Information processing device, determination method, and water treatment system

Abstract

The information processing apparatus of the present application has: an acquisition section (110) that acquires a liquid quality value of discharge of a cleaning liquid used for cleaning an object; a reading section (120) that reads a reference liquid quality value serving as a reference from a database (400); a determination section (130) that determines whether or not a semiconductor device manufactured in a manufacturing process including a process of cleaning the object is a good product based on the liquid quality value acquired by the acquisition section (110) and the reference liquid quality value read by the reading section (120); and a notification section (140) that notifies a determination result determined by the determination section (130).

Description

Technical Field

[0001] This invention relates to an information processing device, a determination method, and a water treatment system. Background Technology

[0002] Whether a manufactured semiconductor device is a qualified product is determined based on the inspection results of a given inspection performed on the semiconductor device. In addition, in the manufacturing of semiconductor devices, in the process of removing the film by contacting the processing liquid with the laminated film of the semiconductor wafer, a method for switching the processing liquid to be contacted based on the concentration of the substance in the contacting processing liquid is considered (for example, see Patent Document 1).

[0003] Existing technical documents Patent documents Patent Document 1: Japanese Patent Application Publication No. 2006-120708 Summary of the Invention

[0004] The technical problem that the invention aims to solve As mentioned above, the process of inspecting semiconductor devices for conformity can take several months to obtain a result. Therefore, if defective products are manufactured in the early stages of the manufacturing process, they will remain on the production line for several months, leading to a decrease in yield. Furthermore, the method described in Patent Document 1 cannot directly determine whether a semiconductor device is conforming based on the processing liquid.

[0005] The purpose of this invention is to provide an information processing device, a judgment method, and a water treatment system capable of early detection of defects in semiconductor devices.

[0006] Technical solutions for solving technical problems The information processing apparatus of the present invention includes: The acquisition unit acquires the liquid quality value of the drained cleaning fluid used to clean the object. The reading unit reads the reference liquid chromatography-mass spectrometry (LC-MS) values ​​from the database as a baseline. The determination unit determines, based on the liquid quality value obtained by the acquisition unit and the reference liquid quality value read by the reading unit, whether the semiconductor device manufactured in the manufacturing process, including the cleaning process of the object, is a qualified product; and The notification department notifies the judgment result determined by the aforementioned judgment department.

[0007] In addition, the determination method of the present invention performs the following processing: Obtain the liquid quality value of the drained cleaning fluid used to clean the object; Read the reference liquid chromatography-mass spectrometry values ​​stored in the database; Based on the obtained liquid chromatography-mass spectrometry (LCS) values ​​and the baseline LCS values ​​retrieved from the aforementioned database, it is determined whether the semiconductor device manufactured in the manufacturing process, including the cleaning of the object, is a qualified product; and The above-mentioned judgment results shall be notified.

[0008] In addition, the water treatment system of the present invention has: A cleaning device that uses pure water treated by a water treatment device to clean the object; A measuring device for measuring the liquid quality value of the drained cleaning fluid used by the cleaning device in cleaning the object. A database that stores reference liquid chromatography values ​​as a baseline; and Information processing device The aforementioned information processing device has: The acquisition unit acquires the liquid quality value measured by the aforementioned measuring instrument; The reading unit reads the aforementioned reference liquid chromatography values ​​from the aforementioned database; The determination unit determines, based on the liquid quality value obtained by the acquisition unit and the reference liquid quality value read by the reading unit, whether the semiconductor device manufactured in the manufacturing process, including the cleaning process of the object, is a qualified product; and The notification department notifies the judgment result determined by the aforementioned judgment department.

[0009] Invention Effects In this invention, defects in semiconductor devices can be detected at an early stage. Attached Figure Description

[0010] Figure 1 This is a diagram illustrating the water treatment system according to the first embodiment.

[0011] Figure 2 It means Figure 1 A diagram showing an example of the positioning of the cleaning device.

[0012] Figure 3 It means Figure 1 A diagram showing an example of the constituent elements of an information processing device.

[0013] Figure 4 It means Figure 1 The diagram shows an example of information stored in a database.

[0014] Figure 5 It is used for explanation Figure 1 A flowchart illustrating an example of the method for setting the reference liquid quality value in the determination method of the information processing device shown.

[0015] Figure 6It is used for explanation Figure 1 The flowchart illustrates an example of a method for determining whether a semiconductor device manufactured using cleaned components as constituent elements is a qualified product, within the determination method of the information processing apparatus shown.

[0016] Figure 7 This is a graph representing an example of time series data, showing the baseline liquid quality value as time series data and the liquid quality values ​​included in the liquid quality information.

[0017] Figure 8 This is a diagram illustrating the water treatment system according to the second embodiment.

[0018] Figure 9 It means Figure 8 A diagram showing an example of the constituent elements of an information processing device.

[0019] Figure 10 It means Figure 8 The diagram shows an example of the input and output of the learning model.

[0020] Figure 11 It is used for explanation Figure 8 A flowchart illustrating an example of the learning phase processing in the determination method of the information processing device shown.

[0021] Figure 12 It is used for explanation Figure 8 A flowchart illustrating an example of the reasoning phase processing in the determination method of the information processing device shown. Detailed Implementation

[0022] Hereinafter, embodiments of the present invention will be described with reference to the accompanying drawings.

