Method and system for constructing electronic electroplating solution preparation model based on CFD (computational fluid dynamics)

Abstract

The invention relates to a CFD-based construction method and system for an electronic electroplating solution preparation model, and belongs to the technical field of fusion of artificial intelligence control and electroplating solution preparation. According to the method, actually-measured plating thickness data of an electronic electroplating module is judged by adopting a plating piece electroplating film thickness threshold value, an electronic electroplating solution preparation proportion is simulated according to the actually-measured data, and the plating thickness, appearance, color and overall morphology characteristics of a semiconductor device after electronic electroplating are monitored, so that an optimal standard of an electronic electroplating solution preparation model of CFD is obtained; the preparation method for accurately and rapidly obtaining the electronic electroplating solution is achieved. An electroplating solution preparation data processing result which is output by the CFD model and meets a threshold range is' available '; and the processing result of the output electroplating solution preparation data exceeding the threshold range is'unavailable ', so that the electroplating solution preparation process flow is strictly managed according to the standard, and an electronic electroplating solution with excellent quality is provided for a high-end electronic electroplating module of a semiconductor electronic element.

Description

Technical Field

[0001] The present invention relates to a method and system for constructing an electronic electroplating solution preparation model based on CFD, and belongs to the technical field of integration of artificial intelligence control and electroplating solution preparation. Background Art

[0002] Metal deposition on semiconductor electronic components is a unique characterization technique sensitive to the local electrochemical-physical environment of metal ions during the electroplating process, and has extensive applications in the fields of physics, electrochemistry, and high-end electronic plating manufacturing. During the operation of electronic plating production equipment, the metal ions and additive components in the electroplating solution, under the electrochemical-physical environmental conditions in which they are located, are influenced by numerous and complex factors on the deposition of metal plating. Consequently, interference from these various factors can lead to various differences in the actual deposited metal coating. In mild cases, excessive additive content can make the coating area difficult to control, resulting in rapid and rough deposition. In severe cases, excessive additive content can lead to a lack of density in the coating, or even the inability to deposit the desired metal coating.

[0003] At present, the preparation method of electronic electroplating solution adopts the common golden section method. Its research and development requires a large amount of solution preparation work. At the same time, it is necessary to have an electronic electroplating module for processing and evaluating electroplating samples. The R&D cycle is long, and it is difficult to adapt to the current semiconductor electronic components and the rapid development of various precision electronic electroplating solutions. Therefore, exploring and creating a systematic method for quickly preparing electronic plating solutions with short cycles and precision has become a very important topic. Summary of the Invention

[0004] In order to improve the preparation accuracy of electronic plating solutions required by high-end semiconductors, the present invention provides a method and system for constructing an electronic plating solution preparation model based on CFD. The technical solution is as follows: A first object of the present invention is to provide a method for constructing an electronic plating solution configuration model, the method comprising: Step 1: Obtain the formula data of the electronic electroplating solution to be prepared and build an initial formula data set; Step 2: Obtain the size data set of the coating area of ​​the semiconductor plated part to be processed and the metal coating information to be electroplated, and calculate the film thickness of the plated part modeling h 建模厚度 , the calculation method is: h 建模厚度 = CI t η k / S ρ in, C is the electrochemical equivalent; I is the current intensity; t is the plating time; ηk is the cathode current efficiency; S is the surface area of ​​the plating area; ρ is the metal density of the electroplating layer; Step 3: Randomly extract electronic electroplating solution formula data from the initial formula data set, use the electronic electroplating module to perform actual electroplating operations on the semiconductor plated parts to be processed, and measure the actual film thickness of the plated parts after the electroplating is completed. 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 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 plating thickness of the plated part model h 建模厚度 The corresponding electronic plating solution formula data is used for training the CFD model system; If the above relationship is not satisfied, the electronic electroplating solution formula data is corrected, and the actual electroplating 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 electronic plating solution formula data is saved for CFD 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 CFD model is trained to classify the electronic plating solution formula data, and a classification model for the electronic plating solution formula data is established.

[0005] Optionally, the formula data of the electronic electroplating solution includes: at least two metal salts; at least two complexing agents, including potassium cyanide, sodium cyanide, ethylenediamine, citric acid, malonic acid, succinic acid, or cyclohexanediic acid; Electrolytes, including: sulfuric acid, sulfamic acid, methanesulfonic acid, or ammonium acetate; additive.

