Synthesis method of high-purity hafnium-zirconium metal complex precursor
The one-pot synthesis of high-purity hafnium-zirconium metal complex precursors solves the problems of low deposition efficiency and insufficient purity of hafnium-zirconium precursors in existing technologies, and achieves efficient and uniform thin film growth, meeting the requirements of high-k gate dielectric layers for ultra-large-scale integrated circuits.
Patent Information
- Application Number
- CN202511701469.2
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-11-19
- Publication Date
- 2026-02-10
AI Technical Summary
Existing hafnium-zirconium precursors have low deposition efficiency, complex synthesis processes, and product purity that is difficult to meet the requirements of ultra-large-scale integrated circuits, resulting in low thin film growth rates and uneven thin film quality, which affects device stability and power consumption.
A one-pot synthesis route was adopted, in which hafnium tetrachloride or zirconium was reacted with dialkylamine, n-butyllithium, and alkyl magnesium halide under an inert atmosphere to generate a high-purity hafnium zirconium metal complex precursor. The operation process was simplified, the reaction temperature was controlled between room temperature and 50°C, and the raw materials were purified by distillation. The optimized molar ratio of the raw materials was 1:(4.0~4.2):(4.0~4.1):(1.0~1.1).
The film growth rate was increased to 0.143 nm/cycle, the film uniformity was ≤0.90%, and the metal purity reached 6N, meeting the high-k gate dielectric layer requirements of ultra-large-scale integrated circuits, reducing energy consumption and simplifying the manufacturing process.
Smart Images
Figure CN121494880A_ABST
Abstract
Description
Technical Field
[0001] This invention belongs to the technical field of high dielectric constant materials for ultra-large-scale integrated circuits, and more specifically, it relates to a method for synthesizing a high-purity hafnium-zirconium metal complex precursor. Background Technology
[0002] With the rapid development of the VLSI industry, the size of individual electronic devices continues to shrink, and the performance of the gate dielectric layer has become a key factor affecting device stability and power consumption. Currently, the silicon dioxide (SiO2) gate dielectric material commonly used in the industry is approaching its physical thickness limit due to the reduction in device feature sizes, leading to intensified tunneling effects and a significant increase in device power consumption, making it difficult to meet the higher stability requirements of the microelectronics industry. Therefore, finding high-k dielectric (high-k) materials to replace traditional SiO2 gate dielectric layers and reducing tunneling effects by increasing the physical thickness of the dielectric layer has become a core technological direction for improving the performance of electronic devices.
[0003] Hafnium dioxide (HfO2) and zirconium dioxide (ZrO2) films are widely recognized as promising new high-k gate dielectric materials due to their moderate dielectric constant (K≈25 for HfO2) and good compatibility with silicon-based integrated circuit processes. These oxide films are typically prepared using chemical vapor deposition (CVD) or atomic layer deposition (ALD) techniques. The performance of the vapor precursor directly determines the film quality and deposition efficiency—an ideal precursor must simultaneously meet the following requirements: sufficient reactivity to ensure film deposition efficiency, good stability to ensure operational safety, suitable vapor pressure for stable supply, and extremely high purity to avoid device problems such as current leakage and threshold voltage drift.
[0004] Existing hafnium-zirconium precursors have significant technical drawbacks:
[0005] 1) When hafnium tetrachloride (HfCl4) is used as a precursor to react with water to deposit hafnium oxide thin films, the film growth rate is only 0.05 nm / cycle, and the deposition efficiency is extremely low;
[0006] 2) Although hafnium tert-butoxide (Hf(OtBu)4) has some applications, it has poor thermodynamic stability and is prone to decomposition, resulting in uneven film composition.
[0007] 3) Although amine precursors such as tetradimethylaminohafnium (Hf(NMe2)4), tetramethylethylaminohafnium (Hf(NEtMe)4), and tetradiethylaminohafnium (Hf(NEt2)4) can increase the thin film growth rate to 0.1 nm / cycle, they still cannot meet the demand for high-efficiency deposition in industrial production.
