A method for testing metal impurities of dichlorodihydrogen silicon

Abstract

The present application relates to dichlorodihydrogen silicon detection technical field, especially in kind, a kind of dichlorodihydrogen silicon metal impurity test method, comprising: dichlorodihydrogen silicon is passed into chemical vapor deposition reaction cavity, and epitaxial layer is formed on the surface of substrate, and sample substrate is obtained;Preparation is not passed into dichlorodihydrogen silicon blank substrate;Sample substrate and blank substrate are respectively immersed in two etching liquids and are completely digested, another one is blank control, three etching liquids are collected and constant volume is made;The metal impurity concentration in sample digestion liquid, blank substrate digestion liquid and reagent blank liquid is detected respectively;The metal impurity mass concentration of epitaxial layer is calculated and substituted into the calibration curve generated in advance, and the concentration of metal impurity in the measured dichlorodihydrogen silicon is obtained;By enriching the metal impurity in gaseous DCS in solid-state epitaxial layer, then the impurity is transferred to liquid phase by acid digestion epitaxial layer, and then the standard method such as ICP-MS is used for detection, which greatly improves the safety and accuracy of detection.

Description

Technical Field

[0001] This invention relates to the field of silicon dichlorosilane detection technology, and in particular to a method for testing silicon dichlorosilane metallic impurities. Background Technology

[0002] Dichlorosilane (SiH2Cl2, DCS) is an electronic specialty gas with significant applications in the semiconductor industry. It is primarily used as a silicon source gas in chemical vapor deposition (CVD) and epitaxial growth processes to deposit silicon thin films or grow single-crystal silicon layers on substrates. Due to its fast deposition rate, relatively low temperature, and ability to improve the uniformity of epitaxial layer thickness, DCS plays an irreplaceable role in the manufacture of advanced process chips (such as those below 3nm). The purity of DCS directly affects the quality of the deposited silicon film, which in turn affects the performance and yield of the final semiconductor device. The presence of trace metal impurities (such as iron, copper, sodium, potassium, zinc, etc.) can introduce deep-level defects, leading to increased leakage current, decreased reliability, and even device failure. Therefore, accurate monitoring of trace metal impurities in DCS is one of the key steps in ensuring semiconductor manufacturing yield. However, DCS itself is highly volatile, flammable and explosive (auto-ignition point in air as low as 44°C), and highly corrosive. Furthermore, it undergoes violent hydrolysis upon contact with water, producing hydrochloric acid and polysiloxanes. These extreme physicochemical properties make direct sampling, storage, and handling of DCS exceptionally difficult and extremely risky. DCS is highly volatile at room temperature, making quantitative sampling and preservation challenging. Traditional direct injection methods may lead to analytical errors due to incomplete sample conversion or changes during storage.

[0003] Currently, the analysis of trace impurities in high-purity electronic gases typically relies on highly sensitive analytical instruments, such as inductively coupled plasma mass spectrometry (ICP-MS) or inductively coupled plasma optical emission spectrometry (ICP-OES). However, for DCS (Distributed Catalytic Reduction System), sample pretreatment remains the biggest technical bottleneck. Existing methods usually attempt to directly introduce cryogenically cooled liquid DCS into the analysis system under strictly controlled temperature, inert gas protection, or vacuum conditions using complex devices. While these methods reduce exposure risks to some extent, they are extremely cumbersome, have poor repeatability, and require specialized equipment, making it difficult to completely avoid the aforementioned matrix interference and instrument corrosion problems. Furthermore, for ultra-low concentration impurities at the ppb or even ppt level, direct injection often faces challenges such as insufficient detection limits and matrix effect interference.

[0004] The information disclosed in this background section is intended only to enhance the understanding of the general background of this disclosure and should not be construed as an admission or in any way implying that the information constitutes prior art known to those skilled in the art. Summary of the Invention

[0005] This invention provides a method for testing dichlorosilane metal impurities. By enriching the metal impurities in the gaseous DCS into a solid epitaxial layer, and then transferring the impurities to the liquid phase by acid digestion of the epitaxial layer, the concentration of metal impurities in the DCS is obtained by standard methods such as ICP-MS, which greatly improves the safety and accuracy of the detection.