[0006] Optionally, the computing system for training the CFD model includes modeling of sensitive condition parameters, and the sensitive condition parameters include: type of electroplating solution, temperature, specific gravity, current intensity, electroplating time, cathode current efficiency, plating area surface area, metal density, current density and solution flow rate.

[0007] Optionally, it also includes constructing a set of parameters for the feeling of sudden change, wherein the parameters for the feeling of sudden change include: electroplating tank, delivery pump, pipeline parameters, type of electroplating power supply, wire, workpiece to be processed, mold and electrolytic anode.

[0008] Optionally, the metal plating layer includes a single metal plating layer and an alloy plating layer.

[0009] A second object of the present invention is to provide a system for constructing an electronic plating solution classification model, for implementing the method for constructing an electronic plating solution configuration model as described in any one of the above items, the system comprising: Memory 110, for storing recipe data and thickness data; The data set processing system 200 is used to process the matching relationship between recipe data and thickness data, and includes: an electronic plating solution recipe module 210, an electronic plating module 220, a coating thickness calculation module 230, and a coating thickness threshold determination module 240; the electronic plating solution recipe module 210 is used to obtain recipe parameters; the electronic plating module 220 is used to perform actual electroplating operations; the coating thickness calculation module 230 is used to calculate the modeled thickness; and the coating thickness threshold determination module 240 is used to determine whether the modeled thickness matches the measured thickness. The CFD computational network training system 300 is used for training and classifying electronic electroplating solution formulas.

[0010] Optionally, the CFD computing network training system 300 includes: Training module 310, for comparison training between modeled film thickness of plated parts and measured film thickness of plated parts; a stability generation module 320 for processing the ordered arrangement and trend of the measured coating thickness; The electronic plating solution composition ratio calculation module 330 is used for simulating the electronic plating solution composition ratio corresponding to the measured coating thickness; The model morphology determination module 340 is used to determine the CFD modeling morphology based on the calculation result of the simulated electronic electroplating solution formula ratio.

[0011] A third object of the present invention is to provide a CFD-based method for preparing an electronic plating solution, wherein the method comprises: Accurately weigh each component; Keep the concentration of the basic component constant and use the additive as a variable parameter; Calculate the modeled thickness and compare it with the measured thickness; The CFD model is used to classify and identify recipes that meet the threshold conditions.

[0012] Optionally, the CFD model identification result includes: Available recipes: Modeling thickness data meets the morphology threshold range; Unusable recipe: The modeling thickness data does not meet the topography threshold range.

[0013] Optionally, an X-ray film thickness analyzer is used to measure the film thickness.

[0014] The beneficial effects of the present invention are: The present invention improves and constructs a CFD network operation model through a computational fluid dynamics simulation model for training. The constructed CFD network can classify and output electronic electroplating solution formula data, namely, classifying them into "usable formula" (modeled coating thickness data is within a threshold range) and "unusable formula" (modeled coating thickness data exceeds a threshold range). This standard is used to strictly control the high-end quality requirements of electronic electroplating solution preparation.

[0015] In addition, a plated coating thickness threshold discrimination module is used to simulate and monitor the appearance, shape, color and overall morphology characteristics of semiconductor devices after electronic electroplating. The accuracy of CFD network training is further improved by meeting the standards of a minimum threshold of -5% of the coating specification value and a maximum threshold of +5% of the coating specification value. The training distribution of the electronic plating solution formula associated with the coating thickness is quickly realized, thereby obtaining the optimal standard of the CFD operation mechanism model, achieving accurate and rapid identification of the screening results of available or unavailable formulas, and obtaining the preparation method of the electronic electroplating solution.

[0016] Based on the screening results of the present invention, the available formula conditions of electronic electroplating solutions of various monomeric metal salts, binary metal salts and ternary or higher metal salts can be quickly established in the actual production process, and then the optimal electronic electroplating control method of semiconductor electronic components can be obtained through the electronic electroplating module. This can not only quickly achieve the accuracy of predicting the local coating film thickness of semiconductor devices and greatly reduce the error with the actual electroplated product coating film thickness, but also achieve further improvement of the electronic electroplating equipment, promote the improvement of the uniformity of the coating film thickness of semiconductor electronic components, and meet the high quality requirements of high-end semiconductor electronic component products. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0018] Figure 1 This is a simplified structural diagram of the CFD-based electronic electroplating solution preparation model of the present invention.

