Method and system for constructing chemical plating solution preparation model based on RNAs

Abstract

The invention relates to a construction method and system of a chemical plating solution preparation model based on RNAS, and belongs to the technical field of artificial intelligence control and chemical plating reduction fusion. The method comprises the following steps: acquiring formula data of a chemical plating solution to be prepared; obtaining a size data set of a plating layer area of a to-be-processed plating piece and electrode potential data of oxidation reduction strength of to-be-chemically plated metal, and calculating the modeling film thickness of the plating piece; actual chemical plating operation is carried out, and the actually measured film thickness of a plated part is measured; the plating piece modeling film thickness and the plating piece actual measurement film thickness which meet the preset relation are used for training of an RNAS model system so as to classify chemical plating solution formula data; according to the method, the formula data of the chemical plating solution is classified into an available formula or an unavailable formula by constructing the RNAS model, the use quantity of a reducing agent in the preparation of the chemical plating solution is strictly managed according to the standard, and the effect of accurately and rapidly screening the formula data of the chemical plating solution is achieved.

Description

Technical Field

[0001] The present invention relates to a method and system for constructing a chemical plating solution formulation model based on RNAS, belonging to the technical field of the integration of artificial intelligence control and chemical plating reduction. Background Art

[0002] The metal coating deposition of semiconductor electronic components is a unique characterization technology that is sensitive to the local electrochemical-physical environment structure of metal ions during the chemical plating process, and has extensive applications in the fields of physics, electrochemistry, and high-end chemical plating manufacturing. During the production operation of the plating production equipment, there are many and complex influencing factors on the metal plating deposition for the metal ions and reducing agent components in the chemical plating solution under the electrochemical-physical environment conditions; therefore, due to the interference of various influencing factors, there will be various differences in the actually deposited metal coatings. In the light case, if the additive or reducing agent component is too high, it will lead to difficult control of the coating area, too fast precipitation and roughness of the coating. In the severe case, if the additive or reducing agent component is too low, it will lead to insufficient density of the coating or even the inability to precipitate the required metal coating.

[0003] Currently, the chemical plating solution formulation method uses the usual golden section method. Its research and development requires a large amount of solution formulation work, and at the same time, it is necessary to have a chemical plating equipment module for processing and evaluating plating samples, with a long research and development cycle, and it is difficult to meet the current demand for the high-speed development of various precision chemical plating solutions for semiconductor electronic components; therefore, exploring and creating a systematic method for quickly formulating chemical plating solutions with a short cycle and high precision has become a very important topic. Summary of the Invention

[0004] In order to improve the formulation accuracy of chemical plating solutions and shorten the research and development cycle, the present invention provides a method and system for constructing a chemical plating solution formulation model based on RNAS, and the technical solutions are as follows: The first object of the present invention is to provide a method for constructing a chemical plating solution configuration model, and the method includes: Step 1: Obtain the formula data of the chemical plating solution to be formulated, and construct an initial formula data set; Step 2: Obtain the size data set of the coating area of the workpiece to be processed and the electrode potential data of the oxidation-reduction strength of the metal to be chemically plated, calculate the mass of the chemical plating layer metal according to the general formula of the oxidation-reduction reaction, and calculate the modeled film thickness of the workpiece in combination with the size data h 建模厚度 ; Step 3: Randomly extract the chemical plating solution formula data from the initial formula data set, use the chemical plating equipment module to perform actual chemical plating operations on the workpiece to be processed, and measure the actual film thickness of the workpiece h 实测厚度 ; Step 4: For the modeled film thickness of the workpieceh 建模厚度 and the actually measured film thickness of the plated part h 实测厚度 Compare them. When the two satisfy the following relationship: h 实测厚度 (100% - 5%) ≤ h 建模厚度 ≤ h 实测厚度 (100% + 5%) Then save the modeled coating thickness of the plated part h 建模厚度 The corresponding electroless plating solution formulation data for training the RNAS model system; If the above relationship is not satisfied, correct the electroless plating solution formulation data, perform the actual electroless plating operation again using Step 3 and measure the actually measured film thickness of the plated part h 实测厚度 ; until the above relationship is satisfied, save the electroless plating solution formulation data for training the RNAS model; Step 5: Using the modeled film thickness of the plated part h 建模厚度 and the actually measured film thickness of the plated part h 实测厚度 as inputs, train the RNAS model to classify the electroless plating solution formulation data and establish a classification model for the electroless plating solution formulation data.