[0023] (First Implementation) Figure 1 This is a diagram illustrating the water treatment system according to the first embodiment. (Example) Figure 1 As shown, the water treatment system 10 according to this embodiment includes an information processing device 100, a cleaning device 200, a measuring device 300, and a database 400. The information processing device 100 and the database 400 can be directly connected. Alternatively, the information processing device 100 and the database 400 can also be connected via a communication network. Furthermore, the database 400 can also be installed within the information processing device 100.

[0024] The cleaning apparatus 200 includes a cleaning section 210 and a discharge pipe 230. The cleaning apparatus 200 is installed on a semiconductor production line in a semiconductor manufacturing plant.

[0025] The cleaning unit 210 cleans the object to be cleaned. The object can be an electronic component (finished product) such as a semiconductor device manufactured in the semiconductor device manufacturing process. Alternatively, the object can be a component 220 constituting the manufactured semiconductor device. In this embodiment, the cleaning unit 210, acting as component 220, cleans wafers (including bare silicon, oxide films, patterned wafers, and semiconductors other than silicon) that are materials used in semiconductor element manufacturing. The cleaning unit 210 performs wet cleaning on component 220 using a chemical solution and pure water. The cleaning method in the cleaning unit 210 can also be a single-tank or multi-tank batch cleaning method where multiple wafers are simultaneously immersed in a processing tank for cleaning. Alternatively, the cleaning method in the cleaning unit 210 can be a single-wafer cleaning method where wafers are rotated one by one and cleaning is performed by spraying processing liquid through nozzles. The cleaning unit 210 uses liquids such as water (including pure water), DHF (Dolute Hydrogen Fluoride), SC1 (a mixed alkaline solution of H2O, ammonia, and H2O2), SC2 (a mixed acid solution of H2O, HCl, and H2O2), IPA (isopropanol), and TMAH (tetramethylammonium hydroxide). Furthermore, the cleaning process performed by the cleaning unit 210 is included in the semiconductor device manufacturing process. This manufacturing process includes multiple cleaning processes.

[0026] The discharge pipe 230 discharges the liquid used for cleaning the component 220 to the outside of the cleaning device 200. The discharge pipe 230 can also separate the liquid used for cleaning the component 220 into organic and inorganic liquids. Furthermore, the discharge pipe 230 can also separate the liquid used for cleaning the component 220 into acidic and alkaline liquids.

[0027] The measuring device 300 measures the liquid quality of the discharged liquid from the discharge pipe 230. The liquid quality measured by the measuring device 300 includes the resistivity, pH, metal concentration, drug concentration, particulate concentration, ion concentration, total organic carbon (TOC) concentration, discharge volume, or any combination thereof. The measuring device 300 can also measure the liquid quality of the discharged liquid flowing in the discharge pipe 230 within the cleaning device 200. Additionally, the measuring device 300 can also measure the liquid quality of the discharged liquid flowing in the discharge pipe 230 outside the cleaning device 200. Furthermore, if the discharge pipe 230 is a discharge pipe that discharges acidic and alkaline liquids separately, the measuring device 300 may also be equipped with measuring devices for measuring the liquid quality of acidic and alkaline liquids. The measuring device 300 sends liquid quality information, including the measured values ​​(liquid quality values), to the information processing device 100. In addition to the measured liquid quality values, the liquid quality information sent by the measuring device 300 may also include information indicating the type of liquid quality measured (such as resistivity, pH, metal concentration, drug concentration, particle concentration, ion concentration, and discharge volume). The measuring device 300 may also send liquid quality information to the information processing device 100, which represents time-series data of the liquid quality measured from the start time of the cleaning member 220.

[0028] Figure 2 It means Figure 1 The figure shows an example of the positioning of the cleaning device 200. In a water treatment system having a pretreatment device 20, a primary pure water production device 30, and a secondary pure water production device 40 as a subsystem, treated water processed by the secondary pure water production device 40 is supplied to the cleaning device 200 for cleaning the components 220.

[0029] The pretreatment unit 20 includes a sand filter, an activated carbon unit, a softening unit, and a decarbonation unit. The structure of the pretreatment unit 20 is not limited to this structure. For example, the pretreatment unit 20 may also be a structure that does not include some of the aforementioned components. Furthermore, the pretreatment unit 20 may also be a structure that includes other components in addition to the aforementioned components. The sand filter uses sand as the filter material to filter the water being treated. The activated carbon unit uses porous activated carbon to further filter the water treated from the sand filter. The softening unit uses an ion exchange resin to remove hardness components from the water treated from the activated carbon unit. The decarbonation unit, for example, uses a decarbonation membrane that allows gas to pass through to degas (remove) carbon dioxide from the water treated from the softening unit.

[0030] The primary pure water production equipment 30 includes: a reverse osmosis (RO) membrane unit 31 for supplying water to be treated, an ultraviolet oxidation unit 32, an ion exchange unit 33, and a membrane degassing unit 34. The structure of the primary pure water production equipment 30 is not limited to this structure. In the primary pure water production equipment 30, for example, the arrangement order of the ultraviolet oxidation unit 32, the ion exchange unit 33, and the membrane degassing unit 34 is not particularly limited. The ion exchange unit 33 is equipped with an SB-P (multilayer bed regenerative ion exchange unit). The membrane degassing unit 34 removes gaseous components such as oxygen contained in the treated water.