[0006] To achieve the above objectives, the technical solution adopted by the present invention is as follows: A method for testing dichlorosilane metallic impurities includes: S1 introduces the silicon dichlorodihydrogen analyte (SDD) into the chemical vapor deposition (CVD) chamber, depositing an epitaxial layer on the substrate surface to obtain the sample substrate. A blank substrate without SSD is prepared under the same conditions. This step utilizes the practical application of SSD in semiconductor manufacturing: CVD, to grow an epitaxial layer on a single-crystal silicon substrate. This process simultaneously transfers and enriches trace metal impurities from the gaseous SSD within the epitaxial layer, transforming the challenge of gas sampling and analysis into a solid sample processing problem. The simultaneously prepared blank substrate undergoes the same treatment under identical process conditions (except for the absence of SSD), providing a benchmark for subsequent subtraction of substrate background and process-introduced interference. The specifications, batches, and sizes of the sample and blank substrates should be consistent to reduce the difference in metal impurities between the two substrates and improve accuracy. S2. Take three identical etching solutions. Immerse the sample substrate and blank substrate in two of the etching solutions for complete digestion, and use the third as a blank control. Collect the three etching solutions and adjust the volume to obtain sample digestion solution, blank substrate digestion solution, and reagent blank solution. In this step, the etching solution is used to completely dissolve the epitaxial layer carrying metal impurities along with the substrate, so that the metal impurities in the solid phase are transferred to the liquid phase. The parallel reagent blank solutions are used to monitor the metal impurities introduced by the digestion reagent itself. Adjust the volume of all digestion solutions to ensure that the sample matrix volume is consistent in subsequent detection steps, laying the foundation for accurate calculations based on concentration data. S3 detects the concentration of metal impurities in the sample digestion solution, blank substrate digestion solution, and reagent blank solution respectively. This step uses inductively coupled plasma mass spectrometry (ICP-MS) or inductively coupled plasma optical emission spectrometry (ICP-OES) to detect the concentration of metal impurities in the sample digestion solution, blank substrate digestion solution, and reagent blank solution, directly obtaining the concentration measurement value of the target metal impurities in each solution. The above measurement method can meet the ultra-low content detection requirements at the ppt level and provide the raw data required for subsequent calculations. S4 calculates the metal impurity mass concentration of the epitaxial layer based on the metal impurity concentrations in the sample digestion solution, blank substrate digestion solution, and reagent blank solution. Specifically, the total mass of metal impurities attributable to the epitaxial layer is obtained by subtracting the mass of metal impurities from the etching solution, ultrapure water, and blank substrate from the total mass of impurities in the sample. The total mass of impurities attributable to the epitaxial layer is determined only by the metal impurity content in the dichlorosilane gas. The total mass of impurities attributable to the epitaxial layer is then correlated with the mass of the epitaxial layer to calculate the metal impurity mass concentration of the epitaxial layer. This calculation process eliminates the interference from the etching solution, ultrapure water, and the substrate itself.

[0007] S5 substitutes the mass concentration of the epitaxial layer metal impurities into the pre-generated calibration curve to obtain the concentration of metal impurities in the silicon dichlorodihydrogen to be tested. This step uses the established calibration curve to convert the mass concentration of the epitaxial layer metal impurities to the concentration of impurities in the silicon dichlorodihydrogen to be tested. The calibration curve is pre-prepared using a standard gas of silicon dichlorodihydrogen with a known metal impurity concentration, processed through the same S1 to S4 processes. This establishes a quantitative relationship between the enrichment degree of impurities in the epitaxial layer and the content of metal impurities in silicon dichlorodihydrogen. Substituting the calculated mass concentration of the epitaxial layer metal impurities in the silicon dichlorodihydrogen to be tested into this relationship, the original concentration of metal impurities in the silicon dichlorodihydrogen to be tested can be determined.

[0008] This invention transforms the challenge of gas analysis into a solid sample analysis problem by enriching impurities in a gaseous DCS onto a substrate through an epitaxial growth process, thus avoiding the direct handling of high-risk gases. It employs an experimental design with parallel sample, blank substrate, and reagent blank setups, and uses mass difference calculations to eliminate interference introduced by the substrate, reagents, and process. Utilizing highly sensitive ICP-MS / ICP-OES detection and calibration curves established based on standard gases, it achieves a complete analysis process for trace metal impurities in DCS, from sample preparation and interference separation to quantitative back-calculation. This provides a safe, reliable, and highly operable solution for the semiconductor industry to monitor the purity of highly reactive silicon source gases.

