set in the condensation zone, obtaining values... The actual vapor pressures, first and second temperatures of the identified chalcogen components were used to calculate the theoretical deposition rate of the material using the Hertz-Knudsen transport equation, yielding the value... In actual process environments with inert gas background pressure, the material volatilization and transport processes dominated by vacuum surface dynamics are inevitably limited by the physical resistance of continuous gas phase collisions. Therefore, when performing forward calculations, the system first obtains the measured inert gas background pressure as the boundary condition for gas phase diffusion resistance. Using the Boltzmann transport integral, which includes gas phase molecule collision cross-sections and mean free path variables in gas dynamics theory, a diffusion attenuation correction coefficient specifically for this background pressure is derived. The system then performs a direct multiplicative attenuation conversion between this correction coefficient and the ideal vacuum maximum volatilization flux calculated unilaterally by the Hertz-Knudsen equation, thereby reasonably and seamlessly mapping the microscopic surface pure dynamics model to the macroscopic continuous gas phase transition flow condition. Since the deviation between the actual material deposition rate and the theoretical material deposition rate is about 1.9%, which does not exceed the preset model reliability threshold, it is set to 5% here. This confirms that the calibration of the first set temperature is effective, and the purification process continues.

[0042] Example 5: Before applying the method to a new vacuum distillation system, an operational baseline model is established. With no material placed in the source and component control zones, the source zone is heated to a first temperature, and the condenser zone is set to a second temperature. The heating power of the condenser zone is monitored to obtain a baseline power value. Pure chalcogenide components are placed in the component control zone to set the first baseline temperature, which is 600°C in this case. The pure material to be purified was placed in the source region and operated under a fixed inert gas background pressure of 50 Pa. The system monitored the actual deposition rate of the material at this time using a quartz crystal microbalance. When the rate stabilized at... At this moment, record the change in heating power in the condensation zone, which is -5.0W, and set this value as the preset target power change; under this stable operating condition, apply +2 to the first temperature. The system records the dynamic response of the second heating power caused by the step thermal disturbance and stores the response curve as a standard response model.

[0043] To establish a standard gain model, a multi-point scan is performed; the source and condensation temperatures are maintained at a first and a second temperature, respectively; the first set temperature is set from 550°C. Start with 10 The step size is incremented sequentially up to 650. During the constant maintenance phase of the first set temperature, the natural fluctuations of the first temperature and the corresponding fluctuations of the second heating power are monitored, based on... Calculate the corresponding dynamic gain Numerical values, among which This represents the cross-power spectral density between the corresponding wave and the natural wave. The self-power spectral density of natural fluctuations. This involves performing an inverse Fourier transform; querying the built-in thermodynamic database for the theoretical vapor pressures of the chalcogen components corresponding to the current first set temperature; and then calculating the vapor pressures of each group. The numerical values ​​and their corresponding theoretical vapor pressure values ​​of chalcogenide components are stored as data points. After scanning all temperature points, a polynomial fit is performed on all stored data points, and the resulting fitted curve is stored as a standard gain model.

[0044] Example 6: When the method of the present invention is used in a vacuum distillation system for purifying high-purity gallium arsenide (GaAs) semiconductor materials, a standardized engineering procedure is established to determine the process window that matches the kinetic rate with thermodynamic suppression. The gallium arsenide crude material contains indium arsenide (InAs) as the main impurity compound and arsenic (As) as the chalcogenide component. A preset target power change is determined. Under the condition that the component control zone is not activated and the first set temperature is set to room temperature, three sets of fixed inert gas background pressures are set to 5 Pa, 50 Pa, and 200 Pa, respectively. The actual deposition rates are measured to be 10.2, 4.8, and 10.2 Pa using a quartz crystal microbalance in the condensation zone. Monitoring records are in At this rate, the stable change in heating power in the condensation zone is -6.2W; X-ray rocking curve tests were performed on the products obtained from the three sets of experiments, and the results showed... The product rocking curve obtained at the lowest rate has the smallest half-width and the highest physical quality; -6.2W is determined and stored as the preset target power change for this specific material and system.