[0031] The secondary pure water production equipment 40 includes: a primary pure water tank 41, a heat exchanger 42, an ultraviolet oxidation device 43, an ion exchange device 44, a membrane degassing device 45, and an ultrafiltration membrane device 46. The treated water (ultrapure water) produced by the secondary pure water production equipment 40 is supplied to the cleaning device 200. The primary pure water tank 41 stores primary pure water produced by the primary pure water production equipment 30. The heat exchanger 42 adjusts the temperature of the primary pure water supplied from the primary pure water tank 41. The ultraviolet oxidation device 43 irradiates the treated water, whose temperature has been adjusted by the heat exchanger 42, with ultraviolet light, decomposing the organic matter contained in the treated water. The ion exchange device 44 removes ionic components from the treated water treated by the ultraviolet oxidation device 43. The ion exchange device 44 is a non-regenerable cartridge filter processor filled with a mixed bed of cation exchange resin and anion exchange resin. The membrane degassing device 45 removes gases such as dissolved oxygen and carbon dioxide contained in the treated water treated by the ion exchange device 44. The filtration membrane device 46 is an ultrafiltration (UF) membrane device that removes particulate matter contained in the treated water after it has been treated by the membrane degassing device 45. Furthermore, the structure of the secondary pure water production equipment 40 is not limited to the above-described structure. For example, a hydrogen peroxide removal device that decomposes hydrogen peroxide generated by the ultraviolet oxidation device 43 may be provided between the ultraviolet oxidation device 43 and the ion exchange device 44. Examples of impurities removed by the secondary pure water production equipment 40 include: metals, ions, TOC (Total Organic Carbon), urea, dissolved oxygen, hydrogen peroxide, and particulate matter.

[0032] Figure 3 It means Figure 1 A diagram illustrating an example of the constituent elements of the information processing apparatus 100 shown. (See diagram for example.) Figure 3 As shown, Figure 1 The information processing device 100 shown includes an acquisition unit 110, a reading unit 120, a determination unit 130, a notification unit 140, a reference liquid quality setting unit 150, and a writing unit 160. Furthermore, in Figure 3 In, it is shown Figure 1 The main elements of the information processing apparatus 100 shown in this embodiment are among the constituent elements.

[0033] The acquisition unit 110 acquires the liquid quality information sent by the measuring device 300. The acquisition unit 110 outputs the acquired liquid quality information to the determination unit 130. The timing of the acquisition unit 110 acquiring the liquid quality information is not specifically defined. Furthermore, the measurement unit for the drained liquid whose liquid quality value is shown in the liquid quality information acquired by the acquisition unit 110 is a given unit in the manufacturing process of component 220. For example, the acquisition unit 110 may also acquire the liquid quality value of the drained liquid measured by the measuring device 300 throughout the entire process of a batch. Additionally, for example, the acquisition unit 110 may also acquire the individual liquid quality values ​​of the drained liquid measured by the measuring device 300 in each of multiple cleaning processes performed in a manufacturing process.

[0034] The reading unit 120 reads the reference liquid quality values ​​stored in the database 400. For example, the reading unit 120 uses information indicating the type of liquid quality included in the liquid quality information acquired by the acquisition unit 110 as a search keyword, and reads the reference liquid quality value corresponding to the information indicating the type of liquid quality from the database 400. The reading unit 120 outputs the read reference liquid quality value to the determination unit 130. If the reference liquid quality values ​​stored in the database 400 are time-series data with the start time of the cleaning member 220 as the starting reference, the reading unit 120 reads the time-series data of the reference liquid quality values ​​as the reference liquid quality values.

[0035] The determination unit 130 determines whether a semiconductor device manufactured using the component 220 cleaned by the cleaning apparatus 200 as a component is a qualified product based on the liquid quality value included in the liquid quality information output from the acquisition unit 110 and the reference liquid quality value output from the reading unit 120. Specifically, if the difference between the liquid quality value included in the liquid quality information output from the acquisition unit 110 and the reference liquid quality value output from the reading unit 120 is below a given threshold, the determination unit 130 determines that the semiconductor device manufactured using the component 220 cleaned by the cleaning apparatus 200 as a component is a qualified product. Conversely, if the difference between the liquid quality value included in the liquid quality information output from the acquisition unit 110 and the reference liquid quality value output from the reading unit 120 exceeds a given threshold, the determination unit 130 determines that the semiconductor device manufactured using the component 220 cleaned by the cleaning apparatus 200 as a component may be a defective product. When the acquisition unit 110 acquires the liquid quality value of each of the multiple cleaning processes, if at least one liquid quality value included in the liquid quality information output from the acquisition unit 110 differs from the reference liquid quality value output from the reading unit 120 by a given threshold, the determination unit 130 may also determine that the semiconductor device manufactured using the component 220 cleaned by the cleaning apparatus 200 as a component may be a defective product. Therefore, if the liquid quality value of the drained liquid after cleaning the component 220 deviates from the given reference range, it can be inferred that the component 220 has some kind of abnormality.

[0036] If the liquid quality information output from the acquisition unit 110 contains liquid quality values ​​that are time series data as described above, and the reference liquid quality values ​​stored in the database 400 are time series data with the start time of the cleaning member 220 as the starting reference, the determination unit 130 compares the approximate curves (hereinafter referred to as liquid quality curves) of each time series data with the start time of cleaning as the starting reference.

[0037] The threshold used by the determination unit 130 is a preset value that determines the reference range for whether a semiconductor device manufactured with component 220 as a constituent is a qualified product. The larger the threshold, the wider the range within which a semiconductor device manufactured with component 220 as a constituent is judged to be a qualified product. Conversely, the smaller the threshold, the narrower the range within which a semiconductor device manufactured with component 220 as a constituent is judged to be a qualified product. This threshold is set according to the type of liquid to be measured. Furthermore, this threshold can also be set according to the usage of the cleaning device 200, the type and properties of component 220, and the type of liquid used for cleaning. The threshold used by the determination unit 130 can also be "0". Additionally, when using the aforementioned liquid quality curve, the threshold represents the degree of deviation (curve) from the liquid quality curve based on the reference liquid quality value. When using a liquid quality curve, the threshold can also be a value that changes as time passes from the start of the measurement. For example, it can be set such that the threshold is set to a larger value closer to the start of the measurement, and becomes a smaller value as time passes from the start of the measurement.

