A preparation process for high-purity cyproconazole technical grade drug

By controlling the closed-loop and open-loop reactions step by step, using sodium alkoxide for salt formation and dehydration under reduced pressure, dynamically adjusting the open-loop temperature, optimizing the mother liquor reuse ratio, and using isopropanol crystallization purification, the problems of impurity generation and quality instability in the preparation of cycloproconazole were solved, achieving the preparation of cycloproconazole with high purity and high yield.

CN123106561APending Publication Date: 2026-11-17PANJIN HONGHE CHEM CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202611491030.6
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2026-09-23
Publication Date
2026-11-17

AI Technical Summary

Technical Problem

In the existing technology for preparing cyproconazole, the generation of regioisomers and related impurities is difficult to control. Improper use of reaction temperature and solvent leads to high impurity levels, improper use of mother liquor causes unstable product quality, and energy consumption is high.

Method used

By controlling the closed-loop and open-loop reactions in steps, using sodium alkoxide to form salts and dehydrating under reduced pressure, dynamically adjusting the open-loop temperature, optimizing the mother liquor reuse ratio, and purifying with isopropanol crystallization, high-purity cyclopropazole was prepared.

Benefits of technology

It effectively inhibits the formation of regioisomers and impurities, reduces reaction cycle time, improves product purity and yield, stabilizes product quality, and reduces solvent residue and energy consumption.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN123106561A_ABST
    Figure CN123106561A_ABST
Patent Text Reader

Abstract

This invention discloses a preparation process for high-purity cyproconazole technical, belonging to the field of chemical pharmaceutical raw material synthesis and purification technology. The process of this invention includes five steps sequentially: Grignard addition, closed-loop purification, solution preparation, variable-temperature ring-opening, and crystallization. Closed-loop purification removes chloride ions and water from the toluene solution of the epoxy intermediate to below a specified level; solution preparation involves preparing a mixed system in which free triazole and sodium triazole coexist in a specified ratio; variable-temperature ring-opening calculates the target temperature of the reaction solution according to a calibrated kinetic model and controls the temperature rise in cascades to ensure that the formation rate of regioisomers does not exceed a set upper limit; crystallization determines the proportion of mother liquor discharged based on the measured impurity content of the mother liquor. This invention inhibits the formation of regioisomers, diols, and diethers, shortens the ring-opening cycle, and improves the purity, batch-to-batch consistency, and overall yield of the technical.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] This invention belongs to the field of synthesis and purification technology of chemical pharmaceutical raw materials, and particularly relates to a preparation process of high-purity cyproconazole technical material. Background Technology

[0002] Ciproconazole is a triazole fungicide chemical raw material. Its molecule contains a quaternary carbon atom linking a chlorophenyl group and a triazole methyl group, and a methine carbon atom adjacent to the cyclopropyl group. Both carbon atoms are chiral centers. The industrial product is a mixture of two diastereomers. The fungicidal activity of ciproconazole derives from the spatial orientation of a specific nitrogen atom on the triazole ring with the tertiary alcohol hydroxyl group. The levels of related substances in the raw material directly affect the stability and registration indicators of the formulation.

[0003] One of the mainstream industrial routes for preparing cyproconazole is the Grignard addition route. This typically involves first preparing a Grignard reagent from a halocyclopropane derivative and then adding it to an α-haloaryl ketone to obtain a chloromethyl-containing tertiary alcohol intermediate. This intermediate is then directly added to a strongly polar aprotic solvent containing a sodium triazole salt, reacted at a high temperature, and then cooled to crystallize, yielding the technical grade cyproconazole.

[0004] In existing technologies, the technical grade drug contains numerous regioisomers with different triazole ring linkage sites. The triazole anion is a binucleophile, meaning both of its bondable nitrogen atoms can attack the electrophilic center. In a system where a chloromethyl-containing tertiary alcohol intermediate coexists with a triazole sodium salt, two competing pathways exist simultaneously. One pathway involves the triazole anion directly undergoing bimolecular nucleophilic substitution of the chloromethyl group. The other pathway involves the tertiary alcohol intermediate first undergoing dehydrohalogenation under alkaline conditions to form an epoxy intermediate, followed by ring-opening by the triazole anion against a sterically less hindered carbon atom on the epoxy ring. In the transition state of the former pathway, the electrophilic carbon is a primary carbon with minimal steric constraint, resulting in a small difference in activation energies between the two nitrogen atoms and poor regioselectivity. In the latter pathway, the transition state is simultaneously constrained by three-membered ring strain and the steric hindrance of the adjacent quaternary carbon, exhibiting a clear preference for attacking the target nitrogen atom. Current techniques combine ring-closing and ring-opening reactions in the same reactor and at the same temperature, without separating and controlling the two pathways. Halogen ions released from the ring-closing reaction are always present in the system. These halogen ions can cause the already formed epoxy intermediate to re-open, returning to the tertiary alcohol intermediate. The pathway with poor regioselectivity is continuously replenished. The aforementioned regioisomers have the same molecular formula and relative molecular mass as the target product, and similar polarity. Their separation coefficients and solubility ratios in commonly used reversed-phase chromatography and crystallization solvent systems are close. Once formed, they are difficult to remove in subsequent purification, requiring repeated recrystallization and resulting in significant yield loss.

[0005] Existing technologies operate at higher temperatures and in highly polar aprotic solvents. Quenching introduces water and water remaining in the system, causing the epoxy intermediate to hydrolyze and generate vicinal diol-type impurities. The alcohol oxy anions generated by ring opening have a long lifetime in highly polar aprotic solvents lacking proton donors, and can continue to undergo nucleophilic ring opening on another molecule of epoxy intermediate, generating diether-type impurities.

[0006] The apparent activation energy of the side reactions that generate regioisomers is higher than that of the main reactions that generate the target product. Lowering the temperature favors selectivity but detrimental to rate, while raising the temperature favors rate but detrimental to selectivity. Existing technologies employ a single constant temperature operation, with all conversions completed under the same selectivity. The process can only strike a trade-off between reaction cycle and impurity level, unable to improve both simultaneously. Existing technologies directly apply the temperature setpoint to the jacket, but the heat transfer temperature difference and dynamic hysteresis between the reaction medium and the jacket further reduce the accuracy of temperature control during the exothermic open-loop phase.

[0007] To recover dissolved products from mother liquor, industrial practice commonly reuses the crystallization mother liquor in the next batch. However, regioisomers accumulate in the mother liquor batch by batch, and once they reach a certain level, they are carried by the crystals, causing related substances in the original drug to exceed the limits. Current technologies rely heavily on experience to determine the proportion of mother liquor discharged, resulting in product loss and difficulty in ensuring batch-to-batch quality consistency. Furthermore, there is a lack of clear handling procedures when fluctuations occur in preceding steps. In addition, the reused mother liquor introduces solvents and water. Current technologies prepare the batches based on the amount of fresh solvent added, resulting in different compositions of the antisolvent in the first batch and subsequent reused batches, making it difficult to stably reproduce the crystal form and particle size distribution.

[0008] In addition, the strong polar aprotic solvents used in the existing technology have high boiling points and are completely miscible with water. The recovery energy consumption is large, and they are easy to remain in the original drug, which is not conducive to the control of solvent residues in chemical pharmaceutical raw materials.

[0009] In summary, there is still a need in the field for a cyproconazole technical preparation process that can suppress the formation of regional isomers and related impurities at the source, reduce impurity levels without prolonging the reaction cycle, and provide executable criteria for the reuse of crystallization mother liquor and the handling of abnormalities. Summary of the Invention

[0010] To solve the above problems, the technical solution of the present invention is: a preparation process for high-purity cyproconazole technical, characterized by comprising the following steps:

[0011] S1 addition: The Grignard reagent obtained by reacting (1-chloroethyl)cyclopropane with magnesium in tetrahydrofuran is added to 2-chloro-1-(4-chlorophenyl)ethyl-1-one. After quenching, separation, and solvent removal, the solution is dissolved in toluene to obtain a toluene solution containing 1-chloro-2-(4-chlorophenyl)-3-cyclopropylbut-2-ol.

[0012] S2 purification: Add sodium hydroxide aqueous solution and tetrabutylammonium bromide to the solution obtained from S1, and close the ring at 28-32℃ until the chlorohydrin intermediate is no more than 0.30%; separate the liquid and wash with water until the chloride ion concentration is no more than 50 mg / kg, and dehydrate azeotropically until the water concentration is no more than 300 mg / kg to obtain a toluene solution containing 2-(4-chlorophenyl)-2-(1-cyclopropylethyl)ethylene oxide;

[0013] S3 solution preparation: 1,2,4-triazole and sodium methoxide methanol solution are salted in toluene, methanol is distilled off under reduced pressure until water content is no more than 400 mg / kg, and n-butanol is added to obtain a mixed system with a molar ratio of free 1,2,4-triazole to sodium 1,2,4-triazole of 3.5 to 5.5.

[0014] S4 ring opening: The solution obtained in S2 is added to the mixture obtained in S3. The current concentrations of the epoxy intermediate and 1,2,4-triazole in the reaction solution are periodically measured. The target temperature of the reaction solution is tracked by cascade control according to the calibrated kinetic model, so that the temperature of the reaction solution is raised from 328-353K to 368K, and the N4-regioisomer formation rate calculated by the model is not higher than the set upper limit. The solution is kept at 368K until the current concentration of the epoxy intermediate is not greater than 0.01mol / L, and the crude product solution is obtained.

[0015] S5 Crystallization: After desolventizing the crude product solution obtained from S4, dissolve it in isopropanol and recycled crystallization mother liquor. Add deionized water according to the total amount of isopropanol and water in the combined system, add seed crystals, and crystallize at a controlled temperature. Filter, rinse, and dry to obtain cyproconazole technical grade. The crystallization mother liquor is discharged at a discharge ratio of 0.10-0.20, and the remainder is returned to the next batch of S5 for reuse. The total dry basis content of the obtained cyproconazole technical grade is not less than 99.0%, and the N4-regioisomer content is not greater than 0.15%.

[0016] Furthermore, S1 includes:

[0017] S11 In a nitrogen atmosphere, magnesium shavings and anhydrous tetrahydrofuran are added to the reactor, and iodine is added as an initiator. The temperature is raised to 40-45°C. The water content of the anhydrous tetrahydrofuran is not greater than 100 mg / kg, and the water content of (1-chloroethyl)cyclopropane and 2-chloro-1-(4-chlorophenyl)ethyl-1-one is not greater than 200 mg / kg.

[0018] S12 is added dropwise with an anhydrous tetrahydrofuran solution of (1-chloroethyl)cyclopropane under nitrogen protection throughout the process, at a controlled temperature of 40–45°C for 1.2–1.8 h. After the addition is complete, the solution is kept at 40–45°C for 0.8–1.2 h to obtain Grignard reagent solution. The molar ratio of (1-chloroethyl)cyclopropane to 2-chloro-1-(4-chlorophenyl)ethyl-1-one is 1.15–1.25, and the molar ratio of magnesium to (1-chloroethyl)cyclopropane is 1.05–1.15.

[0019] S13. The Grignard reagent solution is cooled to -5 to 0°C, and an anhydrous tetrahydrofuran solution of 2-chloro-1-(4-chlorophenyl)ethyl-1-one is added dropwise under nitrogen protection throughout the process. The temperature is controlled at -5 to 0°C, and the dropwise addition time is 1.2 to 1.8 h. After the dropwise addition is completed, the temperature is kept at 0 to 5°C for 0.8 to 1.2 h to ensure that the addition reaction is completed in the range of -5 to 5°C.

[0020] S14 was quenched by adding an 8%–12% ammonium chloride aqueous solution, with the temperature controlled not higher than 15°C. After standing and separating the liquids, the aqueous phase was extracted once with toluene. The organic phases were combined and the tetrahydrofuran was removed under reduced pressure at 38–42°C and 9–13 kPa. Toluene was added to dissolve the solution, yielding a toluene solution containing the chlorohydrin intermediate.

[0021] Furthermore, S2 includes:

[0022] S21 adds a 28%–32% (w / w) sodium hydroxide aqueous solution and tetrabutylammonium bromide to the toluene solution obtained in S1; the molar ratio of sodium hydroxide to the chlorohydrin intermediate is 1.10–1.25, and the molar amount of tetrabutylammonium bromide is 0.3%–0.8% of the molar amount of the chlorohydrin intermediate;

[0023] S22 was reacted at 28–32°C and a stirring paddle tip linear velocity of 1.5–2.5 m / s for 1.5–2.5 h. Samples were taken every 30 min to determine the residual amount of the chlorohydrin intermediate by gas chromatography area normalization method. The residual amount was calculated based on the organic components after deducting the toluene solvent peak, and the closure endpoint was determined when it was no greater than 0.30%.

[0024] S23: Allow the mixture to stand for 20-40 minutes to separate into layers, then discard the lower aqueous phase. The water washing described in S2 is performed by washing with a 4%-6% sodium chloride aqueous solution at 20%-25% of the organic phase volume, stirring for 8-12 minutes each time and letting it stand for 15-25 minutes. The washing should be performed at least twice and continued until the chloride ion content of the organic phase, as determined by ion chromatography, is no more than 50 mg / kg.

[0025] S24 is refluxed at atmospheric pressure, with toluene as an azeotropic agent, and water is separated by a water separator. During the water separation process, the temperature at the top of the column gradually rises from 84°C to 110°C, and the water separation is stopped when the water content determined by Karl Fischer coulometric method is no more than 300 mg / kg, yielding a toluene solution containing an epoxy intermediate; in the toluene solution containing the epoxy intermediate, the normalized content of the epoxy intermediate based on the gas chromatographic area of ​​the organic components after deducting the toluene solvent peak is no less than 98.0%.

[0026] Furthermore, S3 includes:

[0027] S31. 1,2,4-Triazole is dispersed in toluene, and a 30% (w / w) sodium methoxide methanol solution is added. The mixture is stirred at 45–55°C for 25–40 min to form a salt. The amount of sodium 1,2,4-triazole in the system is determined by non-aqueous potentiometric titration until it is not less than 98% of the amount of sodium methoxide added. The molar ratio of 1,2,4-triazole to the epoxy intermediate is 1.6–1.9, and the molar ratio of sodium methoxide to the epoxy intermediate is 0.28–0.38, and the ratio is calculated according to the following formula. The value is 3.5 to 5.5.

[0028] ;

[0029] In the formula, The molar ratio of free 1,2,4-triazole to sodium 1,2,4-triazole is dimensionless. The amount of 1,2,4-triazole added; The amount of 1,2,4-triazole sodium determined by the non-aqueous potentiometric titration method described above;

[0030] S32 distills out methanol and methanol-toluene azeotrope at an absolute pressure of 11-21 kPa and a vessel temperature not exceeding 60°C. Distillation is stopped when the water content determined by Karl Fischer coulometric method is not greater than 400 mg / kg and the methanol mass fraction determined by gas chromatography internal standard method is not greater than 0.50%.

