Pretreatment method for trace metal impurity detection of high-purity electronic-grade silicon-based precursor material

Abstract

The invention discloses a pretreatment method for trace metal impurity detection of a high-purity electronic-grade silicon-based precursor material. The method comprises the following steps: heating and volatilizing a liquid sample in an ultrapure PTFE / PFA beaker, and redissolving by using an electronic-grade inorganic acid-organic acid composite acid system; uniform heating is achieved through a special heating plate, pretreatment volatile organic waste gas is treated in cooperation with a hundred-grade dust-free environment, a clean operation table and a tail gas condensation and collection system, meanwhile, metal impurity concentration of different degrees can be achieved through the proportion of the use amount of a pretreatment sample to the use amount of redissolved inorganic acid, the detection limit is effectively reduced to 1-5 ppt, and the detection efficiency is improved. The recovery rate of the metal impurity detection pretreatment method reaches 85%-115%. The method breaks through the defects of long pretreatment time, silicon and carbon residues, metal loss caused by severe pretreatment reaction and the like of a traditional hydrolysis and digestion method, the treatment time is shortened to 2-4 hours, the G4 / G5 level (single metal 1t, 10ppt) detection requirement is effectively met, and a high-sensitivity and low-background-interference detection scheme is provided for impurity management and control of a high-purity silicon-based precursor in an advanced process.

Description

Technical Field

[0001] This invention relates to the field of high-purity precursor material analysis technology for semiconductor manufacturing, specifically to a pretreatment method for detecting trace metal impurities in high-purity electronic-grade silicon-based precursor materials. Background Technology

[0002] In the miniaturization of semiconductor devices, precursor materials serve as the core film-forming medium, undertaking the construction of key structures such as high-k / low-k dielectric layers, diffusion barrier layers, and metal interconnect layers. Among them, high-purity silicon-based precursors form silicon-oxygen / silicon-nitride passivation layers and dielectric layers through epitaxial growth and deposition processes such as CVD / ALD. The purity of the material directly affects transistor performance and device reliability.

[0003] With the development of semiconductor manufacturing technology and chip miniaturization processes, the requirements for the purity and impurity content of semiconductor materials, especially precursor film-forming materials, are becoming increasingly stringent. In particular, the control of trace metal impurities during the production or application of silicon-based precursor materials is a key focus for material manufacturers and downstream chip manufacturers. The emergence of advanced transistor structures has facilitated further reduction in integrated circuit feature sizes, also placing higher demands on material quality. The metal impurity control standard for precursor materials has exceeded the G3 threshold, and some application scenarios require G4 / G5 level specifications (single metal impurity <10ppt). This stringent standard poses a significant challenge to material production and testing technologies, especially in the field of Low-K silicon-based precursors, where the accurate detection of metal impurities faces multiple technical bottlenecks. In traditional detection procedures, the analysis of metallic impurities in silicon-based materials requires pre-hydrolysis combined with high-purity hydrofluoric acid digestion to avoid the deposition of silicon oxide and carbon residue affecting the ICP-MS system inlet or tubing. Although direct hydrofluoric acid digestion can shorten the processing time, the vigorous reaction kinetics can easily lead to the loss of metal elements, causing the recovery rate to deviate from the detection benchmark.

[0004] However, since most Low-K silicon-based precursor materials undergo extremely slow hydrolysis under mild conditions, existing methods significantly reduce detection efficiency. If samples are directly digested with hydrofluoric acid without prior hydrolysis, the reaction will be too vigorous, resulting in the loss of metal impurities and ultimately affecting the metal impurity recovery rate, failing to meet detection requirements. Furthermore, for Low-K organosilicon precursors, their molecular structure leads to a significant reduction in conventional hydrolysis efficiency, severely restricting detection timeliness. While existing dilute acid digestion schemes can alleviate the intensity of the reaction, they are insufficient to effectively decompose the silicon-based framework and organic components in the material, and the residues can still cause system malfunctions such as plasma instability and carbon buildup in the mass spectrometer cone. Summary of the Invention

[0005] This invention addresses the challenges of high-purity silicon-based precursor metal impurity detection samples being greatly affected by the environment, low hydrolysis efficiency during pretreatment, metal loss due to digestion, and interference from silicon-carbon residues on the instrument. It provides a pretreatment method for detecting trace metal impurities in high-purity electronic-grade silicon-based precursor materials.