[0005] Optionally, the workpieces to be processed include: precision terminal types, lead frame types, wafer chip types, and precision ornaments.

[0006] Optionally, the metal coatings include: Monomeric metal coatings, including Au, Ag, Ni, Sn, Cu, Pd, Rh, Pt; Alloy coatings, including Au-Ni, Pd-Ni, Ni-P, W-Ni, Ag-Sn, Au-Sn, Rh-Ru, and Pt-Rh or ternary and higher alloy metals such as Au-W-Ni.

[0007] Optionally, the electroless plating solution formulation data includes: Metal salts, which can be either monomeric coatings or alloy coatings; Complexing agents, including malonic acid, succinic acid, cyclohexanedicarboxylic acid; Stabilizers, including citric acid, tartaric acid, ethylenediaminetetraacetic acid; Electrolytes, including sodium ethylenediaminetetraacetate, ammonium formate, ammonium acetate; pH regulators, including sulfuric acid, sulfamic acid, methanesulfonic acid; Reducing agents, including sodium borohydride, sodium sulfite, thiourea dioxide.

[0008] Optionally, the operation system for training the RNAS model includes sensitive condition parameter modeling, and the sensitive condition parameters include: types of electroless plating solutions, redox potentials of metal ions, redox potentials of reducing agents, specific gravity of electroless plating solutions, solution temperature, electroless plating time, surface area of the plating area, density of the plating layer metal, and flow rate of electroless plating solutions.

[0009] Optionally, it further includes constructing a dataset of insensitive condition parameters, and the insensitive condition parameters include in the electroless plating equipment module: the tank body, the pump for transporting the electroless plating solution, the length and diameter of the pipeline for transporting the electroless plating solution, the semiconductor electronic workpiece to be processed, and the mold for electroless plating.

[0010] The second object of the present invention is to provide a construction system for an electroless plating solution configuration model, which is used to implement the construction method of the electroless plating solution configuration model described in any one of the above, and the system includes: A memory 310, which is used to store formula data and thickness data; A dataset processing system 200, which is used to process the matching relationship between formula data and thickness data, including: an electroless plating reduction plating operation module 210, an electroless plating equipment module 220, an electroless plating film thickness calculation module 230, and a plating film thickness threshold discrimination module 240; the electroless plating reduction plating operation module 210 is used to obtain formula parameters and calculate the mutual relationship between the electrode potentials of metal ions and reducing agents in the electroless plating solution; the electroless plating equipment module 220 is used to perform actual electroless plating operations; the electroless plating film thickness calculation module 230 is used to calculate the modeling thickness; the plating film thickness threshold discrimination module 240 is used to judge the matching between the modeling thickness and the measured thickness; An RNAS model training system 100, which is used to train and classify electroless plating solution formulas.

[0011] Optionally, the RNAS model training system 100 includes: A training module 110, which is used for comparative training of the modeled film thickness of the workpiece and the measured film thickness of the workpiece; A stability generation module 120, which is used to process the ordered arrangement and trend of the measured coating thickness; An electroless plating reduction component ratio operation module 130, which is used for simulation operation based on the formula ratio of the electroless plating solution corresponding to the measured coating thickness; A model morphology discrimination module 140, which is used for discrimination between the operation result of the simulated electroless plating solution formula ratio and its RNAS modeling morphology.

[0012] The third object of the present invention is to provide an electroless plating solution preparation method based on the RNAS model, which uses the model constructed by any of the above methods for formula discrimination, including: Accurately weighing the weights of each component; Keep the concentration of the basic components unchanged and use the reducing agent as a variable parameter; Calculate the modeled thickness and compare it with the measured thickness; Classify and discriminate the formulations that meet the threshold conditions through the RNAS model.

[0013] Optionally, the discrimination results of the RNAS model include: Available formulation: The modeled thickness data meets the topography threshold range; Unavailable formulation: The modeled thickness data does not meet the topography threshold range.

[0014] Optionally, the thickness of the film layer is measured by an X-ray film thickness analyzer.