[0019] Figure 2 The present invention is a flow chart of a method and system for constructing an electronic electroplating solution preparation model based on CFD.

[0020] Figure 3 It is a schematic diagram of a semiconductor plated test piece to be processed in Example 3 of the present invention.

[0021] Figure 4 Schematic diagram of the copper plating area of ​​each unit in the semiconductor plated workpiece test piece to be processed in Example 3 of the present invention.

[0022] Figure 5 Schematic diagram of the silver-plated area of ​​each unit in the experimental piece of the semiconductor plated component to be processed in the third embodiment of the present invention.

[0023] Figure 6 This is a trend diagram of the variation of the additive concentration of the electronic electroplating Pt-Rh alloy solution and the simulated Pt-Rh alloy coating thickness error obtained by analog quantity processing of the CFD operation network in the third embodiment of the present invention.

[0024] 100-CFD simulation model; 110-memory; 200-dataset processing system; 210-electronic electroplating solution formulation module; 220-electronic electroplating module; 230-plating film thickness calculation module; 240-plating thickness threshold determination module; 300- CFD operation network training system; 310- training module; 320- stability generation module; 330- electronic plating solution component ratio calculation module; 340- model morphology discrimination module; 10-copper plating area; 20-silver plating area; 21-center of the electroplating area. DETAILED DESCRIPTION

[0025] To make the objectives, technical solutions and advantages of the present invention more clear, the embodiments of the present invention will be described in further detail below with reference to the accompanying drawings.

[0026] Example 1: This embodiment provides a method for constructing an electronic electroplating solution configuration model, which classifies configured electronic electroplating solutions as usable or unusable based on the generation of a computational fluid dynamics (CFD) model. The construction process of the configuration model mainly includes two parts: the construction of an electronic plating solution screening dataset and the training of a CFD operation network system. First, the electronic plating solution screening dataset is constructed based on the electronic plating solution formula, electronic plating module, and the semiconductor electronic components to be processed. By collecting data on the components in these electronic plating solution formulas, the feel and dimensions of various components in the electronic plating module, the dimensions of the semiconductor plated parts, the plating area, and other data, the specific steps include: Step 1: Obtain data on the components of the electronic electroplating solution to be prepared, the feel data and dimensions of various components in the electronic electroplating module, the dimensional data of the processed semiconductor electronic components, and the coating thickness and area.

[0027] In this embodiment, the electronic plating solution to be prepared includes the following components in the following concentrations: Table 1 Component data of the electronic plating solution to be prepared

[0028] The stiffness data and dimensions of various components in the electronic electroplating module do not change during the electroplating process and are not listed in detail in this embodiment.

[0029] Step 2: Obtain the size data, coating thickness and area data set of the semiconductor electronic components to be processed, as well as the metal coating information to be electroplated, and calculate the coating thickness of the plated parts modeling h 建模厚度 , the calculation method is: h 建模厚度 = CI t η k / S ρ in, h is the thickness of the electroplating layer; C is the electrochemical equivalent; I is the current intensity; t is the plating time; η k is the cathode current efficiency; S is the surface area of ​​the plating area; ρ is the metal density of the electroplating layer.

[0030] The calculation formula for the thickness of multiple coating layers is: h 建模厚度 =h 镀层1+ h 镀层2 + ...... + h 镀层n Step 3: Randomly select the standard conditions for coating thickness calculated based on the electronic electroplating solution formula, use the prepared platinum-rhodium alloy electronic electroplating solution, and perform actual electroplating operations on the semiconductor plated parts through the electronic electroplating module. After the electroplating is completed, use the X-ray film thickness analyzer to measure the actual film thickness data of the plated parts. h 实测厚度 .

[0031] 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 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 plating thickness of the plated part model h 建模厚度 The corresponding electronic plating solution formula data is used for training the CFD model system; If the above relationship is not satisfied, the electronic plating solution formula is judged as "unusable"; if the above relationship is satisfied and it is "usable", the next step of the CFD training system judgment is carried out.

[0032] After completing the construction of the electronic electroplating solution screening dataset, the dataset was used to train the CFD operation network system.

