Seven stages of chemical technology from creativity to industrialization (Issue 26/Total 100)--Pilot scale-up design

Thread Content

This post was last edited by xiouxingzhe on 2026-6-11 11:43 The seven stages of chemical technology from creativity to industrialization (Issue 26/100 in total) - Technology R&D: Pilot scale-up design Dear Haiyou friends: Hello everyone! In the last issue, we talked about long-cycle operation. The small test continuous device ran stably, a lot of data was accumulated, and the problem list was also drawn up. Next, it’s time to face the core and most heavily invested link in the third phase—amplified verification. As mentioned before, the “pilot test” I usually refer to is in a broad sense, including pilot tests and industrial tests required under certain circumstances. In this issue, we will first talk about the pilot scale-up design, and the relevant content of the industrial test will be discussed separately later. 1. What is the purpose of the pilot test? This question seems simple, but I found that many people have misunderstandings on this issue. Some business managers see that the pilot plant can produce some products, so they treat it as a small production line. They wish to sell the products produced every month to dilute the cost of the pilot plant. This mentality is the fundamental reason why many pilot projects fail. Once you manage the pilot plant as a production plant, your attention will be entirely on the product - how much material was produced today, what is the pass rate, and whether it can be delivered on time. Are you still willing to take the risk of shutting down production just to test a "borderline condition"? Would you still actively adjust the operating parameters to the edge in order to collect a set of envelope data that deviates from the optimal conditions? You won't. As a result, the pilot plant ran smoothly, and products were released batch after batch, but none of the problems that should be exposed most during the pilot stage—amplification effects, boundary fluctuations, and long-term hidden dangers—were not tested. The income from product sales is far from the engineering knowledge you missed. The real purpose of pilot testing is not to produce products, but to do three things. First, verify the amplification law. How will the laws of mass transfer, heat transfer, and mixing in small-scale devices change on a larger scale? Which parameters can be linearly amplified, and which parameters change non-linearly? These issues can only be speculated in the small-scale test stage and must be tested and verified on the pilot plant. Second, expose engineering problems. Laboratory-level equipment and analytical instruments are used in the small-scale trial stage, and many engineering details are not involved. At the pilot stage, equipment that is closer to industrial specifications, longer running time, and more realistic raw material fluctuations must be used to test the process, and problems that will not be exposed in the pilot stage—corrosion, scaling, clogging, and instrument drift—are forced out. Third, form the data packet. The whole-process material balance data, energy balance data, equipment performance data, operating boundary data, and three waste emission data generated during the pilot operation are the core inputs for the subsequent preparation of the process package. Without this data package, the craft package becomes passive water. 2. How to determine the pilot scale? The choice of pilot scale is to find a balance between two conflicting needs. On the one hand, the scale must be large enough. Only when it reaches a certain level can the amplification effect be fully apparent - the inhomogeneity of the flow field, the temperature gradient, and the mass transfer limitations will become observable. A 1-liter small test device is enlarged to 10 liters, and many amplification effects have not yet appeared. ； When enlarged to 200 liters, the effect becomes more obvious. On the other hand, the scale cannot be too large. Too big, and construction and operating costs rise sharply, and risks multiply. After all, pilot plants are exploratory in nature and are not used to stabilize production, so they cannot be applied to the investment logic of industrial plants. Generally speaking, the pilot scale is between 1/10 and 1/100 of the design capacity of the industrial device. Which point to choose depends on several factors: Type of reaction, risk of amplification, feasibility of equipment manufacturing, and budget constraints. Some processes - such as microchannel reactors - have limited amplification of a single channel, and the pilot scale may fall within a small range, and production expansion can be achieved by increasing the number of channels. For some processes - such as large distillation systems - if the pilot scale is too small and the column internals cannot be scaled down, it may be necessary to increase the scale to a larger range. 3. How to choose amplification criteria The selection of amplification criteria is one of the most troublesome issues in chemical amplification. Where is the headache? Because no amplification criterion can simultaneously maintain all physical similarities. You want to maintain the same volumetric power - the stirring effect is maintained, but the Reynolds number may change and the flow field distribution may be different from the pilot test. You want to keep the geometry similar - the reactor shape is the same as the pilot plant, but the heat transfer area per unit volume drops dramatically and the heat transfer capacity is not enough. You want to maintain equal mass transfer coefficients - the stirring speed may be too high to be industrially feasible, or the energy consumption may be too high to be economical. The choice of amplification criteria is essentially a trade-off: For your specific reaction system, which physical quantity is controlling? The way to judge is not to pat your head. My own