[0045] The procedure determines the preferred range of the first set temperature; the system activates feedback regulation to control the inert gas background pressure in a dynamic adjustment mode with a target of -6.2W. This feedback regulation uses a proportional-integral-derivative control algorithm, with the input being the difference between the power change signal and the preset target power change, and the output being the adjustment signal for the inert gas mass flow controller; the system executes a series of purification batches, changing the first set temperature, with the scan range from 500... Up to 700 Analysis of indium arsenide (InAs) impurity content in each batch of products by glow discharge mass spectrometry revealed that at 550... Up to 650 Outside this range, the InAs impurity content is higher than 10 ppmw; within this range, the impurity content can be stably controlled below 5 ppmw. (This 550) Up to 650 The range was determined as the preferred operating range for thermodynamic suppression; a preset target power change of -6.2W and 600 were determined. (From 550) Up to 650 After selecting the first set temperature (within the range), a complete calibration run is performed. Using active thermal perturbation and passive natural fluctuation analysis, data is collected to generate the standard response model and standard gain model of the gallium arsenide purification system. The standard gain model is stored as a set. The polynomial fitting coefficients corresponding to the vapor pressures of chalcogenide components are used for online calibration in subsequent batches.

[0046] Example 7: In the final stage of the gallium arsenide (GaAs) purification process, when the crude material in the source region is about to be exhausted, the rate of the endothermic decomposition reaction begins to drop sharply. The control system detects the premature decrease in the demand for the second heating power (used to maintain the constant temperature of the source region) by monitoring the second heating power, and generates a premature power change signal indicating that the material in the source region is about to be exhausted. If the control system relies solely on the lagging power change signal (which has not yet changed) generated by monitoring the exothermic deposition in the cold NG zone for feedback adjustment, it will incorrectly continue to reduce the inert gas background pressure in an attempt to maintain an impossible material transport rate, leading to system instability. The method in this example adopts a fusion control strategy. When the control system receives the premature power change signal, the weight of the adjustment logic changes. The system no longer targets the lagging power change signal, but instead integrates the warning information of the premature signal to increase the inert gas background pressure, ensuring a smooth shutdown of the purification process during the material exhaustion stage.

[0047] In a specific implementation, both leading and lagging power change signals are integrated and executed in the control unit through preset control logic. For example, a material depletion threshold is set based on the rate of change of the leading power change signal (characterizing heat absorption in the source region). During the steady-state purification stage, the output of the control unit is mainly driven by the lagging power change signal (characterizing heat release in the condensation region) to maintain a stable transport rate. When the control unit detects that the rate of change of the leading power change signal, such as the rate of decrease in power demand, exceeds the material depletion threshold, the weight of the control logic switches, and the leading power change signal dominates the adjustment action, executing a smooth shutdown. The material depletion threshold can be determined by statistical analysis of the source region power data at the end of historical batch processes. When executing the historical data statistical modeling calculation logic, the system data processing unit uniformly retrieves... The transient waveform data of the second heating power in the source area within 30 seconds of the material's actual depletion endpoint from ten batches of identical specifications that have successfully completed purification were used to calculate and derive the overall expected mean and standard deviation of the rate of decline in advanced power demand in historical normal shutdown batches. Simultaneously, based strictly on the engineering statistical assumption that this specific industrial parameter dataset follows a normal Gaussian distribution, the system forcibly solidifies positive high deviation values ​​exceeding three standard deviations above the expected mean—that is, values ​​reaching a 99.7% extreme confidence level in the probability model—as the material depletion judgment threshold for the current batch. Through this exclusive confidence level statistical boundary delineation, occasional pseudo-rate decline interference pulses caused by local coarse material voids or external source temperature and power grid fluctuations are eliminated, providing a clear and unique mathematical logic criterion for triggering weight switching control actions.

[0048] It will be apparent to those skilled in the art that the present invention is not limited to the details of the exemplary embodiments described above, and that the present invention can be implemented in other specific forms without departing from the spirit or essential characteristics of the present invention.

[0049] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention.