[0033] Step 5: Calculate the modeled coating thickness based on the electronic plating solution formula h 建模厚度 and the measured film thickness of the plated parts h 实测厚度 Import the trained CFD operation network system, and the results of the CFD operation network system comparison processing are divided into two categories: Available formula: The modeled coating thickness data processing results obtained by calculating the electronic electroplating solution formula meet the morphology threshold range; Unusable formula: The modeled coating thickness data processing results obtained by calculating the electronic electroplating solution formula do not meet the morphology threshold range.

[0034] The CFD operation network system is composed of sensitive condition parameter modeling, which includes: electroplating solution type, electroplating solution temperature, electroplating solution specific gravity, current intensity, electroplating time, cathode current efficiency, plating area surface area, electroplating layer metal density, current density and electroplating solution flow rate.

[0035] The dataset construction system is composed of the modeling of the jitter condition parameters, which include the following parameters in the electronic electroplating module: the plating tank, the pump for conveying the plating solution, the length and diameter of the pipeline for conveying the plating solution, the type of electroplating power supply, the wire connected to the power supply, the semiconductor plated parts to be processed, the mold for electronic electroplating, and the electrolytic anode.

[0036] This embodiment applies the principle of computational fluid dynamics to calculate the modeling coating thickness data obtained based on the electronic electroplating solution formula. h 建模厚度 The processing results are compared with the actual film thickness data of the plated parts obtained by the actual electroplating operation of the electronic electroplating module on the semiconductor plated parts to be processed. h 实测厚度 The results are compared and trained, and the data set system is used to deeply learn the real data and distribution. The CFD operation system correctly judges the deep learning operation results based on the threshold standard to determine whether they can be used for the CFD operation network model to construct the precise preparation method of each component of the electronic electroplating solution.

[0037] The construction process of the CFD operation system of this embodiment specifically includes the following steps: Set continuous random variable X The logistic distribution means X It has the following distribution and density functions:

[0038]

[0039] in, μ is a positional parameter, γ is the shape parameter.

[0040] When the positional parameter μ tends to zero, γ = 1, we get the common Sigmoid curve:

[0041] The introduction of binomial logistic regression model is a classification model composed of conditional probability P ( Y | X ) represents a parameterized logistic distribution; in this case, the random variable X Taken as a real number, random variable YThe value is 1 or 0, and the model parameters are estimated through supervised learning methods.

[0042] The binomial logistic regression model has the following conditional probability distribution:

[0043]

[0044] in, x ∈R n is the input, y∈{0,1}, w ∈R n Yes and b ∈R is a parameter, w is called the weight vector, b is called bias, w·x yes w and b The inner product of .

[0045] Expand the weight vector and input vector, still denoted as w , x , which is:

[0046]

[0047] Then, the logistic regression model is expressed as:

[0048]

[0049] Next, set the probability of a time odds, which refers to the ratio of the probability of the event occurring to the probability of the event not occurring. If the probability of the event occurring is p , then the probability of the event is:

[0050] The log odds of the event, log odds or log (log odds) = logit, is expressed as:

[0051] For logistic regression, the above formula P ( Y = 1 | x )and P ( Y = 0 | x ) can be obtained:

[0052] In the logistic regression model, the output Y = 1 is the log-odds of the input x A linear function of , or expressed as, output Y = 1 is given by the input x The model represented by the linear function constitutes the preliminary logistic regression model.

[0053] Consider the input x Linear function for classification w · x , its value range w · x ∈R, and x ∈R n+1 , w ∈R n+1 ; Define the linear function through logistic regression w · x Converted into probability, it can be expressed as:

[0054] When the value of the linear function z = w · x → + ∞, then P ( Y = 1 | x ) → 1; when the value of the linear function z = w · x → -∞, then P ( Y = 1 | x ) → 0; the resulting graph is the logistic regression model.

[0055] The training in step 4 is performed using modeling coating thickness parameters that meet the threshold, and specifically includes the following steps: In step 3, for the given training set data T= {( x 1, y 1), ( x 2, y 2)···,( x N , y N )},in,

[0056] y i ∈{0, 1}, the maximum likelihood estimation method is used to predict the model parameters, thus obtaining the iterative logistic regression model.