[0038] The determination unit 130 outputs the determination result to the notification unit 140. For example, if the determination result is "a semiconductor device manufactured using the component 220 cleaned by the cleaning apparatus 200 as a component is a qualified product," the determination unit 130 may also output a value of "0." Conversely, if the determination result is "a semiconductor device manufactured using the component 220 cleaned by the cleaning apparatus 200 as a component may be a defective product," the determination unit 130 may also output a value of "1." In this case, the meaning of the "0" and "1" values ​​output from the determination unit 130 is self-evident to the notification unit 140.

[0039] The notification unit 140 notifies the determination result output from the determination unit 130. The notification method of the notification unit 140 can also be display, sound output, printing, or transmission to other devices. For example, if the determination result output from the determination unit 130 indicates that the semiconductor device manufactured using component 220 as a constituent element may be a defective product, the notification unit 140 can also issue a warning, such as outputting an alarm sound or illuminating an indicator light. By issuing a warning through the notification unit 140, a defect in component 220 can be promptly notified.

[0040] The reference liquid quality setting unit 150 sets reference liquid quality values ​​based on the liquid quality value of the drained cleaning solution used in the manufacturing process of the semiconductor device determined to be a qualified product by the determination unit 130, according to the type of liquid quality (resistivity, pH, metal concentration, chemical concentration, particle concentration, ion concentration, drain volume, etc.). This inspection data refers to data obtained by inspecting whether a semiconductor device manufactured with the cleaned component 220 as a component is a qualified product. The reference liquid quality setting unit 150 may also set the reference liquid quality value as the best value (e.g., low particle concentration and ion concentration) among the liquid quality drained after cleaning the component 220 of a semiconductor device determined to be a defective product, based on past inspection data. Alternatively, the reference liquid quality setting unit 150 may set the reference liquid quality value as the worst value (e.g., high particle concentration and ion concentration) among the liquid quality drained after cleaning the component 220 of a semiconductor device determined to be a qualified product, based on past inspection data. Alternatively, the reference fluid quality setting unit 150 can use the deviation of the fluid quality in the drain after cleaning of the component 220 of the semiconductor device that was determined to be a defective product, based on inspection data from previous inspections, and a given calculation algorithm, to calculate and set the reference fluid quality value. The reference fluid quality setting unit 150 outputs the set reference fluid quality value to the writing unit 160.

[0041] The writing unit 160 writes the reference liquid quality value output from the reference liquid quality value setting unit 150 into the database 400. At this time, the writing unit 160 writes the reference liquid quality value into the database 400 according to the type of liquid quality. By writing the reference liquid quality value set by the reference liquid quality value setting unit 150 based on inspection data and liquid quality value into the database 400, the writing unit 160 can use the reference liquid quality value corresponding to the liquid quality of the drain after cleaning the component 220 as a suitable criterion for determining whether the semiconductor device manufactured with the component 220 as a component is a qualified product.

[0042] Database 400 stores reference liquid quality values ​​related to the drainage after cleaning component 220, according to the type of liquid quality. Figure 4 It means Figure 1 This is a diagram illustrating an example of the information stored in database 400. (See diagram for example.) Figure 4 As shown, in Figure 1The database 400 shown stores the types of measured liquids and their corresponding reference liquid values. The type of measured liquid indicates the type of liquid for which the reference liquid value setting unit 150 has set the reference liquid value. Examples of measured liquid types include, for instance, resistivity, pH, metal concentration, drug concentration, particle concentration, ion concentration, and discharge volume. The reference liquid value is the value set by the reference liquid value setting unit 150 according to the type of measured liquid. This data is written by the writing unit 160.

[0043] The following is about Figure 1 The determination method in the information processing device 100 shown will be explained. Figure 5 It is used for explanation Figure 1 A flowchart illustrating an example of the method for setting the reference liquid quality value in the determination method of the information processing device 100 shown.

[0044] First, the acquisition unit 110 acquires liquid quality information including the measured value (liquid quality value) of the drained liquid measured by the measuring device 300 (step S1). The reference liquid quality value setting unit 150 stores the liquid quality value acquired by the acquisition unit 110. At this time, the reference liquid quality value setting unit 150 pre-stores the liquid quality value so as to be able to identify which component 220 the drained liquid was used to clean when the liquid quality value was measured. For example, the reference liquid quality value setting unit 150 may also assign identification information to the liquid quality value acquired by the acquisition unit 110, which can identify the date and time of acquisition, location, cleaning device, and component 220 to be cleaned, and store it. In addition, the reference liquid quality value setting unit 150 stores the type of liquid quality value acquired by the acquisition unit 110 in correspondence with the liquid quality value. In addition, the reference liquid quality value setting unit 150 may also use the liquid quality value acquired by the acquisition unit 110 as the reference liquid quality value and store it in a time sequence (e.g., a graph showing the time change of the liquid quality value) during the cleaning process of cleaning the component 220. The moment that serves as the reference for this time series data is the start moment of the cleaning process for the component 220 that is being cleaned.