[0031] S33 is added to n-butanol to make the mass ratio of toluene to n-butanol 3.5:1 to 4.5:1, resulting in a mixed system of 1,2,4-triazole and sodium 1,2,4-triazole.

[0032] Furthermore, S4 includes:

[0033] S41: The mixture obtained in S3 is heated to 328–338 K, and the toluene solution obtained in S2 is added at a constant rate over 0.8–1.2 h. After the addition is complete, toluene is added again to make the initial concentration of epoxy intermediate in the reaction solution 1.0–1.4 mol / L and the initial total concentration of 1,2,4-triazole 1.8–2.4 mol / L.

[0034] S42 samples are taken every 10–20 minutes. After quenching with an equal volume of 5% (w / w) glacial acetic acid aqueous solution, the current concentration of the epoxy intermediate in the reaction solution is determined in the same injection using high-performance liquid chromatography with external standard method. With the current total concentration of 1,2,4-triazole The current total concentration is the sum of the concentrations of free 1,2,4-triazole and 1,2,4-triazole sodium.

[0035] S43 calculates the target temperature of the reaction solution using the following formula. :

[0036] ;

[0037] In the formula, This is the calculated value of the target temperature of the reaction solution; The apparent activation energy for the formation of the N4- regioisomer is taken as 101.0 × 10⁻⁶. 3 (J / mol); The molar gas constant; The apparent pre-exponential factor for the N4-regioisomerization reaction; This represents the current concentration (mol / L) of the epoxy intermediate. The current total concentration (mol / L) of 1,2,4-triazole. An upper limit is set for the formation rate of N4-regioisomers;

[0038] S44 take The smaller of 368K and 368K is used as the target temperature for the reaction solution. A cascade control loop, with the reaction liquid temperature as the main loop and the jacket temperature as the slave loop, tracks the reaction. Repeat S42 to S44 until... Stop when the concentration is no greater than 0.01 mol / L, and cool to 331–335 K to obtain a crude cyclopropazole solution with an N4-regioisomer content of no more than 0.40%; the content is calculated by high performance liquid chromatography area normalization method based on the organic components after deducting the solvent peak.

[0039] Furthermore, in S43 Take 8.0 × 10 -4 ~1.6×10 -3 The two consecutive values ​​given in S44 The difference should not exceed 5K, and the heating rate of the reaction solution should not exceed 0.5K / min. At least 368K Take 368K and keep the temperature constant until the endpoint described in S44; and The mixed system obtained from S3 was used when the mass ratio of toluene to n-butanol was 3.5:1 to 4.5:1. The values ​​were measured at no fewer than four isothermal points at a concentration of 3.5–5.5, an initial epoxy intermediate concentration of 1.0–1.4 mol / L, and a temperature of 328–368 K. The results were obtained by regressing the amount of N4-region isomers.

[0040] Furthermore, S5 includes:

[0041] S51 removes toluene and n-butanol from the crude product liquid obtained in S4 under reduced pressure at 328-338K and 6-16kPa to obtain crude product residue;

[0042] S52 determines the mass fraction of isopropanol and water in the recycled crystallization mother liquor; adds the recycled crystallization mother liquor and fresh isopropanol to the crude product residue, so that the ratio of the total mass of isopropanol in the combined system to the mass of cyclopropazole in the crude product residue is 1.1:1 to 1.3:1, and heats to 58 to 62°C to dissolve and clarify it; the first batch does not add recycled crystallization mother liquor;

[0043] S53 determines the total volume of water to be 65%–75% of the total volume of isopropanol in the combined system, and adds deionized water after deducting the water already introduced by the recycled crystallization mother liquor; when the temperature drops to 54–56°C, cyproconazole seed crystals are added, and the temperature is lowered according to the following program:

[0044] ;

[0045] In the formula, For the first time since the addition of the seed crystal Crystallization temperature (°C) per hour; The crystallization temperature is set at 54–56 °C. The final crystallization temperature is 8–12 °C. The time from the addition of the seed crystal, with a value ranging from 0 to 1. ≤ (h); The total duration of the cooling program is 2.5 to 3.5 hours.

[0046] S54 The mixture was kept at a constant temperature for 0.8–1.2 h, filtered, and the filter cake was washed with a mixture of isopropanol and water at a volume ratio of 1:1 (3–7 °C). It was then vacuum dried at 58–62 °C and an absolute pressure of 6–16 kPa for 3.5–4.5 h to obtain cyproconazole technical grade. The amount of cyproconazole seed crystals added was 0.15%–0.25% of the theoretical yield, and the median volume particle size was [missing information]. The thickness is 40-50 μm. The theoretical yield is calculated based on the amount of 2-chloro-1-(4-chlorophenyl)ethyl-1-one added to S1 and its measured content.

[0047] Furthermore, the crystallization mother liquor obtained from S54 filtration is disposed of in the following manner:

[0048] S55 Determines the mass of the crystallization mother liquor. and the mass fraction of N4-regioisomers therein And calculate the lower limit of the mother liquor discharge ratio according to the following formula. :

[0049] ;

[0050] In the formula, This is the lower limit of the proportion of mother liquor discharged, and it is dimensionless. The upper limit for the mass fraction of N4-regioisomers in the crystallization mother liquor is 0.8% to 1.5%. The mass of the crystallization mother liquor obtained from S54 filtration; The mass of N4-regional isomers introduced into the next batch of crude product residue is calculated as the product of the mass of the current batch of crude product residue and the mass fraction of that isomer therein. The measured mass fraction of N4-regioisomers in the crystallization mother liquor obtained by S54 filtration;

[0051] S56 takes 0.10 and The larger of the two values ​​will be used as the discharge ratio for this batch of mother liquor. ,Will and The accumulated crystallization mother liquor is discharged, and the remainder is returned to the next batch of S52 for reuse; when When the concentration is greater than 0.20, stop using the mother liquor and recover ciprofloxacin separately from the mother liquor of this batch of crystallization.

[0052] Furthermore, the total content of cyproconazole technical material, calculated on a dry basis, is not less than 99.0%. The total content of cyproconazole is the sum of the chromatographic peak areas of the two diastereomers of cyproconazole, converted using the external standard method in high-performance liquid chromatography. The dry basis is the mass standard after deducting the loss on drying, measured by constant weight at 105°C. In the cyproconazole technical material, the content of the N4-regioisomer is not greater than 0.15%, the content of 2-(4-chlorophenyl)-3-cyclopropylbut-1,2-diol is not greater than 0.05%, and the content of bis[2-(4-chlorophenyl)-3-cyclopropyl-2-hydroxybutyl] ether is not greater than 0.02%.

[0053] The beneficial effects of this invention are as follows:

[0054] In this invention, step S2 separates the ring-closing and ring-opening reactions in time and space. In a two-phase system under near-room temperature conditions, the chlorohydrin intermediate is quantitatively ring-closed to an epoxy intermediate. Subsequently, the chloride ions released during ring closure are removed from the organic phase through separation and washing. Azeotropic dehydration removes water from the system, and the upper limits of both chloride and water content are included in the material specifications for the toluene solution of the epoxy intermediate. After chloride ion removal to below a predetermined level, the chloride-driven reversible pathway of the epoxy intermediate loses its sustainability, and the poorly regioselective direct nucleophilic substitution pathway is no longer replenished in subsequent ring-opening steps, thus suppressing the formation of regioisomers at the source. After water removal to below a predetermined level, the hydrolysis of the epoxy intermediate is inhibited, and the formation of vicinal diol impurities decreases.

[0055] In this invention, S3 uses sodium alkoxide instead of hydroxide to form a salt with triazole. The byproduct of salt formation is alcohol instead of water. Under reduced pressure, the alcohol and residual water are removed together, ensuring that the low-water state established in S2 is maintained after salt formation. Sodium triazole is not hydrolyzed, and the ratio of free triazole to sodium triazole in the mixed system remains stable. S3 limits the above ratio to a range with upper and lower boundaries. Excess free triazole forms hydrogen-bonded associations with triazole anions, shielding the nitrogen atom, which is spatially most accessible and has a low charge density, and directing the nucleophilic attack towards the target nitrogen atom, thus improving regioselectivity. Excess free triazole also acts as a proton shuttle, immediately protonating the alkoxy anion generated at the moment of ring opening into a tertiary alcohol. The alkoxy anion does not have time to continue attacking the second molecule of epoxy intermediate, and the formation of diether-type impurities is suppressed. When the ratio is too low, both hydrogen bond shielding and proton supply are insufficient; when the ratio is too high, the triazole anion is diluted, prolonging the reaction cycle. S3 also replaces the high-boiling-point, highly polar aprotic solvent used in the prior art with a mixed solvent of alcohol and aromatics. This reduces the energy consumption of solvent recovery and the risk of residual high-boiling-point polar solvent in the active ingredient. The hydroxyl groups of the co-solvent also participate in proton shuttle and form complexes with free triazole.

[0056] In this invention, S4 changes the temperature of the ring-opening step to an amount adjusted according to the conversion process. S42 simultaneously measures the current concentrations of the epoxy intermediate and triazole in the same chromatographic injection, without calculating the concentration based on material balance; deviations introduced by bypass reactions and volume changes are not transmitted to the temperature setpoint. S43 uses the two concentrations as inputs and solves the reaction temperature corresponding to the instantaneous formation rate of the regioisomer being constant at the set upper limit based on the calibrated kinetic parameters. S44 takes the smaller of the calculated value and the upper temperature limit as the target temperature and tracks it using cascade control. In the early stage of the reaction, the concentration product is large, and the main reaction rate is already sufficient; the temperature control provides a lower temperature, and the early conversion, which accounts for most of the total conversion, is completed with high regioselectivity. In the later stage of the reaction, the concentration decreases, and the temperature control provides a higher temperature to complete the reaction on schedule. At this point, there is very little material remaining in the reactor, and the selectivity loss due to temperature increases is limited. The reaction cycle and impurity level are improved simultaneously, and the overall trade-off between the two shifts towards a shorter cycle and lower impurities. Re-selecting the temperature in isothermal operation cannot achieve the same results. After the upper temperature limit takes effect, the reaction liquid temperature is lower than the temperature required by the temperature control system, the instantaneous generation rate is lower than the set upper limit, and the generation rate of regioisomers does not exceed the set upper limit throughout the entire open-loop process. S44 takes the reaction liquid temperature as the controlled object instead of the jacket temperature, so the temperature in the kinetic relationship corresponds to the actual temperature of the reaction medium, and the accuracy of temperature control is guaranteed.

[0057] In this invention, step S5 converts the space freed up by suppressing impurities in the preceding steps into actual yield. Step S52 first measures the solvent and water content in the recycled mother liquor, then adds fresh solvent and antisolvent according to the total amount in the combined system. The antisolvent composition of the first batch is consistent with that of each batch, ensuring stable reproduction of crystal form, particle size distribution, and purity. Step S53 cools the mother liquor according to a cubic cooling curve and uses seed crystals. The supersaturation peak at the initial cooling stage is flattened, and the solute mainly deposits on the seed crystal surface, resulting in larger crystal particles and a narrower particle size distribution. The amount of impurities trapped in the mother liquor and adsorbed on the surface is reduced accordingly. Step S55 uses the measured impurity content of the mother liquor as feedback to calculate the lower limit of the mother liquor discharge ratio. Step S56 takes the larger of the baseline discharge ratio and the lower limit as the actual discharge ratio, and stops recycling when the lower limit exceeds the set upper limit, recovering this batch of mother liquor separately. The lower limit of the discharge ratio is proportional to the loading of regional isomers in the crude product. After the loading is reduced in S2 to S4, the discharge ratio given by the same criterion decreases significantly. The mother liquor can be discharged at a low ratio and reused at a high ratio. Most of the product dissolved in the mother liquor is recovered, and the overall yield is improved. Attached Figure Description

[0058] Figure 1 The charts show the comparison of batch time and crude product impurity content between Examples 1 to 5 and Comparative Examples 1 to 6. (a) is a comparison of the trade-off curve between batch time and N4-regioisomer mass fraction, and (b) is a comparison of N4-regioisomer, diether mass fraction and batch time under the conditions of missing steps and parameter out-of-bounds conditions.

[0059] Figure 2 The diagrams show the temperature trajectory of the reaction solution and the accumulation process of N4-region isomers in Example 1 and Comparative Examples 1-6. (a) is a comparison diagram of the change between the target temperature and the measured temperature of the reaction solution with batch time, and (b) is a comparison diagram of the change between the current concentration of the epoxy intermediate and the cumulative mass fraction of N4-region isomers with batch time.

[0060] Figure 3 This is a comparison chart of the impurity thresholds of epoxy feed solutions in Example 1 and Comparative Examples 9-1 to 9-9 and Comparative Example 3. (a) is a semi-logarithmic relationship between chloride ion content and N4-regioisomer mass fraction in crude product solution, and (b) is a semi-logarithmic relationship between water content and diol mass fraction in crude product solution.

[0061] Figure 4 The following are comparative graphs showing the effects of using the crystallization mother liquor in Examples 6 and 7 with Comparative Examples 7 and 8. In example (a), (b) is a comparative graph showing the evolution of the mass fraction of N4-regional isomers in the mother liquor with the batch number; (c) is a comparative graph showing the change of the mass fraction of N4-regional isomers in the technical grade drug with the batch number; and (d) is a batch-by-batch comparison graph showing the lower limit of the discharge ratio calculated according to S55 and the discharge ratio implemented according to S56.

[0062] Figure 5 This is a process flow diagram for preparing the high-purity cyproconazole technical material of the present invention. Detailed Implementation

[0063] The technical solutions of this invention will be clearly and completely described below. In addition, the forms of the various structures described in the following embodiments are merely illustrative. This invention is not limited to the structures described in the following embodiments. All other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this invention.

[0064] The preparation process of this invention consists of five steps, S1 to S5, in sequence. S1 is Grignard addition, S2 is closed-loop purification, S3 is solution preparation, S4 is temperature-controlled ring opening, and S5 is crystallization. For consistency throughout the text, 1-chloro-2-(4-chlorophenyl)-3-cyclopropylbut-2-ol is referred to as the chlorohydrin intermediate, 2-(4-chlorophenyl)-2-(1-cyclopropylethyl)ethylene oxide as the epoxy intermediate, and 2-(4-chlorophenyl)-3-cyclopropyl-1-(4H-1,2,4-triazol-4-yl)but-2-ol as the N4-regioisomer. The only difference between the N4-regioisomer and the target product cyproconazole is the position of the nitrogen atom involved in bonding in the 1,2,4-triazole ring. Cyproconazole is bonded at the N1 position of the triazole ring, while the N4-regioisomer is bonded at the N4 position. Both have the same molecular formula and relative molecular mass, and are regioisomers rather than stereoisomers. Therefore, their separation coefficient and solubility ratio in commonly used reversed-phase liquid chromatography and commonly used crystallization solvent systems are close to 1. Once formed, they are difficult to remove in subsequent purification.