[0057] The likelihood function is expressed as:

[0058] The log-likelihood function is expressed as:

[0059] Calculate the above formula L ( w ) to obtain the maximum value of w The predicted value of w The predicted value represents the coating thickness.

[0060] Using gradient descent and Newton-like methods, set w The maximum likelihood estimate of is w ∧ , then the learned logistic regression model is:

[0061]

[0062] However, the binomial logistic regression model is only introduced for binary classification problems; when used for multi-classification problems, a multinomial logistic regression model needs to be introduced. The specific steps are: Set discrete random variables Y The value set of is {1, 2, ···, K}, that is, the category label has K The multinomial logistic regression model obtained by analogy is:

[0063]

[0064] Both binomial logistic regression and multinomial logistic regression are widely used classification algorithms. Their principles are based on statistics and probability theory. They model the probability of the combination of input features first, and then use the logistic function to map the linear combination to a probability value between 0 and 1, thereby indicating the probability of belonging to a sample and belonging to a certain category. Step 5 determines the model morphology based on the coating film thickness result obtained in step 4 and compares it with the target threshold. If the target threshold is reached, it is determined to be the optimized electronic electroplating condition, and the optimized electronic electroplating condition is applied to the production and debugging of the electronic electroplating module; if the target value is not reached, the set of training electronic electroplating conditions is discarded and returns to step 4 for retraining.

[0065] Example 2: This embodiment provides a system for constructing an electronic plating solution classification model, such as Figure 2 As shown, the system includes: Memory 110, for storing recipe data and thickness data; The data set processing system 200 is used to process the matching relationship between recipe data and thickness data, and includes: an electronic plating solution recipe module 210, an electronic plating module 220, a coating thickness calculation module 230, and a coating thickness threshold determination module 240. The electronic plating solution recipe module 210 is used to obtain recipe parameters; the electronic plating module 220 is used to perform actual electroplating operations; the coating thickness calculation module 230 is used to calculate the modeled thickness; and the coating thickness threshold determination module 240 is used to determine whether the modeled thickness matches the measured thickness. The CFD computational network training system 300 is used for training and classifying electronic electroplating solution formulas.

[0066] The CFD computing network training system 300 includes: Training module 310, for comparison training between modeled film thickness of plated parts and measured film thickness of plated parts; a stability generation module 320 for processing the ordered arrangement and trend of the measured coating thickness; The electronic plating solution composition ratio calculation module 330 is used for simulating the electronic plating solution composition ratio corresponding to the measured coating thickness; The model morphology determination module 340 is used to determine the CFD modeling morphology based on the calculation result of the simulated electronic electroplating solution formula ratio.

[0067] Example 3: This embodiment, in conjunction with an actual case of an electronic electroplating solution to be prepared, demonstrates in detail the technical effects of the electronic electroplating solution configuration model constructed by the present invention in industrial applications, specifically including the following contents.

[0068] The semiconductor plated test piece to be processed in this embodiment is as follows Figure 3 As shown in the figure, the copper alloy material is 60×60mm and the thickness is 0.127mm; there are 16 basic units in total. The copper plating area of ​​each unit is as follows Figure 4 The size of 10 is 11.5mm×11.5mm, and its area is 132.25mm2; the total area of ​​a single side of a semiconductor copper plated is 16×132.25=2166mm 2 ; In addition, the silver-plated area of ​​each unit is as follows Figure 5 The size of 20 is 7.3mm×7.3mm, and its area is 53.29mm 2 ; The total area of ​​a semiconductor plated with nickel and platinum-rhodium alloy on both sides is 2×16×53.29=1705.28mm 2 .

[0069] The center position of the plating thickness detection of the semiconductor electronic component plating area in this embodiment is as follows: Figure 3 and Figure 5 The coordinates of the center of the electroplating area 21 (black dot ●) of each semiconductor plating unit are: 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; 4 (51.1, 7.3), 3 (51.1, 21.9), 2 (51.1, 36.5), 1 (51.1, 51.1) in column IV.

[0070] Based on Figure 3 The 3D design drawing of the semiconductor plated part shown has the electroplating specifications as follows: 1.5μm≤nickel plating thickness≤3.5μm, and the thickness of the outermost surface 0.3μm≤platinum-rhodium alloy plating≤0.7μm. In the following plating thickness data processing of the plated parts, the thickness data of the bottom nickel plating and the interlayer gold plating are not summarized and discussed, and only the thickness of the surface platinum-rhodium alloy plating is specifically analyzed.