[0045] Next, the acquisition unit 110 acquires characteristic inspection data of the semiconductor device manufactured with component 220 as a constituent element (step S2). Here, the characteristic inspection data of the semiconductor device acquired by the acquisition unit 110 is data representing the result of characteristic inspection performed on the semiconductor device, and is data that at least includes information that can identify whether the semiconductor device is a qualified product. At this time, it is necessary to identify which liquid quality value among the liquid quality values ​​stored in the reference liquid quality value setting unit 150 corresponds to the characteristic inspection data acquired by the acquisition unit 110. For example, when the acquisition unit 110 acquires the liquid quality value, the reference liquid quality value setting unit 150 pre-stores identification information of the component 220 that has been drained and cleaned with that liquid quality value. When the acquisition unit 110 acquires the characteristic inspection data, it acquires the identification information of the component 220 constituting the semiconductor device that is the object of the characteristic inspection data, and the reference liquid quality value setting unit 150 compares these identification information with each other.

[0046] If the correspondence between the liquid quality value and the characteristic inspection data can be identified, the following result can be obtained: whether the semiconductor device manufactured using the component 220, which will be the object of cleaning based on the obtained liquid quality value, as a constituent element is a qualified product. That is, the liquid quality value corresponds to the information indicating whether the semiconductor device is a qualified product. Based on this correspondence, the reference liquid quality value setting unit 150 sets a reference liquid quality value (step S3). Then, the writing unit 160 writes the reference liquid quality value set by the reference liquid quality value setting unit 150 into the database 400 (step S4).

[0047] Figure 6 It is used for explanation Figure 1 This is a flowchart illustrating an example of a method for determining whether a semiconductor device manufactured using the cleaned component 220 as a constituent element is a qualified product, as part of the determination method of the information processing apparatus 100 shown. The processing described below is based on the use of... Figure 5 The flowchart shown illustrates the process performed after the reference liquid mass value is set.

[0048] First, the acquisition unit 110 acquires liquid quality information including the measured value (liquid quality value) of the drained liquid measured by the measuring device 300 (step S11). At this time, the acquisition unit 110 can also acquire liquid quality information including time series data of the measured value (liquid quality value) of the drained liquid measured by the measuring device 300. The time that serves as the reference for the time series data is the start time of the cleaning process of the component 220 to be cleaned. Then, the reading unit 120 uses the type of measured liquid quality contained in the liquid quality information as a search keyword to retrieve and read the reference liquid quality value stored in the database 400 corresponding to the type of measured liquid quality (step S12). Next, the determination unit 130 compares the liquid quality value contained in the liquid quality information acquired by the acquisition unit 110 with the reference liquid quality value read by the reading unit 120 (step S13). The determination unit 130 determines whether the difference between the liquid quality value and the reference liquid quality value is below a threshold (step S14).

[0049] If the difference between the liquid quality value and the reference liquid quality value is below a threshold, the determination unit 130 determines that the semiconductor device manufactured using the component 220 as a constituent element is a qualified product (step S15). On the other hand, if the difference between the liquid quality value and the reference liquid quality value exceeds the threshold, the determination unit 130 determines that the semiconductor device manufactured using the component 220 as a constituent element may be a defective product (step S16). Then, the notification unit 140 notifies the determination unit 130 of the determination result (step S17).

[0050] If the reference liquid quality values ​​stored in the database 400 and the liquid quality values ​​contained in the liquid quality information acquired by the acquisition unit 110 are time series data as described above, in step S13, the determination unit 130 compares the time series data of the reference liquid quality values ​​with the time series data of the liquid quality values ​​contained in the liquid quality information acquired by the acquisition unit 110. Figure 7 This is a graph representing an example of time-series liquid chromatography-mass spectrometry (LCMS) values ​​used as baseline data and the LCMS values ​​included in the LCMS information. Figure 7 In the above, as part of the liquid quality information acquired by the acquisition unit 110, two time-series data of liquid quality values ​​that the determination unit 130 determined to be abnormal (possibly indicating a non-conforming product) are shown. Figure 7 In the example shown, as the cleaning time progresses during the cleaning process, the baseline liquid quality value gradually decreases (the liquid quality gradually improves). Furthermore, in Figure 7 In the text, values ​​below a threshold difference from the reference solution's mass spectrometry value are displayed in gray. Additionally, Figure 7 The vertical axis of the chart shown represents the values ​​of any type of liquid chromatography-mass spectrometry (LCMS), not the actual measured values. The reference point for this time series data is the start time of the cleaning process for component 220, which is the object of cleaning. Figure 7In the anomaly result 1 shown, the liquid chromatography-mass spectrometry (LC-MS) value deviated from the threshold around time 20 seconds. Additionally, in... Figure 7 In the abnormal result 2 shown, the liquid quality value deviates from the threshold during the period from 0 to 7 seconds. In this case, the determination unit 130 determines that the semiconductor device manufactured with component 220 as a constituent element may be a defective product.