[0008] In summary, developing a hafnium-zirconium metal complex precursor with high deposition rate, simple synthesis process, good product stability, and high purity is of great significance for promoting the development of high-k gate dielectric layer technology for ultra-large-scale integrated circuits. Summary of the Invention
[0009] To address the aforementioned technical problems, this invention provides a method for synthesizing high-purity hafnium-zirconium metal complex precursors, which solves the technical problems of low deposition efficiency, complex synthesis process, and difficulty in achieving the required purity of products for ultra-large-scale integrated circuits in existing hafnium-zirconium precursors.
[0010] A method for synthesizing a high-purity hafnium-zirconium metal complex precursor includes the following steps:
[0011] 1): Under an inert atmosphere, hafnium tetrachloride or zirconium tetrachloride is mixed with anhydrous toluene, cooled to 0°C, and tetrahydrofuran is added dropwise. The mixture is then restored to room temperature and stirred for 12 hours. After filtration in anhydrous and oxygen-free environment and washing with n-hexane, a solid product is obtained. The inert atmosphere is high-purity nitrogen or high-purity argon. 2): Disperse the solid product obtained in step 1) with anhydrous n-hexane, cool it to 5°C and add dialkylamine dropwise. After the addition is complete, react at 5°C for 5 hours. The structure of dialkylamine is HNR1R2, where R1 and R2 are both methyl, both are ethyl, or R1 is methyl and R2 is ethyl. 3): Add the n-butyllithium solution dropwise to the reaction system in step 2). The n-butyllithium solution concentration is 2.5 mol / L or 1.6 mol / L n-butyllithium n-hexane solution. After the addition is complete, react at room temperature for 2 hours. 4): Cool the reaction system of step 3) to 5°C for the second time, add a tetrahydrofuran solution of alkyl magnesium halide Grignard reagent dropwise, and heat the reaction at 50°C for 12 hours after the addition is complete. The structure of the alkyl magnesium halide Grignard reagent is R3MgX, where R3 is methyl, ethyl, isopropyl or tert-butyl, and X is chlorine or bromine. 5): After the reaction is complete, the mixture is brought back to room temperature, filtered through anhydrous and oxygen-free filter, washed with n-hexane, and the solvent in the filtrate is evaporated. The crude product is then distilled to obtain a high-purity hafnium zirconium metal complex precursor. The molar ratio of hafnium tetrachloride or zirconium tetrachloride, dialkylamine, n-butyllithium, and alkyl magnesium halide is 1:(4.0~4.2):(4.0~4.1):(1.0~1.1).
[0012] Preferably, in step 1), the ratio of hafnium tetrachloride or zirconium tetrachloride to anhydrous toluene is 0.36 mol: 500 mL, and the amount of tetrahydrofuran added in step 1) is 80 mL.
[0013] Preferably, the dialkylamine mentioned in step 2) is dimethylamine, methyl ethylamine, or diethylamine; when the dialkylamine is dimethylamine, a 20% dimethylamine-hexane solution is used for dropwise addition.
[0014] Preferably, in step 4), the alkyl halide magnesium Grignard reagent is tert-butyl magnesium chloride, isopropyl magnesium chloride, methyl magnesium bromide, ethyl magnesium chloride, tert-butyl magnesium bromide, isopropyl magnesium bromide, or ethyl magnesium bromide.
[0015] Preferably, the metal purity of the high-purity hafnium zirconium metal complex precursor obtained in step 5) is 6N (i.e. 99.9999%).
[0016] Preferably, the anhydrous and oxygen-free filtration and n-hexane washing operations in step 1) and step 5) are both carried out under an inert atmosphere.
[0017] Preferably, in step 4), the concentration of the tetrahydrofuran solution of the alkyl halide magnesium Grignard reagent is 1.0 mol / L or 2.0 mol / L.
[0018] Preferably, in step 5), the distillation operation uses conventional distillation equipment, and the high-purity hafnium-zirconium metal complex precursor obtained after distillation is a pale yellow liquid, and after... 1 HNMR characterization is consistent with the target product structure;
[0019] When the target product is tris(dimethylamino)tert-butylzirconium [Zr(NMe2)3(tBu)], its 1 HNMR (C6D6) data were 2.92 (s, 18H) and 1.27 (s, 9H).