[0071] The thickness of the nickel plating is based on the center value of 2.5μm in the design drawing range, and the thickness of the Pt-Rh alloy plating is based on the center value of 2.5μm in the design drawing range. h Pt-Rh Taking the center value of 0.5μm in the design drawing range as the standard, the electronic plating module is used for actual electroplating treatment. The electroplating conditions are calculated by the following formula: h Pt-Rh = CI t η k / S ρ A (Ampere / dm 2 ) = I / S A = ρ d / t C η k h Pt-Rh = A d d = t C η k / ρ The specific operation steps of this embodiment are as follows: Step 1: After a semiconductor plated piece passes through the alkaline degreasing and acid activation and drying of the electronic plating module 220, a clean semiconductor element is provided for the next electroplating step.

[0072] Step 2: The additive concentration of the electronic electroplating solution in Example 1 is selected to be 0.5 g / L. Using this solution, an electronic electroplating module and the calculated electroplating conditions of various plating types are used to perform electronic nickel plating, gold plating and platinum-rhodium alloy plating treatments one by one to obtain Pt-Rh alloy plating samples of semiconductor components.

[0073] Step 3: Use an X-ray film thickness analyzer to measure the Pt-Rh alloy coating sample of the semiconductor component. The measured thickness of the coating is shown in Table 2.

[0074] Table 2 Pt-Rh alloy coating thickness data

[0075] Step 4: As can be seen from Table 2, the electronic plating solution of Example 1 is used to perform an actual electroplating operation on the semiconductor plated workpiece to be processed through the electronic plating module, and the thickness of the obtained Pt-Rh alloy coating is h 实测厚度 The range is 0.505~0.511μm, which is the center value of the design drawing range, that is, the modeled coating thickness h 建模厚度 Compared with 0.5 μm, the coating thickness error is 0.8% to 2.2%, which meets the modeling threshold conditions compared with the overall quality control threshold of -5.0% to 5.0% for semiconductor plated parts. Therefore, the formula data of the electronic electroplating Pt-Rh alloy solution and its corresponding coating thickness data of Example 1 can be used to construct the CFD model.

[0076] Furthermore, the thickness of the Pt-Rh alloy coating was electroplated in an electroplating solution with an additive concentration of 0.5 ml / L in Example 1. h 实测厚度 and modeling coating thickness h 建模厚度 The 0.5μm data was processed by the simulation of the CFD operation network to obtain the simulated data of the change trend of the additive concentration of the electronic electroplating Pt-Rh alloy solution and the Pt-Rh alloy coating, as shown in Table 3.

[0077] Table 3 Simulated data on the relationship between additive concentration and the change trend of Pt-Rh alloy coating

[0078] Furthermore, the variation trend of the additive concentration of the electronic electroplating Pt-Rh alloy solution and the thickness error of the simulated Pt-Rh alloy coating was obtained through the simulation processing of the CFD operation network. Figure 6 shown.

[0079] Depend on Figure 6 It can be seen that the modeling thickness is between the additive concentration of 0.25ml / L of the simulated electronic plating Pt-Rh alloy solution and the additive concentration of 0.5ml / L of Example 1. h 建模厚度 The critical points of the discriminant formula are as follows: the additive concentrations of 2.5 ml / L and 3.0 ml / L in the simulation example are the critical points of the high concentration region.

[0080] Furthermore, in order to determine the available ratio range of additives for the electronic electroplating Pt-Rh alloy solution, based on the component formula of Example 1, the Pt-Rh alloy solution prepared with the additive dosage of Simulation Example 2 and Simulation Example 7 was also used. The electronic electroplating module was also used to perform actual electroplating operations on the semiconductor plated parts to be processed. The measured thickness h of the Pt-Rh alloy coating is shown in Table 4.

[0081] Table 4 Measured thickness of Pt-Rh alloy coatings in Simulation Examples 2 and 7

[0082] From Table 4, we can see that the thickness of the Pt-Rh alloy coating in Simulation Example 2 and Simulation Example 7 is h 实测厚度 , which is consistent with the standard value of modeled coating thickness h 建模厚度 For comparison with 0.5 μm, the film thickness errors are -6.4% to -5.2% and 5.4% to 7.2% respectively. Therefore, the electronic plating solutions of Simulation Examples 2 and 7 are classified as unusable formulations.

