Thread Content
1.1 Introduction Many of the \"surprises\" encountered during the scale-up of a process can be predicted. If more attention is paid to details during pilot testing, simple experiments are conducted, and data is collected, it will be of great help for subsequent scale-ups. Glass flasks used in experiments generally do not suffer from corrosion problems (glass is not resistant to hydrofluoric acid and compounds that may decompose to produce fluorine, as well as hot concentrated alkalis). However, the compatibility of materials and substances in production must be taken into account; this is also a requirement of GMP for equipment selection. If a corrosion test of the material can be considered during the pilot testing (by adding stainless steel or other material coupons to the reaction system), it will save time when selecting equipment in the future. Simply measuring the bulk density of the filter cake is useful for estimating the volume of the product filter cake and selecting the appropriate equipment in future production processes. The filtration speed, the filtration area, and the thickness of the filter cake are all interrelated. 1.2 Typical scaling problems The most common issue in process scaling is the change in reaction selectivity, which affects the yield and purity of the product; this is mainly due to differences in mixing efficiency and production conditions between pilot-scale experiments and full-scale production. If the effect of rotational speed has already been evaluated in pilot tests, it becomes possible to quickly identify the cause when problems arise. The reaction vessels in the pilot plant are equipped with variable frequency drives, which allow for appropriate adjustments to determine the optimal rotational speed. The emergence of new crystal forms during amplification is also common. During magnification, separation issues may also arise with the product; in production, the washing efficiency of the filter cake is not on par with that achieved in pilot tests, and impurities cannot be completely removed. A three-in-one device for filtration, washing, and drying equipped with a stirrer can replace centrifuges under certain process conditions. With this three-in-one device, it is possible to add a solvent for washing and pulping right after filtration, and the washing effect is better than that of a centrifuge. Another reason for the amplification issue is the impact of production operation time; it is necessary to conduct experiments in pilot scale to examine the effects of extended time on the product. In actual production, the extension of distillation time has led to product decomposition and side reactions occurring on multiple occasions. Expanding on the causes of the problem, a lack of understanding of the reaction mechanism is the most common reason, with crystallization and mixing being the other two most frequent causes. As we can see in the following text, although many issues are related to mixing and heat transfer, the fundamental aspect lies in the understanding of chemistry – namely, what side reactions might occur in addition to the main reaction Under what conditions are side reactions promoted? What changes when zooming in? What impact will these changes have on reaction selectivity? In practical production, the heat transfer conditions in reaction vessels cannot be changed at present (local supercooling/superheating can be reduced by controlling the temperature difference between the heating/cooling medium and the system inside the vessel, as well as the rate of heating/cooling), while mixing can be improved by selecting the rotation speed and paddle design. 2.1 Cooperation within Team 1 is necessary for amplification. Only through close cooperation among professionals from different fields can a stable and scalable manufacturing process be achieved. Chemists have a deep understanding of the impact of various variables on product quality. Process engineers have a better understanding of what types of operations are not feasible or unsafe in production. At the same time, as mentioned earlier, there are many tests that chemists can carry out in the early stages of process development. For example, in experiments involving drying and distillation, record physical parameters of the system such as pressure and density. 2 Principles of Process Development and Operation: Regardless of the size of a company, it is necessary to establish some basic rules to ensure the safe transition of pilot-scale processes to kilogram-scale laboratories and pilot plants. Establish clear and strict procedures as well as the documentation required for pilot tests, and resist pressure to ensure strict compliance even when there are time constraints. These materials include the production operating procedures, along with three batches of pilot tests conducted in accordance with these procedures, at least one of which used raw materials of the same specifications as those used in production. Cleaning validation protocol to avoid the possibility of cross-contamination. Process safety analysis materials. Although the above requirements affect “efficiency,” strict implementation has prevented any serious accidents or batches with amplified failures over the years. 3 Equipment register: Maintain operation and maintenance logs for the main equipment in the kilogram-scale laboratory and pilot plant (reactors, filters, dryers, pumps, etc.). This includes batch records, cleaning records, validation records, and other maintenance records. 