[0065] This invention moves the inhibition of the N4-regioisomer to the formation stage. The 1,2,4-triazole anion is a binucleophile, capable of attacking the electrophilic center at both the N1 and N4 positions. In the system where the chlorohydrin intermediate and sodium 1,2,4-triazole coexist, two competing pathways exist simultaneously. The first pathway involves the direct bimolecular nucleophilic substitution of the chloromethyl group of the chlorohydrin intermediate by the 1,2,4-triazole anion. The second pathway involves the chlorohydrin intermediate first undergoing dehydrochlorination under alkaline conditions to form an epoxy intermediate, followed by ring-opening by the 1,2,4-triazole anion on the sterically less hindered methylene carbon on the epoxy ring. In the transition state of the first pathway, the electrophilic carbon is a primary carbon with little steric constraint, resulting in a small difference in activation energies between the N1 and N4 positions and poor regioselectivity. In the transition state of the second pathway, the three-membered ring strain and the steric hindrance of the adjacent quaternary carbon simultaneously constrain the transition state, leading to a clear preference for attack at the N1 position. This invention separates the closed-loop and open-loop processes in time and space via step S2, and removes chloride ions and water from the system generated during the closed-loop process, thus eliminating the conditions for the first channel to continue. Subsequently, steps S3 and S4 further improve the regioselectivity within a single channel. Finally, step S5 converts the low impurity level obtained in the preceding steps into a high proportion of mother liquor reuse and a high overall yield. The specific implementation process of a high-purity cyproconazole technical material preparation according to this invention is as follows.

[0066] S1 is carried out under a nitrogen atmosphere and includes four sub-steps, S11 to S14. In S11, magnesium shavings and anhydrous tetrahydrofuran are added to the reactor, along with iodine as an initiator. The temperature is then raised to a stable reaction temperature below the reflux temperature of tetrahydrofuran. The role of iodine is to react with the oxide film on the magnesium surface to form magnesium iodide, exposing a fresh metal surface and shortening the induction period of the Grignard reagent. Grignard reagents are extremely sensitive to water and oxygen. Trace amounts of water hydrolyze Grignard reagents into hydrocarbons, and trace amounts of oxygen oxidize Grignard reagents into magnesium alkoxides. Both directly reduce the addition yield. Therefore, there are upper limits on the water content of anhydrous tetrahydrofuran and the water content of the two organic raw materials in S11. It is preferable to control the water content of anhydrous tetrahydrofuran below 100 mg / kg and the water content of (1-chloroethyl)cyclopropane and 2-chloro-1-(4-chlorophenyl)ethyl-1-one below 200 mg / kg. If the water content is too high, it can be dehydrated by standing with molecular sieves or by azeotropic dehydration before feeding.

[0067] In step S12, an anhydrous tetrahydrofuran solution of (1-chloroethyl)cyclopropane was added dropwise to the magnesium tetrahydrofuran suspension obtained in step S11 under nitrogen protection. The dropping rate was maintained at a constant temperature. After the addition was complete, the mixture was kept at this temperature for a period of time to obtain a Grignard reagent solution. The amount of (1-chloroethyl)cyclopropane was slightly in excess relative to 2-chloro-1-(4-chlorophenyl)ethyl-1-one, and the amount of magnesium was also slightly in excess relative to (1-chloroethyl)cyclopropane. The role of the excess magnesium was to consume the residual active hydrogen and trace amounts of water in the system, preventing the Grignard reagent from being gradually quenched. The carbon bonded to chlorine in (1-chloroethyl)cyclopropane is a secondary carbon and is adjacent to the cyclopropyl group. If the magnesium insertion process is carried out at a high temperature, it is easy to generate cyclopropyl carbocation rearrangement products. Therefore, the temperature of step S12 should be maintained in the lower limit range of the magnesium insertion reaction to ensure stable progress, and the exothermic reaction should be controlled by a constant dropping rate.

[0068] S13 involves cooling the Grignard reagent solution obtained in S12 to a low-temperature range near zero degrees Celsius, followed by dropwise addition of anhydrous tetrahydrofuran solution of 2-chloro-1-(4-chlorophenyl)ethyl-1-one. After the addition is complete, the solution is maintained at the same low-temperature range. Adding the ketone to the Grignard reagent dropwise, rather than in the reverse order, aims to ensure a constant excess of Grignard reagent in the system, thus inhibiting proton transfer between the already formed magnesium alkoxide and the unreacted ketone. The low-temperature operation suppresses two side reactions: the first is the abstraction of the α-hydrogen from the ketone by the Grignard reagent, generating magnesium enol, which is recovered as the starting ketone during post-treatment; the second is the direct substitution of the chloromethyl group by the Grignard reagent, generating a coupling product without a hydroxyl group. The addition product is a magnesium salt of a tertiary alcohol, in which the quaternary carbon is simultaneously connected to 4-chlorophenyl, chloromethyl and 1-cyclopropylethyl. The two chiral centers are derived from the quaternary carbon generated by the addition and the methylene carbon on the 1-cyclopropylethyl, respectively. Therefore, this step has determined the ratio of the two diastereomers in the product. Subsequent steps S2 and S4 do not change the configuration of the two chiral centers.

[0069] In S14, an aqueous solution of ammonium chloride is added to the reaction solution obtained in S13 for quenching. The quenching process is temperature-controlled to suppress localized temperature rise. After settling and separation, the aqueous phase is extracted once with toluene, and the organic phases are combined. Subsequently, tetrahydrofuran is removed under reduced pressure and low reactor temperature conditions, and toluene is added to dissolve the intermediate, yielding a toluene solution containing the chlorohydrin intermediate. The use of an aqueous solution of ammonium chloride instead of a strong acid for quenching aims to maintain the pH of the system within a weakly acidic range, preventing the tertiary alcohol from dehydrating to form an olefin or undergoing a cyclopropyl rearrangement via a carbocation under strong acid conditions. Replacing the solvent with toluene from tetrahydrofuran provides a water-immiscible organic phase for the two-phase closed-ring system of S2, and also provides a suitable azeotropic agent for the azeotropic dehydration in S24. The toluene solution obtained in S14 does not need to separate the solid chlorohydrin intermediate; it directly enters S2 in solution form, saving the need for crystallization and drying, and avoiding the dehydration loss of tertiary alcohol intermediates during the drying process.

[0070] S2 comprises four sub-steps, S21 to S24, which quantitatively convert the chlorohydrin intermediate into an epoxy intermediate and remove the stoichiometric chloride ions generated during the ring-closure process, as well as water from the organic phase. Under alkaline conditions, the chlorohydrin intermediate first loses a proton from its hydroxyl group to generate an alkoxide anion. This alkoxide anion then undergoes intramolecular nucleophilic substitution of the chloromethyl group within the same molecule, removing the chloride ion and closing the ring to form a three-membered epoxy intermediate. This intramolecular substitution occurs in a five-membered transition state and can be quantitatively completed at near-room temperature, eliminating the need for the high-temperature conditions required in the ring-opening steps of existing processes. The ring-closure reaction is reversible; chloride ions in the system can nucleophilically open the methylene carbon of the epoxy intermediate, converting it back into a chlorohydrin intermediate. This continuously replenishes the first non-regioselective pathway mentioned earlier in subsequent ring-opening steps. Water in the system hydrolyzes and ring-opens the epoxy intermediate, generating 2-(4-chlorophenyl)-3-cyclopropylbut-1,2-diol. Therefore, S2 takes the degree of chloride ion and water removal as a prerequisite for entering S4.

[0071] In step S21, an aqueous solution of sodium hydroxide and tetrabutylammonium bromide are added to the toluene solution containing the chlorohydrin intermediate obtained in step S1, forming a two-phase system with toluene as the organic phase and the aqueous solution of sodium hydroxide as the aqueous phase. Sodium hydroxide is added in slight excess relative to the chlorohydrin intermediate to compensate for the consumption of hydrogen chloride during ring closure. Tetrabutylammonium bromide acts as a quaternary ammonium salt phase transfer catalyst; its cations can form ion pairs with hydroxide ions in the aqueous phase and enter the organic phase, transporting hydroxide ions to the phase containing the chlorohydrin intermediate. This ensures that the ring-closure reaction maintains a sufficient rate even at low temperatures, avoiding the need for higher temperatures or increased alkali concentrations to improve the rate.

[0072] S22 undergoes a closed-loop reaction at a temperature close to room temperature and a set impeller tip linear velocity. Samples are taken at regular intervals, and the residual amount of the chlorohydrin intermediate is determined by gas chromatography area normalization. The residual amount is calculated based on the organic components after deducting the toluene solvent peak. The closed-loop endpoint is determined when the residual amount falls below the set upper limit. Preferably, the upper limit for the residual amount of the chlorohydrin intermediate is set at 0.30%. If the residual amount is too high, the chlorohydrin intermediate entering S4 will directly generate the N4-regioisomer via a non-regioselective channel. Limiting the impeller tip linear velocity but not the rotational speed aims to allow for the conversion of stirring intensity between laboratory, pilot-scale, and production units based on geometric similarity principles. The impeller tip linear velocity is obtained by multiplying the impeller diameter by the rotational speed and then by pi. The mass transfer rate of the two-phase system is directly related to the impeller tip linear velocity. A low linear velocity results in larger droplet sizes and insufficient phase interface area, while a high linear velocity easily leads to emulsification and prolongs subsequent phase separation time.

[0073] S23 involves allowing the phase to separate after the closed-loop endpoint, discarding the lower aqueous phase containing sodium chloride. The organic phase is then washed with a sodium chloride aqueous solution at a certain ratio based on the volume of the organic phase. Each wash includes two stages: stirring and settling. Washing is performed at least twice, and after the first two washes, washing continues until the chloride ion content in the organic phase, as determined by ion chromatography, drops below a set upper limit. Preferably, the upper limit for the chloride ion content in the organic phase is 50 mg / kg. Using a sodium chloride aqueous solution instead of pure water aims to increase the ionic strength of the aqueous phase and reduce the mutual solubility of the two phases, shortening the phase separation time and reducing the amount of water entrained in the organic phase. The number of washes and the endpoint indicator are included in this step, ensuring clear follow-up actions even if the number of washes reaches the lower limit but the indicator has not yet been met, avoiding an intermediate state without intervention.

[0074] In step S24, the organic phase obtained from step S23 is heated and refluxed under normal pressure. Water is separated using toluene as an azeotropic agent via a water separator. During water separation, the top temperature of the column gradually rises from the boiling point of the toluene-water azeotrope to the boiling point of toluene, until the water content, as determined by the Karl Fischer coulometric method, falls below a set upper limit. Water separation is then stopped, yielding a toluene solution containing the epoxy intermediate. Preferably, the upper limit of the water content is set at 300 mg / kg, and the normalized content of the epoxy intermediate, calculated based on the gas chromatographic area of ​​the organic components after deducting the toluene solvent peak, is not less than 98.0%. The Karl Fischer coulometric method is based on the quantitative reaction of iodine and sulfur dioxide with water in methanol and pyridine media. Iodine is generated by electrolysis, and the water content is calculated according to Faraday's law. It is suitable for the determination of trace amounts of water at the milligram per kilogram level. The upward shift of the column top temperature from the azeotrope boiling point to the toluene boiling point itself is a qualitative indicator of the dehydration process. Using this method in conjunction with the quantitative results of the Karl Fischer coulometric method can reduce the number of samplings required. The toluene solution containing epoxy intermediates obtained from S24 is an intermediate material that can be stored and transported separately. The mass fraction of epoxy intermediates, gas chromatography content, residual amount of chlorohydrin intermediates, content of 2-(4-chlorophenyl)-3-cyclopropylbut-1,2-diol, chloride ion content, and water content together constitute the material specifications entering S4.

[0075] Step S3 comprises three sub-steps, S31 to S33, and its function is to prepare a mixed system in which free 1,2,4-triazole and sodium 1,2,4-triazole coexist in a predetermined ratio and are free of reactive water. In the 1,2,4-triazole anion, the N4 position is located between two nitrogen rings and is spatially most exposed, but its lone pair electron charge density is lower than that of the N1 position. Therefore, when there is an excess of neutral 1,2,4-triazole in the system, the N-H atoms on the neutral molecule preferentially form hydrogen bonds with the N4 position of the anion, shielding the spatially most accessible N4 position and directing the nucleophilic attack towards the N1 position. The excess neutral 1,2,4-triazole also acts as a proton shuttle, immediately protonating the alcohol oxy anion generated at the moment of ring opening into a tertiary alcohol, preventing the alcohol oxy anion from undergoing nucleophilic ring opening on the second epoxide intermediate, thereby inhibiting the formation of bis[2-(4-chlorophenyl)-3-cyclopropyl-2-hydroxybutyl] ether. Both of the above effects require that neutral 1,2,4-triazole and sodium 1,2,4-triazole coexist in a certain ratio. If the ratio is too low, hydrogen bond shielding and proton supply will be insufficient. If the ratio is too high, sodium 1,2,4-triazole will be diluted, the main reaction rate will decrease, the batch time will be prolonged, and the recovery load of unreacted 1,2,4-triazole will increase.

[0076] S31 disperses 1,2,4-triazole in toluene and adds sodium methoxide in methanol. The mixture is stirred in a medium-temperature range to form a salt. The salt-forming reaction is an acid-base reaction between sodium methoxide and 1,2,4-triazole, producing sodium 1,2,4-triazole and methanol. The degree of salt formation is determined by a non-aqueous potentiometric titration method. This method uses glacial acetic acid or isopropanol as the solvent, a perchloric acid-acetic acid solution as the titrant, and an electrode pair consisting of a glass electrode and a reference electrode to indicate the titration endpoint. The measured value is the amount of strongly alkaline triazole anions in the system. Preferably, the measured amount of sodium 1,2,4-triazole is not less than 98% of the amount of sodium methoxide added. The titration value is used directly for proportional calculations instead of extrapolating from the amount of feed, in order to reflect potential content deviations and moisture absorption failures of the sodium methoxide solution in the measured proportional value. The ratio of the amount of free 1,2,4-triazole to sodium 1,2,4-triazole is used. Calculate using the following formula.

[0077] ;

[0078] In the formula, The molar ratio of free 1,2,4-triazole to sodium 1,2,4-triazole is dimensionless. The amount (mol) of 1,2,4-triazole added. This represents the amount (mol) of 1,2,4-triazole sodium determined by non-aqueous potentiometric titration. The minus sign in the numerator indicates the amount of free 1,2,4-triazole remaining in neutral molecular form, obtained by subtracting the amount of salt-formed 1,2,4-triazole from the total amount of added 1,2,4-triazole. The fractional line represents normalization based on the amount of salt-formed 1,2,4-triazole. Preferably, [the following is a more accurate translation of the original text, but the provided text is incomplete and requires further context]. The value is controlled within the range of 3.5 to 5.5, and is further preferably controlled within the range of 4.2 to 4.8.