approach is to do amplified sensitivity analysis based on the data accumulated during the pilot phase. Mix sensitive – Prioritize equal mixing times or equal volume power. Heat transfer sensitive - priority is given to ensuring equal heat transfer area to volume ratio. Residence time distribution is sensitive - priority is given to ensuring equal Pe numbers or equal average residence times. Mass transfer sensitive - priority is given to ensuring equal volume mass transfer coefficients. The accuracy of this judgment directly determines the correctness of the design direction of the pilot plant. The judgment is correct, and the pilot test data can better support the subsequent industrial device design. If the judgment is biased, the data from the pilot test may not have much reference value for the scale-up design - then the pilot test will be in vain. 4. What is the difference between the PID of the pilot plant and the PID of the industrial plant? The PID of the pilot plant also needs to be drawn, but the focus of the PID of the pilot plant and the PID of the industrial plant are different. Pilot PID pays more attention to flexibility. The operating range of industrial equipment is preset, and the control system is configured around the design conditions. However, the pilot plant needs to test different operating conditions, different control strategies, and different equipment combinations - so more switching valves, more sampling interfaces, and more instrument access points need to be reserved on the pilot PID. Pilot PID pays more attention to data collection. The data generated from the pilot plant is the core input for subsequent process packages. Therefore, the PID needs to be clearly marked with the tag number, measuring range, and accuracy level of each instrument to ensure that the data has a traceable source. You cannot use a local thermometer to replace remote temperature measurement, and you cannot install only a pressure gauge without a transmitter at a key location. However, the material grade of pilot PID can be appropriately relaxed - after all, the operation time is limited, unlike industrial equipment that needs to run for decades. The pipe material can be a more cost-effective choice, as long as the data quality is not affected by corrosion during the operating cycle. 5. Safety analysis of pilot plant One mistake that many people make during the pilot stage is: I feel that the scale of the pilot plant is small, the inventory of hazardous chemicals is small, and the safety issues are not as serious as those of industrial plants. The scale is smaller, but the safety risks faced by the pilot plant have their own characteristics. The operator's mentality is a hidden danger that is easily overlooked - when doing small experiments in the laboratory, they are faced with glass bottles, syringes, and fume hoods. When it comes to the pilot plant, we are faced with hundreds of liters of reactors, high-pressure pipelines, and storage tanks for flammable solvents. The hazards of materials were suppressed in the laboratory because of the small quantities. But in the pilot plant, the same materials became dangerous because of the large quantities. However, the safety awareness of operators is still at the laboratory stage to a large extent. Another thing that is easily overlooked is that the operating conditions of the pilot plant often change - running normal operating conditions today, testing boundary conditions tomorrow, and doing amplified sensitivity experiments the day after tomorrow. Frequent changes mean operating procedures are not as standardized as in industrial units, and the risk of misoperation is higher. Therefore, safety analysis at the pilot stage cannot be simplified. The design of safety facilities such as emergency relief, interlocking protection, combustible gas detection, and emergency sprinklers must have all the necessary ones. If hazardous chemical processes are involved in the first industrialization, reaction safety risk assessment and HAZOP analysis are also required. These are regulatory requirements and cannot be skipped just because the scale is small. 6. The dividing line between pilot testing and industrial testing As mentioned earlier, some projects require industrial testing after pilot testing. How to distinguish between the two? The main purpose of the pilot test is to verify the amplification rules and expose engineering problems. The main purpose of industrial testing is to conduct long-term verification under conditions closer to actual industrial devices, provide a more sufficient basis for process package preparation, and also provide batch samples for market development. Not all projects require industrial trials. If the process itself is relatively mature, there is reference experience with similar devices, the magnification is not large, the risks are controllable, and the data after the pilot test is solid, you can directly enter the process package preparation. However, if it is a brand-new process, extremely high magnification, or a hazardous chemical process with clear requirements of regulations, industrial testing needs to be arranged after pilot testing. Whether or not to conduct industrial trials should be initially determined during the risk assessment during the innovation incubation stage, and further confirmation should be made at the pilot scale-up design stage. This decision will directly affect the project budget and cycle, and cannot be decided temporarily until the pilot test is completed. Next issue preview No. 27: Pilot construction and operation - pilot testing according to industrial standards. After the pilot plan is designed, the next step is construction and operation. How to grasp the construction standards of pilot plant? Where must it be as strict as in industrial installations, and where can it be simplified appropriately? After the pilot plant is started, how to organize the operation and record the data? Expand next issue.