Method and apparatus for manufacturing a liquid for semiconductor manufacturing

Abstract

The present application provides a kind of semiconductor manufacturing liquid manufacturing method and manufacturing device, when multiple filling is carried out, even in the synchronous control of the valve of multiple branch flow path, the opening and closing timing deviation occurs, can be filled by inhibiting the generation of bubble, it is helpful to improve the throughput of product container filling liquid. CPU executes the speed control of controlling the filling speed of first branch flow path and second branch flow path by controlling the opening of first valve and second valve. The speed control is as follows: when the filling speed of one of first branch flow path and second branch flow path is increased, it is gradually increased;When the filling speed of the other side is slowed down corresponding to the increase of the filling speed of one of first branch flow path and second branch flow path, it is gradually decreased.

Description

Technical Field

[0001] This invention belongs to the field of semiconductor manufacturing solutions technology, specifically relating to a method and apparatus for manufacturing a semiconductor manufacturing solution. Background Technology

[0002] A method for manufacturing a semiconductor manufacturing solution (hereinafter referred to as the solution) for use in semiconductor manufacturing processes is known. In the semiconductor manufacturing process, circuit patterns, including metal wiring and insulating films, are formed on the surface of a semiconductor wafer by repeatedly performing processes such as film deposition, exposure, etching, and planarization. Planarization is performed, for example, by CMP (Chemical Mechanical Polishing). The solution is, for example, a cleaning solution used to remove residues remaining on the surface of the semiconductor wafer after CMP (see Japanese Patent No. 7530968).

[0003] In recent years, the miniaturization of semiconductor devices has continued to advance, making quality management in semiconductor manufacturing processes extremely stringent. Since the quality of the chemical solutions also affects the quality of semiconductor devices, the quality management of these solutions is therefore crucial.

[0004] In the final stage of the pharmaceutical solution manufacturing process, the manufactured pharmaceutical solution is filled into the product container. During filling, filtration is performed to remove particulate matter, which serves as impurities, from the pharmaceutical solution. The filtration is performed by introducing the pharmaceutical solution taken from the tank (reservoir) into a circulation path, circulating it between the tank and the circulation path, and filtering it using a filter located on the circulation path. A measuring device for measuring the number of impurity particles is installed within the circulation path. For example, when the result of the particle count by the measuring device meets preset quality conditions, the pharmaceutical solution is supplied to the product container via the filling path.

[0005] To increase the throughput of filling liquid medicine into product containers, the inventors considered alternating filling of multiple product containers (hereinafter referred to as multiple filling). In this multiple filling, the filling flow path, for example, uses a multiple filling flow path having a main flow path connected to a circulation flow path and multiple branch flow paths branching from the main flow path. According to the multiple filling flow path, it is possible to open the valve of one branch flow path to fill while closing the valves of other branch flow paths, thereby replacing the filled product container with an empty product container, and thus expecting to increase the throughput of filling liquid medicine into the product container.

[0006] However, when changing product containers, if the timing of closing the valve on the branch flow path connected to the product container where filling has ended is inconsistent with the timing of opening the valve on the branch flow path connected to the product container where filling has begun, the flow rate of the liquid through the main flow path per unit time (hereinafter referred to as the flow velocity of the main flow path) will fluctuate. This fluctuation in the flow velocity of the main flow path will also affect the upstream circulation path, causing pressure fluctuations within the circulation path. Depending on the degree of this pressure fluctuation, it may sometimes cause a temporary generation of a large number of bubbles in the liquid flowing through the circulation path.

[0007] The bubbles that are generated will disappear over time. However, the measuring instrument has difficulty distinguishing between bubbles and impurity particles, and sometimes it will miscount bubbles as impurity particles. If the miscounting of bubbles leads to a misjudgment of insufficient filtration, the filling process will be interrupted or the filtration time will be too long, resulting in a decrease in the throughput of the medicine being filled into the product container.

[0008] Therefore, as a countermeasure to suppress bubble formation, the simultaneous opening and closing of valves in multiple branch flow paths can be considered. For example, when there are two branch flow paths, the timing of switching the valve in one branch flow path from fully closed to a preset opening (hereinafter referred to as the set opening) is synchronized with the timing of switching the valve in the other branch flow path from the set opening to fully closed. Moreover, when the valves in multiple branch flow paths are controlled synchronously in this way, if the opening and closing timings of the valves can be made completely consistent, the flow rate in the main flow path will remain constant and will not be affected by the opening and closing of the valves in each branch flow path. As a result, pressure fluctuations affecting the circulating flow path are suppressed, thereby suppressing bubble formation.

[0009] However, in the aforementioned synchronous control, considering the action delays of the valve drive system (including the delay of the drive signal and the mechanical response delay of the valve drive system to the drive signal) in each branch flow path, achieving perfect consistency in the opening and closing timing of valves in multiple branch flow paths is extremely difficult. In reality, there will be slight time deviations in the opening and closing timings of valves in multiple branch flow paths. Although these time deviations are instantaneous, they may cause sharp pressure fluctuations affecting the circulating flow path. Even if the pressure fluctuations are instantaneous, the disappearance of bubbles takes time, so there is a concern that the throughput may decrease due to miscounting. Therefore, there has always been the expectation that, even with the assumption of deviations in valve opening and closing timings, it is possible to suppress the generation of bubbles caused by pressure fluctuations, thereby increasing the throughput of filling the product container with the liquid medicine.

[0010] Therefore, those skilled in the art need to improve existing methods and apparatus to overcome the aforementioned deficiencies. Summary of the Invention

[0011] This invention provides a method and apparatus for manufacturing a chemical solution for semiconductor manufacturing. Even if the timing of opening and closing of valves in multiple branch flow paths deviates during the synchronous control of multiple filling processes, the generation of bubbles can be suppressed, which helps to improve the throughput of the chemical solution filling the product container.

[0012] To achieve the above objectives, in a first aspect, this application provides a method for manufacturing a semiconductor manufacturing solution, comprising: In the circulating filtration process, the semiconductor manufacturing solution is taken out of the liquid tank, filtered in a circulating flow path connected to the liquid tank, and then returned to the liquid tank. The measurement process involves measuring values ​​related to the concentration of particles contained in the drug solution within the circulating flow path; and In the filling process, when the measured value in the measurement process meets the preset quality conditions, the medicine solution is supplied from the circulation flow path to the filling flow path, and the medicine solution is filled into the product container. The filling flow path uses a multi-stage filling flow path, which includes a main flow path connected to the circulation flow path, multiple branch flow paths branching from the main flow path, and multiple valves for adjusting the flow rate of the liquid medicine flowing through the multiple branch flow paths respectively. The liquid medicine is alternately filled into multiple product containers through the multiple branch flow paths. In the filling process, the following speed control is performed: When the flow rate of the liquid medicine flowing through the multiple branch flow paths per unit time is set as the filling speed for filling the product container with the liquid medicine, And when the flow rate of the drug solution per unit time through the main flow path is set as the flow velocity of the main flow path, The filling speed of each of the multiple branch flow paths is controlled by controlling the opening degree of the multiple valves, so that the flow rate of the main flow path remains constant. Specifically, when the filling speed of at least one of the plurality of branch flow paths is increased, it is gradually increased; when the filling speed of one or more other branch flow paths is slowed down in accordance with the increase in the filling speed, it is gradually decreased.

[0013] Optionally, during a single filling of the liquid medicine into one of the product containers, the absolute value of the acceleration of the filling speed is constant as the filling speed gradually increases and decreases with the opening and closing of the valve.

[0014] Optionally, during a single filling of the liquid medicine into one of the product containers, the absolute value of the acceleration of the filling speed changes as the filling speed gradually increases and decreases with the opening and closing of the valve.

[0015] Optionally, during the filling period, the intervals that gradually increase the filling speed and the intervals that gradually decrease the filling speed respectively include a first interval and a second interval. In the first interval, the absolute value of the acceleration gradually increases from 0 to the maximum value, and in the second interval, the absolute value of the acceleration gradually decreases from the maximum value to 0.

[0016] Optionally, the first interval is the first half of the interval that gradually increases the filling speed and the interval that gradually decreases the filling speed, and the second interval is the second half of the interval that gradually increases the filling speed and the interval that gradually decreases the filling speed.

[0017] Optionally, during the filling period, the intervals that gradually increase the filling speed and the intervals that gradually decrease the filling speed respectively include a relay interval with constant acceleration between the first interval and the second interval.

[0018] Optionally, the filling flow path further includes a drainage flow path for the drug solution. Before filling the first product container with the liquid medicine, the liquid medicine is discharged from the drain path.

[0019] Alternatively, a drain tank may be used instead of the first product container.

[0020] Optionally, measurements taken immediately after the start of filling the first product container with the liquid medicine are excluded.

[0021] Optionally, the multiple branch paths may be two.

[0022] Optionally, the filling speed varies continuously.

[0023] Optionally, the solution contains a surfactant.

[0024] To achieve the above objectives, secondly, this application provides an apparatus for manufacturing a semiconductor manufacturing solution. It includes a processor for controlling various processes, each process comprising: In the circulating filtration process, the semiconductor manufacturing solution is taken out of the liquid tank, filtered in a circulating flow path connected to the liquid tank, and then returned to the liquid tank. The measurement process involves measuring values ​​related to the concentration of particles contained in the drug solution within the circulating flow path; and In the filling process, when the measured value in the measurement process meets the preset quality conditions, the medicine solution is supplied from the circulation flow path to the filling flow path, and the medicine solution is filled into the product container. The filling flow path uses a multi-stage filling flow path, which includes a main flow path connected to the circulation flow path, multiple branch flow paths branching from the main flow path, and multiple valves for adjusting the flow rate of the liquid medicine flowing through the multiple branch flow paths respectively. The liquid medicine is alternately filled into multiple product containers through the multiple branch flow paths. The processor In the filling process, the following speed control is performed: When the flow rate of the liquid medicine flowing through the multiple branch flow paths per unit time is set as the filling speed for filling the product container with the liquid medicine, And when the flow rate of the drug solution per unit time through the main flow path is set as the flow velocity of the main flow path, The filling speed of each of the multiple branch flow paths is controlled by controlling the opening degree of the multiple valves, so that the flow rate of the main flow path remains constant. Specifically, when the filling speed of at least one of the plurality of branch flow paths is increased, it is gradually increased; when the filling speed of one or more other branch flow paths is slowed down in accordance with the increase in the filling speed, it is gradually decreased.

[0025] The present invention provides a method and apparatus for manufacturing a semiconductor manufacturing solution. Compared with the prior art, its advantages are as follows: by controlling the opening degree of multiple valves to control the filling speed of multiple branch flow paths, the flow rate of the main flow path is kept constant. When the filling speed of at least one of the multiple branch flow paths is increased, it is gradually increased; when the filling speed of one or more other branch flow paths is slowed down in accordance with the increase in filling speed, it is gradually decreased. Thus, during multiple filling, even if the opening and closing timing of the valves in the multiple branch flow paths deviates in the synchronous control, the generation of bubbles can be suppressed, which helps to improve the throughput of filling solution into the product container. Attached Figure Description

[0026] The accompanying drawings, which form part of this application, are used to provide a further understanding of the application and to make other features, objects, and advantages of the application more apparent. The illustrative embodiments and descriptions of this application are used to explain the application and do not constitute an undue limitation of the application. In the drawings: Figure 1 This is a diagram illustrating the configuration of a manufacturing apparatus in the prior art and the present invention.

[0027] Figure 2 This is a diagram showing the detailed configuration of the filling portion in the prior art and the technology of the present invention.

[0028] Figure 3 This is a block diagram illustrating the internal structure of the control unit in the prior art and the present invention.

[0029] Figure 4 This is a diagram illustrating the general outline of multiple filling in the prior art and the present invention, in which the liquid medicine is alternately filled into a product container connected to a first branch flow path and a product container connected to a second branch flow path.

[0030] Figure 5 This is a flowchart illustrating the pharmaceutical preparation sequence in the prior art and the present invention.

[0031] Figure 6 This is a diagram illustrating the drive signal output from the CPU to the valve in the prior art.

[0032] Figure 7 It is a graph showing the time variation of the filling speed of the liquid medicine flowing through the first branch flow path and the second branch flow path, the flow rate of the liquid medicine flowing through the main flow path, and the count result of the number of impurity particles measured by the measuring instrument when the opening and closing timing of the valve is completely consistent in the prior art.

[0033] Figure 8 This diagram illustrates a situation where the drive signal output from the CPU to the valve deviates in the prior art.

[0034] Figure 9 This is a diagram illustrating the time variation of the filling speed of the liquid medicine flowing through the first branch flow path and the second branch flow path, the flow rate of the liquid medicine flowing through the main flow path, and the count result of the number of impurity particles measured by the measuring instrument when the valve opening and closing timing is deviated in the prior art.

[0035] Figure 10 This is a diagram illustrating the process in the prior art from the moment the valve opening and closing timing deviates until the filling process is interrupted.

[0036] Figure 11 This is a diagram illustrating the drive signal output from the CPU to the valve in the technology of the present invention.

[0037] Figure 12 This is a graph illustrating the time variation of the filling speed and acceleration of the liquid as it flows through the branch flow path in the technology of the present invention.

[0038] Figure 13 It is a graph showing the time variation of the filling speed of the liquid medicine flowing through the first branch flow path and the second branch flow path, the flow rate of the liquid medicine flowing through the main flow path, and the count result of the number of impurity particles measured by the measuring instrument when the opening and closing timing of the valve is completely consistent in the technology of the present invention.

[0039] Figure 14 This is a diagram illustrating a situation where the drive signal output from the CPU to the valve deviates in the technology of this invention.

[0040] Figure 15It is a graph showing the time changes of the filling speed of the liquid medicine flowing through the first branch flow path and the second branch flow path, the flow rate of the liquid medicine flowing through the main flow path, and the count results of the number of impurity particles measured by the measuring instrument when the valve opening and closing timing is deviated in the technology of the present invention.