[0079] S32 distills methanol and its azeotrope with toluene under reduced pressure and a limited still temperature until the water content (calculated by Karl Fischer coulometric method) and the methanol mass fraction (calculated by gas chromatography internal standard method) both fall below the set upper limit. Sodium methoxide is used instead of sodium hydroxide or potassium hydroxide as the salt-forming alkali because the reaction of hydroxides with 1,2,4-triazole produces water stoichiometrically. This water is sufficient to disrupt the low-moisture state established by S24 and partially hydrolyzes the already formed 1,2,4-triazole sodium back to a neutral molecule. Out of control; when sodium methoxide forms a salt, methanol is generated. Methanol has a lower boiling point than toluene and forms a low-boiling azeotrope with toluene. It can be completely removed under mild reduced pressure without introducing water into the system. The mass fraction of residual methanol is also listed as an endpoint criterion because methanol is a protic solvent and has a certain acidity. When there is too much residual methanol, it will compete with 1,2,4-triazole for proton shuttle interaction and undergo alcoholysis and ring-opening with the epoxy intermediate in S4 to generate methoxy-substituted byproducts. Preferably, the upper limit of water content is set at 400 mg / kg and the upper limit of methanol mass fraction is set at 0.50%.

[0080] In step S33, n-butanol is added to the system obtained in step S32, allowing toluene and n-butanol to coexist at a predetermined mass ratio, resulting in a mixed system of 1,2,4-triazole and sodium 1,2,4-triazole. The role of n-butanol is threefold: first, it increases the solubility of 1,2,4-triazole and sodium 1,2,4-triazole in the organic phase, allowing the ring-opening reaction to proceed in a homogeneous liquid phase and avoiding localized concentration inconsistencies caused by solid-liquid two-phase reactions; second, its hydroxyl group participates in proton shuttle, working with free 1,2,4-triazole to inhibit the formation of diethers; and third, it controls the boiling point range of the system above the temperature range required for the ring-opening reaction, enabling the reaction to proceed under normal pressure. Replacing N,N-dimethylformamide, commonly used in existing processes, with n-butanol simultaneously reduces the energy consumption of solvent recovery and the risk of residual high-boiling-point polar solvents in the active ingredient.

[0081] Step S4 comprises four sub-steps, S41 to S44, whose function is to control the formation of the N4-regioisomer within a single open-ring channel to a predetermined level without prolonging batch time. Both the N4-regioisomer and the target product are generated by the attack of the 1,2,4-triazole anion on the methylene carbon of the epoxy intermediate. Both reactions are first-order reactions with respect to both the epoxy intermediate and total 1,2,4-triazole, exhibiting the same concentration dependence, differing only in rate constants. Calibration experiments showed that the apparent activation energy for the formation of the N4-regioisomer is significantly higher than that for the formation of the target product. Therefore, the ratio of their rate constants increases monotonically with increasing temperature; that is, decreasing temperature favors regioselectivity while increasing temperature favors reaction rate. Under isothermal operation, all conversions are completed with the same selectivity, and the process can only strike a trade-off between batch time and impurity levels. The present invention modifies the reaction solution temperature according to the conversion process. In the early stage of the reaction, when the concentrations of epoxy intermediate and 1,2,4-triazole are high and the main reaction rate is already sufficient, a lower temperature is given so that the early conversion, which accounts for most of the total conversion, is completed under high selectivity. In the later stage of the reaction, when the concentrations of both decrease rapidly, a higher temperature is given to complete the reaction on schedule. At this time, the amount of residual material is very small, and the selectivity loss caused by the temperature increase is limited.

[0082] In step S41, the mixture obtained in S3 is heated to the starting temperature range of the ring-opening reaction. Then, the toluene solution containing the epoxy intermediate obtained in S2 is added at a uniform rate within a set time. After the addition is complete, toluene is added again to adjust the liquid phase volume, so that the initial concentration of the epoxy intermediate and the initial total concentration of 1,2,4-triazole in the reaction solution fall within the set range. The purpose of using uniform feeding instead of adding the entire batch is to distribute the exothermic reaction over the feeding period, avoiding local overheating that could cause early selectivity loss.

[0083] S42 samples are taken from the reaction solution at regular intervals, quenched with an equal volume of glacial acetic acid aqueous solution, and then the current concentration of the epoxy intermediate is simultaneously determined in the same injection using high-performance liquid chromatography with external standard method. With the current total concentration of 1,2,4-triazole ,in This represents the concentration of the sum of free 1,2,4-triazole and 1,2,4-triazole sodium, converted to the total concentration of 1,2,4-triazole. Immediately after sampling, the solution is quenched with glacial acetic acid solution to neutralize the 1,2,4-triazole sodium to a neutral molecule and terminate the ring-opening reaction. This ensures the chromatographic result corresponds to the true concentration at the time of sampling and also merges the free 1,2,4-triazole and 1,2,4-triazole sodium into a single chromatographic peak, facilitating quantification based on the total concentration. Two concentrations are measured simultaneously in the same injection, rather than being extrapolated from the consumption of epoxy intermediates based on material balance. The reason is that, in addition to forming the target product and the N4-regioisomer, the epoxy intermediate is also consumed via bypass pathways such as hydrolysis and alcoholysis. These bypass pathways consume the epoxy intermediate without consuming 1,2,4-triazole at the same stoichiometric rate. Sampling, solvent evaporation, and solvent replenishment also change the liquid phase volume, which is why the equilibrium calculations yield the following results. The aforementioned deviation will be passed on to subsequent temperature settings.

[0084] S43 takes the measurements from S42. and Substitute into the following formula to calculate the target temperature of the reaction solution. .

[0085] ;

[0086] In the formula, The calculated value (K) represents the target temperature of the reaction solution. Apparent activation energy (J / mol) for the reaction that generates the N4-regioisomer; is the molar gas constant, taken as 8.314 (J / (mol·K)); The apparent pre-exponential factor (L / (mol·h)) for the reaction that generates the N4-regioisomer. This represents the current concentration (mol / L) of the epoxy intermediate as measured by S42. The current total concentration (mol / L) of 1,2,4-triazole as measured by S42. Set an upper limit (mol / (L·h)) for the formation rate of N4- regioisomers; ln is the natural logarithm operator; fraction lines represent division; and The values ​​were obtained from the regression of the calibration experiment. The process is selected based on the target impurity level and the target batch time.

[0087] The instantaneous rate of the reaction that produces the N4-regioisomer can be written according to the law of mass action as the rate constant and and The product of these three factors, and the relationship between the rate constant and temperature, can be expressed according to the Arrhenius empirical formula as follows: With negative Divide by The product of temperature and temperature is the product of a natural exponential function. Let the above instantaneous rate be always equal to a set upper limit. Divide both sides of the equation by , and The product of these two equations, taking the natural logarithm of both sides and rearranging, yields the explicit solution for the temperature, resulting in the above equation. Dimensional verification is as follows: [The right-hand numerator...] The unit is J / mol, and the denominator contains The unit of is J / (mol·K), and the unit of the division is K, consistent with the left side; the independent variable of the logarithmic operation is . The unit is L / (mol·h). and The unit is mol / L and The unit mol / (L·h) becomes dimensionless after reduction, satisfying the requirements of logarithmic operations. The above formula is not a theoretically derived relation, nor is it a relation describing the state of the system. Instead, it is a control formula composed of measured kinetic parameters that directly outputs the temperature setpoint. All its input quantities can be obtained by sampling and measurement.

[0088] S44 take The smaller of the two upper and lower temperature limits shall be used as the target temperature for the reaction solution. A cascade control loop, with the reaction liquid temperature as the main loop and the jacket temperature as the slave loop, tracks the reaction. Then repeat S42 to S44 until... Stop when the concentration drops below the set endpoint, then cool to a temperature suitable for entering S5 to obtain the crude solution. Preferably, the upper temperature limit is set to 368 K, the endpoint concentration is set to 0.01 mol / L, and the normalized content of N4-regioisomers in the crude solution (based on the organic components after deducting the solvent peak) is not higher than 0.40%. The reason for setting the upper temperature limit is that, as... and The decrease, given by the above formula Monotonically increasing the temperature will exceed the atmospheric boiling point of the solvent system and the allowable temperature of the equipment. After limiting the temperature, the reaction solution temperature is maintained at the upper limit. At this point, the N4-regioisomer formation rate calculated by the above formula is lower than... Therefore, the generation rate in the unlimited range is equal to The generation rate is lower than [a certain value] within the limited range. The generation rate throughout the entire process is no higher than The controlled object is defined as the temperature of the reaction liquid rather than the jacket temperature because the temperature in the Arrhenius empirical formula represents the temperature of the reaction medium, while there is a heat transfer temperature difference and dynamic hysteresis between the jacket and the reaction liquid. This difference is more pronounced during the exothermic open-loop phase. Therefore, the main loop is constructed using the reaction liquid temperature, and the slave loop using the jacket temperature. The jacket temperature setpoint is used as the controller output instead of being directly substituted into the above formula. To avoid thermal shock and temperature overshoot caused by setpoint jumps, the setpoint values ​​given in adjacent intervals are... The difference between adjacent set values ​​and the heating rate of the reaction solution are both set with upper limits. It is preferable to set the upper limit of the difference between adjacent set values ​​to 5K and the upper limit of the heating rate of the reaction solution to 0.5K / min.

[0089] When operated in the above manner, the cumulative amount of N4-regioisomers, expressed as molar concentration, increases linearly with time within the unlimited range, with a constant growth slope. Within the limited range, the growth slope gradually decreases. Therefore, at the same batch time as the isothermal operation, the accumulation of N4- regioisomers is lower than in any isothermal scheme. However, at the same impurity level as the isothermal operation, the batch time is significantly shorter than in the low-temperature isothermal scheme. These results cannot be obtained by reselecting the temperature in the isothermal process because the batch time and impurity level in the isothermal process change along the same curve. Adjusting the temperature according to the conversion process shifts the curve as a whole towards lower impurities and shorter times.

[0090] and The calibration was obtained through regression testing, and the calibration method is as follows. Using the same solvent composition as S3 and S41, Under the conditions of the initial concentration and the target value, isothermal reactions were carried out at several temperature points covering the operating temperature range of the ring-opening reaction. Samples were taken at each temperature point at regular intervals and measured according to method S42. The rate constant at each temperature point was obtained by linearly regressing the integral of the second-order reaction with time to the amount of N4-regioisomer produced. Then, a linear regression was performed on the natural logarithm of the rate constant against the reciprocal of the temperature. The slope of the regression line was multiplied by the negative... Seek It is obtained from the natural exponent of the intercept. The number of isothermal points should ideally be no less than four, in order to verify the linearity of the regression. and These are apparent parameters obtained under the above calibration conditions; therefore, their applicability is limited to the solvent mass ratio at the time of calibration. The values, initial concentration ranges, and temperature ranges are the limits; if the values ​​exceed these ranges, recalibration is required.

[0091] S5 comprises two parts: crystallization and mother liquor treatment. The crystallization part includes S51 to S54, and the mother liquor treatment part includes S55 and S56. In S51, toluene and n-butanol are removed from the crude product solution obtained in S4 under reduced pressure and a limited reactor temperature, yielding crude product residue. The purpose of the reduced pressure operation is to lower the solvent removal temperature, avoid dehydration of the tertiary alcohol product at high temperatures, and prevent side reactions initiated by the residual 1,2,4-triazole sodium at high temperatures.

[0092] S52 first determines the mass fractions of isopropanol and water in the recycled crystallization mother liquor. Then, the recycled crystallization mother liquor and fresh isopropanol are added to the crude product residue obtained in S51, so that the ratio of the total mass of isopropanol in the combined system to the mass of cyproconazole in the crude product residue falls within a set range. The temperature is then raised to dissolve and clarify the product. The recycled crystallization mother liquor is not added in the first batch. The measurement is based on the total amount in the combined system rather than the amount of fresh isopropanol added because the recycled crystallization mother liquor itself carries a considerable amount of isopropanol and water. If only the amount of fresh isopropanol is added, the solvent composition in the first batch, the initial stage of use, and the steady state of use will be different. The anti-solvent ratio will drift with each batch, and the crystal form, particle size distribution, yield, and purity will be difficult to reproduce stably.

[0093] S53 determines the total volume of water according to the total volume of isopropanol in the combined system and the set ratio. After deducting the water volume brought in by the recycled crystallization mother liquor, deionized water is added. Then, the temperature is lowered to the crystallization temperature and cyproconazole seed crystals are added. The temperature is then lowered according to the following formula.

[0094] ;

[0095] In the formula, For the first time since the addition of the seed crystal Crystallization temperature (°C) per hour; Crystallization temperature (°C); The final temperature of crystallization (°C); The time since the addition of the seed crystal, ranging from 0 to [missing value]. (h); The total cooling time (h) is the total time for the program to cool down; the minus sign indicates the difference between the crystallization temperature and the cooling rate; the difference in parentheses is the total time for the program to cool down. reduce The total cooling rate; fractional Divide by The time is dimensionless relative time; the superscript 3 indicates that the relative time is raised to the cube.

[0096] After adding seed crystals, if the crystals grow at a constant linear growth rate under constant supersaturation, all three characteristic dimensions of the crystals will increase linearly with time, and the crystal mass will increase cubically with time. To keep the supersaturation constant throughout the cooling process, the precipitation rate of the solute in the solution should match the growth rate of the crystal mass. The amount of solute precipitated is determined by the decrease in saturated solubility with temperature; therefore, the decrease in temperature should also change cubically with time. This yields the above formula. Compared to natural cooling and linear cooling, cooling according to the above formula can flatten the supersaturation peak at the initial stage of cooling, suppressing the initial explosive primary nucleation, causing the solute to mainly deposit on the seed crystal surface rather than on newly formed fine crystal nuclei. The resulting crystals have larger particle sizes and narrower particle size distributions, and the amount of impurities trapped in the mother liquor and adsorbed on the surface is reduced accordingly. The amount of seed crystals added and the median particle size per unit volume are related. This also affects the above effects. When the amount of seed crystals is too small, the surface area of ​​the seed crystals is insufficient, and the supersaturation will still exceed the nucleation zone in the early stage of cooling. When the amount of seed crystals is too large, the product particle size is too fine. Therefore, it is advisable to add seed crystals with a narrow particle size distribution in a certain proportion of the theoretical yield. The theoretical yield is calculated based on the amount of 2-chloro-1-(4-chlorophenyl)ethyl-1-one added in S1 and its measured content.