[0009] Preferably, before performing S1, the substrate is pretreated, including: ultrasonic cleaning with ultrapure water, immersion in HF solution, rinsing with ultrapure water, and drying with nitrogen gas in sequence.

[0010] Ultrapure water ultrasonic cleaning removes particulate contaminants adhering to the substrate surface. HF solution is used to dissolve the natural oxide layer on the substrate surface. Ultrapure water rinsing and nitrogen drying remove residual solution and water traces, reducing the amount of additional metal introduced by differences in substrate surface condition. This allows subsequent subtraction calculations using a blank substrate as a reference to more clearly identify metal impurities truly originating from the dichlorosilane gas being tested, thus improving the accuracy of the detection results.

[0011] Preferably, the etching solution is a mixture of hydrofluoric acid and nitric acid, with a volume ratio of hydrofluoric acid to nitric acid of 1:3~10.

[0012] The epitaxial layer is generated by a thermal decomposition chemical reaction of silicon dichlorodihydrogen as the silicon source gas under a high-temperature hydrogen atmosphere, and its main chemical component is silicon. The composition of the above etching solution ensures that the silicon substrate and epitaxial layer are completely dissolved. At the same time, the strong oxidizing property of nitric acid can effectively prevent some metal impurities from forming insoluble fluoride precipitates during the digestion process, thereby ensuring that all target metal impurities can be stably transferred to the liquid phase.

[0013] Preferably, the method for obtaining the sample digestion solution, blank substrate digestion solution, and reagent blank solution includes: Collect the etching solutions containing the sample substrate and the blank substrate after complete digestion, respectively, and rinse the digestion container with ultrapure water. Then, add the rinsing solution to the corresponding etching solution. The above etching solution and the etching solution used as a blank control were transferred to volumetric flasks and diluted to the same volume with ultrapure water.

[0014] The digestion vessel is rinsed with ultrapure water to ensure that the residual etching solution adhering to the vessel wall and the metal impurities dissolved in it are completely collected, avoiding the loss of metal impurities; the volume is adjusted to the same level, so that the subsequent concentration measurements are on the same metrological basis and can be directly used for algebraic calculations; this provides a complete and consistent sample basis for subsequent accurate calculations based on concentration differences.

[0015] Preferably, the method for calculating the mass concentration of metal impurities in the epitaxial layer is as follows: C epi =[(C s -C re )-(C b -C re )]×V / m epi ; Among them, C epi C represents the mass concentration of metal impurities in the epitaxial layer. s C b and C re These represent the concentrations of metal impurities in the sample digestion solution, blank substrate digestion solution, and reagent blank solution, respectively, where V is the constant volume and m is the total volume. epi This refers to the quality of the epitaxial layer.

[0016] C re The detection concentration of the reagent blank solution reflects the metal impurities contained in the etching solution and ultrapure water used in the experiment. Calculate C. s -C re and C b -C reThe purpose is to subtract metal impurities from the etching solution and ultrapure water from the measurements of the sample substrate and the blank substrate. C b -C re The concentration of metal impurities in the blank substrate without DCS deposition, after exposure to metal impurities introduced by the etching solution and ultrapure water, originates from the substrate material and process environment. Since the blank substrate and the sample substrate undergo completely identical processing except for DCS exposure, (C...) s -C re )-(C b -C re This can offset the metal impurities introduced by the substrate, resulting in a metal impurity concentration introduced only by the DCS deposition process; Multiply the concentration difference obtained in the previous step by the constant volume V to obtain the total mass of metal impurities contributed by the epitaxial layer, i.e., DCS. Divide the total mass of metal impurities by the mass of the epitaxial layer itself to complete the normalization and obtain the metal impurity content per unit mass of epitaxial layer. The above calculation method transforms the reagent blank and blank substrate settings in the experiment into algebraic operations, eliminating the metallic interference between the reagent system and the substrate itself, and obtaining C. epi It is a pure parameter that excludes non-target sources, providing accurate and reliable intermediate data for subsequent back-calculation of the original impurity concentration in the DCS gas source through calibration curves.

[0017] Preferably, m epi This represents the quality difference between the sample substrate before and after deposition.