[0006] The technical solution of this invention: A pretreatment method for detecting trace metal impurities in high-purity electronic-grade silicon-based precursor materials is disclosed. The core steps involve a "dry-resolution" technique: ① heating and evaporating 80-120 mL of liquid silicon-based precursor → ② resolution with 5-10 mL of organic-inorganic composite acid → ③ adjusting the volume to 30 mL → ④ simultaneous treatment of a blank sample. This method achieves the concentration and enrichment of metal impurities. It is particularly suitable for liquid silicon-based precursor materials with boiling points of 20-160℃ (such as trichlorosilane, tetramethylsilane, tetramethylcyclotetrasiloxane, octamethylcyclotetrasiloxane, etc.). All steps are performed in a Class 100 cleanroom with an ultra-clean bench.

[0007] A pretreatment method for detecting trace metal impurities in high-purity electronic silicon-based precursor materials includes the following steps: Step 1: Take the electronic-grade silicon-based precursor sample to be tested and heat it on a heating plate in a container until the sample is completely volatilized; Step 2: Add the organic-inorganic composite acid solution to the container after evaporation and reconstitute. Step 3: After redissolution, continue to add the above organic-inorganic composite acid solution and make up to volume. Step 4: Perform the same operation as above for the blank sample. Complete the pretreatment for detecting trace metal impurities in high-purity electronic silicon-based precursor materials.

[0008] Preferably, the high-purity electronic-grade silicon precursor material is a silicon-based precursor that is liquid at room temperature and has a boiling point of 20-160°C, including but not limited to trichlorosilane, tetramethylsilane, tetramethylcyclotetrasiloxane, octamethylcyclotetrasiloxane, tetraethyl orthosilicate, or other silicon-based precursor materials.

[0009] Preferably, the container volume in step 1 is 150-200mL; the volume of the electronic-grade silicon-based precursor sample to be tested is 90-100mL, and the container material is ultrapure polytetrafluoroethylene or soluble polytetrafluoroethylene. The heating plate has a coating of ultrapure polytetrafluoroethylene (PTFE) or soluble polytetrafluoroethylene (PFA).

[0010] Preferably, the heating plate in step 1 has holes that match the size of the container, allowing the container to be embedded for heating; the embedding depth is not less than 5cm to ensure heating efficiency; the heating plate has 8-12 holes, with good temperature uniformity at each hole location and a temperature difference of less than 1℃.

[0011] Preferably, in the heating operation described in step 1, the sample is preheated to 5-10°C above the theoretical boiling point of the sample before being placed on the heating plate. When sporadic bubbles begin to appear in the beaker, the temperature of the heating plate is reduced to the theoretical boiling point of the sample to avoid explosive boiling and sample loss.

[0012] Preferably, the temperature control process in step 1 involves setting a three-stage temperature control program based on preliminary sample boiling point testing: preheating stage (boiling point +10-20 ℃); main evaporation stage (temperature within 5 ℃ above boiling point); and residue removal stage (boiling point +20-40 ℃), achieving a balance between efficient matrix volatilization and impurity retention. Simultaneously, dynamic nitrogen purging at a flow rate of 50-100 mL / min is combined to improve sample evaporation efficiency and effectively solve the problem of residual organic carbon in traditional evaporation. Subsequent comparative experiments have verified that this technology can increase the recovery rate of metal impurities to over 90% (compared to only 70-80% for traditional methods, with some metals recovering less than 50%). First, the sample is preheated to 10-20 ℃ above the theoretical boiling point. When sporadic bubbles begin to appear on the sample surface, the preset temperature of the heating plate is lowered to slightly above the sample boiling point to improve heating efficiency while avoiding sample boiling over and loss. When the sample is nearly dry, the heating plate temperature is increased (boiling point +20-40 ℃), and heating is maintained for 30 min. Preferably, the inorganic acid in the organic-inorganic composite acid solution in step (2) includes, but is not limited to, one of electronic nitric acid, electronic grade hydrochloric acid, or electronic grade hydrofluoric acid; the content of a single metal impurity in the inorganic acid is less than 10 ppt; the sample reconstitution heating temperature in step (2) is 50-60 ℃, and the reconstitution treatment time is 30-60 min, preferably 60 min.