[0097] S54 was kept at the crystallization endpoint temperature for a period of time to fully release the supersaturation, then filtered. The filter cake was washed with a low-temperature mixture of isopropanol and water, and then vacuum dried under reduced pressure and a mild temperature to obtain cyproconazole technical. The washing solution was a mixture of isopropanol and water with a composition similar to that of the crystallization mother liquor, and pre-cooled. Its purpose was to displace the mother liquor entrained on the surface of the filter cake without causing significant dissolution of the crystals. The quality indicators of the obtained technical grade drug include the total content of cyproconazole on a dry basis, the content of N4-regioisomers, the content of 2-(4-chlorophenyl)-3-cyclopropylbut-1,2-diol, the content of bis[2-(4-chlorophenyl)-3-cyclopropyl-2-hydroxybutyl] ether, and the melting point range. The total content of cyproconazole is calculated by converting the sum of the areas of the two diastereomer peaks of cyproconazole using the external standard method in high-performance liquid chromatography (HPLC). The dry basis is the quality standard after deducting the loss on drying, measured at constant weight at 105℃. The contents of the above three impurities are calculated by normalizing the organic components after deducting the solvent peaks using the HPLC area normalization method. Specifying the measurement scope of the total content to the inclusion or exclusion of stereoisomer peaks and the dry basis standard aims to avoid content differences between different batches due to different measurement scopes.

[0098] S55 determines the mass of the crystallization mother liquor obtained from S54 filtration. and the mass fraction of N4-regioisomers therein And calculate the lower limit of the mother liquor discharge ratio according to the following formula. .

[0099] ;

[0100] In the formula, This is the lower limit of the proportion of mother liquor discharged, and it is dimensionless. This represents the upper limit of the permissible mass fraction of N4-regioisomers in the mother liquor for crystallization. Mass (kg) of the crystallization mother liquor obtained by S54 filtration; The mass of N4-regioisomers introduced into the next batch of crude product residue is calculated as the product (kg) of the mass of the current batch of crude product residue and the mass fraction of N4-regioisomers therein. The measured mass fraction of N4-regioisomers in the crystallization mother liquor obtained by S54 filtration; the minus sign and the fraction line represent subtraction and division, respectively. and The product represents the upper limit of the mass of N4-regioisomers allowed to exist in the mother liquor; and The product is the mass of the N4-regioisomer currently present in the mother liquor.

[0101] Assuming that almost all N4-regioisomers remain in the mother liquor during crystallization, the mass of N4-regioisomers in the next batch of mother liquor is equal to the sum of the mass of the mother liquor returned after being discharged from this batch and the mass newly introduced into the crude product residue from the next batch. This is equal to 1 minus the discharge ratio multiplied by [the formula is missing here]. and The accumulation of, plus The sum of the above must not exceed [a certain value]. and By balancing both sides of the inequality and solving the proportional relationship, we obtain the above equation. If each batch follows the same... The calculated values ​​are output, and the continuous application reaches a steady state. Approaching The above equation degenerates into Divide by and The product, that is, the mass of N4-regioisomers carried away by each batch under steady state is equal to the mass newly introduced in each batch; if each batch is discharged according to the basic discharge ratio described below, then the steady state... Below This allows for a margin of safety in product quality. Based on actual measurements... Instead of the allowed upper limit The reason for substituting the upper limit into the denominator is that the upper limit is only a constraint condition and not the actual content of the mother liquor. In the initial stage of application, the actual content of the mother liquor is lower than the upper limit. If the upper limit is substituted, the amount of impurities that can be carried away by a unit mass of mother liquor will be overestimated, making the discharge insufficient to maintain the balance.

[0102] S56 takes a basic discharge ratio and The larger of the two values ​​will be used as the discharge ratio for this batch of mother liquor. ,Bundle and The accumulated crystallization mother liquor is discharged, and the remainder is returned to the next batch of S52 for reuse; when When the set upper limit is exceeded, stop the reuse of mother liquor and recover ciprofloxacin separately from this batch of crystallization mother liquor. Preferably, [the mother liquor is used in this process]. Set it to 1.20%, and set the basic discharge ratio to 0.10. The upper limit is set to 0.20, and makes The mass ratio of the crude product to the residue in the reactor falls within the range of 2.3:1 to 2.7:1. The purpose of setting a basic discharge ratio is to ensure proper initial application. The calculated value may be negative or very small. Discharging according to the basic discharge ratio can maintain the steady-state content of N4-regioisomers in the mother liquor below the allowable upper limit rather than exactly equal to it, thus leaving a margin for product quality; setting The upper limit and corresponding handling methods serve to ensure that when abnormalities occur in the preceding steps or the load of N4-regional isomers in the crude product residue exceeds the design value, the process has a clear exit path, rather than allowing impurities to continuously accumulate in the mother liquor. The value is determined based on the calibration curve between the mass fraction of N4-regioisomers in the mother liquor and the mass fraction of N4-regioisomers in the finished product. Under fixed crystallization conditions, the two show a monotonic relationship. The value is obtained by checking the corresponding content in the mother liquor against the internal control limit of the finished product and leaving a margin. .

[0103] The mother liquor disposal rules in S5 are directly coupled with those in S2 to S4. In the above formula... The mass fraction of N4- regioisomers in the crude product residue is determined by the degree of chloride and water removal from S2 and S3. The value is determined by both the temperature trajectory of S4 and the value of S2. After S2 to S4 suppress the content of the N4-regioisomer, the same inequality gives... As the concentration of impurities decreases, the mother liquor can be discharged at a low proportion and reused at a high proportion, thus recovering most of the cyproconazole dissolved in the mother liquor and increasing the overall yield. Conversely, if the mother liquor disposal rules in S5 are not implemented, the yield margin created by reducing impurities in S2 to S4 cannot be converted into actual yield. In another aspect, S2 removes chloride ions and water from the system to below a set level, while also providing the mixing system in S3 with conditions that prevent hydrolysis. If the system contains water, 1,2,4-triazole sodium will partially hydrolyze into neutral molecules and sodium hydroxide. With the value out of control and the introduction of a strong base, the required ratio for S3 cannot be established. Furthermore, after the combined action of S2 and S3, the formation of the N4-regioisomer occurs only through the ring-opening pathway of the epoxy group. The system is simplified into a pair of parallel competing reactions with the same concentration dependence, which is the premise for the kinetic model used in S4 to hold true. Therefore, the overall effect of this invention is not a simple sum of the effects of each step.

[0104] Each step of this invention can be implemented in a conventional glass-lined or stainless steel batch reactor. Specifically, S1 requires nitrogen purging and a cryogenic refrigerant; S2 and S24 require a water separator and a reflux condenser; S32 requires a distillation column and a vacuum system; S4 requires an external circulation sampling valve and a temperature control loop capable of cascade control; and S5 requires a jacketed temperature control system capable of cooling according to a programmed curve and a filtration and drying device. The calculation for S43 can be automatically completed by a distributed control system according to the above formula. The temperature circuit can be set, or the operator can manually calculate and input the set value based on the sampling results. There is no substantial difference between the two methods in implementing the present invention.

[0105] Example 1: Steps S1 to S5 and their sub-steps of the present invention were carried out in a 2L four-necked flask. In step S11, 31.6g of magnesium shavings and 300mL of anhydrous tetrahydrofuran were added to the reactor under a nitrogen atmosphere, along with 0.10g of iodine as an initiator. The temperature was raised to 42°C. The anhydrous tetrahydrofuran contained 62mg / kg of water, (1-chloroethyl)cyclopropane contained 118mg / kg of water, and 2-chloro-1-(4-chlorophenyl)ethyl-1-one contained 95mg / kg of water. In step S12, a mixture of 125.5g of (1-chloroethyl)cyclopropane and 200mL of anhydrous tetrahydrofuran was added dropwise under nitrogen protection, with the temperature controlled at 40-45°C for 1.5 hours. After the addition was complete, the mixture was kept at this temperature for 1.0 hour to obtain a Grignard reagent solution. The molar ratio of the two organic raw materials was 1.20, and the molar ratio of magnesium to the haloalkanes was 1.08. S13 was cooled to −3℃, and a solution of 189.0 g of 2-chloro-1-(4-chlorophenyl)ethyl-1-one and 200 mL of anhydrous tetrahydrofuran was added dropwise. The temperature was controlled from −5℃ to 0℃, and the addition was carried out over 1.5 h. After the addition was completed, the solution was kept at 0 to 5℃ for 1.0 h. S14 was quenched by adding 400 mL of 10% ammonium chloride aqueous solution, and the temperature was controlled not to exceed 15℃. The solution was allowed to stand and separated. The aqueous phase was extracted once with 150 mL of toluene. The organic phases were combined and tetrahydrofuran was removed at 40℃ and 11 kPa absolute pressure. Toluene was added to a final volume of 900 g of organic phase to obtain a toluene solution containing 248.8 g of chlorohydrin intermediate, with a molar yield of 96.0%. The peak area ratio of the two diastereomers was 57.4 to 42.6.

[0106] S21. Add 153.0 g of a 30% sodium hydroxide aqueous solution and 1.55 g of tetrabutylammonium bromide to the above toluene solution. The molar ratio of sodium hydroxide to the chlorohydrin intermediate is 1.20, and the molar amount of tetrabutylammonium bromide is 0.50% of that of the chlorohydrin intermediate. S22. React at 30°C and a stirrer tip velocity of 1.9 m / s for 2.0 h. The residual amount of the chlorohydrin intermediate is 0.18%, and the ring-closing endpoint is determined. S23. Allow the mixture to stand for 30 min to separate into layers, discard the aqueous phase, and wash three times with a 5% sodium chloride aqueous solution at 22% of the organic phase volume. The chloride ion content of the organic phase is 32 mg / kg. S24 atmospheric pressure reflux is separated by a water separator. The temperature at the top of the column is raised from 84℃ to 110℃. Water separation is stopped when the water content is 210mg / kg, yielding 686.0g of epoxy feed solution, including 209.6g of epoxy intermediate (mass fraction 30.6%), a normalized gas chromatographic area of ​​98.8% after deducting the toluene solvent peak, 0.18% residual chlorohydrin intermediate, 0.11% 2-(4-chlorophenyl)-3-cyclopropylbut-1,2-diol, and a closed-ring molar yield of 98.0%. All indicators fall within the range defined by this invention for epoxy feed solution.

[0107] S31: 113.9 g of 1H-1,2,4-triazole was dispersed in 250 mL of toluene, and 54.0 g of a 30% sodium methoxide methanol solution was added. The mixture was stirred at 50 °C for 30 min to form a salt. Non-aqueous potentiometric titration determined that 0.297 mol of 1,2,4-triazole sodium was present, representing 99.0% of the amount of sodium methoxide added. The ratio was calculated according to the formula in S31 of this invention. The value is 4.56. Distillation of S32 was stopped when 90.5 g of methanol and an azeotrope of methanol and toluene were distilled off at an absolute pressure of 15 kPa and a still temperature of 58 °C, with a water content of 260 mg / kg and a methanol mass fraction of 0.31%. In S33, 172.0 g of n-butanol was added, with a toluene to n-butanol mass ratio of 4.0:1, to obtain a mixed system.

[0108] S41 heats the mixture to 333K, and adds the epoxy solution obtained in S24 at a uniform rate over 1.0h. After the addition is complete, toluene is added to bring the liquid phase to 785mL. It is 1.20 mol / L. The concentration was 2.10 mol / L, and the time of completion of the addition was recorded as the zero point of the batch time. S42 samples were taken every 15 minutes, quenched with an equal volume of 5% (w / w) glacial acetic acid aqueous solution, and determined in the same injection using high-performance liquid chromatography with external standard method. and S43 take 8.0×10 -4 mol / (L·h), calculated according to the temperature control formula in S43 of this invention. The first calculated value is 335.3K. S44 takes... The smaller of 368K as A cascade control loop, with the reaction solution temperature as the main loop and the jacket temperature as the slave loop, tracks the reaction. The difference between adjacent setpoints should not exceed 5K, and the reaction solution heating rate should not exceed 0.5K / min. The batch time is 3.20 hours. The temperature reached 368K, and thereafter remained constant at 368K; batch time was 5.08h. The concentration was set to 0.006 mol / L. The reaction was stopped and the temperature was lowered to 333 K to obtain a crude solution containing 0.25% N4-regioisomer, 0.12% 2-(4-chlorophenyl)-3-cyclopropylbut-1,2-diol, 0.015% bis[2-(4-chlorophenyl)-3-cyclopropyl-2-hydroxybutyl] ether, and 0.14% epoxy intermediate.

[0109] S51 was processed at 60℃ and 11 kPa to remove toluene and n-butanol, yielding 275.9 g of crude product residue, of which 271.9 g was cyproconazole. S52, being the first batch, did not include recycled crystallization mother liquor. 300 mL of isopropanol was added, and the mixture was heated to 60℃ to dissolve and clarify. S53, 210 mL of deionized water (70% of the isopropanol volume) was added, and the mixture was cooled to 55℃. 0.50 g of cyproconazole seed crystals were then added. The median particle size of the seed crystals was... The particle size is 45 μm, and the amount added is 0.18% of the theoretical yield. The product is then cooled according to the programmed cooling method in S53 of this invention. 55℃ 10℃ The temperature was lowered for 3.0 h. S54 was kept at 10℃ for 1.0 h, filtered, and the filter cake was washed with 60 mL of a mixture of isopropanol and water at 5℃ in a 1:1 volume ratio. The mixture was then vacuum dried at 60℃ and 11 kPa absolute pressure for 4.0 h to obtain 234.4 g of cycloproconazole technical, with a total dry basis content of 99.32%, N4-regioisomer 0.09%, 2-(4-chlorophenyl)-3-cyclopropylbut-1,2-diol 0.03%, bis[2-(4-chlorophenyl)-3-cyclopropyl-2-hydroxybutyl] ether less than 0.02%, melting point 106.8 to 108.5℃, loss on drying 0.10%, peak area ratio of the two diastereomers 57.6 to 42.4, and single-pass molar yield based on 2-chloro-1-(4-chlorophenyl)ethyl-1-one 80.3%. S55 yielded 680g of crystallization mother liquor, containing 5.5% cyproconazole and 0.100% N4-regioisomer. It is 0.680g. The weight was 680g, and the ratio of the crude product to the residue in the reactor was 2.46 to 1. Taking 1.20%, we calculate... It is a negative value. S56 takes 0.10 and... The larger one, i.e. The concentration was 0.10, 68g of mother liquor was discharged, and the remaining 612g was returned to the next batch of S52.