[0020] When the target product is tris(dimethylamino)isopropylzirconium [Zr(NMe2)3(iPr)], its 1 HNMR (C6D6) data are 2.92 (s, 18H), 2.41 (m, 1H), and 1.91 (d, 2H).
[0021] When the target product is tris(dimethylamino)methylzirconium [Zr(NMe2)3(Me)], its 1 HNMR (C6D6) data were 2.92 (s, 18H) and -0.12 (s, 3H).
[0022] Compared with the prior art, the present invention has the following beneficial effects:
[0023] This invention employs a one-pot synthesis route, where the reaction of hafnium tetrachloride / zirconium with dialkylamine, n-butyllithium, and alkyl magnesium halides can be completed directly within the same inert atmosphere system without the need to separate intermediate products, simplifying the traditional multi-step synthesis process. The reaction conditions are mild; except for post-processing distillation, the reaction temperature only needs to be between room temperature and 50°C, eliminating the need for high-temperature heating and significantly reducing energy consumption. Furthermore, all operations are based on conventional Schlenk flasks and inert atmosphere systems, requiring no special customized equipment and facilitating the modification of existing chemical production lines.
[0024] This invention optimizes the raw material molar ratio (hafnium tetrachloride / zirconium: dialkylamine: n-butyllithium: alkyl magnesium halide = 1:(4.0~4.2):(4.0~4.1):(1.0~1.1)) and reaction parameters, achieving a stable product yield of 82%~88% (e.g., 88% yield of tris(dimethylamino)methylzirconium), far exceeding the 62% yield of the methyl ethylamine system in the comparative example. Inductively coupled plasma mass spectrometry (ICP-MS) analysis shows that the product has a metal purity of 6N (99.9999%), with no additional metal impurities introduced. This effectively avoids device problems such as current leakage and threshold voltage drift caused by insufficient precursor purity in oxide films, fully meeting the requirements of ultra-large-scale integrated circuits for high-purity precursors.
[0025] The hafnium-zirconium complex precursor synthesized in this invention possesses both excellent vapor pressure and reactivity due to the synergistic effect of amine and alkyl groups in its molecular structure. When applied to atomic layer deposition (ALD), a zirconium dioxide / hafnium film with a thickness of 14.3 nm can be prepared in 100 cycles, corresponding to a growth rate of 0.143 nm / cycle, which is 43% higher than the existing amine precursor's rate of 0.1 nm / cycle, significantly improving deposition efficiency. At the same time, the film inhomogeneity is ≤0.90%, with small global thickness deviation, and excellent density and uniformity, making it suitable for direct use as a high-k gate dielectric layer for very large-scale integrated circuits.
[0026] The raw materials used in this invention (hafnium tetrachloride / zirconium, dimethylamine, n-butyllithium, and alkyl magnesium halide) are all common reagents in the chemical industry, with low procurement costs and stable supply, and no dependence on scarce or highly toxic raw materials; the process has good repeatability, with the yield and purity deviation of precursors synthesized in different batches ≤3%, and the process stability meets the requirements of continuous industrial production, which can quickly realize capacity expansion to meet the growing market demand for high-K semiconductor materials. Attached Figure Description
[0027] Figure 1 This is a schematic diagram of the synthesis reaction equation of the present invention;
[0028] Figure 2 The proton nuclear magnetic resonance spectrum of tris(dimethylamino)tert-butylzirconium in this invention ( 1 HNMR spectrum;
[0029] Figure 3 The nuclear magnetic resonance hydrogen spectrum of the high-purity hafnium zirconium metal complex precursor in this invention (… 1 (HNMR) image. Detailed Implementation
[0030] To clarify the technical solution, effects, and implementation logic of this invention, the following detailed description of the synthesis method for a high-purity hafnium-zirconium metal complex precursor is provided in conjunction with specific experimental examples, comparative examples, and application examples. All operations in this embodiment are based on anhydrous and oxygen-free experimental procedures, and all reagents used are industrial-grade or analytical-grade anhydrous reagents. Inert atmosphere control, product characterization, and performance testing methods are standardized to ensure experimental repeatability.