[0013] More preferably, in step (2), the organic acid in the organic-inorganic composite acid solution is glycolic acid, and the volume ratio of the organic acid is 0.5-1%.

[0014] To address the structural characteristics of Low-K organosilicon precursors, which feature "high organic carbon content and stable silicon-oxygen bonds," an organic-inorganic composite acid system was selected. Hydroxyacetic acid was introduced as both a corrosion inhibitor and a complexing agent: on the one hand, by coordinating hydroxyl groups with silicon atoms, it weakens the silicon-oxygen bond strength and promotes the decomposition of the silicon-based framework by the inorganic acid; on the other hand, it can form stable chelates with metal ions, inhibiting the loss of metal impurities during heating and resolution, and improving the recovery rate.

[0015] Preferably, the blank sample in step (4) is selected from ultrapure organic solvents, ultrapure water or ultrapure inorganic acids; ultrapure organic solvents include, but are not limited to, ultrapure NMP, n-hexane, DMSO or others; the content of single metal impurities in the ultrapure solvents, ultrapure water or ultrapure inorganic acids used in step (4) is less than 10 ppt. Preferably, in the method, the evaporation and reconstitution operations are all performed in a clean bench in a Class 100 cleanroom. The isolation window of the clean bench is not opened throughout the evaporation process to avoid air disturbance and contamination from impurities. The exhaust gas pipeline of the clean bench is connected to a dedicated exhaust gas treatment system.

[0016] Preferably, the exhaust gas treatment system is a multi-stage condensation recovery plus two-stage activated carbon adsorption system. The condensation temperature is 0-5℃. The activated carbon adsorbs the organic waste gas that cannot be condensed, which can effectively prevent the volatilization of organic matter in the sample from being directly emitted into the atmosphere.

[0017] Preferably, in the method, the number of parallel samples is 3-5 groups, and the number of control samples is 3-5 groups. This invention relates to a pretreatment method for detecting trace metal impurities in high-purity electronic silicon-based precursor materials. After the sample is heated and evaporated, it enters the exhaust gas treatment system equipped in the ultra-clean bench: the condensation temperature is 0-5℃, preferably multi-stage condensation recovery coupled with two-stage activated carbon adsorption, and the condensation temperature is preferably 0℃, which can improve the condensation efficiency and effectively prevent the emission of volatile organic compounds. This invention effectively solves the problem of background interference in the detection of trace metals in high-purity silicon-based materials through a three-pronged innovation of sample concentration, equipment anti-pollution design, and environmental control. At the same time, it effectively avoids system failures such as instrument plasma instability and mass spectrometer cone carbon buildup caused by the residue of silicon-based framework and organic components in the sample. The present invention has the following beneficial effects: This invention addresses the challenge of detecting metallic impurities in high-purity silicon-based precursors, and proposes a pretreatment method with the following advantages: (1) Improved the detection sensitivity of high-purity silicon-based precursors and effectively avoided background interference from the environment and blank samples: The evaporation method can select more sample volumes, such as 80-120mL. Subsequent reconstitution and volume adjustment can concentrate the sample, and metal impurities can be enriched by 3-4 times. Combined with the organic=inorganic composite acid system for reconstitution (metal content <10ppt), the detection limit of ICP-MS is optimized from the traditional 10-50ppt to 1-5ppt. At the same time, the recovery rate of metal impurity elements can be improved to meet the detection standards of G4 / G5 grade precursor materials. The metal recovery rate reaches 85%-115%, preferably above 90%. However, if a large amount of sample is selected for strong acid digestion, the gold impurity content of the sample itself is too low, and it is impossible to concentrate the gold impurities in the sample, resulting in low recovery rate or great interference from the environment.

