Wafer classification method and apparatus for wafer rapid thermal processing

Abstract

The application discloses a wafer classification method and device for wafer rapid heating process, acquires several kinds of wafers, heats for a period of time to acquire temperature change data to construct a data set, the temperature change data includes emissivity, temperature value after heating and temperature difference value with resolution in the heating process, trains a support vector machine based on the data set; heats the wafer needing classification for the same time, inputs the acquired temperature change data into the support vector machine after training, obtains the wafer category, and implements the rapid heating process according to the obtained wafer category. The application significantly improves the classification accuracy and production efficiency in the wafer rapid heating process, reduces the production cost, improves the product consistency and quality, and has a wide application prospect in the field of semiconductor manufacturing.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of wafer classification, and particularly relates to a wafer classification method and device for wafer rapid thermal processing. BACKGROUND

[0002] With the advancement of semiconductor manufacturing technology, wafer rapid thermal processing (RTP) is increasingly important in integrated circuit manufacturing. RTP process can quickly heat the wafer to the required temperature, improving manufacturing efficiency and product quality. However, different transmittance and surface coating of wafers have different thermal response characteristics in the process of being heated by tungsten lamp irradiation. Accurate classification of these wafers is crucial to ensure process stability and product consistency.

[0003] Traditional wafer classification methods mainly rely on physical identification and experience judgment, such as etching code, visual identification and RFID tag, etc. These methods have problems such as easy wear, complex identification, and dependence on manual operation, resulting in low efficiency and frequent errors.

[0004] In the domestic semiconductor manufacturing process, the inaccuracy of temperature measurement of wafers at low temperature is a common problem. This is because the transmittance of wafer material affects temperature measurement, leading to inaccurate classification based on temperature judgment, affecting the accuracy of wafer classification and subsequent process control. SUMMARY

[0005] The present application aims to overcome the shortcomings of the prior art and provides a wafer classification method and device for wafer rapid thermal processing.

[0006] The purpose of the present application is achieved by the following technical solutions: in the first aspect, the present application provides a wafer classification method for wafer rapid thermal processing, which comprises the following steps:

[0007] (1) Obtain several types of wafers, heat for a period of time to obtain temperature change data to construct a data set, the temperature change data includes emissivity, temperature value after heating and temperature difference with resolution in the heating process, and train support vector machine based on the data set;

[0008] (2) Heat the wafer to be classified for the same time as in step (1), input the obtained temperature change data into the trained support vector machine, obtain the wafer category, and implement the rapid thermal processing according to the obtained wafer category.

[0009] Further, in step (1), the heating power is controlled to be constant, and the heating time is 10 seconds, wherein the heating time is 5 seconds in the pin up stage, and then the heating time is 5 seconds after pin down.

[0010] Furthermore, the temperature difference with resolution is the temperature difference between the pin-down moment and the temperature 2 seconds after the pin-down in the multi-channel experiment; the temperature value after heating is the temperature value data 5 seconds after the pin-down.

[0011] Furthermore, the data set includes the wafer's emissivity, seven-way temperature differences, and seven-way temperature values, and all data is divided into a training set and a test set with a ratio of 3:1. 5. The wafer classification method for a rapid wafer heating process according to claim 1 is characterized in that the temperature change data in the data set is subjected to median filtering and preprocessing before being input into a support vector machine for training using the corresponding wafer categories as labels.

[0012] Furthermore, the wafer rapid heating process is confirmed based on the thermal response characteristics of different wafer types.

[0013] In a second aspect, the present invention also provides a wafer classification device for a wafer rapid heating process, comprising a memory and one or more processors, wherein the memory stores executable code, and when the processor executes the executable code, it implements the wafer classification method for a wafer rapid heating process.

[0014] In a third aspect, the present invention further provides a computer-readable storage medium having a program stored thereon, which, when executed by a processor, implements the wafer classification method for a rapid wafer heating process.