[0018] More specifically, m epi The method for obtaining it is as follows: Q1. Place the pre-cleaned and fully dried substrate in a temperature and humidity controlled environment, weigh it using a microbalance with an accuracy of not less than 0.01 mg, and record the initial mass. Q2 After the deposition process is completed and the sample cools down, the sample substrate is taken out, rinsed with ultrapure water, and then dried with high-purity nitrogen to remove environmental particles or non-structural adsorbates that may have adhered to the sample surface during the transfer or temporary storage outside the deposition chamber, while avoiding erosion or quality loss of the epitaxial layer. After cleaning and drying, the post-deposition quality is recorded under the same environmental conditions as Q1. The difference between the post-deposition mass and the initial mass is calculated as m. epi ; This invention determines the epitaxial layer quality by directly measuring the physical increment of mass, avoiding geometric measurements of the epitaxial layer thickness and area. The post-deposition cleaning step allows m epi It reflects the quality of the epitaxial layer that is firmly bonded to the substrate, rather than the quality of temporary surface attachments.

[0019] Preferably, the method for generating the calibration curve includes: Using multiple standard gases of dichlorosilane with known metal impurity concentrations, steps S1 to S4 were performed respectively to obtain a series of corresponding epitaxial layer metal impurity mass concentration values. The epitaxial layer metal impurity mass concentration values ​​were used as the x-axis and the corresponding dichlorosilane metal impurity concentrations were used as the y-axis to perform linear fitting.

[0020] More specifically, the calibration curve is generated as follows: R1 selects multiple dichlorosilane standard gases with different concentration gradients of metal impurities. The concentration of the target metal impurity in each dichlorosilane standard gas must be known and verified by an authoritative method, and its concentration range should cover the expected concentration range of the sample to be tested. R2 treats each dichlorosilane standard gas as a separate sample, processing and measuring it independently according to steps S1 to S4, ultimately obtaining a C value corresponding to that dichlorosilane standard gas. epi During this process, each standard gas generates a data point pair: (C epi Value, the known content of metallic impurities in dichlorosilane standard gas); R3 aggregates all the data point pairs obtained above and uses the epitaxial layer impurity mass concentration value C. epi The x-axis is plotted with the concentration of metallic impurities in the corresponding dichlorosilane standard gas as the y-axis, and the data are plotted on a coordinate graph. Statistical methods such as the least squares method are used to perform linear regression fitting on the data points to obtain a calibration curve. The mathematical expression of the curve is typically Y = kX + b, where k is the slope and b is the intercept. After fitting, the linear correlation coefficient (R²) of the calibration curve needs to be evaluated. 2 This was done to confirm the linear reliability of the response relationship within that concentration range.

[0021] The above method is based on the fundamental principle of the standard curve method in analytical chemistry. It provides a data source for the concentration of metal impurities in dichlorodihydrosilicon gas of known metal impurity concentration by subjecting it to a pretreatment and measurement process identical to that of the actual dichlorodihydrosilicon to be tested. Linear fitting handles potential systematic and random errors during the experiment, and the obtained calibration curve serves as the basis for subsequent analysis of the C0 values ​​of the dichlorodihydrosilicon to be tested. epi The value is converted into a quantitative scale for the concentration of metal impurities.

[0022] Preferably, in S1, the flow rate of the dichlorodihydrosilane to be tested is 5~20 sccm, and the reaction temperature is 1100~1200℃.

[0023] By controlling the flow rate of the silicon dichlorodihydrogen to be tested at 5~20 sccm and maintaining the reaction temperature at 1100~1200℃, the epitaxial layer can grow stably and uniformly on the substrate surface in the form of a single crystal. This is beneficial for the effective capture and binding of trace metal impurities in the gas into the lattice of the epitaxial layer, thereby realizing the quantitative transfer of metal impurities from the gas phase to the solid phase.

[0024] Preferably, in S1, the thickness of the epitaxial layer is 30~50μm.

[0025] Controlling the epitaxial layer thickness to 30~50μm ensures that the metallic impurities transferred and enriched from dichlorosilane have sufficient absolute mass to meet the sensitivity requirements of subsequent instruments such as ICP-MS for the detection of trace components. The above thickness range can be stably controlled by adjusting the time under the deposition parameters.

[0026] Preferably, the sample substrate and the blank substrate are cleaned before S2 digestion.