[0015] The beneficial effects of the present invention are: The present invention improves the electro-chemical reaction model and constructs the RNAS network operation model for training. The output of the constructed RNAS network processes the modeled coating thickness data obtained by calculating the electroless plating solution formulation that meets the threshold range as an "available formulation"; while the output of the modeled coating thickness data obtained by calculating the electroless plating solution formulation that exceeds the threshold range is processed as an "unavailable formulation", and strictly controls the high-end quality requirements for the preparation of the electroless plating solution according to this standard.

[0016] In addition, a plating part coating thickness threshold discrimination module is used to simulate and monitor the appearance, shape, color, and overall topography characteristics of the semiconductor device after electroless plating. Further, by meeting the standard with a minimum threshold of -5% of the coating specification value and a maximum threshold of +5% of the coating specification value, the accuracy of the RNAS network training is improved, and the training distribution of the electroless plating solution formulation associated with the coating thickness is quickly realized, so as to obtain the preferred standard of the RNAS operation mechanism model, and achieve the screening results of accurately and quickly identifying available or unavailable formulations and obtaining the preparation method of the electroless plating solution.

[0017] Based on the screening results of the present invention, the available formulation conditions of the electroless plating solutions of various different monomer metal salts, binary metal salts, and ternary or more metal salts can be quickly established during the actual production process. Furthermore, the optimal electroless plating control method for semiconductor electronic components can be obtained through the electroless plating equipment module. It can not only quickly achieve the accuracy of predicting the local coating thickness of semiconductor electronic components, greatly reduce the error between the coating thickness of the actual electroless plating product, but also further improve the electroless plating device technology, promote the improvement of the uniformity of the coating thickness of semiconductor electronic components, and meet the requirements for high quality of high-end semiconductor electronic component products. Description of the Drawings

[0018] To more clearly illustrate the technical solutions in the embodiments of the present invention, the following will briefly introduce the drawings required for the description of the embodiments. Obviously, the drawings in the following description are only some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can be obtained based on these drawings.

[0019] Figure 1 It is a schematic diagram of the structure of the electroless plating solution preparation model based on RNAS of the present invention.

[0020] Figure 2 It is a flowchart of the construction method and system of the electroless plating solution preparation model based on RNAS of the present invention.

[0021] Figure 3 It is a schematic diagram of the experimental wafer of the semiconductor electronic component to be processed in the third embodiment of the present invention.

[0022] Figure 4 It is a schematic diagram of the nickel plating area of each unit in the experimental wafer of the semiconductor electronic component to be processed in the third embodiment of the present invention.

[0023] Figure 5 It is a schematic diagram of the platinum plating area of each unit in the experimental wafer of the semiconductor electronic component to be processed in the third embodiment of the present invention.

[0024] Figure 6 It is a schematic diagram of the change trend of the electroless plating platinum thickness of the RNAS model with the concentration of the reducing agent in the third embodiment of the present invention.

[0025] 100 - RNAS model training system; 110 - training module; 120 - stability generation module; 130 - electroless plating reducing agent component ratio calculation module; 140 - model morphology discrimination module; 200 - data set processing system; 210 - electroless plating reduction plating operation module; 220 - electroless plating equipment module; 230 - electroless plating film thickness calculation module; 240 - plating film thickness threshold discrimination module; 300 - RANS simulation model; 310 - memory; 10 - copper plating area; 20 - silver plating area; 21 - center of the electroless plating area. Detailed implementation manners

[0026] To make the objectives, technical solutions, and advantages of the present invention clearer, the following will further describe the embodiments of the present invention in detail with reference to the drawings.

[0027] Embodiment 1: This embodiment provides a method for constructing a chemical plating solution configuration model. This model is based on the Reynolds Averaged Navier-Stokes (RANS) Turbulence Models. The construction process of the RANS model mainly includes two parts, namely the construction of the chemical oxidation-reduction plating data processing system and the RANS Reynolds-averaged turbulence model training system.

[0028] First, the construction of the chemical oxidation-reduction plating data processing system is based on the chemical plating solution formula, the chemical plating equipment module, and the semiconductor electronic components to be processed. Data on each component in these chemical plating solution formulas, the sensitivity data and dimensions of various components in the chemical plating equipment module, the dimensions of the semiconductor electronic components, the plating layer area, etc. are collected. The specific steps are as follows: Step 1: Obtain data on each component of the chemical plating solution to be prepared, the sensitivity data and dimensions of various components in the chemical plating equipment module, the dimensional data of the processed semiconductor electronic components, the plating layer thickness and area.