[0041] Figure 16 This is a flowchart illustrating the filling process sequence in the technology of the present invention.

[0042] Figure 17 This is a graph showing the time variation of the filling speed and acceleration of the liquid as it flows through the branch flow path in a modified example 1 of the present invention.

[0043] Figure 18 This is a graph showing the time variation of the filling speed of the liquid medicine flowing through the first branch flow path and the second branch flow path, the flow rate of the liquid medicine flowing through the main flow path, and the count result of the number of impurity particles measured by the measuring instrument when the valve opening and closing timing is deviated in Modification 1 of the present invention.

[0044] Figure 19 This is a graph illustrating another example of the time-varying filling speed and acceleration of the liquid as it flows through the branch flow path, as shown in Modification 1 of the present invention.

[0045] Figure 20 This is a diagram illustrating a filling section with three branch flow paths in a modified example 2 of the present invention.

[0046] Figure 21 This is a diagram illustrating the configuration of the main flow path and branch flow paths in a modified example 2 of the present invention.

[0047] Figure 22 This is a graph showing the time variation of the filling speed of the liquid as it flows through the first branch flow path, the second branch flow path, and the third branch flow path in a modified example 2 of the present invention.

[0048] Figure 23 This is a graph showing the time variation of the filling speed of the liquid flowing through the first branch flow path, the flow rate of the liquid flowing through the main flow path, and the count result of the number of impurity particles measured by the measuring instrument when the liquid is first started filling the first product container in the second embodiment of the present invention.

[0049] Figure 24 This is a diagram showing the detailed configuration of the filling portion involved in countermeasure 1 in the second embodiment of the technology of the present invention.

[0050] Figure 25 This is a diagram illustrating the configuration of the main flow path, branch flow paths, and drainage flow paths in a second embodiment of the technology of the present invention.

[0051] Figure 26This is a graph showing the time variation of the filling speed of the liquid flowing through the drain path, the first branch path, and the second branch path, the flow rate of the liquid flowing through the main path, and the count result of the number of impurity particles measured by the measuring instrument when the opening and closing timing of the valves are completely consistent in Countermeasure 1 of the second embodiment of the present invention.

[0052] Figure 27 This is a graph showing the time variation of the filling speed of the liquid medicine flowing through the drain path, the first branch path, and the second branch path, the flow rate of the liquid medicine flowing through the main path, and the count result of the number of impurity particles measured by the measuring instrument when the valve opening and closing timing deviates in Countermeasure 1 of the second embodiment of the present invention.

[0053] Figure 28 This is a diagram showing the configuration of the main flow path, branch flow paths, and drainage flow path in a filling section having three branch flow paths in the second embodiment of the present invention.

[0054] Figure 29 This is a diagram illustrating countermeasure 2 in the second embodiment of the technology of the present invention.

[0055] Figure 30 This is a diagram illustrating countermeasure 3 in the second embodiment of the technology of the present invention.

[0056] The components include: 2. Manufacturing equipment; 10. Raw material supply unit; 11. Mixing tank; 12. Purification unit; 13, 75, 85, 90. Filling unit; 14, 14PA, 14PD. Control unit; 15-17. Storage tank; 18. Supply path; 19, 35. Pump; 20, 40, 41. Valve; 21, 37. Filter; 25. Stirring blade; 29. ​​Circulation path; 30. Conveying path; 31. Return path; 32, 76. Filling path; 34. Middle... 36. Flow path; 38. Switching valve; 39. Branch flow path; 40. Measuring instrument; 41. Counting result; 42. Quality conditions; 53. Main flow path; 54. First branch flow path; 55. Second branch flow path; 56. First valve; 87. Second valve; 58. Product container; 59. Weighing instrument; 50. Measurement result; 61. Storage Storage device; 61, 61PA, 61PD, CPU; 62, Memory; 63, Bus; 65, 70, Drive signal; 80, Third branch flow path; 83, Third valve; 86, Drain flow path; 87, Drain tank; 88, Drain valve; ΔT, Deviation; AC, Acceleration; CS, Liquid; FP, Filling period; FR, Flow rate; FS, Set speed; NFP, Non-filling period; RM1, First raw material; RM2, Second raw material; RM3, Third raw material Material; RS, relay interval; SEC1, first interval; SEC2, second interval; ST100, ST110, ST120, ST130, ST140, ST150, ST1301, ST1302, ST1303, ST1304, ST1305, ST1306, ST1307, ST1308, steps; STI1, first state transition interval; STI2, second state transition interval; ULV, upper limit value. Detailed Implementation

[0057] To enable those skilled in the art to better understand the present application, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present application, and not all embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative effort should fall within the scope of protection of the present application.

[0058] It should be noted that, unless otherwise specified, the embodiments and features described in this application can be combined with each other. This application will now be described in detail with reference to the accompanying drawings and embodiments.

[0059] 1. Manufacturing apparatus In the prior art and the technology disclosed herein, the configuration of the manufacturing apparatus, the overview of multiple filling, and the manufacturing sequence of the pharmaceutical solution are almost identical. Therefore, the configuration of the manufacturing apparatus, the overview of multiple filling, and the manufacturing sequence of the pharmaceutical solution will be described first.

[0060] 1.1 Overall Composition As an example, such as Figure 1 As shown, manufacturing apparatus 2 manufactures a chemical solution CS for semiconductor manufacturing. The chemical solution CS is, for example, a cleaning solution used to remove residues remaining on the surface of semiconductor wafers after CMP (Continuous Processing) in the semiconductor manufacturing process. Manufacturing apparatus 2 includes a raw material supply unit 10, a mixing tank 11, a purification unit 12, a filling unit 13, and a control unit 14 (see reference 14). Figure 2 ).

[0061] The raw material supply unit 10 includes a storage tank 15 for a first raw material RM1, a storage tank 16 for a second raw material RM2, and a storage tank 17 for a third raw material RM3. The first raw material RM1 is a liquid, such as an aqueous solution of a surfactant dissolved in pure water. That is, the first raw material RM1 contains a surfactant. The second raw material RM2 is also a liquid, such as an aqueous solution of an organic acid, such as a carboxylic acid, dissolved in pure water. Furthermore, the third raw material RM3 is also a liquid, such as an aqueous solution of an organic acid, such as a phosphonic acid, dissolved in pure water.

[0062] Supply flow path 18 is connected to one end of the lower part of storage tanks 15, 16 and 17 respectively. The other end of supply flow path 18 is connected to the upper part of mixing tank 11. The first raw material RM1 to the third raw material RM3 are supplied to mixing tank 11 through supply flow path 18. In addition, supply flow path 18 and other flow paths are constructed by piping.

[0063] On the supply flow path 18, a pump 19, a valve 20, and a filter 21 are sequentially arranged from the upstream side. The pump 19 is driven under the control of the control unit 14. The pump 19 delivers the first raw material RM1 to the third raw material RM3 into the mixing tank 11 at a set speed. The set speed is the flow rate per unit time. The valve 20 opens and closes under the control of the control unit 14. The valve 20 adjusts the flow rate of the first raw material RM1 to the third raw material RM3 through the supply flow path 18 by changing its opening degree. The filter 21 removes impurities and particulate matter from the first raw material RM1 to the third raw material RM3 passing through the supply flow path 18.

[0064] A stirring blade 25 is provided inside the mixing tank 11. The stirring blade 25 rotates at a set speed under the control of the control unit 14. Thus, the first raw material RM1 to the third raw material RM3 are mixed in the mixing tank 11 to produce the medicine solution CS.

[0065] The purification unit 12 performs a purification process on the drug solution CS produced in the mixing tank 11 by removing impurity particles contained in the drug solution CS. The purification unit 12 is provided with a circulation path 29, which circulates the drug solution CS between the mixing tank 11 and the purification unit 12 by temporarily discharging the drug solution CS from the mixing tank 11 and returning it to the mixing tank 11.

[0066] The circulation path 29 consists of a conveying path 30 and a return path 31. One end of the conveying path 30 is connected to the lower part of the mixing tank 11. One end of the return path 31 is connected to the upper part of the mixing tank 11. Furthermore, the other ends of the conveying path 30 and the return path 31 are connected to each other, thereby forming the circulation path 29.

[0067] The circulation path 29 is connected to the filling path 32. More specifically, one end of the filling path 32 is connected to the circulation path 29 at the junction of the delivery path 30 and the return path 31.

[0068] Additionally, the circulating flow path 29 is provided with a relay flow path 34. One end of the relay flow path 34 is connected to a switching valve 36 provided on the conveying flow path 30. The other end of the relay flow path 34 is connected to the upstream side of the valve 40 of the return flow path 31.

[0069] On the conveying flow path 30, a pump 35, a switching valve 36, and a filter 37 are sequentially arranged from the upstream side. The pump 35 is driven under the control of the control unit 14. The pump 35 draws in the medicinal liquid CS from the mixing tank 11. Furthermore, the drawn-in medicinal liquid CS is circulated via the circulation flow path 29 or conveyed from the conveying flow path 30 to the filling flow path 32. The switching valve 36 is opened and closed under the control of the control unit 14. The switching valve 36 switches the flow path through which the medicinal liquid CS passes to either the conveying flow path 30 or the relay flow path 34. The filter 37 removes impurities and particles from the medicinal liquid CS passing through the conveying flow path 30.

[0070] Near the connection point of the circulation flow path 29, the conveying flow path 30, and the filling flow path 32, a valve 40 is provided on the circulation flow path 29 side and a valve 41 is provided on the filling flow path 32 side. These valves 40 and 41 are opened and closed under the control of the control unit 14.

[0071] A branch flow path 38 is provided downstream of the filter 37 in the delivery flow path 30. A measuring device 39 is disposed in the branch flow path 38. The measuring device 39 measures a value related to the concentration of impurity particles in the drug solution CS passing through the branch flow path 38 within a unit period (e.g., 1 second). More specifically, the measuring device 39 is a so-called particle counter that counts the number of impurity particles as a measurement value related to the concentration of impurity particles. The measuring device 39 is, for example, a light scattering particle counter that counts the number of times the scattered light generated by irradiating the impurity particles with the measuring light is received as the number of impurity particles. Alternatively, the measuring device 39 may also be a light-shielding particle counter that counts the number of times the measuring light is blocked by the impurity particles as the number of impurity particles. The measuring device 39 displays the count result 45 of the number of impurity particles (refer to...). Figure 2 The output is sent to the control unit 14. In addition, the measurement value related to the concentration of impurity particles can be the concentration of impurity particles themselves, or it can be the number density of impurity particles, etc.

[0072] The control unit 14 selects one of the following three paths as the flow path of the drug solution CS in the purification unit 12 by controlling the opening and closing states of the switching valves 36, 40 and 41.

[0073] First, the first path is a circulating flow path 29 consisting of a delivery flow path 30 and a return flow path 31. When the delivery flow path 30 side of the switching valve 36 is open, the relay flow path 34 side is closed, and valve 40 is open and valve 41 is closed, the circulating flow path 29 is selected as the first path. At this time, the liquid medicine CS flowing out of the mixing tank 11 returns to the mixing tank 11 from the output flow path 30 via the return flow path 31. At this time, the liquid medicine CS does not pass through the relay flow path 34 but passes through the filter 37. Therefore, impurity particles in the liquid medicine CS are removed by the filter 37. Therefore, by repeatedly circulating the liquid medicine CS based on this circulating flow path 29, the removal of impurity particles in the liquid medicine CS can be promoted, thereby improving the purity of the liquid medicine CS.

[0074] When the count result 45 of the impurity particle count of the measuring instrument 39 does not meet the preset quality condition 46 (refer to...) Figure 3 When this occurs, the control unit 14 selects the first path. Specifically, quality condition 46 is the upper limit value (ULV) of the number of impurity particles required for the drug solution CS (refer to...). Figure 7 The situation where the impurity particle count result 45 does not meet quality condition 46 refers to the situation where the impurity particle count result 45 is above the upper limit value ULV.

[0075] The second path is a flow path consisting of a conveying flow path 30 and a filling flow path 32. The second path is selected when the conveying flow path 30 side of the switching valve 36 is open, the relay flow path 34 side is closed, and valve 40 is closed while valve 41 is open. At this time, the liquid medicine CS flowing from the mixing tank 11 is supplied to the product container 55 via the conveying flow path 30 and the filling flow path 32. The control unit 14 selects this second path when the count result 45 of the impurity particle count of the measuring device 39 meets the quality condition 46. The case where the count result 45 of the impurity particle count meets the quality condition 46 means that the count result 45 of the impurity particle count is less than the upper limit value ULV. Therefore, the liquid medicine CS supplied to the filling unit 13 is only the liquid medicine CS that meets the quality condition 46, and not the liquid medicine CS that does not meet the quality condition 46.

[0076] The third path is the path from the delivery flow path 30 through the relay flow path 34 to the return flow path 31. The third path is selected when the delivery flow path 30 side of the switching valve 36 is closed, the relay flow path 34 side is open, and both valves 40 and 41 are closed. At this time, the liquid medicine CS flowing from the mixing tank 11 returns to the mixing tank 11 from the delivery flow path 30 through the relay flow path 34 and the return flow path 31. That is, although different from the circulation flow path 29, the third path from the delivery flow path 30 through the relay flow path 34 to the return flow path 31 also forms a circulation flow path for the liquid medicine CS. However, in the third path, since the liquid medicine CS does not pass through the filter 37, impurity particles in the liquid medicine CS are not removed by the filter 37. When performing maintenance on the filter 37 and / or the measuring device 39, the control unit 14 selects this third path formed by the delivery flow path 30, the return flow path 31, and the relay flow path 34.

[0077] 1.2 Composition of the filling section As an example, such as Figure 2 As shown, the filling flow path 32 has a main flow path 50, a first branch flow path 51 and a second branch flow path 52, as well as a first valve 53 and a second valve 54. The filling section 13 is a part formed by the filling flow path 32.