[0027] More specifically, the cleaning process involves rinsing with ultrapure water and drying with nitrogen to remove environmental particles or non-structural adsorbates that may adhere to the surface of the sample substrate and blank substrate during the transfer from the reaction chamber to the digestion container after the deposition process. If these adsorbates are not removed, they will enter the etching solution during the subsequent digestion process, and the metallic impurities they may contain will be included in the detection value, thereby interfering with the accurate assessment of the impurity content of the substrate and epitaxial layer.

[0028] The technical solution of this invention can achieve the following technical effects: This invention transforms the challenge of gas analysis into a solid sample analysis problem by enriching impurities in a gaseous DCS onto a substrate through an epitaxial growth process, thus avoiding the direct handling of high-risk gases. It employs an experimental design with parallel sample, blank substrate, and reagent blank setups, and uses mass difference calculations to eliminate interference introduced by the substrate, reagents, and process. Utilizing highly sensitive ICP-MS / ICP-OES detection and calibration curves established based on standard gases, it achieves a complete analysis process for trace metal impurities in DCS, from sample preparation and interference separation to quantitative back-calculation. This provides a safe, reliable, and highly operable solution for the semiconductor industry to monitor the purity of highly reactive silicon source gases. Attached Figure Description

[0029] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0030] Figure 1 This is a schematic flowchart of the method for testing dichlorosilane metal impurities according to the present invention. Detailed Implementation

[0031] The technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments.

[0032] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains. The terminology used in this specification is for the purpose of describing particular embodiments only and is not intended to be limiting of the invention. The term "and / or" as used herein includes any and all combinations of one or more of the associated listed items.

[0033] The above description is only an overview of the technical solution of this application. In order to better understand the technical means of this application and to implement it in accordance with the contents of the specification, and to make the above and other objects, features and advantages of this application more obvious and understandable, specific embodiments of this application are given below. Example 1:

[0034] This embodiment provides a method for testing metallic impurities in dichlorosilane (DCS). Taking iron as an example, the specific steps are as follows: S1 takes two N-type monocrystalline silicon substrates of the same specification from the same wafer: crystal orientation <100> The resistivity is >10000Ω·cm, the size is 10mm×10mm, the thickness is 0.5mm, and the total metal impurity content is <10ppt as characterized. The two substrates are pretreated sequentially as follows: At room temperature, it was immersed in ultrapure water with a resistivity greater than 18.2 MΩ·cm and ultrasonically cleaned for 10 minutes; then, it was immersed in 0.5% electronic grade hydrofluoric acid (HF) solution for 30 seconds to remove the natural oxide layer on the surface; then, it was rinsed 5 times with ultrapure water; finally, the surface was dried with high-purity nitrogen gas with a purity greater than 5N; one piece was marked as the sample substrate and the other as the blank substrate. The sample substrate and the blank substrate were placed in an environment with controlled temperature and humidity of 23±1℃. Their masses were weighed using a microbalance with an accuracy of 0.01mg. The initial mass of the sample substrate was recorded as 230.45mg and the initial mass of the blank substrate was recorded as 230.38mg. Subsequently, the sample substrate and the blank substrate were placed in a chemical vapor deposition reaction chamber. High-purity hydrogen (6N) was introduced into the reaction chamber as a carrier gas at a flow rate of 1200 sccm. The reaction temperature was stabilized at 1150℃. The DCS gas to be measured was then introduced into the reaction chamber at a flow rate of 10 sccm, and the deposition time was controlled at 80 minutes. During this process, only the surface of the sample substrate grew an epitaxial layer. The blank substrate underwent the exact same temperature, hydrogen atmosphere, and time treatment, except that no DCS gas was introduced. After deposition, once the reaction chamber has cooled to room temperature, the sample substrate and blank substrate are removed, rinsed with ultrapure water, and dried with high-purity nitrogen. Under the same temperature and humidity conditions, the mass of the sample substrate after deposition is weighed again using the same microbalance; the mass is 230.85 mg. The epitaxial layer mass m is calculated. epi =0.40mg; S2 prepared an etching solution with a hydrofluoric acid to nitric acid volume ratio of 1:5. Three 30 mL portions of the etching solution were placed in three polytetrafluoroethylene (PTFE) digestion vessels. The sample substrate was immersed in the first portion of the etching solution, the blank substrate was immersed in the second portion, and the third portion served as a reagent blank, without any substrate. The three digestion vessels were sealed and allowed to stand at room temperature until the substrate was completely dissolved. The three portions of digested liquid were collected, and the inner walls of each digestion vessel were rinsed three times with ultrapure water. The rinsing liquid was combined with the corresponding solution. Finally, the three solutions were transferred to 50 mL PTFE volumetric flasks, diluted to the mark with ultrapure water, and shaken well to obtain the sample digestion solution, the blank substrate digestion solution, and the reagent blank solution, respectively.