[0051] Furthermore, multiple thresholds can be set as described above. In this case, for example, "threshold A," "threshold B," and "threshold C" can be set such that they satisfy the relationship of threshold A < threshold B < threshold C, and the determination unit 130 can also compare the difference between the liquid quality value included in the liquid quality information obtained from the acquisition unit 110 and the reference liquid quality value read by the reading unit 120 with thresholds A to C respectively. If the difference is below threshold A, the determination unit 130 can determine that the semiconductor device manufactured using component 220 as a constituent element is a qualified product. In addition, if the difference is greater than threshold A and less than threshold B, the determination unit 130 can determine that the probability of the semiconductor device manufactured using component 220 as a constituent element is a defective product is 30%. In addition, if the difference is greater than threshold B and less than threshold C, the determination unit 130 can determine that the probability of the semiconductor device manufactured using component 220 as a constituent element is a defective product is 70%. Furthermore, if the difference exceeds the threshold C, the determination unit 130 can determine that the probability of the semiconductor device manufactured using the component 220 as a constituent element being a defective product is 90%. In this way, multiple thresholds of different values ​​can also be set, and the determination unit 130 can determine the probability of the semiconductor device being a qualified product or a defective product based on the result of comparing the difference with each threshold.

[0052] In this method, the quality value of the liquid drained after cleaning component 220 is used to determine whether a semiconductor device manufactured using component 220 as a constituent element is a qualified product. A pre-set reference liquid quality value is used in the determination. This reference liquid quality value is set using the liquid quality value of the drained liquid after cleaning component 220 and the results of a characteristic inspection of the semiconductor device. If the difference between the liquid quality value of the drained liquid after cleaning component 220 and the reference liquid quality value is below a threshold, the semiconductor device manufactured using component 220 as a constituent element is determined to be a qualified product. Conversely, if the difference between the liquid quality value of the drained liquid after cleaning component 220 and the reference liquid quality value exceeds a threshold, the semiconductor device manufactured using component 220 as a constituent element is determined to be potentially a defective product. Therefore, it is possible to detect semiconductor device defects early without waiting for the results of a characteristic inspection performed on the semiconductor device after cleaning.

[0053] (Second Implementation) Figure 8This is a diagram illustrating the water treatment system according to the second embodiment. (Example) Figure 8 As shown, the water treatment system 11 according to this embodiment includes an information processing device 101, a cleaning device 200, a measuring device 300, and a learning model 500. The cleaning device 200 and the measuring device 300 are the same as in the first embodiment. The information processing device 101 and the learning model 500 can also be directly connected. Alternatively, the information processing device 101 and the learning model 500 can be connected via a communication network. Furthermore, the learning model 500 can also be housed within the information processing device 101.

[0054] Figure 9 It means Figure 8 A diagram illustrating an example of the constituent elements of the information processing apparatus 101 shown. (See diagram for example.) Figure 9 As shown, Figure 8 The information processing apparatus 101 shown includes an acquisition unit 110, a determination unit 131, a notification unit 140, and a learning model generation unit 171. The acquisition unit 110 and the notification unit 140 are the same as in the first embodiment. Furthermore, in Figure 9 In, it is shown Figure 8 The main elements of the information processing apparatus 101 shown in this embodiment are among the constituent elements.

[0055] The learning model generation unit 171 stores the liquid chromatography-mass spectrometry (LCS) values ​​corresponding to the inspection data obtained from checking whether a semiconductor device manufactured using the cleaned component 220 as a constituent element is a qualified product, as training data in advance, and performs machine learning on the LCS values ​​to generate a learning model 172. The correspondence between the inspection data and the LCS values ​​is the same as that in the first embodiment. The learning model generation unit 171 is used in the so-called learning phase (training phase) of machine learning.

[0056] Learning model 500 is a learning model generated by learning model generation unit 171. Learning model 500 uses training data generated by learning model generation unit 171 to output a judgment result indicating whether a semiconductor device is a qualified product. The semiconductor device is manufactured using a component 220 that has undergone drain cleaning as a constituent element, and the drain has a liquid quality value. Learning model 500 may, for example, have a neural network structure composed of multiple neurons interconnected. A neuron is a component that performs a given operation on multiple inputs and outputs a value as the result of the operation. Learning model 500 is stored in a storage device (not shown). The learning method in learning model 500 can use general methods for generating learning models. Figure 10 It means Figure 8 The diagram shows an example of the input and output of the learning model 500. Figure 10 As shown, Figure 8The learning model 500 shown is as follows: if a liquid quality value is input, the output indicates whether the semiconductor device manufactured using component 220 as a constituent element is a qualified product. The learning model 500 is set according to the type of liquid quality measured and the type of cleaning fluid. Alternatively, the learning model 500 can also be set according to the usage of the cleaning device 200, the type of component 220, and its properties.

[0057] The determination unit 131 inputs the liquid quality value contained in the liquid quality information acquired by the acquisition unit 110 into the learning model 500, and obtains a determination result from the learning model 500. This determination result indicates whether the semiconductor device manufactured using the component 220, which has been cleaned by draining liquid with the acquired liquid quality value as a constituent element, is a qualified product. The determination unit 131 is used in the so-called inference stage of machine learning.

[0058] The following is about Figure 8 The determination method in the information processing device 101 shown will be explained. Figure 11 It is used for explanation Figure 8 A flowchart illustrating an example of the learning phase processing in the determination method of the information processing device 101 shown.

[0059] First, the acquisition unit 110 acquires liquid quality information including the measured value (liquid quality value) of the discharged liquid measured by the measuring device 300 (step S21). The learning model generation unit 171 stores the liquid quality value acquired by the acquisition unit 110. At this time, the learning model generation unit 171 pre-stores the liquid quality value so as to be able to identify which component 220 was cleaned by the discharged liquid when the liquid quality value was measured. For example, the learning model generation unit 171 may also assign identification information to the liquid quality value acquired by the acquisition unit 110, which can identify the date and time of acquisition, location, cleaning device, and component 220 cleaned, and store it. In addition, the learning model generation unit 171 stores the type of liquid quality value acquired by the acquisition unit 110 in association with the liquid quality value.