[0078] The main flow path 50 connects to the circulation flow path 29 at the junction of the delivery flow path 30 and the return flow path 31. A first branch flow path 51 and a second branch flow path 52 branch from the main flow path 50, forming two paths. A first valve 53 is installed on the first branch flow path 51, and a second valve 54 is installed on the second branch flow path 52. The first valve 53 and the second valve 54 open and close under the control of the control unit 14. By changing their opening degrees, the first valve 53 and the second valve 54 regulate the flow rate of the liquid medicine CS through the first branch flow path 51 and the second branch flow path 52, thereby regulating the filling speed. The filling speed refers to the flow rate of the liquid medicine CS through the first branch flow path 51 and the second branch flow path 52 per unit time.

[0079] The first branch flow path 51 and the second branch flow path 52 are interchangeably connected to the product container 55. The chemical solution CS is filled into the product container 55 through the first branch flow path 51 and the second branch flow path 52. The product container 55 filled with the chemical solution CS is shipped as a product to a factory that uses the chemical solution CS for semiconductor manufacturing. Furthermore, hereinafter, the product container 55 connected to the first branch flow path 51 is sometimes referred to as product container 55A, and the product container 55 connected to the second branch flow path 52 is sometimes referred to as product container 55B.

[0080] Furthermore, when the capacity of the product container 55 is, for example, 210 kg, the filling speed is preferably 1 to 100 kg / min (more than 1 kg / min and less than 100 kg / min), more preferably 5 to 50 kg / min (more than 5 kg / min and less than 50 kg / min), and even more preferably 10 to 30 kg / min (more than 10 kg / min and less than 30 kg / min). Moreover, when the capacity of the product container 55 is, for example, 210 kg, the filling speed is typically preferably 24 kg / min.

[0081] The control unit 14 controls the opening and closing of valves 40 and 41 based on the counting result 45 of the measuring device 39. Additionally, the control unit 14 controls the closing of the first valve 53 and the second valve 54 based on the measurement result 57 of the weighing instrument 56, which measures the weight of the product container 55 containing the liquid CS. More specifically, the weighing instrument 56 includes: a weighing instrument 56A, which measures the weight of the product container 55A containing the liquid CS; and a weighing instrument 56B, which measures the weight of the product container 55B containing the liquid CS. Weighing instrument 56A outputs a measurement result 57A, and weighing instrument 56B outputs a measurement result 57B. When the measurement result 57A reaches the specified filling amount, the control unit 14 closes the first valve 53. Furthermore, when the measurement result 57B reaches the specified filling amount, the control unit 14 closes the second valve 54. The closing timing of the first valve 53 and the second valve 54 is described later. Figure 6 As shown, it is indicated by a solid inverted triangle mark. The specified filling amount is the weight of the liquid medicine CS when the product container 55 is fully filled. Alternatively, instead of the weight of the product container 55, the water level (liquid level) of the liquid medicine CS in the product container 55 can be measured. When the water level reaches the specified filling amount, the first valve 53 and the second valve 54 are closed.

[0082] 1.3 Composition of the Control Unit As an example, such as Figure 3 As shown, the control unit 14 includes a storage device 60, a CPU (Central Processing Unit) 61, and a memory 62. These storage devices 60, CPU 61, and memory 62 are interconnected via a bus 63.

[0083] Storage device 60 may be, for example, a hard disk drive built into the computer constituting control unit 14, or connected to the computer via cable or network. Storage device 60 stores control programs such as the operating system, various application programs, and various data associated with these programs. Specifically, storage device 60 stores the aforementioned quality conditions 46 and a specified fill quantity for weight. Furthermore, a solid-state drive can be used instead of a hard disk drive.

[0084] Memory 62 is the working memory used by CPU 61 to perform processing. CPU 61 loads the program stored in storage device 60 into memory 62 and executes the processing according to the program. Thus, CPU 61 uniformly controls all parts of manufacturing device 2. In addition, memory 62 can be built into CPU 61.

[0085] 1.4 Overview of Multiple Filling As an example, such as Figure 4 As shown, the manufacturing apparatus 2 performs multiple fillings, that is, it alternately fills the product container 55A connected to the first branch flow path 51 and the product container 55B connected to the second branch flow path 52 with liquid CS. Specifically, first, for example, liquid CS is filled into product container 55A. After the filling of product container 55A with liquid CS is completed, liquid CS is filled into product container 55B. Then, while filling product container 55B with liquid CS, the already filled product container 55A is replaced with an empty product container 55A. Next, liquid CS is filled into product container 55A. Then, while filling product container 55A with liquid CS, the already filled product container 55B is replaced with an empty product container 55B. Since liquid CS is filled into one of the product containers 55A and 55B simultaneously, the throughput of liquid CS filling can be increased compared to the case where one product container 55 is connected to a filling flow path 32 and liquid CS is filled into it. Furthermore, the filling of the drug solution CS into the product container 55A ends when the measurement result 57A reaches the specified filling amount. Similarly, the filling of the drug solution CS into the product container 55B ends when the measurement result 57B reaches the specified filling amount.

[0086] 1.5 Order of Pharmaceutical Solution Preparation The manufacturing sequence of the CS liquid is as follows: Figure 5 As shown in the flowchart. First, under the control of the control unit 14 (CPU 61), the pump 19 is driven and the valve 20 is opened, supplying the first raw material RM1 to the third raw material RM3 to the mixing tank 11 through the supply flow path 18 (raw material supply process, step ST100). In the mixing tank 11, the first raw material RM1 to the third raw material RM3 are mixed by the stirring blades 25 to prepare the medicine solution CS (raw material mixing process, step ST100).

[0087] The control unit 14 selects the circulation path 29 as the first path, causing the medicine solution CS in the mixing tank 11 to circulate in the circulation path 29. During this process, the medicine solution CS is circulated and filtered using the filter 37 (circulation filtration step, ST110). At this time, the number of impurity particles in the medicine solution CS flowing through the branch path 38 is counted by the measuring device 39 (measurement step). The count result 45 of the number of impurity particles is output from the measuring device 39 to the control unit 14.

[0088] In the control unit 14, the count result 45 is compared with the upper limit value ULV of the quality condition 46 (step ST120). When the count result 45 is above the upper limit value ULV and does not meet the quality condition 46 ("No" in step ST120), the cyclic filtration process in step ST110 continues. On the other hand, when the count result 45 is less than the upper limit value ULV and meets the quality condition 46 ("Yes" in step ST120), the process proceeds to step ST130.

[0089] In step ST130, the control unit 14 selects a second path and fills the product container 55 with liquid medicine CS via the filling flow path 32. Specifically, it performs the following... Figure 4 The multiple filling process (filling process) is shown. Next, the number of impurity particles in the liquid medicine CS flowing through the branch flow path 38 is counted using measuring device 39 (measuring process).

[0090] Similar to step ST120, in the control unit 14, the count result 45 is compared with the upper limit value ULV of quality condition 46 (step ST140). When the count result 45 is above the upper limit value ULV and does not meet quality condition 46 (No in step ST140), the filling process is interrupted, and the circulating filtration process and measurement process in step ST110 are restarted. On the other hand, when the count result 45 is less than the upper limit value ULV and still meets quality condition 46 (Yes in step ST140), the filling process and measurement process in step ST130 continue as long as the filling of the drug solution CS into the product container 55 is not finished (No in step ST150).

[0091] 2. Existing Technology 2.1 Speed ​​Control The following describes the speed control of the drug solution CS filling rate in the prior art of multiple filling. Furthermore, as an example in the prior art, such as... Figure 6 As shown, the control unit 14 is referred to as control unit 14PA, and the CPU 61 is referred to as CPU 61PA.

[0092] exist Figure 6In this process, CPU 61PA outputs drive signal 65 to the first valve 53 and the second valve 54. Drive signal 65 is a signal that instantaneously switches the opening of the first valve 53 or the second valve 54 from a fully closed state to a set opening. Furthermore, drive signal 65 is a signal that maintains the opening of the first valve 53 or the second valve 54 at the set opening during a single filling period FP when filling the product container 55A or 55B with liquid medicine CS. Further, drive signal 65 is a signal that instantaneously switches the opening of the first valve 53 or the second valve 54 from the set opening to a fully closed state. Therefore, drive signal 65 is a rectangular wave. In addition, the set opening is an opening of approximately 80% to 95% when the fully open state is 100%. The fully open state can also be set as the set opening. Furthermore, the filling period FP is the period during which a predetermined filling amount of liquid medicine CS is filled into the empty product container 55.

[0093] Here, the first valve 53 and the second valve 54 each have mechanical structures for opening and closing the flow path. Furthermore, actuators must be actuated to drive the first valve 53 and the second valve 54. Therefore, switching from a fully closed state to a set opening degree and vice versa naturally requires time. The term "instantaneous switching of the opening degrees of the first valve 53 and the second valve 54" means switching the opening degree in the shortest possible time possible, given the specifications of each valve. Specifically, "instantaneous" refers to a time sufficiently short compared to the filling period FP. Although it depends on the performance of the first valve 53 and the second valve 54, the term "instantaneous" is, for example, from several milliseconds to hundreds of milliseconds.

[0094] CPU61PA achieves multiple filling of the drug solution CS into product containers 55A and 55B by alternately outputting drive signals 65 to the first valve 53 and the second valve 54. At this time, as... Figure 6 As shown, CPU 61PA performs synchronous control with the goal of ensuring that the start timing of the first valve 53 switching from a set opening to a fully closed state is exactly the same as the start timing of the second valve 54 switching from a fully closed state to a set opening state. Furthermore, CPU 61PA performs synchronous control with the goal of ensuring that the start timing of the second valve 54 switching from a set opening to a fully closed state is exactly the same as the start timing of the first valve 53 switching from a fully closed state to a set opening state. In addition, in Figure 6 For ease of explanation, the change in the drive signal 65 is shown after a certain period of time since the initial filling of the first product container 55 with the drug solution CS. Subsequent... Figure 8 , Figure 11 , Figure 14 The same applies. The process of starting to fill the first product container 55 with the liquid medicine CS will be explained later in "3.2 Second Embodiment".

[0095] Figure 7 This illustrates the synchronous control of the first valve 53 and the second valve 54, such as... Figure 6Under the condition that the opening and closing timings are exactly the same, the time changes of the filling rate of the liquid medicine CS flowing through the first branch flow path 51 and the second branch flow path 52, the flow rate of the liquid medicine CS flowing through the main flow path 50, and the count result 45 of the number of impurity particles measured by the measuring device 39 are analyzed. Furthermore, in Figure 7 In the diagram, dashed circular markers indicate positions where the opening and closing timings of the first valve 53 and the second valve 54 are completely synchronized. Additionally, in... Figure 7 In, with Figure 6 The same situation is illustrated, showing the time variation after a certain period of time has elapsed since the initial filling of the first product container 55 with the drug solution CS. Subsequent... Figure 9 , Figure 13 , Figure 15 The same applies to others.

[0096] When the drive signal 65 from CPU 61PA instantaneously switches the first valve 53 from a fully closed state to a set opening, the filling speed of the liquid medicine CS into the product container 55A via the first branch flow path 51 instantaneously changes from 0 to the set speed FS. Then, while the weight measurement result 57A of the product container 55A measured by the weighing instrument 56A has not yet reached the specified filling amount, the opening of the first valve 53 is maintained at the set opening by the drive signal 65, thereby maintaining the filling speed of the liquid medicine CS into the product container 55A at the set speed FS. When the weight measurement result 57A of the product container 55A measured by the weighing instrument 56A reaches the specified filling amount, and the first valve 53 is instantaneously switched from the set opening to a fully closed state by the drive signal 65, the filling speed of the liquid medicine CS into the product container 55A instantaneously changes from the set speed FS to 0. During the filling of the product container 55B with the liquid medicine CS, the operator replaces the product container 55A filled with the liquid medicine CS with an empty product container 55A. Figure 7 In the example shown, a solid-line square mark indicates when product container 55A was replaced.

[0097] Similarly, when the drive signal 65 from CPU 61PA instantaneously switches the second valve 54 from a fully closed state to a set opening, the filling speed of the liquid medicine CS into the product container 55B via the second branch flow path 52 instantaneously changes from 0 to the set speed FS. Then, while the weight measurement result 57B of the product container 55B measured by the weighing instrument 56B has not yet reached the specified filling amount, the opening of the second valve 54 is maintained at the set opening by the drive signal 65, thereby maintaining the filling speed of the liquid medicine CS into the product container 55B at the set speed FS. When the weight measurement result 57B of the product container 55B measured by the weighing instrument 56B reaches the specified filling amount, and the second valve 54 is instantaneously switched from the set opening to a fully closed state by the drive signal 65, the filling speed of the liquid medicine CS into the product container 55B instantaneously changes from the set speed FS to 0. During the filling of the product container 55A with the liquid medicine CS, the operator replaces the product container 55B filled with the liquid medicine CS with an empty product container 55B. Figure 7 In the case of product container 55A, a solid-line square mark is used to indicate when product container 55B is replaced.

[0098] In the synchronous control of the first valve 53 and the second valve 54, such as Figure 6 When the opening and closing timings shown are exactly the same, such as Figure 7 As shown, the flow velocity of the main path 50 is constant at FR. Therefore, as shown in the graph of the count result 45 at the bottom, no bubbles will be generated due to the turbulence of the flow velocity in the main path 50, as described later.

[0099] 2.2 Research Topic However, in the synchronous control of the first valve 53 and the second valve 54, even with Figure 6 The opening and closing control shown is for controlling the target, but achieving perfect consistency in the timing of opening and closing is extremely difficult, almost impossible. Therefore, in practice, as an example, such as Figure 8 As shown, the opening and closing timings of the first valve 53 and the second valve 54 will deviate by ΔT. Figure 8 The example illustrates a case where the opening and closing timing of the second valve 54 is later than that of the first valve 53. Conversely, the same applies when the opening and closing timing of the first valve 53 is later than that of the second valve 54.

[0100] Thus, when the opening and closing timings of the first valve 53 and the second valve 54 deviate by ΔT, the speed control of the filling rate is not as expected. Figure 7 As shown, but as an example, such as Figure 9 As shown. Moreover, as an example, it occurs as follows: Figure 10 The phenomenon shown.