[0031] Before detailing the embodiments, it is necessary to clarify the basic conditions that need to be uniformly controlled during the synthesis process of this invention to avoid deviations in results due to differences in operational details, as follows:
[0032] Inert atmosphere control: All reactions (including raw material mixing, reagent addition, filtration and washing) are carried out in a high-purity inert atmosphere. The inert gas used is high-purity nitrogen (purity ≥99.999%) or high-purity argon (purity ≥99.999%). The reaction system is made anhydrous and oxygen-free (water content ≤1ppm, oxygen content ≤1ppm) through a Schlenk flask vacuum-gas circulation system.
[0033] Reagent specifications:
[0034] Metal source: Zirconium tetrachloride (ZrCl4, purity 99.99%), hafnium tetrachloride (HfCl4, purity 99.99%), which are vacuum dried (120℃, 2h) to remove moisture before use;
[0035] Amine reagents: dimethylamine (analytical grade, prepared as a 20% hexane solution), methyl ethylamine (analytical grade), and diethylamine (analytical grade), all dried using molecular sieves;
[0036] Alkali metal reagent: n-Butyllithium (n-BuLi), prepared as a 2.5 mol / L or 1.6 mol / L hexane solution, to be used immediately;
[0037] Grignard reagents: tert-butyl magnesium chloride, isopropyl magnesium chloride, methyl magnesium bromide, etc., are all prepared into 1.0 mol / L or 2.0 mol / L tetrahydrofuran (THF) solutions. The THF is dried under reflux with sodium benzophenone and then distilled before use.
[0038] Solvents: Toluene and n-hexane are both anhydrous (water content ≤5ppm), and are dried by standing for more than 72 hours using molecular sieves.
[0039] Detection and characterization methods:
[0040] Purity testing: The content of metal impurities in the product was determined by inductively coupled plasma mass spectrometry (ICP-MS). The metal purity is expressed as "N grade" (e.g., 6N is 99.9999%).
[0041] Structural characterization: Hydrogen nuclear magnetic resonance (HMR) spectroscopy was used. 1 H NMR, using deuterated benzene (C6D6, 99.9% purity) as solvent and tetramethylsilane (TMS) as internal standard, confirmed the molecular structure of the product;
[0042] Thin film performance testing: The thickness and non-uniformity of the oxide thin film were measured using an ellipsometry. The test range covered the entire silicon substrate (4 inches in diameter), and the average value of 5 measurement points was taken.
[0043] Examples of synthesis of high-purity hafnium zirconium metal complex precursors:
[0044] Examples 1-3 below are all based on the "one-pot" synthesis route of the present invention, only changing the type of alkyl magnesium halide Grignard reagent to verify the effect of different R3 groups on product yield and purity; in all examples, the molar ratio of raw materials is strictly controlled as zirconium tetrachloride / hafnium:dialkylamine:n-butyllithium:alkyl magnesium halide = 1:(4.0~4.2):(4.0~4.1):(1.0~1.1).
[0045] Example 1: Synthesis of tris(dimethylamino)tert-butylzirconium [Zr(NMe2)3(tBu)]
[0046] In this embodiment, tert-butylmagnesium chloride is used as a Grignard reagent, and the target product is a zirconium complex containing tert-butyl (tBu). The specific steps are as follows:
[0047] Step 1: Metal source pretreatment and preliminary reaction:
[0048] In a 1000 mL Schlenk flask under high-purity nitrogen protection, zirconium tetrachloride (ZrCl4, 83.9 g, 0.36 mol) and anhydrous toluene (500 mL) were added, and the solid was dispersed by magnetic stirring. The system was cooled to 0 °C (temperature controlled by an ice-salt bath), and anhydrous tetrahydrofuran (80 mL) was slowly added dropwise. During the dropwise addition, the system gradually changed from a white suspension to a yellow turbid liquid. After the dropwise addition was completed, the ice-salt bath was removed, and the system was allowed to return to room temperature (25 °C) naturally. Stirring was continued for 12 h, at which point the system turned into a white turbid liquid (a coordination intermediate between ZrCl4 and THF was formed).