[0035] S3 used inductively coupled plasma mass spectrometry (ICP-MS) to detect the above sample digestion solution, blank substrate digestion solution, and reagent blank solution. The instrument was calibrated using a standard solution containing the same concentration of mixed acid matrix; the measurements were as follows: The concentration of iron C in the sample digestion solution s =12.5ppt The iron concentration C in the blank substrate digestion solution b =2.8ppt The concentration of iron C in the reagent blank solution re =0.5ppt; S4 substitutes the above measured values ​​into the formula to calculate the iron impurity mass concentration in the epitaxial layer: C epi =[(C s -C re )-(C b -C re )]×V / m epi =[(12.5-0.5)-(2.8-0.5)]×50 / 0.4=1212.5ppb; S5 calculates the iron impurity concentration C in the epitaxial layer. epi =1212.5ppb, substituted into the pre-generated iron impurity calibration curve, which was obtained as follows: using ten DCS standard gases with known iron impurity concentrations (1, 5, 10, 20, 40, 60, 80, 120, 160, and 200 ppt, respectively), steps S1 to S4 were repeated to obtain ten corresponding C epi The values ​​and the resulting data point pairs are shown in Table 1. Table 1 Data Point Pairs With C epi Using the x-axis as the horizontal axis and the corresponding known iron impurity concentration as the y-axis, a linear regression fitting was performed using the least squares method to obtain the calibration curve equation Y = 0.0501X + 0.12, where the linear correlation coefficient R0 is... 2 >0.999; C epi Substituting 1212.5 into the calibration curve equation, we obtain the concentration of iron impurities in the DCS gas as 60.9 ppt. Example 2:

[0036] This embodiment provides a method for testing metallic impurities in dichlorosilane (DCS). Taking copper as an example, the specific steps are as follows: S1 takes two N-type monocrystalline silicon substrates of the same specification from the same wafer: crystal orientation <100> The substrates had a resistivity >10000 Ω·cm, dimensions of 10 mm × 10 mm, and a thickness of 0.5 mm. Characterization showed a total metal impurity content <10 ppt. Two substrates were pretreated sequentially as follows: at room temperature, they were immersed in ultrapure water with a resistivity greater than 18.2 MΩ·cm and ultrasonically cleaned for 10 minutes; subsequently, they were immersed in a 0.5% electronic-grade hydrofluoric acid solution for 30 seconds to remove the natural oxide layer on the surface; then, they were rinsed 5 times with ultrapure water; finally, the surface was dried with high-purity nitrogen gas with a purity greater than 5N. One substrate was labeled as the sample substrate, and the other as the blank substrate. The sample substrate and the blank substrate were placed in an environment with controlled temperature and humidity of 23±1℃. Their masses were weighed using a microbalance with an accuracy of 0.01mg. The initial mass of the sample substrate was recorded as 230.42mg and the initial mass of the blank substrate was recorded as 230.35mg. Subsequently, the sample substrate and the blank substrate were placed in a chemical vapor deposition reaction chamber. High-purity hydrogen (6N) was introduced into the reaction chamber as a carrier gas at a flow rate of 1200 sccm. The reaction temperature was stabilized at 1150℃. The DCS gas to be measured was then introduced into the reaction chamber at a flow rate of 10 sccm, and the deposition time was controlled at 80 minutes. During this process, only the surface of the sample substrate grew an epitaxial layer. The blank substrate underwent the exact same temperature, hydrogen atmosphere, and time treatment, except that no DCS gas was introduced. After deposition, once the reaction chamber has cooled to room temperature, the sample substrate and blank substrate are removed, rinsed with ultrapure water, and dried with high-purity nitrogen. Under the same temperature and humidity conditions, the mass of the sample substrate after deposition is weighed again using the same microbalance; the mass is 230.82 mg. The epitaxial layer mass m is calculated. epi =0.40mg; S2 prepared an etching solution with a hydrofluoric acid to nitric acid volume ratio of 1:5. Three 30 mL portions of the etching solution were placed in three PTFE digestion vessels. The sample substrate was immersed in the first portion of the etching solution, the blank substrate was immersed in the second portion, and the third portion of the etching solution was used as a reagent blank without any substrate. The three digestion vessels were sealed and allowed to stand at room temperature until the substrate was completely dissolved. The three portions of digested liquid were collected, and the inner walls of each digestion vessel were rinsed three times with ultrapure water. The rinsing liquid was combined with the corresponding solution. Finally, the three solutions were transferred to 50 mL PTFE volumetric flasks, diluted to the mark with ultrapure water, and shaken well to obtain the sample digestion solution, the blank substrate digestion solution, and the reagent blank solution, respectively. S3 used ICP-MS to detect the above sample digestion solution, blank substrate digestion solution, and reagent blank solution. The instrument was calibrated using a standard solution containing the same concentration of mixed acid matrix; the results were as follows: The concentration of copper C in the sample digestion solution s =8.2ppt The concentration of copper in the blank substrate digester, C b =1.5ppt The concentration of copper C in the reagent blank solution re =0.3ppt; S4 substitutes the above measured values ​​into the formula to calculate the copper impurity mass concentration in the epitaxial layer: C epi =837.5ppb; S5 calculates the copper impurity concentration C in the epitaxial layer. epi =837.5ppb, substituted into the pre-generated copper impurity calibration curve, which was obtained as follows: using ten DCS standard gases with known copper impurity concentrations (1, 5, 10, 20, 40, 60, 80, 120, 160, and 200 ppt, respectively), steps S1 to S4 were repeated to obtain ten corresponding C epiThe values ​​and the resulting data point pairs are shown in Table 2. Table 2 Data Point Pairs With C epi Using the x-axis as the horizontal axis and the corresponding known copper impurity concentration as the y-axis, a linear regression fitting was performed using the least squares method to obtain the calibration curve equation Y = 0.0632X + 0.08, where the linear correlation coefficient R0 is... 2 >0.999; C epi Substituting 837.5 into the calibration curve equation, the concentration of copper impurities in the DCS gas is found to be 53.0 ppt.