[0018] (2) The sample evaporation-resolution treatment effectively avoids the influence of silicon and organic matter residues on the instrument: the evaporation method bypasses inefficient hydrolysis and strong acid digestion (the treatment time is shortened from 8-24h to 2-4h), eliminates the interference of silicon and carbon residues on the mass spectrometry system, and avoids the danger or loss of the gold and impurities to be tested caused by violent reaction during acid digestion.

[0019] (3) The exhaust gas condensation system effectively realizes the recovery and treatment of samples, avoids the large-scale volatilization of organic matter and environmental pollution, and the VOCs condensation recovery rate is >95%. With the help of two-stage activated carbon adsorption, the exhaust gas treatment efficiency can be increased to over 99%.

[0020] (4) It provides an effective means for the analysis of metal impurities in high-purity silicon-based materials used in semiconductor manufacturing, while also being efficient and environmentally friendly, and can provide key technical support for advanced semiconductor manufacturing. Detailed Implementation The technical solutions of the present invention will be clearly and completely described below with reference to specific embodiments. Obviously, the described embodiments are only a part of the embodiments of the present invention, and not all of them. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0021] Example 1 (Organic-Inorganic-Compound Acid System) Detection of trace metal impurities in electronic-grade tetramethylsilane Materials and Equipment: Sample: Electronic grade tetramethylsilane (boiling point 26℃, declared metal impurity content <50ppt) Instruments: PTFE beaker (200mL), PTFE-coated heating plate (5cm hole depth, temperature uniformity ±0.5℃), electronic-grade nitric acid (30% volume concentration, metal impurities <10ppt). Environment: Class 100 cleanroom with ultra-clean bench, equipped with a two-stage condensation exhaust gas recovery system, condensation temperature 0℃.

[0022] Operating steps: 1. Pour 90 mL of tetramethylsilane sample into a PTFE beaker and place it in the holes of a heating plate. The heating plate has holes that match the size of the beaker, allowing the beaker to be embedded for heating. The embedding depth should be no less than 5 cm to ensure heating efficiency. The heating plate has 12 holes, and the temperature uniformity of each hole is good, with a temperature difference of less than 1 °C. Before placing the sample on the heating plate, preheat it to 5-10 °C above the theoretical boiling point of the sample. When sporadic bubbles begin to appear in the beaker, lower the temperature of the heating plate to the theoretical boiling point of the sample to avoid explosive boiling and sample loss.

[0023] The heating plate was set to 40℃. When the sample began to boil, the heating plate temperature was adjusted to 30℃ and heated until complete evaporation (about 1.5h). When the sample was nearly dry, the heating plate temperature was adjusted to 60℃ and heated for another 30 min. At the same time, nitrogen was dynamically purged (flow rate 80mL / min). A total of 3 parallel samples were used. 2. Add 5 mL of organic-inorganic composite acid (organic acid is glycolic acid, and inorganic acid is HNO3 and HF, with a volume ratio of glycolic acid, HNO3 and HF of 1:5:4) and redissolve at 50℃ for 60 min. Then add the composite acid to make up the volume to 30 mL. 3. The blank sample is an electronic-grade composite acid, which is processed simultaneously with the test sample. There are 3 sets of blank samples. 4. Elements such as Na, Mg, Al, K, Ca, Cr, Fe, Cu, and Zn were detected by ICP-MS (Agilent 8900). The mass numbers used are shown in the table below: Table 1

[0024] Test results: Table 2

[0025] Example 2 (Organic-Inorganic Composite Acid System) Detection of trace metal impurities in electronic-grade tetramethylcyclotetrasiloxane Materials and Equipment: Sample: Electronic grade tetramethylcyclotetrasiloxane (boiling point 134℃, declared metal impurity content <50ppt) Instruments: PTFE beaker (200mL), PTFE-coated heating plate (hole depth 5cm, temperature uniformity ±0.5℃), electronic grade nitric acid (metallic impurities <10ppt). Environment: Class 100 cleanroom with ultra-clean bench, equipped with a two-stage condensation exhaust gas recovery system, condensation temperature 0℃.