[0029] Among them, the chemical plating solution to be prepared includes components with the following concentrations: Table 1 Data on Components of the Chemical Plating Solution to be Prepared

[0030] The sensitivity data and dimensions of various components in the chemical plating equipment module do not change during the chemical plating process and are not specifically listed.

[0031] Step 2: Obtain the quantity of each component of the chemical plating solution to be prepared. Among them, the usage quantities of items 1 to 5 remain unchanged, and only the sixth item, sodium borohydride as a reducing agent, is a variable parameter.

[0032] Step 3: Based on the usage quantity parameters of the platinum chemical plating solution prepared from each group in Step 1, calculate the modeled plating layer thickness h 建模厚度 , and the calculation method is based on the electrode potential E (Pt n+ / Pt) of the platinum metal electrode, with the unit of V (volt).

[0033] Its general formula is expressed as: Pt(IV) → (II)[PtCl 6 2- +2e - =[PtCl 4 2- +2Cl - E = 0.68 V Pt(II) → (0)[PtCl 4 2- + 2e - ​​​= Pt + 4Cl - E = 0.755 V The electrode potential E (B n+ / B) of the reducing agent sodium borohydride is expressed in volts (V).

[0034] B(IV) → (Ⅲ)[BH ４ - —e - →B(OH) 3 +2H 2 E = -1.3 V The general formula for its chemical reduction reaction is: Na 2 (PtCl 6 ) + Na(BH 4 ) + 6H 2 O→Pt + B(OH) 3 + 6HCl In the formula, sodium hexachloroplatinate and sodium borohydride are based on an oxidation-reduction reaction. The metal ions in the solution are reduced by the sodium borohydride reducing agent and plated and deposited on the surface of the semiconductor electronic component to obtain the modeled coating thickness h 建模厚度 .

[0035] Step 4: Randomly select the standard conditions of the coating thickness obtained according to the electroless plating solution formula. Using the prepared platinum electroless plating solution, perform actual electroless plating operations on the semiconductor workpiece to be processed through the electroless plating module. After the electroless plating is completed, measure the actual coating thickness data of the workpiece using an X-ray film thickness analyzer h 实测厚度 .

[0036] Step 5: Compare the modeled coating thickness of the workpiece h 建模厚度 and the actual coating thickness of the workpiece h 实测厚度 . When the two satisfy the following relationship: h 实测厚度 (100% - 5%) ≤ h 建模厚度 ≤ h 实测厚度 (100% + 5%) Then save the electroless plating solution formula data corresponding to the modeled coating thickness of the workpiece h 建模厚度 for training the RNAS model system.

[0037] If the above relationship is not satisfied, adjust the formula data of the standard formulated electroless plating solution according to the actual coating thickness of the workpiece h ​实测厚度 Make corrections and calculate the standard conditions of the formula data for the corrected electroless plating solution, then return to step 3 to re-run the simulation modeling of the film thickness of the plated part. h 建模厚度 And re-obtain new measured film thickness data of the plated film according to the electroless plating conditions for modeling through the electroless plating equipment module. h 实测厚度 And further determine the modeled film thickness of the corrected plated part. h 建模厚度 Whether it meets the threshold range until the modeled film thickness of the plated part h 建模厚度 Meets the above relationship, then this set of data can be used in the RNAS training system.

[0038] Step 6: Import the modeled coating thickness h 建模厚度 Calculated according to the electroless plating solution formula and the measured film thickness of the plated part h 实测厚度 Into the trained RNAS operation system. The results of the comparison and processing by the RNAS operation system are divided into two categories: Usable formula: The processing result of the modeled coating thickness data calculated according to the electroless plating solution formula meets the morphology threshold range; Unusable formula: The processing result of the modeled coating thickness data calculated according to the electroless plating solution formula does not meet the morphology threshold range.

[0039] The Navier-Stokes equation of the RNAS model is a mathematical equation describing turbulent flow. It is an equation obtained by averaging the fluid flow and can describe the motion law of turbulence. The equation is as follows:

[0040] Among them, Is the term describing the change of flow rate with time, which is 0 in steady-state calculation (steady calculation); Represents the term of the flow effect carried by the flow itself; Is the term representing the influence of the pressure difference on the flow; Represents the term of the effect of smoothing the flow distribution due to the fluid viscosity (diffusion), which has the function of stable numerical analysis calculation; Is the term describing the external force, which is 0 if there is no external force.