4 Sample Database: Establish a sample database to collect and organize data for each sample (product, wet filter cake, distillate, process by-products). Including production batch number, collection time, analysis results, etc. These data are of great reference value, and collecting and organizing them can ensure that the data is not lost. For research and regulatory purposes, it is often necessary to re-analyze these samples for confirmation, while also performing calculations to ensure quality consistency. 5 Sample storage: As a sample database is established, it is necessary to set up a dedicated sample room for storing samples, paying attention to keeping them dry, away from light, and at low temperatures. It is important to establish a system that allows for quick access to the required samples when needed. 6 Fixed process: Identify and resolve related issues before trial production. Changing the process at the last moment before amplification is dangerous. It may lead to accidents and safety issues. 7 Process risk assessment (HAZOP analysis): A risk assessment must be conducted before the pilot production of a new process. An evaluation team composed of personnel from different departments should be formed to conduct a thorough review of the overall process safety and preventive measures. There is no 100% safe process, but based on the evaluation results, appropriate measures can be taken to prevent or minimize safety incidents. 8 Determining reaction energy: A lack of awareness or failure to identify the hazards associated with exothermic reactions can be a major cause of serious injuries and accidents. In production, the heat transfer area per unit volume of the reaction vessel is much smaller than that of a lab flask. The heat transfer area of a 500ml flask is approximately 0.02㎡, whereas a 4000L reactor has only a heat transfer area of 10.7㎡. Therefore, safe amplification reactions require calorimetry or similar experiments. We are gradually paying more attention to work in this area. Although no serious accidents have occurred, material shattering during production happens from time to time. During the pilot scale stage, it is necessary to avoid using compounds containing high-energy functional groups (such as those with multiple amine groups, tetrazoles, and hydrazine hydrate). Reactions that may generate free radicals or gases must be given due attention, and these aspects should be clearly described during the process transfer. 9 Establish production operation procedures. The importance of production operation procedures goes without saying; it is crucial to ensure that these procedures remain up-to-date and to minimize any textual errors. While using “copy–paste” during document editing brings convenience, it can also lead to unintentional errors. This is also the importance of conducting repeated experiments by trainees in accordance with the operating procedures. 10 Raw materials: Industrial-grade raw materials were used in the experiments; prior to scale-up, small-scale tests were conducted on all raw materials. In this way, if magnification is not successful, the cause in the raw material can be eliminated directly. In addition to the chemical purity of the raw materials, physical properties also have an impact on the reaction, such as the particle size of solid materials. 11 Seize the opportunity – Preparing for pilot production requires a great deal of labor, time, and money. However, in many cases, only limited data is collected. This is why it is important to make use of every production batch as an opportunity to learn*. A detailed sampling and analysis plan can help complete the mass balance calculations, identify any unexpected by-products, and address other potential amplification issues. Every process stream (including waste) must be weighed and sampled. There is no other opportunity to collect so much data like in pilot testing! All observed phenomena should be recorded, and the isolated intermediates and samples should be kept as backups. Take the opportunity to collect as many enlarged images as possible, conduct a detailed summary and analysis of production, and prepare a report for future reference. Verification is required when transferring production; prior to this, trial production is needed to get familiar with the process and to establish the process procedures and operating protocols. If the trial production goes smoothly, the time required for verification will also be reduced. The success of the pilot production depends on the preliminary pilot research and the depth of investigation into process issues during the pilot-scale stage. 