[0026] Operating steps: 1. Pour 90 mL of tetramethylcyclosilane sample into a PTFE beaker and place it in the holes of a heating plate. Set the heating plate to 150 °C. When the sample begins to boil, adjust the heating plate temperature to 135 °C and heat until complete volatilization (about 1.5 h). When the sample is nearly dry, adjust the heating plate temperature to 160 °C (residue removal stage) and continue heating for 30 min, while simultaneously purging with nitrogen gas (flow rate 80 mL / min). A total of 3 parallel samples were prepared. Everything else is the same as in Example 1.

[0027] Table 3

[0028] Test results: Table 4

[0029] Comparative Example 1 The difference from Example 1 is that the acid used for resolution is an electronic-grade inorganic acid (the volume ratio of inorganic acid HNO3 to HF is 5:4), and the blank sample is an electronic-grade inorganic acid. Otherwise, it is the same as Example 1.

[0030] The elements Na, Mg, Al, K, Ca, Cr, Fe, Cu, and Zn were detected by ICP-MS (Agilent 8900). The mass fractions used are shown in Table 1, and the detection results are shown in Table 5. Table 5

[0031] Comparative Example 2: Detection of trace metal impurities in electronic-grade tetramethylcyclotetrasiloxane The difference from Example 1 is that the acid used for resolution is only electronic-grade glycolic acid, and the blank sample is electronic-grade glycolic acid; otherwise, it is the same as Example 1.

[0032] The test results are shown in Table 6 below: Table 6

[0033] Comparative Example 3 Sample: Electronic grade tetramethylcyclotetrasiloxane (boiling point 134℃, declared metal impurity content <50ppt) Instruments: PTFE beaker (200mL), PTFE-coated heating plate (5cm hole depth, temperature uniformity ±0.5℃), electronic-grade nitric acid (metallic impurities <10ppt) Environment: Class 100 cleanroom with ultra-clean bench, equipped with a two-stage condensation exhaust gas recovery system, condensation temperature 0℃.

[0034] Operating steps: 1. Pour 10 mL of tetramethylcyclotetrasiloxane sample into a PTFE beaker, place it in the hole of a heating plate, set the heating plate to 60°C, and slowly add 49% electronic grade hydrofluoric acid dropwise to avoid violent reaction and sample boiling. After the addition is complete, continue heating for 30 min to ensure the sample is fully digested. Prepare a total of 3 parallel samples. 2. Add electronic-grade nitric acid to bring the volume to 30 mL; 3. The blank sample is an equal volume of electronic-grade nitric acid, which is processed simultaneously with the test sample. There are 3 sets of blank samples. 4. Elements including Na, Mg, Al, K, Ca, Cr, Fe, Cu, and Zn were detected by ICP-MS (Agilent 8900). The mass fractions used are shown in Table 1, and the detection results are shown in Table 7. Table 7

[0035] By comparing and analyzing the detection data of the above embodiments and comparative examples, it can be found that the method of the present invention, after pretreatment of high-purity metal precursors, allows for precise analysis of the content of trace metal impurities using instruments, and the spiked recovery rate is maintained between 90-110%, with minimal sample loss and environmental pollution. However, in the comparative examples, the detection results using a single inorganic acid system (Comparative Examples 1 and 2) show that, due to the lack of chelation by organic acids, the recovery rates of some metal elements such as Zn, Ga, As, Ta, and Pb are less than 80%, with only a few elements maintaining a recovery rate in the 90-110% range. Furthermore, the strong acid digestion pretreatment (Comparative Example 3), due to its long digestion time and vigorous reaction, resulted in the loss of trace metal impurities such as Na, K, Cu, Zn, Ge, As, Sr, and Zr, or environmental pollution, which is directly reflected in the recovery rate not meeting the detection requirements (exceeding the spiked recovery rate range of the embodiments). In summary, the pretreatment method of the present invention, which involves heating with a deep-space heating plate to evaporate and redissolve, can effectively improve the pretreatment efficiency, while achieving the concentration of trace metal impurities. It effectively avoids sample splashing and loss or environmental and human contamination caused by violent reactions during digestion. Compared with strong acid digestion, the pretreatment operation of the present invention is more reliable.