[0101] More specifically, such as Figure 9 and Figure 10As shown in phenomenon (1), when the opening and closing timings of the first valve 53 and the second valve 54 deviate by ΔT, the flow velocity in the main flow path 50 becomes turbulent, thereby causing a sharp pressure fluctuation in the main flow path 50. Figure 9 In the diagram, solid circles indicate the positions where the opening and closing timings of the first valve 53 and the second valve 54 deviate by ΔT. When compared with... Figure 7 When comparing the positions indicated by the dashed circular markers, it can be seen that... Figure 9 The position indicated by the solid circle mark in the middle shows a deviation ΔT.

[0102] Specifically, the sharp pressure fluctuations in the main flow path 50 caused by the turbulent flow velocity are as follows: When the first valve 53 switches from a fully closed state to a set opening, at the locations indicated by phenomenon (2_1) and the shaded triangle, the flow velocity in the main flow path 50 increases sharply, and the pressure in the main flow path 50 decreases sharply. Conversely, when the first valve 53 switches from a set opening to a fully closed state, at the locations indicated by phenomenon (2_2) and the shaded triangle, the flow velocity in the main flow path 50 decreases sharply, and the pressure in the main flow path 50 increases sharply.

[0103] The sharp pressure fluctuations in the main flow path 50 also affect the circulating flow path 29. As shown in phenomenon (3), flow velocity disturbances occur in the circulating flow path 29, causing sharp pressure fluctuations. Then, as shown in phenomenon (4), a large number of bubbles are temporarily generated in the liquid CS flowing through the circulating flow path 29. As shown in phenomenon (5), the generated bubbles are miscounted as impurity particles in the measuring device 39. As a result, due to the miscounting of bubbles, the counting result 45 may be higher than the upper limit value ULV of quality condition 46. In this case, as shown in phenomenon (6), CPU61PA misjudges that the counting result 45 does not meet quality condition 46. Figure 9 In the diagram, the locations where phenomena (4) and (5) occur are indicated by shaded hexagonal markers.

[0104] like Figure 10 As shown in phenomenon (7), CPU61PA switches the flow path of the liquid CS from the second path to the first path, interrupts the filling process, and restarts the circulating filtration process. In addition, bubbles will disappear over time, but the time required for them to disappear is long enough compared to the deviation ΔT.

[0105] Here, in Figure 9 The triangular waveform representing bubble generation in the count result 45 graph is a conceptual representation and may not necessarily reflect the actual waveform shape. Furthermore, although the minimum value of the count result 45 for the number of impurity particles measured by the measuring instrument 39 is set to 0 in the count result 45 graph, trace impurity particles that were not completely filtered out by the filter 37 are actually counted. The same applies to subsequent graphs. Furthermore, in... Figure 9In the middle, the following was omitted. Figure 7 The solid-line square mark attached is to indicate when product container 55 should be replaced. (Subsequent...) Figure 15 , Figure 18 The same applies to others.

[0106] Thus, in the prior art, the sharp pressure fluctuations caused by the opening and closing timing deviation ΔT between the first valve 53 and the second valve 54 can lead to the generation of bubbles, which may cause a decrease in the throughput of the liquid medicine CS filled into the product container 55.

[0107] 3. The technology of this invention The technology of the present invention will now be described. The manufacturing apparatus 2 involved in the technology of the present invention is basically the same as... Figure 1 The manufacturing apparatus 2 shown is the same, but the control functions of the control unit 14 are different. Hereinafter, descriptions of the same basic structure will be omitted, and the focus will be on the control functions of the control unit 14, which have differences. Furthermore, in the technology of this invention, as an example, such as... Figure 10 As shown, the control unit 14 is referred to as control unit 14PD, and the CPU 61 is referred to as CPU 61PD. The manufacturing apparatus 2 equipped with this control unit 14PD is an example of a "manufacturing apparatus for semiconductor manufacturing solutions" according to the technology of this invention. In addition, the manufacturing method of the solution CS performed by the manufacturing apparatus 2 equipped with the control unit 14PD is an example of a "manufacturing method for semiconductor manufacturing solutions" according to the technology of this invention.

[0108] Furthermore, the mixing tank 11 is an example of a "liquid tank" according to the technology of this invention. The filling flow path 32 is an example of a "multiple filling flow path" according to the technology of this invention. The first branch flow path 51 and the second branch flow path 52 are examples of "branch flow paths" according to the technology of this invention. The first valve 53 and the second valve 54 are examples of "valves" according to the technology of this invention. The CPU 61PD is an example of a "processor" according to the technology of this invention.

[0109] The process of counting the number of impurity particles using the measuring device 39 is an example of a "measuring process" according to the present invention. The process of selecting a first path and gradually increasing the purity of the drug solution CS using the filter 37 is an example of a "circulating filtration process" according to the present invention. Furthermore, the process of selecting a second path and supplying the drug solution CS to the filling section 13 is an example of a "filling process" according to the present invention.

[0110] To suppress bubble formation caused by pressure fluctuations, it is crucial to suppress flow velocity disturbances in the main flow path 50. Therefore, in the technology of this invention, CPU 61PD performs speed control by controlling the opening degrees of the first valve 53 and the second valve 54 to control the filling speed of the first branch flow path 51 and the second branch flow path 52, so that the flow velocity of the main flow path 50 is kept constant at FR.

[0111] Here, the control of "keeping the flow rate of the main flow path 50 constant" means controlling the flow rate of the main flow path 50 with the goal of keeping it constant, and does not mean controlling the flow rate of the main flow path 50 to be completely constant. Furthermore, if a reduction in bubble generation can be expected compared to the case of random opening and closing of the first valve 53 and the second valve 54, and the case of a deviation ΔT in the synchronous control of instantaneous opening and closing of the first valve 53 and the second valve 54 in the prior art, then the flow rate of the main flow path 50 does not need to be completely constant and can fluctuate slightly. Therefore, "constant" as used herein means not only completely constant, but also constant within the range of errors (e.g., fluctuation errors of approximately ±10%) that are generally acceptable in the technical field to which this invention pertains.

[0112] 3.1 First Implementation Method 3.1.1 Speed ​​Control The following describes the speed control of the filling speed of the multi-filled Chinese medicine liquid CS according to the first embodiment of the present invention.

[0113] As an example, such as Figure 11 As shown, CPU 61PD outputs drive signal 70 to the first valve 53 and the second valve 54. Drive signal 70 is used to gradually switch the opening of the first valve 53 or the second valve 54 from a fully closed state to a set opening. Additionally, drive signal 70 is used to maintain the opening of the first valve 53 or the second valve 54 at the set opening. Furthermore, drive signal 70 is used to gradually switch the opening of the first valve 53 or the second valve 54 from the set opening to a fully closed state. Therefore, drive signal 70 is a trapezoidal wave. Thus, CPU 61PD gradually switches the opening of the first valve 53 and the second valve 54. In other words, CPU 61PD performs speed control, intentionally making the switching time between the set opening and the fully closed state of the first valve 53 and the second valve 54 later than the minimum switching time within the specification. Hereinafter, the interval in which the first valve 53 or the second valve 54 switches from the fully closed state to the set opening is denoted as the first state transition interval STI1. In addition, the interval in which the first valve 53 or the second valve 54 switches from the set opening degree to the fully closed state is denoted as the second state transition interval STI2.

[0114] The first state transition interval STI1 and the second state loop interval STI2 are sufficiently long compared to the "instantaneous" time of the instantaneous opening and closing of the first valve 53 and the second valve 54 in the prior art. The first state transition interval STI1 and the second state transition interval STI2 have the same duration, for example, several seconds to several minutes. Furthermore, the proportion of the first state transition interval STI1 and the second state transition interval STI2 in the FP during the filling period is, for example, a few percent to several tens of percent. The first state transition interval STI1 is an example of the "interval in which the filling speed gradually increases" according to the technology of this invention. Furthermore, the second state transition interval STI2 is an example of the "interval in which the filling speed gradually decreases" according to the technology of this invention.

[0115] In actual synchronous control, the CPU 61PD controls the opening and closing of the first valve 53 and the second valve 54 based on the weight measurement result 57 of the product container 55 measured by the weighing instrument 56. For example, when the weight measurement result 57 of the product container 55 measured by the weighing instrument 56 reaches the preset first switching threshold (indicated by a dashed inverted triangle), the CPU 61PD fine-tunes the opening of the first valve 53 or the second valve 54 to make it exactly reach the set opening, and ends the first state transition interval STI1. In addition, when the weight measurement result 57 of the product container 55 measured by the weighing instrument 56 reaches the preset second switching threshold (indicated by a solid inverted triangle), the CPU 61PD begins to close the opening of the first valve 53 or the second valve 54 from the set opening to the fully closed state, thereby initiating the second state transition interval STI2. Then, when the weight measurement result 57 of the product container 55 measured by the weighing instrument 56 reaches the specified filling amount (indicated by a shaded inverted triangle), the CPU 61PD fine-tunes the opening of the first valve 53 or the second valve 54 to bring it to a fully closed state, ending the second state transition interval STI2, and thus ending the filling period FP. Alternatively, instead of the weight measurement result 57 of the product container 55 measured by the weighing instrument 56, the opening and closing of the first valve 53 and the second valve 54 can be controlled solely based on time.

[0116] CPU 61PD achieves multiple fillings of the drug solution CS in product containers 55A and 55B by alternately outputting drive signals 70 to the first valve 53 and the second valve 54. At this time, CPU 61PD performs synchronous control with the goal of ensuring that the timing of the first valve 53 switching from a set opening to a fully closed state is exactly the same as the timing of the second valve 54 switching from a fully closed state to a set opening state. Furthermore, CPU 61PD also performs synchronous control with the goal of ensuring that the timing of the second valve 54 switching from a set opening to a fully closed state is exactly the same as the timing of the first valve 53 switching from a fully closed state to a set opening state.

[0117] When the CPU61PD outputs a drive signal 70 to the first valve 53 or the second valve 54, the time change in the filling speed of the liquid medicine CS flowing through the first branch flow path 51 or the second branch flow path 52 is shown as an example. Figure 12 As shown. That is, when the first valve 53 or the second valve 54 is gradually switched from a fully closed state to a set opening degree by the drive signal 70 from the CPU 61PD, the filling speed of the liquid medicine CS into the product container 55A via the first branch flow path 51 or the second branch flow path gradually changes from 0 to the set speed FS. Then, while the weight measurement result 57 of the product container 55 measured by the weighing instrument 56 has not yet reached the second switching predetermined amount, the opening degree of the first valve 53 or the second valve 54 is maintained at the set opening degree by the drive signal 70, thereby maintaining the filling speed of the liquid medicine CS into the product container 55 at the set speed FS. When the weight measurement result 57 of the product container 55 measured by the weighing instrument 56 reaches the second switching predetermined amount, and the first valve 53 or the second valve 54 is gradually switched from the set opening degree to a fully closed state by the drive signal 70, the filling speed of the liquid medicine CS into the product container 55 gradually changes from the set speed FS to 0. Thus, the CPU61PD gradually increases the filling speed when the filling speed is increased, and gradually decreases it when the filling speed is decreased.

[0118] As the filling speed gradually increases with the opening and closing of the first valve 53 or the second valve 54, the acceleration of the filling speed remains constant at AC. Similarly, as the filling speed gradually decreases with the opening and closing of the first valve 53 or the second valve 54, the acceleration of the filling speed also remains constant at AC (in this case, -AC). Therefore, in the first state transition interval STI1 and the second state transition interval STI2, the filling speed increases and decreases linearly. Furthermore, the term "constant" acceleration as used herein means not only completely constant, but also constant within the range of errors generally acceptable in the technical field to which this invention pertains (e.g., fluctuation errors of approximately ±10%).

[0119] Furthermore, when the capacity of the product container 55 is, for example, 210 kg, the acceleration AC is preferably 5 to 300 kg / min² (5 kg / min² or more and 300 kg / min² or less), more preferably 10 to 200 kg / min² (10 kg / min² or more and 200 kg / min² or less), and even more preferably 20 to 100 kg / min² (20 kg / min² or more and 100 kg / min² or less). Further, when the capacity of the product container 55 is, for example, 210 kg, the acceleration AC is typically preferably 72 kg / min².

[0120] Figure 13The diagram shows the time variation of the filling speed of the liquid medicine CS flowing through the first branch flow path 51 and the second branch flow path 52, the flow rate of the liquid medicine CS flowing through the main flow path 50, and the count result 45 of the number of impurity particles measured by the measuring device 39, under the condition that the opening and closing timings of the first valve 53 and the second valve 54 are completely consistent.

[0121] When the filling speed of the liquid medicine CS in one of the first valves 53 and the second valve 54 is increased, the CPU61PD gradually increases it. Then, when the filling speed of the other liquid medicine CS in the first valve 53 and the second valve 54 is decreased in accordance with the increase in filling speed, the CPU61PD gradually decreases it. Therefore, the filling period FP of liquid medicine CS into product container 55A and the filling period FP of liquid medicine CS into product container 55B overlap in the first state transition interval STI1 and the second state transition interval STI2. Therefore, during the first state transition interval STI1 and the second state transition interval STI2, liquid medicine CS is filled into product containers 55A and 55B in parallel.

[0122] 3.1.2 Effects like Figure 13 As shown, when the opening and closing timings of the first valve 53 and the second valve 54 are completely synchronized, it is consistent with the prior art. Figure 7 As shown, the flow velocity of the main path 50 is constant at FR. Therefore, as shown in the chart of count results 45 at the bottom, no bubbles are generated due to flow velocity disturbances in the main path 50.