2.2 Things to Avoid in Scaling Up 1. Avoid complexity: In process development and scaling up, it is necessary to keep things as simple as possible; the simpler it is, the fewer opportunities there are for process errors. In practice, the more complex a process is, the harder it is for operators to master, and it is also difficult to describe in detail through standard operating procedures. Simplification is not only for safety reasons, but it can also reduce the production cycle and minimize waste. Avoid reactions involving the use of highly specialized equipment. Or reactions that are highly dangerous and require safety facilities, such as nitration, hydrogenation, etc. The simplification of the process stems from the simplification of the reaction route; generally, the route with the fewest reaction steps is the best one. During the process development stage, it is necessary to consider whether it is possible to avoid the separation of intermediates, combine reactions, and reduce the types and quantities of solvents used. 2 Avoid heating after adding all the raw materials. One of the most dangerous operations in this process is to add all the reactants together and then raise the temperature to initiate the reaction. Similarly, do not add the catalyst after adding the materials at the end. The danger is that once the reaction mixture reaches the reaction temperature and the reaction begins, there is no way to stop it. Some reactions are highly exothermic and will raise their temperature on their own to increasingly higher levels. If the boiling point of the mixture is reached, it will boil and even cause material to spill over. Some raw materials degrade at higher temperatures, and this degradation accelerates on its own; the heat release is even more intense than that of the reaction itself. When the reaction requires emergency cooling, switching and cooling are not as convenient and fast as in pilot tests. Of course, for reactions that have already been understood and confirmed to be safe, it is acceptable to use the raw materials in one batch. However, it should be prohibited for the first amplification process. The most common method for controlling exothermic reactions is to add a reagent drop by drop, with the addition time depending on the heat generated by the reaction and the heat transfer capacity of the reactor. During the addition, it is necessary to avoid the accumulation of the reactants, which could lead to sudden reactions; therefore, the addition should be carried out at an appropriate temperature to ensure that the added reactants can react immediately, such as in a nucleophilic substitution reaction. 3 Avoid heating without stirring. In production, heat transfer inside the reaction vessel relies mainly on stirring. In addition to ensuring safety, proper mixing can reduce the temperature difference inside the tank, resulting in more accurate temperature readings. Generally, the temperature of the reactor wall is higher than that at the center of the system, which leads to localized overheating, causing the product to decompose or coking, and ultimately affecting yield and quality. The stirring also cannot be stopped until the reaction is complete and cooled to a safe temperature. For example, an accident can occur due to the highly exothermic reaction between organic amines and sulfuric acid. According to the process, the amine needs to be slowly added to hot sulfuric acid under vigorous stirring; the reaction takes place in a biphasic system. One day, the operator performed a replacement, and without turning on the mixer, amine was started to be added; the amine settled at the bottom of the reactor and did not react. Later, another employee noticed that the mixer was not running, so they started it; at that moment, all the materials reacted instantly, resulting in an explosion. 4 Do not ignore potential degradation reactions. Do not carry out the reaction at temperatures within 50°C, the known degradation temperature of the reactants, to avoid loss of control over the reaction. In addition to the calorimetric assessment of exothermic reactions, self-accelerating degradation reactions also need to be examined. This requires additional experiments, such as adiabatic reaction calorimetry (ARC); if analysis suggests that the reaction may produce potentially unstable and labile products, corresponding calorimetric tests should be conducted. Some degradation reactions may proceed very slowly and cannot be detected by conventional tests. Even below the onset temperature, the heat release from the reaction still increases at a slow rate; by the time a significant rise in temperature is detected, the decomposition reaction has already taken place. At a domestic active pharmaceutical ingredient factory, during the diazotization reaction, the operators closed the steam valve and left their posts to eat. Due to a leak in the steam valve, the reaction temperature rose, causing the diazonium salt to decompose. Due to the lack of supervision, the rise in temperature was not detected in time. By the time the workers on duty returned and noticed the abnormal increase in temperature, it was already too late to control the situation, resulting in an explosion that destroyed the entire workshop. 5 Avoid adding solids to the reaction mixture. Do not add solids to a reaction mixture that is under reflux or hot. This is a common operation in pilot tests, but difficult to implement in production. Batch addition is used to control the reaction, and it is easy to achieve in pilot tests. However, adding solid materials during production necessarily requires opening the manhole, and the solvent vapor already present in the tank will combine with air to form an explosive mixture. If the material reacts rapidly, it can cause the liquid mixture to spray out of the manhole (a similar accident occurred when sodium hydride was added). One improvement is to consider adding the solid first, followed by the solvent. However, changing the order of feed addition may affect the selectivity of the reaction. Additionally, the solid can be dissolved and added, or even turned into a slurry before being pressed into the reactor. If it is not possible to change the process, engineering considerations must be given to how to add the substance under sealed conditions. Attempts and improvements have been ongoing in this area, such as the use of vacuum feeding, but so far there is no economical, effective, and widely applicable solution, especially for solid raw materials and intermediates that are corrosive, highly toxic, or have poor fluidity. 