[0110] Examples 2 through 5 are identical to Example 1 and will not be repeated. Example 2 is a lower limit parameter example. In S31, 1.505 mol of 1H-1,2,4-triazole and 0.335 mol of sodium methoxide were added, and titration revealed 0.334 mol of sodium 1,2,4-triazole. The concentration is 3.51, with a toluene to n-butanol mass ratio of 3.5:1; S23 is washed until chloride ions reach 48 mg / kg, and S24 is dehydrated to 295 mg / kg; S41... 1.00 mol / L It is 1.80 mol / L; S43's 5.0×10 -4mol / (L·h); S53 54℃ 12℃ The duration was 2.5 hours. Example 3 is an example with the upper limit parameter. In S31, 1.786 mol of 1H-1,2,4-triazole and 0.277 mol of sodium methoxide were added, and the titration yielded 0.276 mol of sodium 1,2,4-triazole. The concentration was 5.48, with a toluene to n-butanol mass ratio of 4.5:1; S41's... 1.40 mol / L The concentration was 2.40 mol / L, and the feeding temperature was 338 K; S43 2.4 × 10 -3 mol / (L·h), the first The temperature is 343.0K, which is 5.0K different from the feeding temperature, and does not exceed the upper limit of the difference between adjacent set values; S53 takes... 56℃ 8℃ The time was 3.5 hours. Example 4 was a pilot-scale amplification example, scaled up 60 times according to the ratio of Example 1 and implemented in a 500L glass-lined reactor. S42 used external circulation sampling with a sampling interval of 20 minutes. S43... 1.2×10 -3 mol / (L·h), The temperature is automatically controlled by the distributed control system and then automatically sent to the main circuit of the reaction liquid. Example 5 is a substitute material example. Toluene in S1 and S2 is replaced with xylene, and the temperature range at the top of the column in S24 is changed from 92°C to 140°C. Tetrabutylammonium bromide in S21 is replaced with an equal amount of tetrabutylammonium chloride. Sodium methoxide methanol solution in S31 is replaced with a 21% sodium ethoxide ethanol solution and ethanol is distilled off. n-Butanol in S33 is replaced with an equal mass of n-pentanol. The isopropanol and water system in S52 and S53 is replaced with an ethanol and water system. In S43… 1.6×10 -3 mol / (L·h). The main process parameters and test results of Examples 1 to 5 are compared in Table 1.

[0111] Table 1. Main parameters and results of Examples 1 to 5:

[0112]

[0113] Table 1 shows Examples 1 and 4. Similarly, the batch times and crude product impurity levels in both cases were consistent with the results calculated using the kinetic parameters in Table 5; Examples 2, 3, and 5 were... Different values ​​or different solvent systems can lead to deviations from the results calculated using the same set of kinetic parameters. The direction of these deviations differs from that in Experimental Example 1. The effects on reaction rate and regioselectivity are consistent, and the calibrated kinetic parameters are within the limits defined in this invention. Use the apparent parameters within the range.

[0114] The melting points of the technical materials obtained in Examples 2 to 5 all fall between 106.0 and 109.0 °C, the content of 2-(4-chlorophenyl)-3-cyclopropylbut-1,2-diol is not greater than 0.04%, the content of bis[2-(4-chlorophenyl)-3-cyclopropyl-2-hydroxybutyl] ether is less than 0.02%, and N,N-dimethylformamide was not detected in the residual solvent. All indicators fall within the range defined by this invention for the technical materials.

[0115] Example 6 is a continuous application of the mother liquor example, using the apparatus of Example 4 according to Example 1. The feed was continuously fed for 30 batches. Each batch of crude product had a residue of 72.0 kg in the reactor, including 71.0 kg of cyproconazole, 0.25% N4-regioisomer, and crystallization mother liquor. The weight was 180 kg, and the ratio of the crude product to the residue in the reactor was 2.5:1. The mass fraction of cyproconazole in the mother liquor was 5.5%. Take 1.20%. The percentage for each batch is calculated according to S55. All are negative values, take S56. The yield was 0.10. The first batch had a yield of 80.2% of qualified products because there was no mother liquor to reuse; this yield stabilized at 91.8% from the second batch onwards.

[0116] Example 7 illustrates the handling of an abnormal batch. The mother liquor obtained from the 12th batch of filtration in Example 6 contained 0.718% N4-regioisomers. Two subsequent batches were then processed. In the 13th batch, the water washing of S23 was intentionally interrupted, causing the chloride ion concentration in the epoxy feed solution to rise to 8600 mg / kg, and the N4-regioisomer content in the crude product residue to rise to 1.40%. The concentration was 1.008 kg; the N4-regioisomer content in this batch of mother liquor increased to 1.206%, only slightly higher than... The percentage was 0.006 percentage points higher, and the N4-regioisomer content in the resulting technical grade drug was 0.141%, still within the internal control limit of 0.15%; calculated according to S55... If the concentration is 0.469 and exceeds 0.20, the mother liquor reuse should be stopped according to S56, and the mother liquor from this batch should be recovered separately for ciprofloxacin. For batch 14, S23 and S24 should be restored to normal, and the N4-regioisomer content in the crude product residue should return to 0.25%. Since the reuse of the mother liquor from the previous batch has been stopped, this batch was crystallized with fresh isopropanol and deionized water, and the N4-regioisomer content in the mother liquor was 0.100%. Negative value Take 0.10 and reuse the mother liquor as usual.

[0117] Except for the changes listed, the raw materials, proportions, equipment, operations, and testing methods of each comparative example are consistent with those of Example 1, and all comparative examples are practically feasible. Comparative Example 1 is a isothermal open-loop series, in which the temperature control of S43 and S44 is eliminated. S4 is kept at a constant temperature of seven points from 338.15K to 368.15K throughout the process, numbered sequentially from Comparative Example 1-1 to Comparative Example 1-7. The reaction endpoint is still taken as... The concentration was not greater than 0.01 mol / L, and the rest was the same as in Example 1. The results of ring opening at seven isothermal points for Comparative Examples 1-1 to 1-7 are shown in Table 2.

[0118] Table 2. Isothermal open-loop results of Comparative Examples 1-1 to 1-7:

[0119]

[0120] Comparative Example 2 is the closest to the prior art. S2 and S3 are omitted. The toluene solution containing the chlorohydrin intermediate obtained in S14 is desolvated and directly reacted with 1.65 mol of 1,2,4-triazole sodium in 785 mL of N,N-dimethylformamide at 100 °C for 1.5 h, followed by cooling and crystallization. Comparative Example 3 is a comparison example omitting S23 and S24. After determining the ring-closing endpoint in S22, no separation, water washing, or azeotropic dehydration is performed; the solution is directly added to S3. The resulting solution contains 8600 mg / kg chloride ions and 9200 mg / kg water. Comparative Examples 4 to 6 have the sodium methoxide dosage in S31 changed to 1.650 mol, 0.660 mol, and 0.183 mol, respectively. The values ​​were 0, 1.50, and 8.00, respectively, with the rest being the same as in Example 1. Table 3 shows the results of Comparative Examples 2 to 6 under the conditions of missing steps and parameter out-of-bounds conditions.

[0121] Table 3, Results of Comparative Examples 2 to 6:

[0122]

[0123] The diol in Table 3 is 2-(4-chlorophenyl)-3-cyclopropylbut-1,2-diol, and the diether is bis[2-(4-chlorophenyl)-3-cyclopropyl-2-hydroxybutyl] ether. The contents of both diols and the N4-regioisomer are calculated using the high-performance liquid chromatography area normalization method after deducting the solvent peak. In Example 1, the crude product solution at the same diameter contained 0.12% diol and 0.015% diether. Comparative Example 2 was run according to the same mother liquor disposal rules. The yield was 1.166 kg. Under steady-state conditions, the discharge ratio needed to be increased to 0.54, and the total yield of qualified products under steady-state conditions was 79.9%.

[0124] Comparative Examples 7 and 8 are comparative examples of mother liquor discharge ratios. The apparatus, ingredients, and preliminary steps are the same as in Example 6, except that the discharge ratios in S56 are fixed at 0.05 and 0.30 respectively, and run continuously for 30 batches. 0.05 is lower than the baseline discharge ratio specified in S56, and 0.30 is higher than the upper limit specified in S56. Comparative Example 9 is a series of chloride and water spiked epoxy feed solutions. The epoxy feed solution obtained in S24 of Example 1 is used. Chloride ions are provided by a toluene solution of tetrabutylammonium chloride without introducing water, and water is provided by deionized water. The five levels of chloride ion spike are sequentially numbered as Comparative Examples 9-1 to 9-5, and the four levels of water spike are sequentially numbered as Comparative Examples 9-6 to 9-9. The remaining steps are carried out according to S3 and S4 of Example 1. Tetrabutylammonium chloride was chosen instead of an inorganic chloride because tetrabutylammonium bromide already exists as a phase transfer catalyst in S21. Spiking with a similar quaternary ammonium salt does not introduce new species into the system, and it is soluble in toluene and does not introduce water of crystallization. A comparison table of the spiked series of Comparative Example 9 with Examples 1 and 3 is shown in Table 4.

[0125] Table 4. Comparison of the spiked series of Comparative Example 9 with Examples 1 and 3:

[0126]

[0127] In the following experimental examples, the total dry basis content of cyproconazole was determined by reversed-phase high-performance liquid chromatography (RP-HPLC) using the external standard method, calculated by the sum of the peak areas of the two diastereomers of cyproconazole, and the dry basis content was reduced by the drying loss measured at 105℃ constant weight method; the contents of N4-regioisomers, diols, and diethers were determined by the area normalization method under the same chromatographic conditions and the solvent peaks were subtracted; the contents of chlorohydrin intermediates and epoxy intermediates, as well as residual solvents, were determined by gas chromatography flame ionization detection, using the area normalization method and the internal standard method, respectively; trace moisture was determined by coulometric Karl Fischer titration; organic phase chloride ions were determined by ion chromatography, with the sample first extracted with deionized water at a quantitative ratio by shaking and then the aqueous phase was injected; the amount of 1,2,4-triazole sodium was determined by non-aqueous potentiometric titration; the melting point was determined by capillary method; and the median particle size of the seed crystals was determined by laser diffraction particle size analysis. The analysis and determination of each data point were performed in triplicate, and the arithmetic mean was taken. The relative standard deviation of the analysis and determination did not exceed 1.0%. In Experiment 3, three additional batches of experiments were conducted in parallel for each content level. Figure 3 The error bars represent the standard deviation of the measured values ​​of the above three batches. Their relative standard deviations range from 2.5% to 4.7%, which is higher than the relative standard deviation of a single analysis. The difference comes from the fluctuations in feeding and operation between batches.

[0128] Experimental Example 1: This experimental example uses the methods of Examples 1 to 5 and Comparative Examples 1-1 to 1-7, and uses the methods of Comparative Examples 2 to 6 as controls for missing steps and parameter out-of-range conditions. First, under the S3 mixture system and initial concentration conditions of S41 in Example 1, the reaction was carried out at constant temperatures of 328.15 K, 338.15 K, 348.15 K, 358.15 K, and 368.15 K, respectively. These five temperature points cover all possible reaction solution temperatures within the parameter range defined by this invention for S4. Samples were taken every 15 minutes according to S42 for measurement. The rate constants at each temperature were obtained by linearly regressing the integral of the second-order reaction with time to determine the amount of N4-regioisomers produced. Then, the apparent activation energy and apparent pre-exponential factor were obtained by linearly regressing the natural logarithm of the rate constant with the reciprocal of the temperature. The rate constants of the two competing reactions in the ring-opening step at five calibration temperatures are shown in Table 5.

[0129] Table 5. Rate constant determination results for the two competing reactions in the ring-opening step:

[0130]

[0131] Table 5 The rate constant for the reaction that produces cycloprozil. The rate constant for the reaction that produces the N4-regioisomer is given by the reciprocal of temperature, which is the reciprocal of the thermodynamic temperature. and A linear regression of the natural logarithm of the gas on the reciprocal of temperature was performed. The coefficients of determination of both regression lines were not less than 0.999. The result was obtained by taking the negative sign of the product of the slope of the line and the molar gas constant. 62.0 kJ / mol It is 101.0 kJ / mol, obtained from the natural exponent of the intercept. 1.00×10 9 L / (mol·h), 1.72×10 12 L / (mol·h); where and These represent the apparent activation energy and apparent pre-exponential factor for the reaction that produces cyproconazole, respectively. and This is the value used in the temperature control system of S43. Subsequently, with batch time as the x-axis and the mass fraction of N4-region isomers in the crude product liquid as the y-axis, the measured points of the constant temperature series and the variable temperature series were listed separately, and comparative examples 2 to 6 were compared with Example 1 in terms of impurities and time.

[0132] Experimental results are as follows Figure 1 As shown. Figure 1In (a), the horizontal axis represents the batch time, the vertical axis represents the mass fraction of N4- regioisomers in the crude product liquid, the hollow squares represent the isothermal measurement points of Comparative Examples 1-1 to 1-7 with the comparative example number and isothermal temperature marked next to each point, the gray solid line represents the isothermal process curve recalculated according to the kinetic parameters in Table 5, the solid dots represent the variable temperature measurement points of Examples 1 to 5 with the example number marked next to each point, the yellow inverted triangle represents the control level obtained by interpolation of the isothermal measurement points at the same batch time as each example, and the dotted line connects the paired example points with the control level. Figure 1 In (b), the horizontal axis represents Comparative Examples 2 to 6 and Example 1, respectively. The left vertical axis shows the mass fraction of N4-regioisomers in the crude product solution in blue bars and the mass fraction of diethers in the crude product solution in orange bars. The right vertical axis shows the batch time in green broken lines.

[0133] from Figure 1 In (a), two non-overlapping curves can be read. The seven points of the isothermal series fall on the same monotonically decreasing curve. The gray recalculated curve matches the seven measured points, indicating that the batch time and impurity level are simultaneously determined by the same temperature parameter under isothermal operation. Shortening the time will inevitably increase impurities, and suppressing impurities will inevitably prolong the time. The five points of Examples 1 to 5 are all located to the lower left of the above curve. At 5.1h, Example 1 is 0.25% while the control level is 0.37%, a reduction of 32%; at 5.8h, Example 2 is 0.28% while the control level is 0.34%, a reduction of 18%; at 3.9h, Example 3 is 0.35% while the control level is 0.44%, a reduction of 20%; at 4.4h, Example 4 is 0.29% while the control level is 0.40%, a reduction of 28%; at 4.1h, Example 5 is 0.33% while the control level is 0.42%, a reduction of 21%. Reading the graph from a different angle, the impurity level of Example 1 is comparable to that of Comparative Examples 1-3, which were kept at an isothermal temperature of 348.15 K, while the batch time was reduced from 9.9 h to 5.1 h. This indicates that the present invention does not simply change the operating point on the isothermal curve, but rather shifts the entire trade-off curve to the lower left. This shift is related to the kinetic facts upon which S43 is based. The regression results in Table 5 show that the apparent activation energy for the formation of the N4-regioisomer is 101.0 kJ / mol, higher than the 62.0 kJ / mol for the formation of cyclopropazole. Therefore, the ratio of the rate constants of the two competing reactions increases monotonically with increasing temperature. The isothermal operation places all conversions under the same selectivity, while the temperature control method of S43 provides a lower temperature in the early stage when the concentration product is large and the main reaction rate is already sufficient. This allows the early conversion, which accounts for the majority of the total conversion, to be completed under high selectivity. The temperature is then raised again in the later stage when the concentration decreases to finish the process. At this point, there is very little material remaining in the reactor, and the selectivity loss caused by the temperature increase is limited. Thus, both time and impurities are improved simultaneously.