[0123] But in reality, such as Figure 14 As shown, the opening and closing timings of the first valve 53 and the second valve 54 deviate by ΔT. However, as an example, as... Figure 15 As shown, by gradually switching the opening degrees of the first valve 53 and the second valve 54, and... Figure 9 Compared to existing technologies, the flow velocity disturbance (represented by solid squares) in the main flow path 50 caused by deviation ΔT is suppressed. Therefore, as shown in the graph of the count result 45 at the bottom, the amount of bubbles generated due to the flow velocity disturbance in the main flow path 50 caused by deviation ΔT is also relatively small, and the count result 45 is less than the upper limit value ULV of quality condition 46. Therefore, the possibility of the CPU 61PD misjudging that the count result 45 does not meet quality condition 46 can be reduced. Figure 14 In, with Figure 8 The same applies to the case where the opening and closing timing of the second valve 54 is later than that of the first valve 53. Conversely, the same applies to the case where the opening and closing timing of the first valve 53 is later than that of the second valve 54.

[0124] As described above, CPU61PD performs speed control by controlling the opening degrees of the first valve 53 and the second valve 54 to control the filling speed of the first branch flow path 51 and the second branch flow path 52 respectively. Figures 11 to 15 As shown, the speed control is as follows: when the filling speed of one of the first branch flow path 51 and the second branch flow path 52 is increased, it is gradually increased; when the filling speed of the other branch flow path is slowed down in accordance with the increase of the filling speed of one of the first branch flow path 51 and the second branch flow path 52, it is decelerated.

[0125] Based on this speed control, such as Figure 15 As shown, this reduces the amount of bubbles generated when the opening and closing timings of the first valve 53 and the second valve 54 deviate by ΔT. That is, in the synchronous control of the first valve 53 and the second valve 54, even if the opening and closing timings of the first valve 53 and the second valve 54 deviate by ΔT, bubble generation can be suppressed. Suppressing bubble generation in this way helps to increase the throughput of the pharmaceutical solution CS filled into the product container 55. As mentioned above, in the synchronous control of the first valve 53 and the second valve 54, such as... Figure 11 As shown, achieving perfect synchronization of opening and closing timings is extremely difficult in practice. In synchronous control, in reality... Figure 14 As shown, the opening and closing timings of the first valve 53 and the second valve 54 almost always exhibit a time deviation ΔT. The technology of this invention is a countermeasure proposed to address this practical problem and has very high effectiveness.

[0126] like Figure 12 As shown, during a single filling process FP of a product container 55 filled with liquid medicine CS, as the filling speed gradually increases and decreases with the opening and closing of the first valve 53 and the second valve 54, the absolute value of the acceleration of the filling speed remains constant. Therefore, speed control can be easily implemented, and thus the opening and closing control of the first valve 53 and the second valve 54 can be easily implemented.

[0127] like Figure 2 As shown, there are two branch flow paths: a first branch flow path 51 and a second branch flow path 52. Therefore, the configuration of the filling section 13 can be simplified, and the installation space for the filling section 13 can be reduced. Furthermore, the component costs required for the filling section 13 can be reduced. Moreover, speed control can be easily implemented.

[0128] like Figure 1As shown, both the first raw material RM1 and the prepared pharmaceutical solution CS contain surfactants. Factors that may generate bubbles in the pharmaceutical solution CS include: cavitation (causing bubbles to form in the liquid due to local pressure reduction), release of dissolved gases accompanying pressure reduction, bubble formation at the interface between solids (pipes, etc.) and the liquid, and detachment of bubbles attached to pipes, etc. When the pharmaceutical solution CS contains surfactants, the gas-liquid interface stability of bubbles generated by these factors may lead to bubbles that are difficult to dissolve or break down, resulting in residues. Consequently, in the circulation path 29, the measuring instrument 39 may misjudge residual bubble impurities as impurity particles, leading to a temporary increase in the measured value. Furthermore, when the pharmaceutical solution CS is aqueous, due to the gas-liquid interface characteristics (surface tension, interfacial viscoelasticity, etc.) and dissolved gas conditions, tiny bubbles may remain, which can also easily cause false counting (in liquids where the solvent is primarily water). Therefore, in the manufacturing apparatus 2 for processing pharmaceutical solutions CS containing surfactants, this technology can effectively suppress false counting caused by foaming, demonstrating significant advantages.

[0129] 3.1.3 Sequence of filling process Next, as an example, we will refer to... Figure 16 The flowchart shown illustrates the sequence of the filling process in the present invention.

[0130] In the filling process, the control unit 14PD performs the following: Figure 11 Synchronous control, including speed control as shown. Specifically, the control unit 14PD gradually switches one of the first valve 53 and the second valve 54 from a fully closed state to a set opening degree, thereby increasing the filling speed of one of the first branch flow paths 51 and the second branch flow path 52. Furthermore, in response to the increase in the filling speed of one of the first branch flow paths 51 and the second branch flow path 52, the control unit 14PD gradually switches the other of the first valve 53 and the second valve 54 from a set opening degree to a fully closed state, thereby slowing down the filling speed of the other of the first branch flow path 51 and the second branch flow path 52.

[0131] First, consider the case where liquid medicine CS is filled into product container 55A via the first branch flow path 51 (step ST1301). While the weight measurement result 57A of product container 55A measured by weighing instrument 56A has not yet reached the second switching specified amount ("No" in step ST1302), continue to fill liquid medicine CS into product container 55A via the first branch flow path 51.

[0132] When the weight measurement result 57A of product container 55A, as measured by weighing instrument 56A, reaches the second switching predetermined amount ("Yes" in step ST1302), under the control of control unit 14PD, the first valve 53 is gradually switched from the set opening to the fully closed state. At the same time, the second valve 54 is gradually switched from the fully closed state to the set opening (step ST1303). The opening of the second valve 54 is maintained at the set opening. Thus, the operation of filling liquid CS into product container 55A via the first branch flow path 51 ends. In addition, liquid CS is filled into product container 55B via the second branch flow path 52. Afterwards, the operator replaces the filled product container 55A with an empty product container 55A (step ST1304). During the period before the weight measurement result 57B of product container 55B, as measured by weighing instrument 56B, reaches the second switching predetermined amount ("No" in step ST1305), the opening of the second valve 54 continues to be maintained at the set opening.

[0133] When the weight measurement result 57B of product container 55B, as measured by weighing instrument 56B, reaches the second switching predetermined amount ("Yes" in step ST1305), under the control of control unit 14PD, the second valve 54 is gradually switched from the set opening to the fully closed state. At the same time, the first valve 53 is gradually switched from the fully closed state to the set opening (step ST1306). The opening of the first valve 53 is maintained at the set opening. Thus, the operation of filling liquid CS into product container 55B via the second branch flow path 52 ends. In addition, liquid CS is filled into product container 55A via the first branch flow path 51. Afterwards, the operator replaces the filled product container 55B with an empty product container 55B (step ST1307). During the period before the weight measurement result 57A of product container 55A, as measured by weighing instrument 56A, reaches the second switching predetermined amount ("No" in step ST1308), the opening of the first valve 53 continues to be maintained at the set opening. By repeating this process, the drug solution CS is alternately filled into product containers 55A and 55B.

[0134] 3.1.4 Variation Example 1 Figure 17 In the variation example 1 shown, the gradual increase and decrease of the filling speed are similar to... Figure 12 The situation is different. Specifically, the CPU61PD makes the fill speed continuously increase and decrease gradually. Here, "continuous" means that the change in fill speed over time is smooth. In other words, it means that the graph representing the change in fill speed over time is curved. This curve can be, for example, part of a quadratic curve such as a parabola or hyperbola, part of an exponential curve, or part of a logarithmic curve, etc. Mathematically, "continuous" means that the graph representing the change in fill speed over time does not contain non-differentiable discontinuous points.

[0135] Furthermore, when the first valve 53 and the second valve 54 are pulsed, the filling speed exhibits a step-like variation from a microscopic perspective. Therefore, strictly speaking, it is not continuous. However, this microscopic step-like variation is included within the error range of "continuous" in the technology of this invention. However, as long as the amount of bubble generation can be suppressed to a level that will not cause the CPU 61PD to misjudge the counting result 45 as not meeting the quality condition 46, the filling speed can be made to vary in a step-like manner.

[0136] exist Figure 17 In Variation 1, the first state transition interval STI1 and the second state transition interval STI2 are each divided into a first interval SEC1 and a second interval SEC2. The first interval SEC1 and the second interval SEC2 have the same time length. Moreover, by performing two inversion operations with the filling velocity axis (vertical axis) and the time axis (horizontal axis) as the axes of symmetry, the curve change of the filling velocity in the first interval SEC1 is consistent with the curve change of the filling velocity in the second interval SEC2. Therefore, the change of the acceleration of the filling velocity is mirror-symmetrical between the first interval SEC1 and the second interval SEC2. The first interval SEC1 is the first half of the first state transition interval STI1 and the second state transition interval STI2 after being equally divided, and the second interval SEC2 is the second half of the first state transition interval STI1 and the second state transition interval STI2 after being equally divided.

[0137] In the first interval SEC1 of the first state transition interval STI1, where the fill speed gradually increases in a curved manner, the acceleration of the fill speed gradually increases from 0 to a maximum value AC. Conversely, in the second interval SEC2, the acceleration of the fill speed gradually decreases from the maximum value AC to 0. Similarly, in the first interval SEC1 of the second state transition interval STI2, where the fill speed gradually decreases in a curved manner, the acceleration of the fill speed gradually increases from 0 to a maximum value AC (in this case, -AC). On the other hand, in the second interval SEC2, the acceleration of the fill speed gradually decreases from the maximum value AC (in this case, -AC) to 0.

[0138] By precisely controlling the dynamic change of the absolute value of the filling rate acceleration—that is, by allowing the absolute value of the acceleration to change smoothly and gradually from 0 to its maximum value and then back to 0—the impact caused by flow rate changes during valve opening and closing can be effectively reduced. Furthermore, because the acceleration change is gradual, even if there are timing deviations in valve operation, flow velocity fluctuations in the main channel can be minimized. Therefore, when switching material storage tanks, no pressure fluctuations occur in the main channel, and the amount of bubble generation can be controlled to an extremely low, almost negligible level.

[0139] From another perspective, the synchronous control of multiple filling processes eliminates the need for valves to achieve 100% precise synchronization, thereby reducing the requirements for hardware configuration and control accuracy. Simultaneously, chemical solutions containing surfactants are highly prone to generating bubbles, but this invention minimizes bubble formation by eliminating pressure fluctuations in the main flow channel. Therefore, this system is suitable for harsh operating conditions, such as the transport of special fluids like chemical solutions used in high-end semiconductor processes (e.g., cleaning solutions after chemical mechanical polishing).

[0140] As an example, such as Figure 18 As shown, in Modification 1, compared to the prior art, the flow velocity turbulence in the main flow path 50 is suppressed when the opening and closing timings of the first valve 53 and the second valve 54 deviate by ΔT. Furthermore, compared to... Figure 15 Compared to the example shown, flow velocity turbulence in the main path 50 was also suppressed. Therefore, compared with existing technologies and Figure 15 Compared to the example shown, the amount of bubbles generated due to the deviation ΔT is also significantly reduced, and the count result 45 is far below the upper limit value ULV of quality condition 46. Therefore, the possibility of the CPU 61PD misjudging the count result 45 as not meeting quality condition 46 can be further reduced.

[0141] Thus, in Modification 1, during a single filling FP of the liquid medicine CS into a product container 55, as the filling speed gradually increases and decreases with the opening and closing of the first valve 53 and the second valve 54, the absolute value of the acceleration of the filling speed is changed, thereby causing the filling speed to change continuously. Therefore, with... Figure 12 Compared to the example shown where the absolute value of the acceleration of the filling speed is set constant, pressure fluctuations in the circulating flow path 29 caused by turbulence in the main flow path 50 when a deviation ΔT occurs can be further suppressed. As a result, bubble formation can be further suppressed. This can further help to improve the throughput of filling liquid CS into the product container 55.

[0142] Furthermore, the first state transition interval STI1, which is the interval that gradually increases the filling speed, and the second state transition interval STI2, which is the interval that gradually decreases the filling speed, each include a first interval SEC1 and a second interval SEC2. Moreover, in the first interval SEC1, the absolute value of acceleration gradually increases from 0 to its maximum value AC, and in the second interval SEC2, the absolute value of acceleration gradually decreases from its maximum value to 0. Therefore, with... Figure 12 Compared to the example shown where the absolute value of the acceleration of the filling speed is set constant, pressure fluctuations in the circulating flow path 29 caused by turbulence in the main flow path 50 when a deviation ΔT occurs can be further suppressed. As a result, bubble formation can be further suppressed. This can further help to improve the throughput of filling liquid CS into the product container 55.

[0143] The first interval SEC1 is the first half of both the first state transition interval STI1 and the second state transition interval STI2, and the second interval SEC2 is the second half of both the first state transition interval STI1 and the second state transition interval STI2. Therefore, compared to not setting the first interval SEC1 and the second interval SEC2 as equally divided intervals (i.e., first and second halves), speed control can be performed more simply.

[0144] As an example, such as Figure 19 As shown, the first state transition interval STI1, which is the interval that gradually increases the filling speed, and the second state transition interval STI2, which is the interval that gradually decreases the filling speed, can each include a relay interval RS between the first interval SEC1 and the second interval SEC2. The relay interval RS is the same as the first state transition interval STI1 and the second state transition interval STI2 in the first embodiment described above; it is an interval where the acceleration of the filling speed is constant at AC or -AC. Therefore, in the relay interval RS, the filling speed increases and decreases linearly. By including this relay interval RS, speed control can be performed more simply compared to the case without a relay interval RS.

[0145] 3.1.5 Variation Example 2 Although a filling portion 13 with two branch flow paths, namely a first branch flow path 51 and a second branch flow path 52, is illustrated, it is not limited thereto. There may be more than three branch flow paths.