[0049] Step 2: Intermediate separation and amination reaction:
[0050] Under nitrogen protection, the above white turbid liquid was filtered anhydrous and oxygen-free (filter membrane pore size 0.22 μm), and the filter cake was washed with anhydrous n-hexane (100 mL × 3). The washed white solid was transferred to a 3000 mL L Schlenk flask, and anhydrous n-hexane (500 mL) was added and ultrasonically dispersed. The system was cooled to 5 °C (temperature controlled by an ice-water bath), and a 20% (w / w) dimethylamine n-hexane solution (332.7 g, 1.476 mol, corresponding to ZrCl4:dimethylamine = 1:4.1) was slowly added dropwise. No significant exothermic reaction occurred during the dropwise addition. After the dropwise addition was completed, the reaction was stirred for 5 h.
[0051] Slowly add a 2.5 mol / L n-butyllithium hexane solution (583.2 mL, 1.458 mol, corresponding to ZrCl4:n-butyllithium = 1:4.05). The system becomes slightly turbid during the addition, so the dropping rate needs to be controlled (1~2 drops / second) to avoid local overheating. After the addition is complete, return to room temperature and stir the reaction for 2 hours. The amination reaction generates Zr(NMe2)4 intermediate.
[0052] Step 3: Grignard reagent substitution and preparation of the target product:
[0053] Cool the system to 5℃ and add dropwise a 1.0 mol / L tetrahydrofuran solution of tert-butyl magnesium chloride (378 mL, 0.378 mol, corresponding to ZrCl4: tert-butyl magnesium chloride = 1: 1.05). After the addition is complete, raise the external temperature of the system to 50℃ (oil bath temperature control) and heat and stir the reaction for 12 h (substitution reaction occurs: Zr(NMe2)4 + tBuMgCl → Zr(NMe2)3(tBu) + Mg(NMe2)Cl).
[0054] Post-processing and product purification:
[0055] After the reaction was completed, the mixture was cooled to room temperature and filtered to remove the Mg salt precipitate in anhydrous and oxygen-free conditions. The filter cake was washed with anhydrous n-hexane (100 mL × 2). The filtrates were combined and the solvent was evaporated to dryness under a vacuum of 0.08–0.1 MPa and an external temperature of 60–70 °C to obtain a pale yellow crude product. The crude product was purified by vacuum distillation (the distillation column packing was glass spring, the vacuum was 1–5 Pa, the pot temperature was 120–130 °C, and the fraction temperature was 85–90 °C) to finally obtain 85.9 g of tris(dimethylamino)tert-butylzirconium.
[0056] Product performance data:
[0057] Yield: 85% (based on ZrCl4);
[0058] Metal purity: 6N (ICP-MS test, total impurity content ≤1ppm);
[0059] Structural characterization ( 1H NMR, C6D6: δ=2.92 (s, 18H, corresponding to the methyl hydrogen in -NMe2), δ=1.27 (s, 9H, corresponding to the methyl hydrogen in -tBu), which perfectly matches the target molecule structure.
[0060] Example 2: Synthesis of tris(dimethylamino)isopropylzirconium [Zr(NMe2)3(iPr)]
[0061] The only difference between this embodiment and Example 1 is that the tert-butyl magnesium chloride tetrahydrofuran solution is replaced with an equivalent amount of isopropyl magnesium chloride tetrahydrofuran solution. All other raw material amounts, reaction conditions, and post-treatment steps remain completely identical. Specific adjustments are as follows:
[0062] Grignard reagent: 189 mL of 2.0 mol / L isopropyl magnesium chloride tetrahydrofuran solution (0.378 mol, molar ratio of ZrCl4 1:1.05);
[0063] Reaction and post-processing: Same as in Example 1, reacted at an external temperature of 50°C for 12 hours, and obtained a pale yellow liquid product after distillation.
[0064] Product performance data:
[0065] Yield: 82% (based on ZrCl4);
[0066] Metal purity: 6N (ICP-MS detection);
[0067] Structural characterization ( 1 HNMR, C6D6: δ=2.92 (s, 18H, -NMe2 hydrogen), δ=2.41 (m, 1H, -iPr methine hydrogen), δ=1.91 (d, 2H, -iPr methyl hydrogen), consistent with the target structure.