[0134] Figure 1(b) explains that the above improvements are based on S2 and S3. The N4-regioisomer in Comparative Example 2, which eliminates S2 and S3, is 1.62%, while in Comparative Example 3, which only eliminates S23 and S24, it is 0.98%. The comparative example with a value of 0 was 0.79% in case 4. The comparative example 5, which falls below the lower limit, is 0.48%. Comparative Example 6, which achieved the upper limit, had a yield of 0.31% but a batch time extended to 8.7 hours. Example 1, which implemented S2, S3, and S4 simultaneously, had a yield of 0.25% but a batch time of 5.1 hours. The orange bars show the diether at... The proportion of 0% in the comparative group was as high as 0.21%, with... Increase and decrease and in After reaching a level not lower than 3.5%, it fell below 0.05%. When chloride ions and water were not removed, the epoxy intermediate was re-ring-opened by chloride ions, returning to the chlorohydrin intermediate. The direct substitution channel with poor regioselectivity was thus reactivated, and the single-channel model based on S43 no longer matched the actual process. This explains why the impurity level in Comparative Example 3 was significantly higher than that in Example 1. When free 1,2,4-triazole was insufficient, the N4 position of the triazole anion lacked hydrogen bond shielding, and the alcohol oxy anion generated by ring opening also lacked a proton donor and continued to attack the second molecule of epoxy intermediate. This explains why the N4-regioisomer and diether were both higher in Comparative Example 4. When free 1,2,4-triazole was in excess, the triazole anion was diluted, and the main reaction rate decreased. This explains why the batch time in Comparative Example 6 was longer. The upper and lower limits are thus based on their respective grounds.

[0135] Experimental Example 2: This experimental example uses the methods of Example 1 and Comparative Examples 1-6. In Example 1, samples were taken and measured every 15 minutes according to S42. Calculated according to S43 Press S44 to retrieve The smaller of 368K as The measured temperature of the reaction solution was recorded, and the cumulative mass fraction of the N4- regioisomer was determined simultaneously; samples from Comparative Examples 1-6 were taken at the same intervals at a constant temperature of 363.15 K for determination. The cumulative mass fraction of N4-regioisomers. Batch times for both are calculated from the completion of epoxy feed. The point-by-point measured data for the target temperature of the reaction solution, the concentration of the epoxy intermediate, and the cumulative amount of N4-regioisomers in Example 1 are shown in Table 6.

[0136] Table 6. Measured data of temperature variation trajectory and accumulated impurities in Example 1:

[0137]

[0138] Table 6 shows the batch time 3.20h and subsequent times. The calculated values ​​are all no less than 368K, after being clipped using S44. Take 368K; the last row of the table, 5.08h, is the reaction endpoint time. Table 1 and... Figure 1 The time is calculated as 5.1 hours.

[0139] Experimental results are as follows Figure 2 As shown. Figure 2 In (a), the horizontal axis represents batch time, the vertical axis represents temperature, and the red solid line represents Example 1 after being limited by S44. The red hollow circle represents the measured temperature of the reaction solution in Example 1, and the red dashed line represents the temperature before amplitude limiting. The calculated values ​​are shown below. The black dotted line represents the 368K limiting line, and the blue solid line represents the constant temperature conditions of Comparative Examples 1-6. The two curves are labeled with Example 1 and Comparative Examples 1-6, respectively. Figure 2 In (b), the horizontal axis represents the batch time, the left vertical axis represents the current concentration of the epoxy intermediate, the red solid line and the red hollow circle represent the concentration model curve and the measured point of Example 1, respectively, and the blue dashed line and the blue hollow square represent the concentration model curve and the measured point of Comparative Examples 1-6, respectively; the right vertical axis represents the cumulative mass fraction of N4-regional isomers, the green solid circle line represents Example 1, and the purple hollow triangle line represents Comparative Examples 1-6. Each curve is labeled with its corresponding number.

[0140] from Figure 2 As can be seen in (a), the target temperature of Example 1 monotonically increases from 335.3 K, intersects the limiting line at 3.20 h, and then... The calculated value continued to rise to 403.7K, while Maintain 368K until the endpoint; the deviation between the measured temperature of the reaction solution and the set value should not exceed 0.4K; the temperature of Comparative Examples 1-6 remains a horizontal line throughout. Figure 2 As can be seen in (b), the cumulative curve of Example 1 is a straight line from 0 to 3.20h, with a constant increment of 0.033 percentage points every 0.5h, and a slope of 0.067% / h. Take 8.0 × 10 -4 mol / (L·h) divided by The quotient of 1.20 mol / L was consistent; after 3.20 h, the increments every 0.5 h decreased successively to 0.029, 0.014, 0.006, and 0.003 percentage points. The cumulative curves of Comparative Examples 1-6 reached 0.264% within the first 0.5 h, accounting for 63% of their final value of 0.418%, and then rapidly leveled off.

[0141] The two cumulative curves show significant differences in shape because the S43 temperature-controlled system treats temperature as a quantity regulated by concentration feedback. The temperature-controlled system assumes that the instantaneous formation rate of the N4-regioisomer is constant. The temperature is obtained by inverse solution, so as long as Without being limited, the instantaneous generation rate is equal to The cumulative amount increases linearly with time; after the limit is applied, the temperature of the reaction solution is lower than the temperature required by the temperature control system, the rate constant is correspondingly smaller, and the instantaneous generation rate is lower than... Therefore, the generation rate throughout the entire process is no higher than Under isothermal operation, the concentration product is largest in the initial stage of the reaction, resulting in the highest reaction rate. At the selectivity corresponding to 363.15 K, most impurities are generated in a short time, hence the steep initial shape. S44 controls the temperature of the reaction solution rather than the jacket temperature, ensuring that the temperature in the temperature control program corresponds to the actual temperature of the reaction medium. The deviation between the measured temperature and the set value is no more than 0.4 K, thus guaranteeing the accuracy of the control.

[0142] Experimental Example 3: This experimental example uses the methods of Comparative Examples 9-1 to 9-9, Example 1, and Comparative Example 3. In the chloride ion spiked series, the water content was fixed at 210 mg / kg, and in the water spiked series, the chloride ion content was fixed at 32 mg / kg. Each point was repeated in triplicate. The index was the mass fraction of N4- regioisomer and diol in the crude product solution.

[0143] Experimental results are as follows Figure 3 As shown. Figure 3 In (a), the horizontal axis represents the chloride ion content of the epoxy feed solution, with a logarithmic scale. The vertical axis represents the mass fraction of N4- regioisomers in the crude product solution. The blue solid circles with error bars represent comparative examples 9-1 to 9-5, with the comparative example number marked next to each point. The error bars represent the standard deviation of 3 batches. The red pentagram represents the baseline point of the purified feed solution in Example 1. The green solid triangle represents comparative example 3, where chloride ions and water were not removed simultaneously. The black vertical dashed line represents the limit of 50 mg / kg. Figure 3 In (b), the horizontal axis represents the water content of the epoxy feed solution and is logarithmically scaled, the vertical axis represents the mass fraction of diol in the crude product solution, the purple solid squares with error bars represent comparative examples 9-6 to 9-9 and each point is marked with a comparative example number, the red pentagram represents the baseline point of the feed solution after dehydration in Example 1, the green solid triangle represents comparative example 3, and the black vertical dashed line represents the limit of 300 mg / kg.

[0144] from Figure 3 As can be seen in (a), the N4-regioisomer is highly sensitive to chloride ions. The chloride ion concentration increased from 0.25% to 0.41% when the concentration increased from 32 mg / kg in Example 1 to 150 mg / kg in Comparative Example 9-2, and reached 0.96% when the concentration increased to 8600 mg / kg in Comparative Example 9-5. The limit was 0.28% in Comparative Example 9-1, which translates to 0.10% of the original drug, still within the 0.15% limit. Figure 3As shown in Figure (b), the diol content mainly changes with water content. When water content increases from 210 mg / kg in Example 1 to 1000 mg / kg in Comparative Examples 9-7, the diol content increases from 0.12% to 0.24%. The limit is 0.15% in Comparative Example 9-6, which is equivalent to 0.04% of the original drug, also within the limit. Meanwhile, the effect of increased water content on the N4-regioisomer is relatively limited; in Comparative Example 9-9, it is only 0.38% at 9200 mg / kg. The green triangles in the two figures corroborate each other. In Comparative Example 3, when chloride ions and water are simultaneously uncontrolled, the N4-regioisomer content is 0.98% and the diol content is 0.58%, consistent with the readings of the two spiking curves at the corresponding content levels.

[0145] Each of the two indicators dominates a class of impurities, consistent with the chemical nature of the two side reaction pathways. The stoichiometric release of chloride ions from the ring closure of S22 leads to nucleophilic ring-opening of the methylene carbon in the epoxy intermediate, reconverting it back into a chlorohydrin intermediate. This ensures that the poorly regioselective direct substitution pathway is continuously replenished in S4. Therefore, washing chloride ions to below 50 mg / kg in S23 is a prerequisite for suppressing the N4-regioisomer. Water, on the other hand, hydrolyzes and ring-opens the epoxy intermediate to generate a diol and partially hydrolyzes 1,2,4-triazole sodium, thus disturbing the reaction. Therefore, S24's removal of water from the azeotropic solution to below 300 mg / kg is a prerequisite for reducing the diol. The two limits are respectively set at positions where the original drug value falls exactly within the limit after conversion between the two types of impurities; thus, the endpoints are experimentally based, and the two indicators are not interchangeable.

[0146] Experimental Example 4: This experimental example uses the methods of Examples 6, 7, Comparative Examples 7 and 8, and references the discharge ratio of Comparative Example 2. The indicators are the mass fraction of N4-regioisomers in the crystallization mother liquor, the mass fraction of N4-regioisomers in the technical grade drug, and the values ​​calculated according to S55. With S56 The total yield of qualified products, and the correlation between the mass fraction of N4-regioisomers in the technical grade and the mass fraction of the aforementioned isomers in the mother liquor were obtained by spiking under fixed crystallization conditions. When the mass fraction of the aforementioned isomers in the mother liquor was 0.10%, 0.40%, 0.80%, 1.20%, 1.60%, 2.00%, and 2.40%, the mass fraction of the aforementioned isomers in the technical grade was 0.09%, 0.10%, 0.11%, 0.14%, 0.20%, 0.29%, and 0.42%, respectively. Based on the internal control limit of 0.15% for the technical grade, the upper limit of the mother liquor was approximately 1.27%. After leaving a margin... Take 1.20%. The total yield of qualified products is calculated as the product of the chemical yields of S1 to S4 and the yield of the crystallization process. The yield of the crystallization process is determined by the mass fraction of cyproconazole in the mother liquor and the discharge ratio. The key node data of 30 consecutive batches of Example 6, Comparative Examples 7 and 8 under three mother liquor discharge ratios are compared in Table 7.

[0147] Table 7. Key node data for 30 consecutive batches under three discharge ratios:

[0148]

[0149] Experimental results are as follows Figure 4 As shown. Figure 4 In (a), the horizontal axis represents the batch number, the vertical axis represents the mass fraction of N4- regioisomers in the crystallization mother liquor, the red solid circles represent Example 6, the blue hollow squares represent Comparative Example 7, the green hollow triangles represent Comparative Example 8, and the black horizontal dashed line represents... 1.20%. Figure 4 In (b), the horizontal axis represents the batch number, and the vertical axis represents the mass fraction of N4-regioisomers in the technical grade drug. The colors and markings of the three curves are consistent with... Figure 4 (a) is consistent with the black horizontal dashed line, which represents the internal control limit of 0.15% for the original drug. Figure 4 In the middle (c), the horizontal axis represents the batch number, and the left vertical axis, represented by red bars, shows the batches 1 to 12 of Example 6 and batches 13 to 14 of Example 7 executed according to S56. The right vertical axis is represented by a blue diamond-shaped line, showing the values ​​calculated according to S55 for the same batch. The black horizontal dashed lines indicate the basic discharge ratio of 0.10 and the upper limit of 0.20, respectively.

[0150] from Figure 4 As can be seen in (a), the impurities in the mother liquor of Example 6 monotonically converged to 0.958% within 30 batches, remaining consistently below 1.20%; Comparative Example 7 exceeded 1.20% in the 18th batch and continued to rise to 1.571%; Comparative Example 8 quickly converged to 0.333%. Figure 4 As shown in Table (b), the N4-regioisomer of the technical grade drug in Example 6 remained stable between 0.090% and 0.122%, while in Comparative Example 7, the content exceeded 0.15% from the 18th batch onwards, rising to 0.196%, and in Comparative Example 8, it remained stable at 0.098%. Referring to Table 7, Comparative Example 8 had the lowest impurity level, but the loss of ciprofloxacin per batch of mother liquor increased from 0.99 kg to 2.97 kg, and the overall steady-state yield of qualified product decreased from 91.8% in Example 6 to 89.2%. Although the yield of Comparative Example 7 was 92.5%, the product was substandard. Figure 4 As can be seen in (c), batches 1 to 12 All values ​​were negative, and the actual discharge ratio was taken as the baseline value of 0.10; the 13th batch was discharged because the crude impurity load increased to 1.40%. The concentration jumped to 0.469 and exceeded 0.20, at which point the columnar growth stopped, and the reuse of the mother liquor was discontinued. The 14th batch was crystallized with fresh solvent. The value dropped to negative, and the discharge ratio returned to 0.10.

[0151] The different trends of the three curves can be seen from the batch-by-batch balance of the N4-regioisomers in the mother liquor. The mass fraction of the above isomers in each batch of mother liquor is denoted as... Then, the following formula applies between adjacent batches.