[0083] Similarly, the electronic plating solution electronic plating module of simulation example 3 and simulation example 6 is used to perform actual electroplating operation on the semiconductor plated workpiece to be processed, and the thickness of the obtained Pt-Rh alloy coating is h 实测厚度 As shown in Table 5.

[0084] Table 5 Measured thickness of Pt-Rh alloy coatings in simulation examples 3 and 6

[0085] From Table 5, we can see that the thickness of the Pt-Rh alloy coating in Simulation Example 3 and Simulation Example 6 is h 实测厚度 , which is consistent with the standard value of modeled coating thickness h 建模厚度 Compared with 0.5μm, the film thickness errors are 1.8% to -3.2% and 3.8% to 5.0% respectively; therefore, the corresponding electronic electroplating solutions belong to the usable formula.

[0086] In summary, the additive concentrations in all electronic plating solution formulations that meet the morphology threshold range form an orderly usable trend range, the results of which are shown in Table 6.

[0087] Table 6 Trend of additive concentrations meeting the morphology threshold range

[0088] As can be seen from Table 6, the present invention adopts the preferred standard of the CFD electronic electroplating solution preparation model obtained by the CFD operation system, which can achieve accurate and rapid acquisition of the preparation method of the electronic electroplating solution.

[0089] The preparation of electronic plating solutions in the prior art mostly adopts the golden section experimental method, and the actual operation of electronic plating is carried out according to the applicable range of different components. However, when screening the concentration of additives, it is necessary to prepare and test samples of each comparative example 1 to 9 with the additive range of 0.15 to 3.5 g / L in order to obtain the corresponding coating thickness and its comparison with the standard value of the management coating thickness. h 管理厚度 The error of 0.5μm is compared, and the results are shown in Table 7.

[0090] Table 7 Additives, coating thickness and error in Comparative Examples 1 to 9

[0091] As can be seen from Table 7, the existing technology can only rely on actual testing of electronic plating samples with various concentrations of additives to obtain the corresponding coating thickness, and further calculate the coating thickness and the management coating thickness standard value. h 管理厚度 The error results for comparison are 0.5μm.

[0092] Therefore, the existing electronic electroplating solution preparation method has the following defects: since the actual electronic electroplating operation requires the preparation of electroplating samples with all additive concentrations, a large amount of raw materials for each component is required; the preparation of the above-mentioned electronic electroplating solution requires a lot of manpower and time; the use of the prepared electronic electroplating solution to perform actual electronic electroplating operations requires a large amount of manpower, time, water, electricity and gas energy; the electroplating processing efficiency of semiconductor electronic products in the electronic electroplating production line is very low, and therefore the production cost remains high.

[0093] However, the method and system for constructing a CFD-based electroplating solution preparation model of the present invention outputs a processing result of electroplating solution preparation data that meets the threshold range as "usable" through the constructed CFD model; and the processing result of electroplating solution preparation data that exceeds the threshold range is "unusable". This standard is used to strictly manage the electroplating solution preparation process flow, which not only overcomes the defects of the existing technology for preparing electronic electroplating solutions, but also provides high-quality electronic electroplating solutions for high-end electronic electroplating modules of semiconductor electronic components. At the same time, by using the plated film thickness threshold to judge the measured coating thickness data of the electronic electroplating module, and using this measured data to simulate the electronic plating solution preparation ratio, the coating thickness, appearance, color and overall morphology characteristics of the semiconductor device after electronic electroplating are monitored, thereby obtaining the preferred standard of the CFD electronic plating solution preparation model, and achieving the goal of accurately and quickly obtaining the preparation method of the electronic plating solution.

[0094] Some steps in the embodiments of the present invention may be implemented using software, and the corresponding software program may be stored in a readable storage medium, such as a CD or a hard disk.