[0060] Next, the acquisition unit 110 acquires the characteristic inspection data of the semiconductor device (step S22). Here, the characteristic inspection data of the semiconductor device acquired by the acquisition unit 110 is data representing the result of a characteristic inspection performed on the semiconductor device, and is data that at least includes information that can identify whether the semiconductor device is a qualified product. At this time, it is necessary to identify which liquid quality value among the liquid quality values ​​stored in the learning model generation unit 171 corresponds to the characteristic inspection data acquired by the acquisition unit 110. For example, when the acquisition unit 110 acquires the liquid quality value, the learning model generation unit 171 pre-stores the identification information of the component 220 constituting the semiconductor device that has been drained and cleaned by the liquid quality value. When the acquisition unit 110 acquires the characteristic inspection data, it acquires the identification information of the component 220 constituting the semiconductor device that is the object of the characteristic inspection data, and the learning model generation unit 171 compares these identification information with each other.

[0061] If the correspondence between the liquid quality value and the characteristic inspection data can be identified, the following result can be obtained: whether the semiconductor device manufactured using the component 220, which is the object to be cleaned based on the obtained liquid quality value, as a constituent element is a qualified product. That is, the liquid quality value corresponds to the information indicating whether the semiconductor device manufactured using the component 220 as a constituent element is a qualified product. The learning model generation unit 171 causes the learning model 500 to learn this correspondence to generate a learning model (step S23). At this time, the learning model generation unit 171 causes the learning model 500 corresponding to the type of liquid quality value to learn. In the case where the learning model 500 is not set separately according to the type of liquid quality value, but is a learning model that determines the qualification of the semiconductor device based on the liquid quality value and the type of liquid quality value, the learning model generation unit 171 causes the learning model 500 to learn the liquid quality value, the type of liquid quality value, and the information indicating whether the semiconductor device is a qualified product.

[0062] Figure 12 It is used for explanation Figure 8 A flowchart illustrating an example of the reasoning stage processing in the determination method of the information processing apparatus 101 shown. The processing described below is performed after... Figure 11 The learning phase, as illustrated in the flowchart, will proceed afterward.

[0063] First, the acquisition unit 110 acquires liquid quality information including the measured value (liquid quality value) of the discharged liquid measured by the measuring device 300 (step S31). Next, the determination unit 131 inputs the liquid quality value contained in the liquid quality information into the learning model 500 corresponding to the type of liquid quality value (step S32). If the learning model 500 is not set separately according to the type of liquid quality value, the determination unit 131 inputs the liquid quality value along with the type of liquid quality value into the learning model 500. If the learning model 500 outputs a result with the input liquid quality value, the determination unit 131 acquires the result output by the learning model 500, which is the determination result (step S33). This determination result indicates whether the semiconductor device is a qualified product. Next, the notification unit 140 notifies the determination unit 131 of the determination result acquired (step S34).

[0064] Alternatively, the determination result notified in step S34 and the result of the subsequent characteristic check of the semiconductor device can be used to enable the learning model generation unit 171 to relearn the learning model 500 in order to improve the determination accuracy.

[0065] Alternatively, a learning model can be set up such that, based on the characteristic inspection results of the semiconductor device and the measured liquid quality value, the semiconductor device manufactured with component 220 as a constituent element learns and responds to the detected defect under certain liquid quality values. In this case, a system can also be constructed that infers the content of the semiconductor device's defects based on the measured liquid quality value.

[0066] In this method, the quality value of the liquid drained after cleaning the components 220 constituting the semiconductor device is used to determine whether the semiconductor device is a qualified product. A learning model 500 is used in this determination. When the learning model 500 is input with the quality value of the liquid drained after cleaning the components 220 constituting the semiconductor device, it outputs a determination result indicating whether the semiconductor device is a qualified product. Therefore, it is possible to detect semiconductor device defects early without waiting for the results of characteristic checks performed on the semiconductor device after cleaning. Furthermore, by using the learning model 500, it is possible to handle patterns of multiple quality values ​​and obtain useful determination results.

[0067] Although the above description illustrates how each component undertakes its respective function (processing), this allocation is not limited to the above method. Furthermore, the above method is merely an example regarding the structure of the components and is not limited thereto. Additionally, the above embodiments can be combined in any arbitrary combination.

[0068] The processing performed by the information processing devices 100 and 101 can also be carried out by logic circuits made separately according to their respective purposes. Alternatively, a computer program (hereinafter referred to as a program) describing the processing content can be recorded on a recording medium that can be read by the information processing devices 100 and 101, allowing the information processing devices 100 and 101 to read and execute the program recorded on that recording medium. The recording media that the information processing devices 100 and 101 can read include portable recording media such as floppy disks, optical disks, DVDs (Digital Versatile Discs), CDs (Compact Discs), Blu-ray discs, and USB (Universal Serial Bus) memory, as well as memory such as ROMs (Read Only Memory), RAMs (Random Access Memory), and HDDs (Hard Disc Drives) respectively built into the information processing devices 100 and 101. The program recorded on the recording medium is read by a CPU (not shown) respectively installed in the information processing devices 100 and 101, and the same processing as described above is performed under the control of the CPU. Here, the CPU operates as a computer that executes the program read from the recording medium containing the program.

[0069] Some of the above embodiments may also be described as follows, but are not limited to the following description.