[0041] By using the Navier-Stokes turbulent flow operation equation of the RNAS model, for the prepared electroless platinum plating solution, simulate and calculate the flow condition in the electroless plating equipment module to obtain accurate turbulent simulation results, and combine with the electrode potential of the metal used in the electroless plating solution to accurately calculate and obtain the data of the modeled electroless plating metal layer thickness under different concentrations of reducing agents.h 建模厚度 Processing result

[0042] Example Two This example provides a construction system for an electroless plating solution configuration model, which is used to implement the construction method of the electroless plating solution configuration model described in Example One. The system of this example includes: A memory 310 for storing recipe data and thickness data; A data set processing system 200 for processing the matching relationship between recipe data and thickness data, including: an electroless plating reduction plating operation module 210, an electroless plating equipment module 220, an electroless plating film thickness calculation module 230, and a plated film thickness threshold discrimination module 240; the electroless plating reduction plating operation module 210 is used to obtain recipe parameters and calculate the mutual relationship between the electrode potentials of metal ions and reducing agents in the electroless plating solution; the electroless plating equipment module 220 is used to perform actual electroless plating operations; the electroless plating film thickness calculation module 230 is used to calculate the modeled thickness; the plated film thickness threshold discrimination module 240 is used to judge the matching of the modeled thickness and the measured thickness; An RNAS model training system 100 for training and classifying electroless plating solution recipes.

[0043] The RNAS model training system 100 includes: A training module 110 for comparative training of the modeled film thickness of the workpiece and the measured film thickness of the workpiece; A stability generation module 120 for processing the ordered arrangement and trend of the measured coating thickness; An electroless plating reduction component ratio operation module 130 for simulation operations based on the recipe ratio of the electroless plating solution corresponding to the measured coating thickness; A model morphology discrimination module 140 for discriminating based on the operation result of the simulated electroless plating solution recipe ratio and its RNAS modeled morphology.

[0044] Example Three This example introduces the technical effects of the present invention in combination with the actual electroless plating solution configuration process.

[0045] The experimental wafer of the semiconductor electronic component to be processed in this example is as Figure 3 shown, made of copper alloy material, 60×60mm, thickness 0.127mm; there are 16 basic units in total. The copper plating area of each unit is as Figure 4 shown in 10 with a size of 11.5mm×11.5mm and an area of 132.25mm 2 ; then the total double-sided area of nickel plating for one semiconductor electronic component is 16×132.25×2 = 4232mm 2 ; in addition, the silver plating area of each unit is asFigure 5 The size of the 20 in 2 is 7.3 mm × 7.3 mm, and its area is 53.29 mm 2 .

[0046] The center position of the coating thickness detection in the coating area of the semiconductor electronic component in this embodiment is determined according to Figure 3 and Figure 5 the center black dots ● in the electroless plating area of each semiconductor plating unit in

[0047] The coordinates are as follows: 4 (14.6, 7.3), 3 (14.6, 21.9), 2 (14.6, 36.5), 1 (14.6, 51.1) in column I; 4 (21.9, 7.3), 3 (21.9, 21.9), 2 (21.9, 36.5), 1 (21.9, 51.1) in column II; 4 (36.5, 7.3), 4 (36.5, 21.9), 2 (36.5, 36.5), 1 (36.5, 51.1) in column III, and 4 (51.1, 7.3), 3 (51.1, 21.9), 2 (51.1, 36.5), 1 (51.1, 51.1) in column IV.

[0047] According to the 3D design drawing of the semiconductor plating part as shown in Figure 3 , the electroless plating specifications are: 1.5 μm ≤ nickel plating thickness ≤ 3.5 μm, and 0.3 μm ≤ platinum plating thickness ≤ 0.7 μm on the outermost surface; in the following data processing of the coating thickness of the plating part, the nickel plating data at the bottom layer is not summarized and discussed, and only the surface platinum plating thickness is specifically analyzed.