[0095] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. A method for constructing an electronic plating solution configuration model, characterized in that: The method comprises: Step 1: Obtain the formula data of the electronic electroplating solution to be prepared and build an initial formula data set; Step 2: Obtain the size data set of the coating area of ​​the semiconductor plated part to be processed and the metal coating information to be electroplated, and calculate the film thickness of the plated part modeling h 建模厚度 , the calculation method is: h 建模厚度 = CI t η k / S ρ in, C is the electrochemical equivalent; I is the current intensity; t is the plating time; η k is the cathode current efficiency; S is the surface area of ​​the plating area; ρ is the metal density of the electroplating layer; Step 3: Randomly extract electronic electroplating solution formula data from the initial formula data set, use the electronic electroplating module to perform actual electroplating operations on the semiconductor plated parts to be processed, and measure the actual film thickness of the plated parts after the electroplating is completed. 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 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 plating thickness of the plated part model h 建模厚度 The corresponding electronic plating solution formula data is used for training the CFD model system; If the above relationship is not satisfied, the electronic electroplating solution formula data is corrected, and the actual electroplating 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 electronic plating solution formula data is saved for CFD 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 CFD model is trained to classify the electronic plating solution formula data, and a classification model for the electronic plating solution formula data is established; The formula data of the electronic plating solution includes: at least two metal salts; at least two complexing agents, including potassium cyanide, sodium cyanide, ethylenediamine, citric acid, malonic acid, succinic acid, or cyclohexanediic acid; Electrolytes, including: sulfuric acid, sulfamic acid, methanesulfonic acid, or ammonium acetate; additive.

2. The method for constructing an electronic plating solution configuration model according to claim 1, wherein: The computing system for training the CFD model includes modeling of sensitive condition parameters, which include: type of electroplating solution, temperature, specific gravity, current intensity, electroplating time, cathode current efficiency, plating area surface area, metal density, current density and solution flow rate.

3. The method for constructing an electronic plating solution configuration model according to claim 1, wherein: It also includes constructing a set of parameters for the feeling of suddenness condition, wherein the parameters for the feeling of suddenness condition include: an electroplating tank, a delivery pump, pipeline parameters, a type of electroplating power supply, a conductor, a workpiece to be plated, a mold, and an electrolytic anode.

4. The method for constructing an electronic plating solution configuration model according to claim 1, wherein: The metal plating layer includes a single metal plating layer and an alloy plating layer.

5. A system for constructing a classification model for electronic plating solutions, characterized in that: A method for constructing an electronic plating solution configuration model according to any one of claims 1 to 4, the system comprising: A memory (110) for storing recipe data and thickness data; A data set processing system (200) is used to process the matching relationship between formula data and thickness data, comprising: an electronic electroplating solution formula module (210), an electronic electroplating module (220), a coating thickness calculation module (230), and a coating thickness threshold determination module (240); the electronic electroplating solution formula module (210) is used to obtain formula parameters; the electronic electroplating module (220) is used to perform actual electroplating operations; the coating thickness calculation module (230) is used to calculate modeling thickness; and the coating thickness threshold determination module (240) is used to determine the matching between the modeling thickness and the measured thickness. A CFD computing network training system (300) is used for training and classifying electronic electroplating solution formulas.

6. The system for constructing an electronic plating solution classification model according to claim 5, characterized in that: The CFD operation network training system (300) includes: A training module (310) is used for comparative training of the modeled film thickness of the plated part and the measured film thickness of the plated part; a stability generation module (320) for processing the ordered arrangement and trend of the measured coating thickness; An electronic electroplating solution composition ratio calculation module (330) is used for simulating calculations based on the electronic electroplating solution composition ratio corresponding to the measured coating thickness; The model morphology discrimination module (340) is used to discriminate the CFD modeling morphology based on the calculation result of the simulated electronic electroplating solution formula ratio and its CFD modeling morphology.

7. A method for preparing an electronic plating solution based on CFD, characterized in that: The model constructed using the method according to any one of claims 1 to 4 is used to perform recipe identification, comprising: Accurately weigh each component; Keep the concentration of the basic component constant and use the additive as a variable parameter; Calculate the modeled thickness and compare it with the measured thickness; The CFD model is used to classify and identify recipes that meet the threshold conditions.

8. The method for preparing an electronic plating solution according to claim 7, wherein: The CFD model identification results include: Available recipes: Modeling thickness data meets the morphology threshold range; Unusable recipe: The modeling thickness data does not meet the topography threshold range.

9. The method for preparing an electronic plating solution according to claim 7, wherein: The film thickness was measured using an X-ray film thickness analyzer.

Images