6 Avoid evaporation to dryness. Using rotary evaporation to concentrate the solution until it dries is a common practice in pilot testing. But in most reactors in the workshop, there is a minimum stirring volume of about 10–20%. When the liquid mixture is concentrated to its final stage, it is inevitable that the material will be heated in the absence of proper stirring. The hazards of heating without stirring have been mentioned earlier. This can lead to safety and quality issues. When evaporation is carried out to replace the solvent, by evaporating the mixture to a certain volume and then adding the second solvent and repeating this process, it is possible to avoid concentrating it to dryness; however, this approach depends on the relative volatility of the two solvents and whether they are azeotropic. A more efficient method is to use \"constant-volume distillation\". The process of concentrating to dryness is common in our production; if the pilot-scale process indicates that concentration to dryness is feasible, it will be carried out in production as well. The process of concentrating until dryness is achieved may seem simple, but it is uncontrollable; there are no standards by which to measure it, and it is only when problems arise in subsequent yield and quality that it becomes apparent that the previous steaming step may not have been thorough enough. There have been cases of broken stirring shafts, reduced yields due to incomplete replacement of the solvent, and poor quenching of the material after concentration as a result of inadequate stirring. Therefore, it is necessary to consider whether there are better processes available during development. 7 Avoid underestimating process time. For those new to process scaling, the biggest surprise might be that all operations take so much time. It is important to conduct stability assessments for all raw materials, intermediates, and products prior to amplification. Avoid reactions that require immediate quenching and separation after the reaction occurs. 8 Avoid neglecting solvent usage issues; in pilot tests, solvents with good solvency and ease of distillation for recovery may be used. But some of them need to be avoided in production. This includes all solvents of that category, those with a flash point below -18°C. Hexane has a flash point of -23°C and poor electrical conductivity; as a result, foreign companies prohibit its use in production and opt instead for heptane. However, due to cost issues, it is still used in some processes. The toxicity of dichloromethane is lower than that of other chlorinated alkane solvents (chloroform, dichloroethane, etc.), but its use should still be avoided as much as possible; in some processes, methyl tert-butyl ether or toluene can be used as alternatives. 9 Avoid neglecting quenching and extraction; many amplification problems stem from the post-treatment process. Therefore, it should receive the same level of attention as the reaction. The upper layer that prevents stratification is the waste liquid. Extraction is often the step that requires the largest volume of solvent; to increase the production capacity per unit volume, the amount of solvent used for extraction should be minimized as much as possible. As the amount of solvent increases, the time required for stratification and discharge also lengthens, and emulsification may intensify. In weak acid/base environments, the prolonged extraction and separation times may lead to the hydrolysis of compounds containing easily hydrolyzable functional groups. When an API intermediate is scaled up to a capacity of 80 tons per year, one of the extraction steps requires 2000 L of solvent to be used twice, resulting in a total of 4000 L of solvent; the transfer of this solvent takes three to four hours. As a common extraction solvent, ethyl acetate is prone to hydrolysis in acidic or alkaline environments to produce acetic acid, thereby increasing the acidity of the system. A more stable isopropyl acetate or butyl acetate can be used as a substitute; solvents with a lower water saturation content are easier to recover. 10 Safety Safety is the most important. What is emphasized here is to avoid the risk of using up all the limited raw materials at once in a single reaction. Be prepared for possible failures, especially with new processes. It can be carried out in two batches or smaller batches to avoid project delays due to the exhaustion of all raw materials. At the same time, smaller batch sizes result in a higher heat transfer area, which reduces mixing problems and accordingly lowers the scaling factor. 3 Summary: Experience is very important in a timely and successful amplification process. At the same time, there are also many excellent references worth studying and adopting. Although it is not possible to predict all unforeseen issues during the first scale-up, the precautions listed above will guide further considerations for process development and scaling up; meanwhile, being aware of the importance of collaboration will increase the chances of success.