[0048] The thickness of the nickel coating is based on the central value of 2.5 μm within the design drawing range, and the thickness of the platinum coating is based on the central value of 0.5 μm within the design drawing range. The actual plating process is carried out using an electroless plating equipment module. The plating conditions are calculated based on the quantity parameters of the platinum electroless plating solution prepared for each component to model the coating thickness. h 建模厚度 , and the calculation method is represented by the electrode potential E (Pt n+ / Pt) of metallic platinum, with the unit of V (volt).

[0049] Its general formula is: Pt(IV) → (II)[PtCl 6 2- +2e - =[PtCl 4 2- +2Cl - E = 0.68 V Pt(II) → (0)[PtCl4 2- + 2e - = Pt + 4Cl - E = 0.755 V The electrode potential E (B n+ / B) of the reducing agent sodium borohydride is expressed in volts (V).

[0050] B(IV) → (Ⅲ)[BH ４ - —e - →B(OH) 3 +2H 2 E = -1.3 V Its general chemical reduction reaction formula is: Na 2 (PtCl 6 ) + Na(BH 4 ) + 6H 2 O→Pt + B(OH) 3 + 6HCl In the formula, sodium hexachloroplatinate and sodium borohydride are based on an oxidation-reduction reaction, and their strong reducing agents are used to reduce metal ions in the solution and deposit them by plating on the surface of semiconductor electronic components to obtain the thickness of the modeling coating h 建模厚度 .

[0051] The electroless platinum plating samples of semiconductor electronic components are measured by an X-ray film thickness analyzer, and the coating thickness data h 实测厚度 are shown in Table 2.

[0052] Table 2 Coating Thickness Data of Electroless Platinum Plating Samples

[0053] As can be seen from Table 1, the electroless plating solution of Example 1 performs actual electroless plating operations on the semiconductor electronic components to be processed through an electroless plating equipment module, and the thickness of the obtained platinum coating h 实测厚度 ranges from 0.505 to 0.509 μm, and the central value of the design drawing range, that is, the thickness of the modeling coating h 建模厚度 is 0.5 μm. Comparing the coating thickness error is 0.8% - 1.8%. Comparing with the overall quality control threshold of semiconductor electronic components of -5.0% to 5.0%, it meets the modeling threshold conditions of the modeling conditions. Therefore, the formulation data of the electroless platinum plating solution of Example 1 and its corresponding coating thickness data can be used for the construction of the RNAS model.

[0054] ​​Further, for the electroless plating solution with an additive concentration of 0.5 g / L in Example 1, the thickness of the electrolessly plated platinum coating h 实测厚度 and the modeled coating thickness h 建模厚度 data of 0.5 μm were processed by the analog quantity processing of the operation network of the RNAS model to obtain the simulation data of the relationship between the reducing agent concentration of the electroless platinum plating solution and the change trend of the electrolessly plated platinum coating as shown in Table 3.

[0055] Table 3 Relationship between reducing agent concentration and change trend of electrolessly plated platinum coating

[0056] The change trend of the relationship between the reducing agent concentration of the electroless platinum plating solution and the error of the simulated platinum coating thickness was obtained by the analog quantity processing of the operation equation of the RNAS model as Figure 6 shown. It can be seen from Figure 6 that the critical point of the discriminant for the simulated reducing agent concentration of 0.5 g / L of the electroless platinum plating solution and the additive of 1.0 g / L in Example 1 is the modeled thickness h 建模厚度 and the critical point of the high-concentration region for the simulated reducing agent concentration of 6.5 g / L.

[0057] To confirm the available ratio range of the reducing agent concentration of the electroless platinum plating solution, based on the component formulation of Example 1, the electroless platinum plating solution prepared with the reducing agent dosages of Simulation Example 2 and Simulation Example 8 was used to perform actual electroless plating operations on the semiconductor electronic components to be processed using the same electroless plating equipment module, and the thickness of the plated platinum coating obtained h 实测厚度 is shown in Table 4.

[0058] Table 4 Measured thickness of the plated platinum coating in Simulation Example 2 and Simulation Example 8

[0059] It can be seen from Table 4 that the thickness of the electroless platinum coatings in Simulation Example 2 and Simulation Example 8 h 实测厚度 , when compared with the standard value of the modeled coating thickness h 建模厚度 of 0.5 μm, did not meet the threshold range. Therefore, the electroless platinum plating solutions in Simulation 2 and Simulation 8 belong to the unusable formulations.