[0037] Comparative Example 1: This comparative example employs a typical method known in the art: cryogenic cold trap adsorption / transfer-inductively coupled plasma mass spectrometry (ICP-MS) coupled with other techniques to perform parallel detection of the analyte dichlorosilane gas from the same source as in Examples 1 and 2. The specific method is as follows: T1 introduces the DCS gas to be tested into a stainless steel or Monel alloy cold trap maintained at -80°C at a constant flow rate of 50 sccm. The DCS gas condenses into a liquid state and accumulates on the inner wall of the cold trap, while the metal impurities in the gas are adsorbed or co-condensed. After T2 enrichment, the cold trap is slowly heated to room temperature to completely vaporize the condensed DCS and purge it with high-purity inert gas into a PTFA container pre-filled with a small amount of dilute nitric acid. During this process, the DCS undergoes violent hydrolysis upon contact with water, and metallic impurities are transferred to the acid solution. To ensure complete transfer of impurities, the inner wall of the cold trap needs to be rinsed multiple times with a small amount of dilute acid, and the rinsing solutions are combined. T3 brought the combined acid solution to a final volume of 30 mL with ultrapure water, and then analyzed it using ICP-MS. The instrument was calibrated using a standard solution that matched the acid matrix. T4 calculates the original concentration of metal impurities in the DCS gas based on the sampling volume, constant volume, and impurity concentration measured by ICP-MS.

[0038] After three tests, the iron content in the dichlorosilane tested in Example 1 was 52.1 ppt, 71.8 ppt, and 45.3 ppt, respectively, with an average value of 56.4 ppt and a relative deviation of -7.4%. After three tests, the copper content in the dichlorosilane tested in Example 2 was 61.5 ppt, 42.2 ppt, and 57.2 ppt, respectively, with an average value of 57.2 ppt and a relative deviation of +7.9%. The average concentrations of iron and copper impurities measured in the examples and comparative examples are on the same order of magnitude, indicating that the results obtained by the test method of the present invention are basically consistent with the measurement center trend of the prior art method, and there is no deviation in order of magnitude.

[0039] Existing technologies require operating high-risk DCS gases at low temperatures and involve violent hydrolysis steps, which places high demands on the corrosion resistance of equipment and are complex to operate. This invention completely avoids the direct processing of active gases and significantly improves safety.