[0146] As an example, such as Figure 20 and Figure 21 As shown, the filling section 75 of Modified Example 2 has a main flow path 77 and three branch flow paths as filling flow paths 76, namely a first branch flow path 78, a second branch flow path 79, and a third branch flow path 80. The first branch flow path 78, the second branch flow path 79, and the third branch flow path 80 branch from the main flow path 77 to form three paths. In addition, the filling section 75 has a first valve 81, a second valve 82, and a third valve 83. The filling flow path 76 is an example of a "multiple filling flow path" according to the technology of this invention. The first branch flow path 78, the second branch flow path 79, and the third branch flow path 80 are examples of a "branch flow path" according to the technology of this invention. The first valve 81, the second valve 82, and the third valve 83 are examples of a "valve" according to the technology of this invention.

[0147] A first valve 81 is located on the first branch flow path 78, a second valve 82 is located on the second branch flow path 79, and a third valve 83 is located on the third branch flow path 80. The first valve 81, the second valve 82, and the third valve 83 open and close under the control of the control unit 14PD. By changing their opening degrees, the first valve 81, the second valve 82, and the third valve 83 regulate the flow rate of the liquid medicine CS passing through the first branch flow path 78 and the second branch flow path 79, thereby regulating the filling speed.

[0148] Product container 55 is interchangeably connected to the first branch flow path 78, the second branch flow path 79, and the third branch flow path 80, respectively. Furthermore, the product container 55 connected to the first branch flow path 78 is sometimes referred to as product container 55A, the product container 55 connected to the second branch flow path 79 as product container 55B, and the product container 55 connected to the third branch flow path 80 as product container 55C.

[0149] Weighing instruments 56A, 56B, and 56C respectively measure the weight of product containers 55A, 55B, and 55C containing liquid medicine CS, and output the measurement results 57A, 57B, and 57C (only measurement result 57C is shown in the figure) to the control unit 14PD. The control unit 14PD controls the opening and closing of the first valve 81, the second valve 82, and the third valve 83 based on the measurement results 57A to 57C.

[0150] Viewing the filling section 75 from above Figure 21 In this configuration, the ends of the first branch flow path 78, the second branch flow path 79, and the third branch flow path 80, opposite to the side connected to the product container 55, are connected to the main flow path 77. The first branch flow path 78, the second branch flow path 79, and the third branch flow path 80 extend horizontally at 120° intervals from their ends connected to the main flow path 77, and then bend at right angles towards the product container 55 at the same length. Thus, the first branch flow path 78, the second branch flow path 79, and the third branch flow path 80 are equally spaced and have the same length and shape. Therefore, deviations in the filling speed of the liquid medicine CS in the first branch flow path 78, the second branch flow path 79, and the third branch flow path 80 can be suppressed.

[0151] The time variation of the filling velocity of the liquid medicine CS flowing through the first branch flow path 78, the second branch flow path 79, and the third branch flow path 80 in the filling section 75 is as follows: Figure 22As shown. Specifically, when the weight measurement result 57C of product container 55C reaches the second specified switching amount, CPU 61PD gradually switches the third valve 83 from a set opening to a fully closed state, gradually reducing the filling speed from the set speed FS to 0, thereby ending the operation of filling liquid CS into product container 55C via the third branch flow path 80. Simultaneously, CPU 61PD gradually switches the first valve 81 from a fully closed state to a set opening, gradually reducing the filling speed from 0 to the set speed FS, thereby filling liquid CS into product container 55A via the first branch flow path 78. Afterwards, the operator replaces the filled product container 55C with an empty product container 55C.

[0152] When the weight measurement result 57A of product container 55A reaches the second switching predetermined amount, CPU 61PD gradually switches the first valve 81 from the set opening to the fully closed state, gradually reducing the filling speed from the set speed FS to 0, thereby ending the operation of filling product container 55A with liquid medicine CS via the first branch flow path 78. Simultaneously, CPU 61PD gradually switches the second valve 82 from the fully closed state to the set opening, gradually changing the filling speed from 0 to the set speed FS, thereby filling product container 55B with liquid medicine CS via the second branch flow path 79. Afterwards, the operator replaces the filled product container 55A with an empty product container 55A.

[0153] Next, when the weight measurement result 57B of product container 55B reaches the second switching predetermined amount, CPU 61PD gradually switches the second valve 82 from the set opening to the fully closed state, gradually changing the filling speed from the set speed FS to 0, thereby ending the operation of filling liquid CS into product container 55B via the second branch flow path 79. Simultaneously, CPU 61PD gradually switches the third valve 83 from the fully closed state to the set opening, gradually changing the filling speed from 0 to the set speed FS, thereby filling liquid CS into product container 55C via the third branch flow path 80. Afterwards, the operator replaces the filled product container 55B with an empty product container 55B. By repeating this process, liquid CS is alternately filled into product containers 55A, 55B, and 55C. Furthermore, in Figure 22 The diagram illustrates the case where no deviation ΔT occurs, omitting the time variation of the flow rate of the liquid CS flowing through the main flow path 77 and the count result 45 of the impurity particle number measured by the measuring device 39. Additionally, in Figure 22 The diagram shows a linearly increasing and decreasing filling speed. Figure 12 One example is the filling speed, but a variation of Example 1, in which the filling speed gradually increases and decreases in a curved manner, can also be applied.

[0154] 3.2 Second Implementation Method 3.2.1 Problems encountered when initially filling the first product container with the solution. In the first embodiment described above, such as Figure 13 and Figure 15 As shown, this illustrates the time variation after a certain period of time has elapsed since the first filling of the first product container 55 with the drug solution CS. However, as an example, such as Figure 23 As shown, when the first product container 55 begins to be filled with the liquid CS, the initial filling speed is zero. Therefore, fluctuations in the filling speed inevitably cause turbulence in the main flow path 50, and bubbles are generated due to pressure fluctuations. These bubbles are miscounted as impurity particles in the measuring device 39, causing the count result 45 to exceed the upper limit value (ULV) of quality condition 46. In this case, the CPU 61PD mistakenly determines that the count result 45 does not meet quality condition 46. Here, the first product container 55 refers to the product container 55 that is first filled with the liquid CS after selecting the second path. Figure 23 In the example, product container 55A, which is filled with liquid medicine CS through branch flow path 51, is shown as the first product container 55.

[0155] The second embodiment is an implementation that adds countermeasures to the first embodiment to address the following problem: when the first product container 55 is filled with liquid CS, bubbles are generated, causing the CPU 61PD to misjudge the count result 45 as not meeting quality condition 46. That is, in the second embodiment, any one of countermeasures 1 to 3 as shown below is implemented. Furthermore, in any one of countermeasures 1 to 3, the processing after the operation of filling the first product container 55 with liquid CS is completed is the same as in the first embodiment.

[0156] 3.2.2 Countermeasure 1 As an example, such as Figure 24 As shown, the filling portion 85 is configured to be in Figure 2 The filling section 13 shown is provided with a drainage flow path 86. The drainage flow path 86, like the first branch flow path 51 and the second branch flow path 52, branches off from the main flow path 50. That is, the first branch flow path 51, the second branch flow path 52, and the drainage flow path 86 branch off from the main flow path 50, forming three paths. The drainage flow path 86 is connected to a drainage tank 87. Furthermore, the drainage flow path 86 has a drainage valve 88. The drainage valve 88 is opened and closed under the control of the control unit 14PD.

[0157] Viewing the filling section 85 from above Figure 25In this configuration, the ends of the first branch flow path 51, the second branch flow path 52, and the drain flow path 86, opposite to the side connected to the product container 55, are connected to the main flow path 77. The first branch flow path 51, the second branch flow path 52, and the drain flow path 86 extend horizontally at 120° intervals from the ends connected to the main flow path 50, and then bend at right angles in the vertical direction towards the product container 55 and the drain tank 87 at the same length. Thus, the first branch flow path 51, the second branch flow path 52, and the drain flow path 86 are equally spaced and have the same length and shape. Therefore, deviations in the filling speed of the drug solution CS in the first branch flow path 51, the second branch flow path 52, and the drain flow path 86 can be suppressed.

[0158] As an example, such as Figure 26 As shown, CPU61PD first gradually switches the drain valve 88 from a fully closed state to a set opening degree. Thus, as... Figure 23 The situation is similar to the initial opening of the first valve 53 and the start of filling the product container 55A with the drug solution CS, where bubbles are generated due to pressure fluctuations. However, the bubbly drug solution CS is discharged into the drain tank 87 through the drain path 86 instead of being filled into the product container 55. Furthermore, the time for the drain valve 88 to gradually change from a fully closed state to a set opening degree is, for example, the first state transition interval STI1.

[0159] CPU 61PD uses the monitoring count result 45 to determine whether the time required from setting the drain valve 88 to the set opening until the bubbles disappear has elapsed. When it is determined that the time required from setting the drain valve 88 to the set opening until the bubbles disappear has elapsed, CPU 61PD spends the second state transition interval STI2 time to gradually switch the drain valve 88 from the set opening to the fully closed state. At the same time, CPU 61PD spends the first state transition interval STI1 time to gradually switch the first valve 53 from the fully closed state to the set opening. This ends the operation of discharging the liquid medicine CS to the drain tank 87 via the drain flow path 86. In addition, the liquid medicine CS is filled into the product container 55A via the first branch flow path 51. The period of gradually switching the drain valve 88 from the fully closed state to the set opening and then from the set opening to the fully closed state is the non-filling period NFP when the liquid medicine CS is not filled into the product container 55.

[0160] Furthermore, since drainage is necessary regardless, the drain valve 88 can be switched instantaneously from a fully closed state to a set opening, rather than gradually switching from a fully closed state to a set opening. Additionally, when the count result 45 reaches a value less than the upper limit ULV, the CPU 61PD can gradually switch the drain valve 88 from a set opening to a fully closed state, and gradually switch the first valve 53 from a fully closed state to a set opening, without waiting for the time required from setting the drain valve 88 to the set opening until the bubbles disappear.

[0161] exist Figure 26 The example illustrates a scenario where the opening and closing timings of the first valve 53, the second valve 54, and the drain valve 88 are perfectly synchronized. However, as repeatedly stated earlier, achieving perfect synchronization of opening and closing timings in the synchronous control of multiple valves is extremely difficult. In fact, as an example, such as... Figure 27 As shown, for example, the opening and closing timing of the first valve 53 and the drain valve 88 deviates by ΔT.

[0162] However, by gradually switching the opening degrees of the first valve 53 and the drain valve 88, the flow rate disturbance in the main flow path 50 caused by the deviation ΔT was suppressed. Therefore, as shown in the graph of the count result 45 at the bottom, the amount of bubbles generated due to the flow rate disturbance in the main flow path 50 caused by the deviation ΔT is also relatively small, and the count result 45 is less than the upper limit value ULV of the quality condition 46. Therefore, the possibility of the CPU 61PD misjudging that the count result 45 does not meet the quality condition 46 can be reduced.

[0163] The count result 45, measured immediately after the filling of the first product container 55 with the liquid CS, is relatively low in reliability because it is more likely that bubbles caused by pressure fluctuations at the start of filling the first product container 55 with the liquid CS will be mistakenly counted as impurity particles. Even so, the possibility that impurity particles are being counted instead of bubbles cannot be ruled out. Therefore, before filling the first product container 55 with the liquid CS, the liquid CS is discharged from the drain path 86 to the drain tank 87. This prevents liquid CS that is difficult to determine whether it is a miscount of bubbles or miscounted impurity particles from being filled into the product container 55.

[0164] Furthermore, when a drainage flow path 86 is added to the filling portion 75 in the modified example 2 of the first embodiment described above, as an example, it is preferable to use... Figure 28 The filling section 90 shown is configured as described above. Specifically, the first branch flow path 78, the second branch flow path 79, the third branch flow path 80, and the drain flow path 86 are connected to the main flow path 77 at their ends opposite to the side connected to the product container 55 and the drain tank 87. The first branch flow path 78, the second branch flow path 79, the third branch flow path 80, and the drain flow path 86 extend horizontally at 90° intervals from their ends connected to the main flow path 77, and bend at right angles in the vertical direction towards the product container 55 and the drain tank 87 at the same length. Thus, the first branch flow path 78, the second branch flow path 79, and the third branch flow path 80 are equally spaced and have the same length and shape. Therefore, deviations in the filling speed of the drug solution CS in the first branch flow path 78, the second branch flow path 79, the third branch flow path 80, and the drain flow path 86 can be suppressed.

[0165] Furthermore, the location of the drainage flow path 86 is not limited to... Figure 24The location is illustrated in the diagram. The main flow path 50 can be branched off downstream of valve 41 and upstream of the first branch flow path 51 and the second branch flow path 52 to form a drain flow path 86.

[0166] 3.2.3 Countermeasure 2 As an example, such as Figure 29 As shown, in countermeasure 2, for example, a drain tank 87 is first installed on the first branch flow path 51 to replace the product container 55A. The CPU 61PD gradually switches the first valve 53 from a fully closed state to a set opening degree and discharges the foaming liquid CS into the drain tank 87 via the first branch flow path 51.

[0167] CPU 61PD monitors the count result 45 to determine whether the time required from setting the first valve 53 to the set opening until the bubbles disappear has elapsed. When it is determined that the time required from setting the first valve 53 to the set opening until the bubbles disappear has elapsed, CPU 61PD spends the second state transition interval STI2 time to gradually switch the first valve 53 from the set opening to the fully closed state. At the same time, CPU 61PD spends the first state transition interval STI1 time to gradually switch the second valve 54 from the fully closed state to the set opening. This ends the operation of discharging the drug solution CS to the drain tank 87 via the first branch flow path 51. In addition, the drug solution CS is filled into the product container 55B via the second branch flow path 52. Afterwards, the operator replaces the drain tank 87 with the product container 55A. In this case, the period of gradually switching the first valve 53 from the fully closed state to the set opening state and then from the set opening state to the fully closed state in order to discharge the drug solution CS into the drain tank 87 is the non-filling period (NFP). Furthermore, in this case, the first branch flow path 51 also serves as the drain flow path 86, and the first valve 53 also serves as the drain valve 88. Since there is no need to install the drain flow path 86 and the drain valve 88, it helps to reduce the cost and save space of the filling section 13 and even the manufacturing apparatus 2.