[0060] Similarly, the electroless platinum plating solutions in Simulation Example 3 and Simulation Example 7 were applied to the electroless plating equipment module to perform actual electroless plating operations on the semiconductor workpieces to be plated, and the thickness of the electrolessly plated platinum obtained h 实测厚度 is shown in Table 5.

[0061] Table 5 Measured Thickness of Platinum Coating in Simulation Example 3 and Simulation Example 7

[0062] As can be seen from Table 4, the coating thickness of the electroless platinum solutions in Simulation 3 and Simulation 7 h 实测厚度 , when compared with the standard value of the modeled coating thickness h 建模厚度 of 0.5 μm, the film thickness errors are 2.0% - 2.6% and 4.4% - 5.0% respectively; therefore, the corresponding electroless plating solutions belong to the available formulations.

[0063] In summary, the reducing agent concentrations in all the electroless plating solution formulations that meet the morphology threshold range form an ordered available trend range, and the results are shown in Table 6.

[0064] Table 6 Range of Reducing Agent in Electroless Platinum Solution

[0065] As can be seen from Table 6, the operation system of the RNAS model adopted in the present invention can obtain the preferred standards for preparing electroless plating solutions, enabling accurate and rapid acquisition of the preparation methods of electroless plating solutions.

[0066] For the preparation of electroless plating solutions in the prior art, the golden section experiment method is mostly used for the actual operation of electroless plating for the applicable ranges of different components. However, when screening for the reducing agent concentration, samples need to be prepared and tested for each of the reducing agent variation ranges of 0.5 - 7.5 g / L in Comparative Examples 1 - 10 to obtain the corresponding platinum coating thickness and its error compared with the standard value of the management coating thickness h 管理厚度 of 0.5 μm, and the results are shown in Table 7.

[0067] Table 7 Relationship between Error and Additive Concentration in Comparative Examples 1 - 10

[0068] As can be seen from Table 7, since the existing electroless plating solution preparation technology can only rely on actual testing of electroless plating samples with various reducing agent concentrations to obtain the corresponding coating thickness, and further calculate the error results of the platinum coating thickness compared with the standard value of the management coating thickness h 管理厚度 of 0.5 μm.

[0069] The construction method and system of an electroless plating solution formulation model based on an RNA simulation operation and training system according to the present invention use the electroless plating solution formulation data processing results that meet the threshold range output by the constructed RNA model as "usable formulations"; while the electroless plating solution formulation data processing results that exceed the threshold range are output as "unusable formulations". By strictly managing the electroless plating solution formulation process according to this standard, it can not only overcome the deficiencies of the existing electroless plating solution formulation technology, but also provide electroless plating solutions of excellent quality for high-end electroless plating equipment modules of semiconductor electronic components. At the same time, by using the threshold of the electroless plating film thickness of semiconductor electronic components to judge the measured coating thickness data of the electroless plating equipment module, and simulating the electroless plating solution formulation ratio based on this measured data, monitoring the coating thickness, appearance, color and overall morphological characteristics of the semiconductor electronic components after electroless plating, the preferred standard of the electroless plating solution formulation model of the RNA model is obtained, so as to achieve an accurate and rapid method for obtaining the formulation of the electroless plating solution.

[0070] Some steps in the embodiments of the present invention can be implemented by software, and the corresponding software program can be stored in a readable storage medium, such as an optical disc or a hard disk.

[0071] The above are only the preferred embodiments of the present invention and are not intended to limit the present invention. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A method for constructing a chemical plating solution configuration model, characterized in that: The method comprises: Step 1: Obtain the formula data of the chemical plating solution to be prepared and construct an initial formula data set; Step 2: Obtain the dimension data set of the plated area of ​​the workpiece to be processed and the electrode potential data of the oxidation-reduction strength of the metal to be chemically plated, calculate the metal mass of the chemically plated layer according to the general formula of the oxidation-reduction reaction, and calculate the film thickness of the plated model based on the dimension data h 建模厚度 ; Step 3: Randomly extract chemical plating solution formula data from the initial formula data set, use the chemical plating equipment module to perform actual chemical plating operation on the plated part to be processed, and measure the actual film thickness of the plated part h 实测厚度 ; Step 4: Model the film thickness of the plated part h 建模厚度 And the measured film thickness of the plated parts h 实测厚度 When the two satisfy the following relationship: h 实测厚度 （100%-5%）≤ h 建模厚度 ≤ h 实测厚度 （100%+5%） Then save the modeling coating thickness of the plated part h 建模厚度 The corresponding chemical plating solution formula data is used for training the RNAS model system; If the above relationship is not satisfied, the chemical plating solution formula data is corrected, and the actual chemical plating operation is performed again using step 3 to measure the actual film thickness of the plated part. h 实测厚度 ; Until the above relationship is satisfied, the chemical plating solution formula data is saved for RNAS model training; Step 5: Model the film thickness of the plated part h 建模厚度 And the measured film thickness of the plated parts h 实测厚度 As input, the RNAS model is trained to classify the chemical plating solution formula data, and a classification model for the chemical plating solution formula data is established.