[0036] The above are merely preferred embodiments of the present invention and are not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A pretreatment method for detecting trace metal impurities in high-purity electronic silicon-based precursor materials, characterized in that, Includes the following steps: Step 1: Take the electronic-grade silicon-based precursor sample to be tested and heat it on a heating plate in a container until the sample is completely volatilized; Step 2: Add the organic-inorganic composite acid solution to the container after evaporation and reconstitute. Step 3: After redissolution, continue to add the above organic-inorganic composite acid solution and make up to volume. Step 4: Perform the same operation as above for the blank sample. Complete the pretreatment for detecting trace metal impurities in high-purity electronic silicon-based precursor materials.

2. The method according to claim 1, characterized in that, The high-purity electronic-grade silicon precursor material is a silicon-based precursor that is liquid at room temperature and has a boiling point of 20-160℃, including but not limited to trichlorosilane, tetramethylsilane, tetramethylcyclotetrasiloxane, octamethylcyclotetrasiloxane, tetraethyl orthosilicate, or other silicon-based precursor materials.

3. The method according to claim 1, characterized in that, The container volume in step 1 is 150-200mL; the volume of the electronic-grade silicon-based precursor sample to be tested is 90-100mL, and the container material is ultrapure polytetrafluoroethylene or soluble polytetrafluoroethylene. The heating plate has a coating of ultrapure polytetrafluoroethylene or soluble polytetrafluoroethylene.

4. The method according to claim 1, characterized in that, The heating plate described in step 1 has holes that match the size of the container, allowing the container to be embedded for heating; the embedding depth is not less than 5cm; the heating plate has 8-12 holes, with a temperature difference of less than 1℃.

5. The method according to claim 1, characterized in that, In the heating operation described in step 1, the sample is preheated to 5-10°C above the theoretical boiling point of the sample before being placed on the heating plate. When sporadic bubbles begin to appear in the container, the temperature of the heating plate is reduced to the theoretical boiling point of the sample.

6. The method according to claim 1, characterized in that, The temperature control process in step 1 is as follows: based on the preliminary boiling point test of the sample, a three-stage temperature control program is set: preheating stage: boiling point +10-20℃; main evaporation stage: temperature within 5℃ above the boiling point; residue removal stage: boiling point +20-40℃; combined with dynamic nitrogen purging at a flow rate of 50-100mL / min.

7. The method according to claim 1, characterized in that, The inorganic acid in the organic-inorganic composite acid solution in step 2 includes, but is not limited to, one of electronic nitric acid, electronic grade hydrochloric acid, or electronic grade hydrofluoric acid; the content of a single metal impurity in the inorganic acid is less than 10 ppt; in step (2), 5-10 mL of organic-inorganic composite acid solution is added to the container after evaporation; the sample reconstitution heating temperature is 50-60 ℃, and the reconstitution treatment time is 30-60 min.

8. The method according to claim 1, characterized in that, In step (2), the organic acid in the organic-inorganic composite acid solution is glycolic acid, and the volume ratio of organic acid is 0.5-1%; the blank sample selection in step (4) includes ultrapure organic solvent, ultrapure water or ultrapure inorganic acid; ultrapure organic solvent includes but is not limited to ultrapure NMP, n-hexane, DMSO or others; the content of single metal impurities in the ultrapure solvent, ultrapure water or ultrapure inorganic acid used in step (4) is less than 10ppt.

9. The method according to claim 1, characterized in that, In the method described, the evaporation and resolution operations are all performed in a clean bench in a Class 100 cleanroom, and the clean bench isolation window is not opened throughout the evaporation process; the exhaust gas pipeline of the clean bench is connected to a dedicated exhaust gas treatment system; the exhaust gas treatment system is a multi-stage condensation recovery plus two-stage activated carbon adsorption system, with a condensation temperature of 0-5℃.

10. The method according to claim 1, characterized in that, In the method described, the number of parallel samples is 3-5 groups, and the number of control samples is 3-5 groups.

Images