[0152] ;

[0153] In the formula, For the first Mass fraction of N4- regioisomers in batch crystallization mother liquor; For the first The proportion of mother liquor discharged from the batch, dimensionless; For the first Mass fraction of N4- regioisomers in batch crystallization mother liquor; The mass of N4-regioisomers introduced into the reactor residue of this batch of crude product, in kg; Let be the mass of each batch of mother liquor for crystallization, in kg. The steady-state solution of the above equation is: Divided by the discharge ratio and The product of the steady-state solution and Compared to the lower limit formula for the discharge ratio of S55, when the discharge ratio is too low, the steady-state solution exceeds... Impurities in the mother liquor were carried away by the crystals, making the original drug unqualified. Therefore, Comparative Example 7 became ineffective after the 18th batch. When the discharge ratio was too high, although the steady-state solution was low, the cyproconazole dissolved in the mother liquor was lost with the discharge stream, so the yield of Comparative Example 8 decreased. S56 uses a basic discharge ratio of 0.10 as the lower limit and 0.20 as the upper limit to avoid the costs at both ends.

[0154] In steady-state solutions The mass fraction of N4- regioisomers in the crude product residue is determined by the degree of chloride and water removal from S23 and S24, and the mass fraction of S31. The values ​​and the temperature trajectories of S43 and S44 are jointly determined. In Comparative Example 2, the crude impurity was 1.62%, requiring the discharge ratio to be increased to 0.54 under the same rules, resulting in a steady-state yield of only 79.9% for qualified products. In Example 6, the crude impurity was 0.25%, requiring only 0.10, resulting in a steady-state yield of 91.8% for qualified products, a difference of 11.9 percentage points. The yield space freed up by reducing impurities in the preceding steps is converted into the actual yield by S55 and S56. S55 uses the measured mass fraction of impurities in the mother liquor as feedback, ensuring clear action even when abnormalities occur in the preceding steps. In the 13th batch of Example 7, the mother liquor impurities were pushed to 1.206% while the active ingredient remained at 0.141%, and then the process was exited according to S56, returning to normal in the 14th batch.

[0155] Although embodiments of this application have been shown and described above, it is understood that the above embodiments are exemplary and should not be construed as limiting this application. Those skilled in the art can make changes, modifications, substitutions and variations to the above embodiments within the scope of this application.

Claims

1. A process for the preparation of high purity cyprosulfamide technical, characterized in that, Includes the following steps: S1 addition: The Grignard reagent obtained by reacting (1-chloroethyl)cyclopropane with magnesium in tetrahydrofuran is added to 2-chloro-1-(4-chlorophenyl)ethyl-1-one. After quenching, separation, and solvent removal, the solution is dissolved in toluene to obtain a toluene solution containing 1-chloro-2-(4-chlorophenyl)-3-cyclopropylbut-2-ol. S2 purification: Add sodium hydroxide aqueous solution and tetrabutylammonium bromide to the solution obtained from S1, and close the ring at 28-32℃ until the chlorohydrin intermediate is no more than 0.30%; separate the liquid and wash with water until the chloride ion concentration is no more than 50 mg / kg, and dehydrate azeotropically until the water concentration is no more than 300 mg / kg to obtain a toluene solution containing 2-(4-chlorophenyl)-2-(1-cyclopropylethyl)ethylene oxide; S3 solution preparation: 1,2,4-triazole and sodium methoxide methanol solution are salted in toluene, methanol is distilled off under reduced pressure until water content is no more than 400 mg / kg, and n-butanol is added to obtain a mixed system with a molar ratio of free 1,2,4-triazole to sodium 1,2,4-triazole of 3.5 to 5.

5. S4 ring opening: Add the solution obtained from S2 to the mixed system obtained from S3, periodically measure the current concentration of epoxy intermediate and 1,2,4-triazole in the reaction solution, calculate according to the calibrated kinetic model and track the target temperature of the reaction solution with cascade control, so that the temperature of the reaction solution rises from 328-353K to 368K, so that the N4-regioisomer formation rate calculated by the model does not exceed the set upper limit; The solution was kept at 368K until the current concentration of the epoxy intermediate was no greater than 0.01 mol / L, and then a crude solution was obtained. S5 Crystallization: After desolventizing the crude product solution obtained from S4, dissolve it in isopropanol and recycled crystallization mother liquor. Add deionized water according to the total amount of isopropanol and water in the combined system, add seed crystals, and crystallize at a controlled temperature. Filter, rinse, and dry to obtain cyproconazole technical grade. The crystallization mother liquor is discharged at a discharge ratio of 0.10-0.20, and the remainder is returned to the next batch of S5 for reuse. The total dry basis content of the obtained cyproconazole technical grade is not less than 99.0%, and the N4-regioisomer content is not greater than 0.15%.

2. The manufacturing process according to claim 1, characterized in that, S1 includes: S11 In a nitrogen atmosphere, magnesium shavings and anhydrous tetrahydrofuran are added to the reactor, and iodine is added as an initiator. The temperature is raised to 40-45°C. The water content of the anhydrous tetrahydrofuran is not greater than 100 mg / kg, and the water content of (1-chloroethyl)cyclopropane and 2-chloro-1-(4-chlorophenyl)ethyl-1-one is not greater than 200 mg / kg. S12 is added dropwise with an anhydrous tetrahydrofuran solution of (1-chloroethyl)cyclopropane under nitrogen protection throughout the process, at a controlled temperature of 40–45°C for 1.2–1.8 h. After the addition is complete, the solution is kept at 40–45°C for 0.8–1.2 h to obtain Grignard reagent solution. The molar ratio of (1-chloroethyl)cyclopropane to 2-chloro-1-(4-chlorophenyl)ethyl-1-one is 1.15–1.25, and the molar ratio of magnesium to (1-chloroethyl)cyclopropane is 1.05–1.

15. S13. The Grignard reagent solution is cooled to -5 to 0°C, and an anhydrous tetrahydrofuran solution of 2-chloro-1-(4-chlorophenyl)ethyl-1-one is added dropwise under nitrogen protection throughout the process. The temperature is controlled at -5 to 0°C, and the dropwise addition time is 1.2 to 1.8 h. After the dropwise addition is completed, the temperature is kept at 0 to 5°C for 0.8 to 1.2 h to ensure that the addition reaction is completed in the range of -5 to 5°C. S14 was quenched by adding an 8%–12% ammonium chloride aqueous solution, with the temperature controlled not higher than 15°C. After standing and separating the liquids, the aqueous phase was extracted once with toluene. The organic phases were combined and the tetrahydrofuran was removed under reduced pressure at 38–42°C and 9–13 kPa. Toluene was added to dissolve the solution, yielding a toluene solution containing the chlorohydrin intermediate.

3. The manufacturing process of claim 1, wherein S2 include: S21 adds a 28%–32% (w / w) sodium hydroxide aqueous solution and tetrabutylammonium bromide to the toluene solution obtained in S1; the molar ratio of sodium hydroxide to the chlorohydrin intermediate is 1.10–1.25, and the molar amount of tetrabutylammonium bromide is 0.3%–0.8% of the molar amount of the chlorohydrin intermediate; S22 was reacted at 28–32°C and a stirring paddle tip linear velocity of 1.5–2.5 m / s for 1.5–2.5 h. Samples were taken every 30 min to determine the residual amount of the chlorohydrin intermediate by gas chromatography area normalization method. The residual amount was calculated based on the organic components after deducting the toluene solvent peak, and the closure endpoint was determined when it was no greater than 0.30%. S23: Allow the mixture to stand for 20-40 minutes to separate into layers, then discard the lower aqueous phase. The water washing described in S2 is performed by washing with a 4%-6% sodium chloride aqueous solution at 20%-25% of the organic phase volume, stirring for 8-12 minutes each time and letting it stand for 15-25 minutes. The washing should be performed at least twice and continued until the chloride ion content of the organic phase, as determined by ion chromatography, is no more than 50 mg / kg. S24 is refluxed at atmospheric pressure, with toluene as an azeotropic agent, and water is separated by a water separator. During the water separation process, the temperature at the top of the column gradually rises from 84°C to 110°C, and the water separation is stopped when the water content determined by Karl Fischer coulometric method is no more than 300 mg / kg, yielding a toluene solution containing an epoxy intermediate; in the toluene solution containing the epoxy intermediate, the normalized content of the epoxy intermediate based on the gas chromatographic area of ​​the organic components after deducting the toluene solvent peak is no less than 98.0%.

4. The preparation process according to claim 1, characterized in that, S3 include: S31. 1,2,4-Triazole is dispersed in toluene, and a 30% (w / w) sodium methoxide methanol solution is added. The mixture is stirred at 45–55°C for 25–40 min to form a salt. The amount of sodium 1,2,4-triazole in the system is determined by non-aqueous potentiometric titration until it is not less than 98% of the amount of sodium methoxide added. The molar ratio of 1,2,4-triazole to the epoxy intermediate is 1.6–1.9, and the molar ratio of sodium methoxide to the epoxy intermediate is 0.28–0.38, and the ratio is calculated according to the following formula. The value is 3.5 to 5.

5. ; In the formula, The molar ratio of free 1,2,4-triazole to sodium 1,2,4-triazole is dimensionless. The amount of 1,2,4-triazole added; The amount of 1,2,4-triazole sodium determined by the non-aqueous potentiometric titration method described above; S32 distills out methanol and methanol-toluene azeotrope at an absolute pressure of 11-21 kPa and a vessel temperature not exceeding 60°C. Distillation is stopped when the water content determined by Karl Fischer coulometric method is not greater than 400 mg / kg and the methanol mass fraction determined by gas chromatography internal standard method is not greater than 0.50%. S33 is added to n-butanol to make the mass ratio of toluene to n-butanol 3.5:1 to 4.5:1, resulting in a mixed system of 1,2,4-triazole and sodium 1,2,4-triazole.

5. The preparation process according to claim 1, characterized in that, S4 include: S41: The mixture obtained in S3 is heated to 328–338 K, and the toluene solution obtained in S2 is added at a constant rate over 0.8–1.2 h. After the addition is complete, toluene is added again to make the initial concentration of epoxy intermediate in the reaction solution 1.0–1.4 mol / L and the initial total concentration of 1,2,4-triazole 1.8–2.4 mol / L. S42 samples are taken every 10–20 minutes. After quenching with an equal volume of 5% (w / w) glacial acetic acid aqueous solution, the current concentration of the epoxy intermediate in the reaction solution is determined in the same injection using high-performance liquid chromatography with external standard method. With the current total concentration of 1,2,4-triazole The current total concentration is the sum of the concentrations of free 1,2,4-triazole and 1,2,4-triazole sodium. S43 calculates the target temperature of the reaction solution using the following formula. : ; In the formula, This is the calculated value of the target temperature of the reaction solution; The apparent activation energy for the formation of the N4- regioisomer is taken as 101.0 × 10⁻⁶. 3 (J / mol); The molar gas constant; The apparent pre-exponential factor for the N4-regioisomerization reaction; This represents the current concentration (mol / L) of the epoxy intermediate. The current total concentration (mol / L) of 1,2,4-triazole. An upper limit is set for the formation rate of N4-regioisomers; S44 take The smaller of 368K and 368K is used as the target temperature for the reaction solution. A cascade control loop, with the reaction liquid temperature as the main loop and the jacket temperature as the slave loop, tracks the reaction. Repeat S42 to S44 until... Stop when the concentration is no greater than 0.01 mol / L, and cool to 331–335 K to obtain a crude cyclopropazole solution with an N4-regioisomer content of no more than 0.40%; the content is calculated by high performance liquid chromatography area normalization method based on the organic components after deducting the solvent peak.

6. The preparation process according to claim 5, characterized in that, S43 Take 8.0 × 10 -4 ~1.6×10 -3 The two consecutive values ​​given in S44 The difference should not exceed 5K, and the heating rate of the reaction solution should not exceed 0.5K / min. At least 368K Take 368K and keep the temperature constant until the endpoint described in S44; and The mixed system obtained from S3 was used when the mass ratio of toluene to n-butanol was 3.5:1 to 4.5:

1. The values ​​were measured at no fewer than four isothermal points at a concentration of 3.5–5.5, an initial epoxy intermediate concentration of 1.0–1.4 mol / L, and a temperature of 328–368 K. The results were obtained by regressing the amount of N4-region isomers.

7. The preparation process according to claim 1, characterized in that, The crystals in S5 include: S51 removes toluene and n-butanol from the crude product solution obtained in S4 under reduced pressure at 328-338K and 6-16kPa to obtain crude product residue. S52 determines the mass fraction of isopropanol and water in the recycled crystallization mother liquor; adds the recycled crystallization mother liquor and fresh isopropanol to the crude product residue, so that the ratio of the total mass of isopropanol in the combined system to the mass of cyclopropazole in the crude product residue is 1.1:1 to 1.3:1, and heats to 58 to 62°C to dissolve and clarify it; the first batch does not add recycled crystallization mother liquor; S53 determines the total volume of water to be 65%–75% of the total volume of isopropanol in the combined system, and adds deionized water after deducting the water already introduced by the recycled crystallization mother liquor; when the temperature drops to 54–56°C, cyproconazole seed crystals are added, and the temperature is lowered according to the following program: ; In the formula, For the first time since the addition of the seed crystal Crystallization temperature (°C) per hour; The crystallization temperature is set at 54–56 °C. The final crystallization temperature is 8–12 °C. The time from the addition of the seed crystal, with a value ranging from 0 to 1. ≤ (h); The total duration of the cooling program is set to 2.5–3.5 hours. S54 at The mixture was kept at a constant temperature for 0.8–1.2 h, filtered, and the filter cake was washed with a mixture of isopropanol and water at a volume ratio of 1:1 (3–7 °C). It was then vacuum dried at 58–62 °C and an absolute pressure of 6–16 kPa for 3.5–4.5 h to obtain cyproconazole technical grade. The amount of cyproconazole seed crystals added was 0.15%–0.25% of the theoretical yield, and the median volume particle size was [missing information]. The thickness is 40-50 μm. The theoretical yield is calculated based on the amount of 2-chloro-1-(4-chlorophenyl)ethyl-1-one added to S1 and its measured content.

8. The preparation process according to claim 7, characterized in that, The crystallization mother liquor obtained from S54 filtration shall be disposed of in the following manner: S55 Determines the mass of the crystallization mother liquor. and the mass fraction of N4-regioisomers therein And calculate the lower limit of the mother liquor discharge ratio according to the following formula. : ; In the formula, This is the lower limit of the proportion of mother liquor discharged, and it is dimensionless. The upper limit for the mass fraction of N4-regioisomers in the crystallization mother liquor is 0.8% to 1.5%. The mass of the crystallization mother liquor obtained from S54 filtration; The mass of N4-regional isomers introduced into the next batch of crude product residue is calculated as the product of the mass of the current batch of crude product residue and the mass fraction of that isomer therein. The measured mass fraction of N4-regioisomers in the crystallization mother liquor obtained by S54 filtration; S56 takes 0.10 and The larger of the two values ​​will be used as the discharge ratio for this batch of mother liquor. ,Will and The accumulated crystallization mother liquor is discharged, and the remainder is returned to the next batch of S52 for reuse; when When the concentration is greater than 0.20, stop using the mother liquor and recover cyproconazole separately from the mother liquor of this batch of crystallization.