2. The method for constructing a chemical plating solution configuration model according to claim 1, characterized in that: The chemical plating solution formula data includes: metal salt, complexing agent, stabilizer, electrolyte, pH regulator and reducing agent.

3. The method for constructing a chemical plating solution configuration model according to claim 1, characterized in that: The computing system for RNAS model training includes modeling of sensitive condition parameters, and the sensitive condition parameters include: type of chemical plating solution, redox potential of metal ions, redox potential of reducing agent, specific gravity of chemical plating solution, solution temperature, chemical plating time, surface area of ​​plating area, metal density of coating layer and flow rate of chemical plating solution.

4. The method for constructing a chemical plating solution configuration model according to claim 1, characterized in that: It also includes constructing a set of parameters for the sudden feeling condition, wherein the parameters for the sudden feeling condition include: a tank body, a pump for conveying the chemical plating solution, a length and diameter of a pipeline for conveying the chemical plating solution, semiconductor electronic plated parts to be processed, and a mold for chemical plating in the chemical plating equipment module.

5. The method for constructing a chemical plating solution configuration model according to claim 1, characterized in that: Metal plating includes single metal plating and alloy plating.

6. A system for constructing a chemical plating solution configuration model, characterized in that: A method for constructing a chemical plating solution configuration model according to any one of claims 1 to 5, the system comprising: A memory (310) for storing formula data and thickness data; A data set processing system (200) is used to process the matching relationship between recipe data and thickness data, comprising: a chemical plating reduction plating operation module (210), a chemical plating equipment module (220), a chemical plating film thickness calculation module (230), and a plating film thickness threshold determination module (240); the chemical plating reduction plating operation module (210) is used to obtain recipe parameters and calculate the relationship between the electrode potentials of metal ions and reducing agents in a chemical plating solution; the chemical plating equipment module (220) is used to perform actual chemical plating operations; the chemical plating film thickness calculation module (230) is used to calculate modeling thickness; and the plating film thickness threshold determination module (240) is used to determine the matching between the modeling thickness and the measured thickness; The RNAS model training system (100) is used for training and classifying chemical plating solution formulas.

7. The system for constructing a chemical plating solution configuration model according to claim 6, characterized in that: The RNAS model training system (100) comprises: A training module (110) is used for comparative training between the modeled film thickness of the plated part and the measured film thickness of the plated part; A stability generation module (120) for processing the ordered arrangement and trend of the measured coating thickness; A chemical plating reduction component ratio calculation module (130), used for simulating the chemical plating solution formula ratio corresponding to the measured coating thickness; The model morphology discrimination module (140) is used to discriminate the RNAS modeling morphology based on the calculation result of the simulated chemical plating solution formula ratio and its RNAS modeling morphology.

8. A method for preparing a chemical plating solution based on an RNAS model, characterized in that: The model constructed by any one of the methods described in claims 1 to 5 is used to identify the recipe, comprising: Accurately weigh each component; Keep the concentration of the basic component constant and take the reducing agent as a variable parameter; Calculate the modeled thickness and compare it with the measured thickness; The RNAS model is used to classify and identify recipes that meet the threshold conditions.

9. The method for preparing a chemical plating solution according to claim 8, characterized in that: The RNAS model discrimination results include: Available recipes: Modeling thickness data meets the topography threshold range; Unusable recipe: Modeling thickness data does not meet the topography threshold range.

10. The method for preparing a chemical plating solution according to claim 8, characterized in that: The film thickness was measured using an X-ray film thickness analyzer.

Images