[0040] Although this application has been described in conjunction with specific features and embodiments, it is obvious that various modifications and combinations can be made thereto without departing from the spirit and scope of this application. Accordingly, this specification and drawings are merely exemplary illustrations of the application as defined herein, and are to be considered as covering any and all modifications, variations, combinations, or equivalents within the scope of this application. Clearly, those skilled in the art can make various alterations and modifications to this application without departing from its scope. Thus, if such modifications and modifications fall within the scope of this application and its equivalents, this application intends to include such modifications and modifications.

Claims

1. A method for testing dichlorosilane metallic impurities, characterized in that, include: S1 introduces the silicon dichlorodihydrogen to be tested into the chemical vapor deposition reaction chamber, deposits an epitaxial layer on the substrate surface, and obtains the sample substrate; A blank substrate without silicon dichlorosilane was prepared under the same conditions; S2 Take three identical etching solutions, immerse the sample substrate and blank substrate in two of the etching solutions for complete digestion, and use the other part as a blank control. Collect the three etching solutions and adjust the volume to obtain sample digestion solution, blank substrate digestion solution and reagent blank solution. S3 detects the concentration of metal impurities in the sample digestion solution, blank substrate digestion solution, and reagent blank solution, respectively. S4 calculates the mass concentration of metal impurities in the epitaxial layer based on the concentration of metal impurities in the sample digestion solution, blank substrate digestion solution, and reagent blank solution; S5 substitutes the mass concentration of the metal impurities in the epitaxial layer into the pre-generated calibration curve to obtain the concentration of metal impurities in the dichlorosilane to be tested.

2. The method for testing dichlorosilane metal impurities according to claim 1, characterized in that, Before S1, the substrate is pretreated, including: ultrasonic cleaning with ultrapure water, immersion in HF solution, rinsing with ultrapure water, and drying with nitrogen.

3. The method for testing dichlorosilane metal impurities according to claim 1, characterized in that, The etching solution is a mixture of hydrofluoric acid and nitric acid, with a volume ratio of hydrofluoric acid to nitric acid of 1:3~10.

4. The method for testing dichlorosilane metal impurities according to claim 1, characterized in that, The methods for obtaining the sample digestion solution, blank substrate digestion solution, and reagent blank solution include: Collect the etching solutions containing the sample substrate and the blank substrate after complete digestion, respectively, and rinse the digestion container with ultrapure water. Then, add the rinsing solution to the corresponding etching solution. The above etching solution and the etching solution used as a blank control were transferred to volumetric flasks and diluted to the same volume with ultrapure water.

5. The method for testing dichlorosilane metal impurities according to claim 1, characterized in that, The method for calculating the mass concentration of metal impurities in the epitaxial layer is as follows: C epi =[(C s -C re )-(C b -C re )]×V / m epi ; Among them, C epi C represents the mass concentration of metal impurities in the epitaxial layer. s C b and C re These represent the concentrations of metal impurities in the sample digestion solution, blank substrate digestion solution, and reagent blank solution, respectively, where V is the constant volume and m is the total volume. epi This refers to the quality of the epitaxial layer.

6. The method for testing dichlorosilane metal impurities according to claim 5, characterized in that, The m epi The difference in mass between the sample substrate before and after deposition is denoted as .

7. The method for testing dichlorosilane metal impurities according to claim 1, characterized in that, The method for generating the calibration curve includes: Using multiple standard gases of dichlorosilane with known metal impurity concentrations, steps S1 to S4 were performed respectively to obtain a series of corresponding epitaxial layer metal impurity mass concentration values. The epitaxial layer metal impurity mass concentration values ​​were used as the x-axis and the corresponding dichlorosilane metal impurity concentrations were used as the y-axis to perform linear fitting.

8. The method for testing dichlorosilane metal impurities according to claim 1, characterized in that, In S1, the flow rate of the dichlorodihydrosilane to be tested is 5~20 sccm, and the reaction temperature is 1100~1200℃.

9. The method for testing dichlorosilane metal impurities according to claim 1, characterized in that, In S1, the thickness of the epitaxial layer is 30~50μm.

10. The method for testing dichlorosilane metal impurities according to claim 1, characterized in that, The sample substrate and blank substrate were cleaned before S2 digestion.

Images