[0068] Example 3: Synthesis of Tris(dimethylamino)methylzirconium [Zr(NMe2)3(Me)]
[0069] The only difference between this embodiment and Example 1 is that the tert-butyl magnesium chloride tetrahydrofuran solution is replaced with an equivalent amount of methyl magnesium bromide tetrahydrofuran solution, while the other conditions remain unchanged. The specific adjustments are as follows:
[0070] Grignard reagent: 378 mL of 1.0 mol / L methyl magnesium bromide tetrahydrofuran solution (0.378 mol, molar ratio of ZrCl4 1:1.05);
[0071] Reaction and post-processing: Same as in Example 1, reacted at an external temperature of 50°C for 12 hours, and obtained a pale yellow liquid product after distillation.
[0072] Product performance data:
[0073] Yield: 88% (based on ZrCl4, the highest among the three examples, due to the minimal steric hindrance of the methyl group, resulting in a more complete substitution reaction);
[0074] Metal purity: 6N (ICP-MS detection);
[0075] Structural characterization ( 1 HNMR, C6D6: δ=2.92 (s, 18H, -NMe2 hydrogen), δ=-0.12 (s, 3H, -Me hydrogen), perfectly matching the target structure.
[0076] Example of application of precursor in oxide thin film deposition (Example 4):
[0077] This embodiment verifies the application effect of the synthesized high-purity zirconium complex precursor in the preparation of zirconium dioxide (ZrO2) thin films by atomic layer deposition (ALD). The specific steps are as follows:
[0078] Substrate pretreatment: Take an undoped silicon wafer (4 inches in diameter, 500 μm thick) and perform the following cleaning steps in sequence:
[0079] Acetone ultrasonic cleaning for 15 minutes (to remove organic oil stains);
[0080] Ultrasonic cleaning with anhydrous ethanol for 15 minutes (to remove acetone residue);
[0081] Ultrasonic cleaning with deionized water (resistivity ≥ 18.2 MΩ·cm) for 20 min;
[0082] Immerse in RCA solution (NH4OH:H2O2:H2O=1:1:5, 80℃) for 10 minutes (to remove the oxide layer and metal impurities on the silicon wafer surface).
[0083] Clean again with deionized water for 20 minutes, and then with anhydrous ethanol for 15 minutes.
[0084] Dry with high-purity nitrogen (flow rate 5L / min), then place in a 120℃ oven to dry for 2 hours, and set aside for later use.
[0085] ALD deposition process parameters: Place the pretreated silicon wafer into the ALD chamber and set the following parameters:
[0086] Chamber temperature: 230~250℃ (optimized temperature, balancing precursor decomposition activity and film density).
[0087] Chamber vacuum level: 5×10 -3 Pa;
[0088] Precursor supply: Zr(NMe2)3(tBu) prepared in Example 1 was heated to 150°C to evaporate it, and then introduced into the chamber by high-purity argon gas (carrier gas flow rate 20 sccm) in a 0.2s pulse time.
[0089] Cleaning steps: After the precursor pulse, continue to purge with argon gas for 10 seconds to remove residual precursors and byproducts from the chamber;
[0090] Oxidant supply: H2O is used as the oxidant and is introduced into the chamber in a 0.3s pulse time through argon carrier gas (flow rate 20sccm). Then, argon gas is passed through for 10s to clean the residual H2O and reaction byproducts.
[0091] Number of cycles: Repeat the above "precursor pulse-cleaning-H2O pulse-cleaning" process 100 times.
[0092] Thin film performance test results: After deposition, the ZrO2 thin film performance was tested using an ellipsometry.
[0093] Film thickness: 14.3 nm (average thickness, corresponding to a growth rate of 0.143 nm / cycle for 100 cycles, significantly higher than the existing technology's 0.05~0.1 nm / cycle).
[0094] Thin film non-uniformity: 0.90% (thickness deviation at 5 measurement points across the entire area ≤ 0.13 nm, proving excellent thin film uniformity and meeting the requirements for semiconductor gate dielectric layer).
[0095] Comparative experiments and results analysis (Comparative Examples 1-2):
[0096] To verify the effect of the "type of dialkylamine" on the reaction in the technical solution of this invention, the following comparative example was set up. The amount of raw materials and reaction conditions were basically the same as in Example 1, except that the type of dialkylamine was changed.