[0168] 3.2.4 Countermeasure 3 like Figure 30 As indicated by the label box, as countermeasure 3, the CPU61PD does not use the count result 45 measured immediately after the start of filling the first product container 55 with the drug solution CS, i.e., the count value of impurity particles measured by the measuring device 39. Here, "immediately after the start" refers to the period from the start of filling the first product container 55 until the latest completion of filling.

[0169] The term "not using the count result 45" means performing any of the following processes: not accepting the count result 45 from the measuring device 39; accepting but discarding the count result 45 from the measuring device 39; accepting and not discarding the count result 45 from the measuring device 39, but not determining whether the count result 45 meets the quality condition 46; and accepting and not discarding the count result 45 from the measuring device 39, determining whether the count result 45 meets the quality condition 46, but not using the determination result. During the period when the count result 45 is not used, a period is set based on the time it takes for the bubbles generated when the first valve 53 is first opened to begin filling the product container 55A with the liquid medicine CS to disappear. Thus, by setting the count result 45, which is relatively unreliable and measured immediately after the start of filling the first product container 55 with the liquid medicine CS, to not use the count result 45, which is relatively unreliable, it is possible to prevent the filling process from being interrupted due to the low reliability of the count result 45, thereby suppressing the decrease in the throughput of filling the product container 55 with the liquid medicine CS.

[0170] 3.3 Other As a surfactant, there are no particular limitations as long as it is a compound that has both hydrophilic and hydrophobic (lipophilic) groups in one molecule. Examples include nonionic surfactants and anionic surfactants.

[0171] From the viewpoint of superior cleaning performance, the CS solution preferably contains a surfactant. Surfactants generally have at least one hydrophobic group selected from the group consisting of aliphatic hydrocarbon groups, aromatic hydrocarbon groups, and groups combining these. The total number of carbon atoms in the surfactant is preferably 16 to 100.

[0172] Examples of nonionic surfactants include ester-type nonionic surfactants, ether-type nonionic surfactants, and ester-ether-type nonionic surfactants. Among these, ether-type nonionic surfactants are preferred.

[0173] As nonionic surfactants, compounds illustrated, for example, in paragraph 0126 of International Publication No. 2022 / 044893 may also be cited, the contents of which are incorporated herein by reference.

[0174] Examples of anionic surfactants include phosphate ester surfactants with phosphate ester groups, sulfonic acid surfactants with sulfonium groups, phosphonic acid surfactants with phosphonic acid groups, carboxylic acid surfactants with carboxyl groups, and sulfate ester surfactants with sulfate ester groups.

[0175] As anionic surfactants, compounds illustrated, for example, in paragraphs 0116 to 0123 of International Publication No. 2022 / 044893 may also be cited, the contents of which are incorporated herein by reference.

[0176] The surfactant content relative to the total mass of the CS solution is preferably 0.0001 to 1.0% by mass, more preferably 0.001 to 0.8% by mass. A single surfactant or two or more surfactants may be used. When two or more surfactants are used, their combined content is preferably within the above range.

[0177] The solvent is not particularly limited as long as it does not affect the semiconductor substrate, and examples such as pure water, distilled water, or ion-exchanged water can be used. From the viewpoint of minimizing the impact on the semiconductor substrate, pure water or ion-exchanged water is preferred. The water content is only the balance that can be contained in the pharmaceutical solution CS. The water content relative to the total mass of the pharmaceutical solution CS is preferably 50.0 to 99.99% by mass, more preferably 60.0 to 98.0% by mass, and even more preferably 65.0 to 85.0% by mass.

[0178] From the viewpoint of improving the removal performance of metal-containing substances, the CS solution preferably contains an organic acid. An organic acid is an organic compound having an acidic functional group. Examples of acidic functional groups include carboxyl, phosphonic, sulfonic, phenolic hydroxyl, and thiol groups. Furthermore, in this specification, it is assumed that the organic acid does not contain compounds that function as the aforementioned anionic surfactants.

[0179] As an organic acid, there is no particular limitation, and examples include carboxylic acids (organic carboxylic acids) having a carboxyl group in the molecule, phosphonic acids (organic phosphonic acids) having a phosphonic acid group in the molecule, and sulfonic acids (organic sulfonic acids) having a sulfonic group in the molecule. Among these, carboxylic acids or phosphonic acids are preferred.

[0180] The number of functional groups in the organic acid is not particularly limited, but is preferably 1 to 4, more preferably 1 to 3. Furthermore, from the viewpoint of improving cleaning performance, the organic acid is preferably a compound that has the function of chelating with metals contained in the residue, and more preferably a compound having two or more functional groups (ligands) within the molecule that form coordinate bonds with metal ions. Examples of ligands include the aforementioned acidic functional groups, preferably carboxylic acid groups or phosphonic acid groups.

[0181] The carboxylic acid can be a monocarboxylic acid having one carboxyl group or a polycarboxylic acid having two or more carboxyl groups. From the viewpoint of superior cleaning performance, a polycarboxylic acid having two or more (more preferably 2 to 4, and even more preferably 2 or 3) carboxyl groups is preferred.

[0182] Examples of carboxylic acids include aminopolycarboxylic acids, amino acids, hydroxycarboxylic acids, and aliphatic carboxylic acids.

[0183] Amino polycarboxylic acids are compounds that have one or more amino groups and two or more carboxyl groups as ligands within their molecules. Examples of amino polycarboxylic acids include aspartic acid, glutamic acid, butanediaminetetraacetic acid (DTPA), diethylenetriaminepentaacetic acid (EDTA), ethylenediaminetetrapropionic acid (EDTA), triethylenetetraaminehexaacetic acid (TTA), 1,3-diamino-2-hydroxypropane-N,N,N',N'-tetraacetic acid (DTA), propylenediaminetetraacetic acid (PDTA), ethylenediaminetetraacetic acid (EDTA), and trans-1,2-cyclohexanediaminetetraacetic acid (CyDTA). 1,6-Hexamethylenediaminetetraacetic acid (HDA), ethylenediaminedipropionic acid (EDTA), 1,6-hexamethylenediamine-N,N,N',N'-tetraacetic acid (HDI), N,N-bis(2-hydroxybenzyl)ethylenediamine-N,N-diacetic acid (DHA), diaminopropanetetraacetic acid (DPA), 1,4,7,10-tetraazacyclododecanetetraacetic acid (DDA), diaminopropanoltetraacetic acid (DPA), (hydroxyethyl)ethylenediaminetriacetic acid (EDTA), and iminodiacetic acid (IDA). Preferably, DTPA, EDTA, CyDTA, or IDA are used.

[0184] Amino acids are compounds having one carboxyl group and one or more amino groups within a molecule. Examples of amino acids include glycine, serine, α-alanine (2-aminopropionic acid), β-alanine (3-aminopropionic acid), lysine, leucine, isoleucine, cystine, cysteine, methionine, ethionine, threonine, tryptophan, tyrosine, valine, histidine, histidine derivatives, asparagine, glutamine, arginine, proline, phenylalanine, compounds described in paragraphs 0021 to 0023 of Japanese Patent Application Publication No. 2016-086094, and their salts. Furthermore, as histidine derivatives, compounds described in Japanese Patent Application Publication Nos. 2015-165561 and 2015-165562 are cited, the contents of which are incorporated herein by reference. Additionally, as salts, examples include alkali metal salts such as sodium and potassium salts, ammonium salts, carbonates, and acetates.

[0185] Preferably, the amino acid is histidine, a histidine derivative, or a sulfur-containing amino acid containing a sulfur atom, and more preferably histidine or a sulfur-containing amino acid. Examples of sulfur-containing amino acids include cystine, cysteine, ethionine, and methionine, with cystine or cysteine ​​being the most preferred.

[0186] Hydroxycarboxylic acids are compounds having one or more hydroxyl groups and one or more carboxyl groups within their molecules. Examples of hydroxycarboxylic acids include malic acid, citric acid, glycolic acid, gluconic acid, heptanoic acid, tartaric acid, and lactic acid, with gluconic acid, glycolic acid, malic acid, tartaric acid, or citric acid being preferred, and gluconic acid or citric acid being more preferred.

[0187] Examples of aliphatic carboxylic acids include oxalic acid, malonic acid, succinic acid, glutaric acid, adipic acid, pimelic acid, sebacic acid, and maleic acid. Among these, adipic acid is preferred from the viewpoint of improving the effectiveness and cleaning performance of the invention.

[0188] Examples of carboxylic acids other than the aforementioned aminopolycarboxylic acids, amino acids, hydroxycarboxylic acids, and aliphatic carboxylic acids include monocarboxylic acids. Examples of monocarboxylic acids include formic acid, acetic acid, propionic acid, and butyric acid, which are lower (1 to 4 carbon atoms) aliphatic monocarboxylic acids.

[0189] As a carboxylic acid, it is preferably an amino acid, a hydroxycarboxylic acid, or an aliphatic carboxylic acid, more preferably cystine, cysteine, histidine, gluconic acid, glycolic acid, malic acid, tartaric acid, citric acid, or adipic acid, and even more preferably cysteine, gluconic acid, citric acid, or adipic acid.

[0190] Carboxylic acids can be used alone or in combination of two or more. The content of carboxylic acids is not particularly limited, but is preferably 0.1–35.0% by mass relative to the total mass of the CS solution, more preferably 1.0–35.0% by mass.

[0191] Phosphonic acids can be monophosphonic acids having one phosphonic acid group or polyphosphonic acids having two or more phosphonic acid groups. From the viewpoint of achieving superior cleaning performance, polyphosphonic acids having two or more phosphonic acid groups are preferred.

[0192] As polyphosphonic acids, compounds of general formulas 1 to 3 described in paragraphs 0013 to 0023 of International Publication No. 2013 / 162020, compounds described in paragraphs 0026 to 0036 of International Publication No. 2018 / 020878, and compounds described in paragraphs 0031 to 0046 of International Publication No. 2018 / 030006 (copolymers) may be cited, the contents of which are incorporated herein by reference.

[0193] Examples of polyphosphonic acids include ethylidene diphosphonic acid, 1-hydroxyethylidene-1,1'-diphosphonic acid (HEDPO: 1-hydroxyethane-1,1-diphosphonic acid), 1-hydroxypropylidene-1,1'-diphosphonic acid, 1-hydroxybutylidene-1,1'-diphosphonic acid, ethylaminobis(methylenephosphonic acid), dodecylaminobis(methylenephosphonic acid), nitrilotris(methylphosphonic acid) (NTPO: nitrilotris(methylphosphonic acid)), ethylenediaminebis(methylenebisphosphonic acid) (EDDPO: ethylenebisimino(methylene bisphosphonic acid)), 1,3-propanediaminebis(methylenephosphonic acid), ethylenediaminetetra(methylenephosphonic acid) (EDTPO: ethylene diamine tetra(methylene phosphonic acid)), ethylenediaminetetra(ethylidenephosphonic acid), and 1,3-propanediaminetetra(methylenephosphonic acid) (PDTMP: propylenediaminetetra(methylene phosphonic acid)). HEDPO is preferred. The following are listed: 1,2-diaminopropanetetra(methylenephosphonic acid), 1,6-hexanediaminetetra(methylenephosphonic acid), diethylenetriaminepenta(methylenephosphonic acid), diethylenetriaminepenta(methylenephosphonic acid), triethylenetetraaminehexa(methylenephosphonic acid), and triethylenetetraaminehexa(methylenephosphonic acid).

[0194] The number of phosphonic acid groups in the phosphonic acid is preferably 2 to 5, more preferably 2 to 4, and even more preferably 2 or 3. Furthermore, the number of carbon atoms in the phosphonic acid is preferably 12 or less, more preferably 10 or less, and even more preferably 8 or less. There is no particular limitation on the lower limit, but 1 or more is preferred.

[0195] Phosphonic acid can be used alone or in combination with two or more. The content of phosphonic acid is not particularly limited, but is preferably 0.001–5.0% by mass relative to the total mass of the CS solution, more preferably 0.01–2.0% by mass.

[0196] The organic acid is preferably of low molecular weight. More specifically, the molecular weight of the organic acid is preferably 600 or less, more preferably 450 or less, and even more preferably 300 or less. The lower limit of the molecular weight of the organic acid is not particularly limited, but is preferably 85 or more. Furthermore, the number of carbon atoms in the organic acid is preferably 15 or less, more preferably 12 or less, and even more preferably 8 or less. The lower limit of the number of carbon atoms in the organic acid is not particularly limited, but is preferably 1 or more.

[0197] Organic acids can be used alone or in combination of two or more. The content of organic acids is not particularly limited, but is preferably 0.1–40% by mass, more preferably 1.0–35% by mass, relative to the total mass of the CS solution.

[0198] From the perspective of superior cleaning performance, CS solution preferably contains carboxylic acid and phosphonic acid.

[0199] The pharmaceutical solution CS may contain other components besides those mentioned above. Examples of other components include pH adjusters (e.g., basic and acidic compounds), organic solvents (e.g., alcohol solvents, glycol solvents, glycol ether solvents, and ketone solvents), water-soluble polymers (e.g., the water-soluble polymers described in paragraphs 0043 to 0047 of Japanese Patent Application Publication No. 2016-171294), and oxidizing agents (e.g., peroxides, persulfides, percarbonates, their acids, and their salts).

[0200] The properties of the CS liquid are described in detail below.

[0201] The CS solution can be either alkaline or acidic. The pH of the CS solution is preferably between 0.10 and 4.00. The pH of the CS solution can be adjusted using the aforementioned pH adjuster. The pH of the CS solution can be measured using a known pH meter according to the method in JIS (Japanese Industrial Standard) Z8802-1984. The measurement temperature is set to 25°C.

[0202] The content (measured as ion concentration) of metals (e.g., Fe, Co, Na, Cu, Mg, Mn, Li, Al, Cr, Ni, Zn, Sn, and Ag) contained as impurities in the CS solution is preferably 5 ppm by mass or less, more preferably 1 ppm by mass or less. In the manufacture of cutting-edge semiconductor devices, higher purity CS solutions are expected to be required. Therefore, the metal content is further preferably less than 1 ppm by mass, i.e., less than ppb by mass, particularly preferably less than 100 ppb by mass, and most preferably less than 10 ppb by mass. As a lower limit, 0 is preferred.