[0070] (Appendix 1) An information processing device, comprising: The acquisition unit acquires the liquid quality value of the drained cleaning fluid used to clean the object. The learning model generation unit pre-stores the liquid quality values ​​of the inspection data as training data, performs machine learning on the liquid quality values, and generates a learning model that takes the liquid quality values ​​as input and outputs a judgment result indicating whether the semiconductor device is a qualified product. The inspection data is obtained by checking whether the semiconductor device manufactured in the manufacturing process including the cleaning process of the object after cleaning is a qualified product. The determination unit inputs the liquid quality value obtained by the acquisition unit into the learning model and obtains a determination result from the learning model. This determination result indicates whether the semiconductor device, after draining and cleaning based on the obtained liquid quality value, is a qualified product. The notification department notifies the determination results obtained by the aforementioned determination department.

[0071] Although the present invention has been described above with reference to embodiments, the present invention is not limited to the above embodiments. Various modifications that can be understood by those skilled in the art regarding the structure and details of the present invention can be made within the scope of the present invention.

[0072] This application claims priority based on Japanese Patent Application No. 2024-042241, filed on March 18, 2024, the entire disclosure of which is incorporated herein by reference.

[0073] Explanation of reference numerals in the attached figures 10, 11 Water Treatment Systems 20 Pretreatment Unit 30 Pure water production equipment 31 Reverse osmosis membrane unit 32, 43 Ultraviolet Oxidation Device 33, 44 Ion exchange devices 34, 45 Membrane degassing device 40 Secondary Pure Water Production Equipment 41 Pure water tank 42 Heat Exchanger 46 Membrane Filtration Device Information processing devices 100 and 101 110 Acquisition Department 120 Reading Department Judgment Departments 130 and 131 140 Notification Department 150 Standard Fluid Quality Setting Unit 160 Writing Department 171 Learning Model Generation Department 200 Cleaning Device 210 Cleaning Department 220 components 230 Discharge Pipe 300 measuring instrument 400 Database 500 learning model.

Claims

1. An information processing device, comprising: The acquisition unit acquires the liquid quality value of the drained cleaning fluid used to clean the object. The reading unit reads the reference liquid chromatography-mass spectrometry (LC-MS) values ​​from the database as a baseline. The determination unit determines, based on the liquid quality value obtained by the acquisition unit and the reference liquid quality value read by the reading unit, whether the semiconductor device manufactured in the manufacturing process, including the cleaning process of the object, is a qualified product; and The notification department notifies the determination result determined by the determination department.

2. The information processing apparatus according to claim 1, wherein, If the difference between the liquid quality value and the reference liquid quality value is below a given threshold, the determination unit determines that the semiconductor device is a qualified product; if the difference between the liquid quality value and the reference liquid quality value exceeds the threshold, the determination unit determines that the semiconductor device manufactured in this manufacturing process may be a defective product.

3. The information processing apparatus according to claim 1 or 2, wherein, The information processing device has: The reference liquid quality setting unit sets the reference liquid quality value based on the liquid quality value of the drained cleaning fluid used in the manufacturing process of the semiconductor device determined to be a qualified product by the determination unit; and The writing unit writes the reference fluid mass value set by the reference fluid mass value setting unit into the database.

4. The information processing apparatus according to claim 1 or 2, wherein, The manufacturing process includes multiple cleaning steps. The acquisition unit acquires the liquid quality value of the discharged liquid measured by the measuring device in each of the plurality of cleaning processes. If, in each of the plurality of cleaning processes, the difference between at least one liquid quality value of the drain fluid measured by the measuring instrument and the reference liquid quality value exceeds a threshold, the determination unit determines that the semiconductor device may be a defective product.

5. The information processing apparatus according to claim 1 or 2, wherein, The liquid quality value obtained by the acquisition unit is measured in the drain solution used for cleaning the wafers or patterned wafers used in the semiconductor device.

6. The information processing apparatus according to claim 2, wherein, If the determination unit determines that the semiconductor device may be a defective product, the notification unit issues a given warning.

7. The information processing apparatus according to claim 1 or 2, wherein, The acquisition unit acquires time-series data of the liquid quality values ​​based on the start time of the cleaning process. The reading unit reads the time-series data of the reference liquid chromatography-mass spectrometry values ​​from the database. The determination unit compares the time series data of the liquid quality value obtained by the acquisition unit with the time series data of the reference liquid quality value read from the database by the reading unit, and determines whether the semiconductor device manufactured in the manufacturing process, including the cleaning process of the object, is a qualified product based on the comparison result.

8. A determination method, wherein the determination method performs the following processing: Obtain the liquid quality value of the drained cleaning fluid used to clean the object; Read the reference liquid chromatography-mass spectrometry values ​​stored in the database; Based on the obtained liquid chromatography-mass spectrometry (LCS) value and the baseline LCS value read from the database, it is determined whether the semiconductor device manufactured in the manufacturing process, including the cleaning of the object, is a qualified product; and The notification will be issued based on the aforementioned determination result.

9. A water treatment system, comprising: A cleaning device that uses pure water treated by a water treatment device to clean the object; A measuring device for measuring the liquid quality value of the drained cleaning fluid used by the cleaning device in cleaning the object. A database that stores benchmark liquid chromatography values ​​as a reference. as well as Information processing device The information processing device has: The acquisition unit acquires the liquid quality value measured by the measuring instrument; The reading unit reads the reference liquid chromatography value from the database; The determination unit determines whether a semiconductor device manufactured in a manufacturing process, including the cleaning process of the object, is a qualified product based on the liquid quality value obtained by the acquisition unit and the reference liquid quality value read by the reading unit. as well as The notification department notifies the determination result determined by the determination department.

Images