[0097] Comparative Example 1: Synthesis of tris(methylethylamino)tert-butylhafnium using methylethylamine as a dialkylamine
[0098] Dialkylamine: Methylethylamine (87.2 g, 1.476 mol, molar ratio of methylethylamine to HfCl4 1:4.1).
[0099] Metal source: Hafnium tetrachloride (HfCl4, 115.3 g, 0.36 mol);
[0100] Other conditions: Same as in Example 1 (378 mL of tert-butylmagnesium chloride, 0.378 mol, reaction at 50°C for 12 h).
[0101] result:
[0102] After the first reaction, take a small amount of the reaction solution for... 1HNMR characterization revealed a large amount of Hf(NEtMe)4 intermediate remaining (proving that the substitution reaction was incomplete).
[0103] An equivalent amount of tert-butylmagnesium chloride (378 mL, 0.378 mol) was added, and the reaction was continued for 12 h. Finally, 91.5 g of the product was obtained by distillation.
[0104] Product performance: Yield 62% (significantly lower than 85% in Example 1), metal purity 5N (ICP-MS detection, impurity content is higher than in the example);
[0105] Cause analysis: The steric hindrance of methyl ethylamine (NEtMe) is greater than that of dimethylamine (NMe2), which increases the difficulty of Grignard reagent substitution reaction, requires the addition of reagents, and the reaction is still incomplete, resulting in a decrease in yield and purity.
[0106] Comparative Example 2: Synthesis of the target product using diethylamine as the dialkylamine:
[0107] Dialkylamine: Diethylamine (108g, 1.476mol, molar ratio of ZrCl4 1:4.1).
[0108] Other conditions: Same as in Example 1 (ZrCl4 0.36mol, tert-butylmagnesium chloride 756mL, 0.756mol, reaction at 50℃ for 24h).
[0109] result:
[0110] After the reaction was completed, the crude product was analyzed. 1 HNMR characterization revealed that only the signal of tetra(diethylamino)zirconium (Zr(NEt2)4) intermediate was detected, and the characteristic peak of the target product was not detected.
[0111] Analysis of the cause: Diethylamine (NEt2) has extremely large steric hindrance. The alkyl halide magnesium Grignard reagent cannot overcome the steric hindrance to undergo a substitution reaction with the intermediate, and ultimately the target product cannot be generated. This proves that dialkylamine should be selected with less steric hindrance, such as dimethylamine (preferred) or methyl ethylamine (second choice). Diethylamine cannot be used.
[0112] Based on the above embodiments, application examples, and comparative examples, the following conclusions can be drawn:
[0113] The one-pot synthesis route of the present invention (zirconium tetrachloride / hafnium → reaction with dialkylamine and n-butyllithium → reaction with alkyl magnesium halide → distillation purification) is simple to operate, requires no separation of intermediates, has low energy consumption, and the product yield can reach 82%~88%, with the metal purity stable at 6N, meeting the requirements of high-purity semiconductor precursors.
[0114] The choice of dialkylamine is crucial: dimethylamine is preferred due to its low steric hindrance, which allows for efficient substitution reactions and yields the best results and purity; methyl ethylamine is the next best choice, but requires additional reagents and results in lower yields; diethylamine is unusable due to its excessive steric hindrance.
[0115] When the synthesized precursor is applied to ALD deposition, the growth rate reaches 0.143 nm / cycle and the film uniformity is ≤0.90%, which solves the problems of "low growth rate and poor film uniformity" in the existing technology and can be used for the preparation of gate dielectric layers of ultra-large scale integrated circuits.
[0116] Good process stability: All examples use industrial-grade raw materials and mild reaction conditions (room temperature to 50°C), which are suitable for large-scale production in factories and can meet the growing market demand for semiconductor materials.
[0117] The embodiments of the present invention are given for illustrative and descriptive purposes only, and are not intended to be exhaustive or to limit the invention to the forms disclosed. Many modifications and variations will be apparent to those skilled in the art. The embodiments were chosen and described in order to better illustrate the principles and practical application of the invention, and to enable those skilled in the art to understand the invention and to design various embodiments with various modifications suitable for a particular purpose.