[0203] The pharmaceutical solution CS may contain coarse particles, but it is preferable that their content is low. Coarse particles refer to particles with a diameter (particle size) of 0.03 μm or more when the particle shape is considered as a sphere. The coarse particles contained in the pharmaceutical solution CS refer to particles such as dust, dirt, organic solid matter, and inorganic solid matter contained as impurities in the raw materials; as well as particles such as dust, dirt, organic solid matter, and inorganic solid matter introduced as contaminants during the preparation of the pharmaceutical solution CS, which are equivalent to substances that ultimately do not dissolve in the pharmaceutical solution CS and exist as particulate matter.

[0204] Regarding the content of coarse particles in the pharmaceutical solution CS, the content of particles with a diameter of 0.1 μm or larger per 1 mL of pharmaceutical solution CS is preferably 10,000 or less, more preferably 5,000 or less. Regarding the lower limit, it is preferably 0 or more per 1 mL of pharmaceutical solution CS, more preferably 0.01 or more. The content of coarse particles present in the pharmaceutical solution CS can be determined in the liquid phase using a commercially available measuring device that uses a laser as a light scattering method for particle determination in liquid.

[0205] Materials for filters 21 and 37 include, for example, polyamides such as 6-nylon and 6,6-nylon, polyethylene, polypropylene, polystyrene, polyimide, polyamide-imide, and fluoropolymers. Polyimides and polyamide-imides may also have at least one group selected from the group consisting of carboxyl groups, salt-type carboxyl groups, and -NH- bonds. Fluoropolymers, polyimides, and polyamide-imides are excellent in terms of solvent resistance. Furthermore, from the viewpoint of adsorbing metal ions, polyamides such as 6-nylon and 6,6-nylon are preferred, and nylon is more preferred.

[0206] Filters 21 and 37 may also have structures that include both metal ion adsorption filters and organic impurity adsorption filters. As a metal ion adsorption filter, a filter capable of ion exchange is preferred. Examples of metal ion adsorption filters include, for instance, the polyimide and / or polyamide-imide porous membranes described in Japanese Patent Application Publication No. 2016-155121.

[0207] There are no particular limitations on organic impurity adsorption filters, and known organic impurity adsorption filters can be cited. From the viewpoint of improving the adsorption performance of organic impurities, it is preferable that the surface of the organic impurity adsorption filter has an organic framework that can interact with the organic impurities; in other words, the surface is modified by an organic framework that can interact with the organic impurities. Alternatively, filters in which activated carbon is fixed to nonwoven fabric, as described in Japanese Patent Application Publication Nos. 2002-273123 and 2013-150979, can also be used for organic impurity adsorption.

[0208] As for filters 21 and 37, there are no particular limitations as long as they are filters conventionally used for filtration purposes. Examples include filters made of fluoropolymers such as polytetrafluoroethylene (PTFE) and tetrafluoroethylene perfluoroalkyl vinyl ether copolymers, polyamide resins such as nylon, and polyolefin resins (including high-density or ultra-high molecular weight) such as polyethylene and polypropylene (PP). Among these materials, those selected from the group consisting of polyethylene, polypropylene (including high-density polypropylene), fluoropolymers (including PTFE and PFA), and polyamide resins (including nylon) are preferred, and fluoropolymers are more preferred.

[0209] There are no particular limitations on the semiconductor components used in the application of CS (Chemical Solid State) solutions. More specifically, examples of semiconductor components include logic LSI (Large Scale Integration), ASIC (Application Specific Integrated Circuit), FPGA (Field Programmable Gate Array), ASSP (Application Specific Standard Product), and microprocessors (such as CPU and GPU).In addition, examples include memory (such as DRAM (Dynamic Random Access Memory), SRAM (Static Random Access Memory), HMC (Hybrid Memory Cube), MRAM (Magnetic RAM), PCM (Phase Change Memory), ReRAM (Resistive RAM), FeRAM (Ferroelectric RAM), flash memory, etc.), LED (Light Emitting Diode), power devices, analog ICs (Integrated Circuits), such as DC (Direct Current) to DC converters, Insulated Gate Bipolar Transistors (IGBTs), such as accelerometers, pressure sensors, vibrators, gyroscopes, etc., MEMS (Micro-Electro-Mechanical Systems), such as GPS (Global Positioning System), FM (Frequency Modulation), NFC (Near Field Communication), RFEM (RF Expansion). Modules, including radio frequency expansion modules, MMIC (Monolithic Microwave Integrated Circuit), WLAN (Wireless Local Area Network), discrete components, BSI (Back Side Illumination), CIS (Contact Image Sensor), camera modules, passive components, SAW (Surface Acoustic Wave) filters, RF (Radio Frequency) filters, RFIPD (Radio Frequency Integrated Passive Devices), and BB (Broadband), etc.

[0210] There are no particular limitations on the composition of the semiconductors that constitute semiconductor devices. Examples of semiconductor compositions include diamond, silicon (Si), germanium (Ge), silicon-germanium (SiGe), silicon carbide (SiC), gallium arsenide (GaAs), gallium nitride (GaN), gallium oxide, and silicon-on-insulator (SOI).

[0211] In each of the above embodiments, each process is executed by any computer. Furthermore, any computer can execute these processes via a processor as hardware, a program as software, or a combination thereof. In this case, the processor is configured to cooperate with the program in executing the various processes in the above embodiments, and may also function as a unit or means in the above embodiments. Additionally, the order in which the processor executes the processes is not limited to the stated order and can be appropriately changed. Any computer can be a general-purpose computer, a special-purpose computer, a workstation, or other system capable of executing the various processes.

[0212] A processor can be composed of one or more hardware components, and the type of hardware is not limited. For example, a processor can be composed of programmable logic devices such as the exemplified CPU61 (CPU61PA and 61PD), MPU (Micro Processing Unit), FPGA (Field Programmable Gate Array), dedicated circuits for performing specific processes such as ASIC (Application Specific Integrated Circuit), GPU (Graphics Processing Unit), or NPU (Neural Processing Unit). Furthermore, the hardware can be a combination of different types of hardware. When multiple hardware components are configured to execute one or more processes of a processor, these multiple hardware components can exist in physically separate devices or in the same device. Additionally, in any embodiment, the order in which the processor executes the various processes is not limited to the above order and can be appropriately changed. Moreover, the hardware is composed of circuits composed of combined semiconductor elements and other circuitry.

[0213] Furthermore, the program can be software such as firmware or microcode. Additionally, the program can be, for example, a group of program modules, each of which can be implemented by a processor configured to perform the corresponding function. The program can be program code or multiple code segments stored in one or more non-transitory computer-readable media (e.g., storage media or other storage devices). The program can also be segmented and stored in multiple non-transitory computer-readable media existing in physically separate devices. Program code or code segments can represent any combination of procedures, functions, subroutines, routines, subroutines, modules, software packages, classes, or instructions, data structures, or program statements. Program code or code segments can be connected to other code segments or hardware circuitry by sending or receiving information, data, variables, parameters, or memory contents.

[0214] The technology of the present invention can also be appropriately combined with the various embodiments and / or variations described above. Furthermore, the present invention is not limited to the embodiments described above; various structures can be employed as long as they do not depart from the spirit of the invention. In addition to programs, the technology of the present invention also relates to storage media for non-transitory storage of programs, and computer program products including programs.

[0215] The foregoing descriptions and illustrations are detailed explanations relating to the parts of this invention and are merely one example of the invention. For example, the descriptions related to the structure, function, operation, and effect described above are examples of the structure, function, operation, and effect relating to the parts of this invention. Therefore, it is possible to delete unnecessary parts, add new elements, or replace them in the foregoing descriptions and illustrations without departing from the spirit of the invention. In addition, to avoid complexity and facilitate understanding of the parts of this invention, descriptions related to technical common sense that do not require special explanation in the foregoing descriptions and illustrations have been omitted.

[0216] In this specification, "A and / or B" is synonymous with "at least one of A and B". That is, "A and / or B" means that it can be only A, only B, or a combination of A and B. In addition, in this specification, when "and / or" is used to refer to more than three items, the same concept applies as "A and / or B".

[0217] All documents, patent applications and technical standards described herein are incorporated herein by reference to the same extent as if they were specifically and separately described and incorporated by reference to separate documents, patent applications and technical standards.

[0218] The above description is merely a preferred embodiment of this application and is not intended to limit this application. Various modifications and variations can be made to this application by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the protection scope of this application.

Claims

1. A method for manufacturing a pharmaceutical solution for semiconductor manufacturing, characterized in that, include: In the circulating filtration process, the semiconductor manufacturing solution is taken out of the liquid tank, filtered in a circulating flow path connected to the liquid tank, and then returned to the liquid tank. The measurement process involves measuring values ​​related to the concentration of particles contained in the liquid medicine within the circulating flow path. as well as In the filling process, when the measured value in the measurement process meets the preset quality conditions, the medicine solution is supplied from the circulation flow path to the filling flow path, and the medicine solution is filled into the product container. The filling flow path uses a multi-stage filling flow path, which includes a main flow path connected to the circulation flow path, multiple branch flow paths branching from the main flow path, and multiple valves for adjusting the flow rate of the liquid medicine flowing through the multiple branch flow paths respectively. The liquid medicine is alternately filled into multiple product containers through the multiple branch flow paths. In the filling process, the following speed control is performed: When the flow rate of the liquid medicine flowing through the multiple branch flow paths per unit time is set as the filling speed for filling the product container with the liquid medicine, And when the flow rate of the drug solution per unit time through the main flow path is set as the flow velocity of the main flow path, The filling speed of each of the multiple branch flow paths is controlled by controlling the opening degree of the multiple valves, so that the flow rate of the main flow path remains constant. Specifically, when the filling speed of at least one of the plurality of branch flow paths is increased, it is gradually increased; when the filling speed of one or more other branch flow paths is slowed down in accordance with the increase in the filling speed, it is gradually decreased.

2. The method for manufacturing a semiconductor manufacturing solution as described in claim 1, characterized in that: During a single filling of the liquid medicine into one of the product containers, the absolute value of the acceleration of the filling speed is constant as the filling speed gradually increases and decreases with the opening and closing of the valve.

3. The method for manufacturing a semiconductor manufacturing solution as described in claim 1, characterized in that: During a single filling of the liquid medicine into one of the product containers, the absolute value of the acceleration of the filling speed changes as the filling speed gradually increases and decreases with the opening and closing of the valve.

4. The method for manufacturing a semiconductor manufacturing solution as described in claim 3, characterized in that, During the filling period, the intervals that gradually increase the filling speed and the intervals that gradually decrease the filling speed respectively include a first interval and a second interval. In the first interval, the absolute value of the acceleration gradually increases from 0 to the maximum value, and in the second interval, the absolute value of the acceleration gradually decreases from the maximum value to 0.

5. The method for manufacturing a semiconductor manufacturing solution as described in claim 4, characterized in that: The first interval is the first half of the interval that gradually increases the filling speed and the interval that gradually decreases the filling speed, and the second interval is the second half of the interval that gradually increases the filling speed and the interval that gradually decreases the filling speed.

6. The method for manufacturing a semiconductor manufacturing solution as described in claim 4, characterized in that: During the filling period, the intervals that gradually increase the filling speed and the intervals that gradually decrease the filling speed also include the relay intervals with constant acceleration between the first interval and the second interval.

7. The method for manufacturing a semiconductor manufacturing solution as described in claim 1, characterized in that: The filling flow path also includes a drainage flow path for the drug solution. Before filling the first product container with the liquid medicine, the liquid medicine is discharged from the drain path.

8. The method for manufacturing a semiconductor manufacturing solution as described in claim 1, characterized in that: Use a drain tank instead of the first product container.

9. The method for manufacturing a semiconductor manufacturing solution as described in claim 1, characterized in that: The measurements taken immediately after the start of filling the first product container with the liquid medicine are excluded.

10. The method for manufacturing a semiconductor manufacturing solution as described in claim 1, characterized in that: The multiple branch flow paths mentioned above are two in number.

11. The method for manufacturing a semiconductor manufacturing solution as described in claim 1, characterized in that: The filling speed varies continuously.

12. A method for manufacturing a semiconductor manufacturing solution according to any one of claims 1-11, characterized in that: The drug solution contains a surfactant.

13. An apparatus for manufacturing a semiconductor manufacturing solution, characterized in that: It includes a processor for controlling various processes, each process comprising: In the circulating filtration process, the semiconductor manufacturing solution is taken out of the liquid tank, filtered in a circulating flow path connected to the liquid tank, and then returned to the liquid tank. The measurement process involves measuring values ​​related to the concentration of particles contained in the drug solution within the circulating flow path; and In the filling process, when the measured value in the measurement process meets the preset quality conditions, the medicine solution is supplied from the circulation flow path to the filling flow path, and the medicine solution is filled into the product container. The filling flow path uses a multi-stage filling flow path, which includes a main flow path connected to the circulation flow path, multiple branch flow paths branching from the main flow path, and multiple valves for adjusting the flow rate of the liquid medicine flowing through the multiple branch flow paths respectively. The liquid medicine is alternately filled into multiple product containers through the multiple branch flow paths. The processor In the filling process, the following speed control is performed: When the flow rate of the liquid medicine flowing through the multiple branch flow paths per unit time is set as the filling speed for filling the product container with the liquid medicine, And when the flow rate of the drug solution per unit time through the main flow path is set as the flow velocity of the main flow path, The filling speed of each of the multiple branch flow paths is controlled by controlling the opening degree of the multiple valves, so that the flow rate of the main flow path remains constant. Specifically, when the filling speed of at least one of the plurality of branch flow paths is increased, it is gradually increased; when the filling speed of one or more other branch flow paths is slowed down in accordance with the increase in the filling speed, it is gradually decreased.

Images