Claims

1. A vacuum distillation purification method for high-purity chalcogenide semiconductor materials, the chalcogenide semiconductor materials being composed of metal components and chalcogenide components, the method being carried out in a vacuum system comprising a source region, a condensation region, and a component control region, wherein the temperature of the condensation region is lower than the temperature of the source region, characterized in that... The method includes the following steps: Step 101: Set the crude material to be purified in the source area; Step 102: Set pure chalcogenide components in the component control area; Step 103: Heat the source region to the first temperature and set the condensation region to the second temperature; Step 104: Set and maintain the temperature of the component control zone at a constant first set temperature to establish a preset chalcogenide vapor pressure. The chalcogenide vapor pressure asymmetrically inhibits the decomposition and volatilization of impurity compounds with stronger chemical bond energy in the source region. The combination of chalcogenide vapor pressure and second temperature is used to regulate the stoichiometry of the deposited products. Step 105: Introduce an inert gas background pressure into the vacuum system; Step 106: Monitor the heating power required to maintain the condensation zone at a constant second temperature; Step 107: Based on the change in heating power caused by the exothermic deposition reaction of the volatile components of the chalcogenide semiconductor material in the condensation zone, a power change signal characterizing the transport rate of the chalcogenide semiconductor material is generated. Step 108: Based on the difference between the power change signal and the preset target power change, while keeping the first set temperature, the first temperature and the second temperature constant, the background pressure of the inert gas is adjusted accordingly.

2. The vacuum distillation purification method of a high-purity chalcogenide semiconductor material according to claim 1, characterized by, Step 106 also includes monitoring the second heating power required to maintain the source region at a constant first temperature; Step 107 specifically involves: generating a leading power change signal based on the change in the second heating power caused by the endothermic decomposition of the coarse material; generating a lagging power change signal based on the change in heating power; and the feedback adjustment in Step 108 is performed by integrating the leading power change signal and the lagging power change signal.

3. The vacuum distillation purification method of a high-purity chalcogenide semiconductor material according to claim 2, characterized by, The method further includes: periodically applying a preset thermal disturbance to a first temperature; monitoring the dynamic response of the second heating power caused by the thermal disturbance; identifying the actual value of the chalcogenide vapor pressure based on the difference between the dynamic response and the standard response model; and calibrating the first set temperature based on the deviation between the actual value of the chalcogenide vapor pressure and the preset chalcogenide vapor pressure.

4. The vacuum distillation purification method of a high-purity chalcogenide semiconductor material according to claim 2, characterized by, The method also includes: continuously monitoring the natural fluctuations of the first temperature and the corresponding fluctuations of the second heating power; and calculating the dynamic gain between the natural fluctuations and the corresponding fluctuations through cross-correlation analysis. : ,in, This represents the cross-power spectral density between the corresponding wave and the natural wave. The self-power spectral density of natural fluctuations. Inverse Fourier transform; based on dynamic gain By comparing with the standard gain model, the actual value of the chalcogenide vapor pressure is passively identified; based on the deviation between the actual value of the chalcogenide vapor pressure and the preset chalcogenide vapor pressure, the first set temperature is calibrated.

5. The vacuum distillation purification method for high-purity chalcogenide semiconductor materials according to claim 3, characterized in that, The method also includes: monitoring the actual deposition rate of the material in the condensation zone; identifying the actual value of the vapor pressure of the chalcogen components, the first temperature, and the second temperature, and calculating the theoretical deposition rate of the material in a forward direction; when the deviation between the actual deposition rate and the theoretical deposition rate of the material exceeds the preset model confidence threshold, stopping the calibration of the first set temperature based on the actual value of the vapor pressure of the chalcogen components.

6. The vacuum distillation purification method of a high-purity chalcogenide semiconductor material according to claim 4, characterized by, The method also includes: monitoring the actual deposition rate of the material in the condensation zone; identifying the actual value of the vapor pressure of the chalcogen components, the first temperature, and the second temperature, and calculating the theoretical deposition rate of the material in a forward direction; when the deviation between the actual deposition rate and the theoretical deposition rate of the material exceeds the preset model confidence threshold, stopping the calibration of the first set temperature based on the actual value of the vapor pressure of the chalcogen components.

7. The vacuum distillation purification method of a high-purity chalcogenide semiconductor material according to claim 1, characterized by, The chalcogenide semiconductor material is cadmium telluride, the chalcogenide component is tellurium, and the impurity compound is zinc telluride.

8. The vacuum distillation purification method of a high-purity chalcogenide semiconductor material according to claim 1, characterized by, The chalcogenide semiconductor material is gallium arsenide, the chalcogenide component is arsenic, and the impurity compound is indium arsenide.

9. The vacuum distillation purification method of a high-purity chalcogenide semiconductor material according to claim 1, characterized by, The first set temperature ranges from 550 to 650 .

10. The vacuum distillation purification method of a high-purity chalcogenide semiconductor material according to claim 1, characterized by, The inert gas background pressure can be adjusted from 1 Pa to 500 Pa.

Images