Claims
1. A method for synthesizing a high-purity hafnium-zirconium metal complex precursor, characterized in that, Includes the following steps: 1): Under an inert atmosphere, hafnium tetrachloride or zirconium tetrachloride is mixed with anhydrous toluene, cooled to 0°C, and tetrahydrofuran is added dropwise. The mixture is then restored to room temperature and stirred for 12 hours. After filtration in anhydrous and oxygen-free environment and washing with n-hexane, a solid product is obtained. The inert atmosphere is high-purity nitrogen or high-purity argon. 2): Disperse the solid product obtained in step 1) with anhydrous n-hexane, cool it to 5°C and add dialkylamine dropwise. After the addition is complete, react at 5°C for 5 hours. The structure of dialkylamine is HNR1R2, where R1 and R2 are both methyl, both are ethyl, or R1 is methyl and R2 is ethyl. 3): Add the n-butyllithium solution dropwise to the reaction system in step 2). The n-butyllithium solution concentration is 2.5 mol / L or 1.6 mol / L n-butyllithium n-hexane solution. After the addition is complete, react at room temperature for 2 hours. 4): Cool the reaction system of step 3) to 5°C for the second time, add a tetrahydrofuran solution of alkyl magnesium halide Grignard reagent dropwise, and heat the reaction at 50°C for 12 hours after the addition is complete. The structure of the alkyl magnesium halide Grignard reagent is R3MgX, where R3 is methyl, ethyl, isopropyl or tert-butyl, and X is chlorine or bromine. 5): After the reaction is complete, the mixture is brought back to room temperature, filtered through anhydrous and oxygen-free filter, washed with n-hexane, and the solvent in the filtrate is evaporated. The crude product is then distilled to obtain a high-purity hafnium zirconium metal complex precursor. The molar ratio of hafnium tetrachloride or zirconium tetrachloride, dialkylamine, n-butyllithium, and alkyl magnesium halide is 1:(4.0~4.2):(4.0~4.1):(1.0~1.1).
2. The synthesis method according to claim 1, characterized in that, In step 1), the ratio of hafnium tetrachloride or zirconium tetrachloride to anhydrous toluene is 0.36 mol: 500 mL, and the amount of tetrahydrofuran added in step 1) is 80 mL.
3. The synthesis method according to claim 1, characterized in that, The dialkylamine mentioned in step 2) is dimethylamine, methyl ethylamine, or diethylamine; when the dialkylamine is dimethylamine, it is added dropwise using a 20% dimethylamine-hexane solution.
4. The synthesis method according to claim 1, characterized in that, In step 4), the alkyl magnesium halide Grignard reagent is tert-butyl magnesium chloride, isopropyl magnesium chloride, methyl magnesium bromide, ethyl magnesium chloride, tert-butyl magnesium bromide, isopropyl magnesium bromide, or ethyl magnesium bromide.
5. The synthesis method according to claim 1, characterized in that, The metal purity of the high-purity hafnium zirconium metal complex precursor obtained in step 5) is 6N (i.e. 99.9999%).
6. The synthesis method according to claim 1, characterized in that, The anhydrous and oxygen-free filtration and hexane washing operations in step 1) and the anhydrous and oxygen-free filtration and hexane washing operations in step 5) are both carried out under an inert atmosphere.
7. The synthesis method according to claim 1, characterized in that, In step 4), the concentration of the tetrahydrofuran solution of the alkyl halide magnesium Grignard reagent is 1.0 mol / L or 2.0 mol / L.
8. The synthesis method according to claim 1, characterized in that, In step 5), the distillation operation uses conventional distillation equipment. The high-purity hafnium-zirconium metal complex precursor obtained after distillation is a pale yellow liquid, and after further processing... 1 HNMR characterization is consistent with the target product structure; When the target product is tris(dimethylamino)tert-butylzirconium [Zr(NMe2)3(tBu)], its 1 HNMR (C6D6) data were 2.92 (s, 18H) and 1.27 (s, 9H). When the target product is tris(dimethylamino)isopropylzirconium [Zr(NMe2)3(iPr)], its 1 HNMR (C6D6) data are 2.92 (s, 18H), 2.41 (m, 1H), and 1.91 (d, 2H). When the target product is tris(dimethylamino)methylzirconium [Zr(NMe2)3(Me)], its 1 HNMR (C6D6) data were 2.92 (s, 18H) and -0.12 (s, 3H).