[0015] In a fourth aspect, the present invention further provides a computer program product, comprising a computer program, which, when executed by a processor, implements the wafer classification method for a rapid wafer heating process.

[0016] Beneficial effects of the present invention:

[0017] (1) High-precision classification: Using the SVM algorithm and temperature change data, high-precision identification of different types of wafers can be achieved.

[0018] (2) Automated identification: Integrate with automated equipment to achieve fully automated operation and reduce human intervention and errors.

[0019] (3) Real-time response: Real-time classification based on temperature change data without waiting for the complex identification process.

[0020] (4) Wide applicability: Applicable to various wafer materials, including silicon wafers and compound semiconductor wafers.

[0021] (5) Solve the problem of inaccurate low-temperature measurement: By collecting temperature change data within 10 seconds before heating, the influence of wafer transmittance on low-temperature measurement is avoided, and the classification accuracy is improved.

[0022] The present invention significantly improves the classification accuracy and production efficiency in the wafer rapid heating process, reduces production costs, and improves product consistency and quality, and has broad application prospects in the field of semiconductor manufacturing. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. 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.

[0024] Figure 1 It is a schematic diagram of the linear SVM principle;

[0025] Figure 2 This is the temperature response curve of the bare wafer at an air pressure of 5 torr, an inner circle heating voltage of 15%, and an outer circle heating voltage of 5%.

[0026] Figure 3 This is the temperature response curve of a patterned wafer at a pressure of 5 torr, an inner heating voltage of 15%, and an outer heating voltage of 5%.

[0027] Figure 4 This is a flow chart of the training process;

[0028] Figure 5 This is a schematic diagram of the 5-fold cross validation results;

[0029] Figure 6 This is a structural diagram of a wafer sorting device used in a wafer rapid heating process according to the present invention. DETAILED DESCRIPTION

[0030] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described below with reference to the accompanying drawings and examples. It should be understood that the specific examples described herein are only used to explain the present invention and are not intended to limit the present invention.

[0031] like Figure 1 As shown, the present invention provides a wafer classification method for a wafer rapid heating process, and the specific implementation process of the method is as follows:

[0032] 1. Technical principle:

[0033] The basic concept of the Support Vector Machine (SVM) algorithm can be derived from a simple linear binary classification problem: Assuming that the goal of the algorithm is to use a single straight line to distinguish two types of sample points of different colors, it is necessary to use a straight line passing through the edge points as a reference so that the two sample sets are strictly separated by two straight lines, such as Figure 1 In the actual application of the algorithm model, in order to achieve the best classification effect, it is necessary to minimize the distance between the two dividing lines through methods such as rotation and select the classification hyperplane between the two dividing lines to avoid the situation where the collection point slightly deviates from the sample set and thus causes misclassification.

[0034] Therefore, the above geometric problem can be transformed into a mathematical problem of finding the maximum value of the minimum function, which can be described by the formula derived from the Lagrangian function method:

[0035]

[0036] Among them, α i represents the non-negative Lagrange multiplier, y i Represents the decision value of the sample point, such as Figure 1 In the example, the decision value of the purple sample point is set to 1, and the decision value of the green sample point is -1. At the same time, due to the basic assumption of the Lagrangian method for saddle points: the function is calculated for each variable separately. The value of each partial derivative function is 0 at the optimal solution. It can be deduced that α i y i The weighted sum of Φ(x i ) represents the kernel function of the input vector. In conventional linear problem solving, Φ(x) = x is generally taken. Different kernel functions can also be selected to correct the weights of the input vector for nonlinear problems.

[0037] From the above introduction, it is not difficult to see that SVM is an algorithm with strict mathematical theory support. It is good at finding key sample points as support vectors for classification. Moreover, since the classification judgment process is only linked to a small number of support vectors, complex calculations are avoided when calling, making it very convenient to use.

[0038] 2. Implementation Cases:

[0039] (1) Set the recipe, such as Figure 2 and Figure 3As shown, different wafers exhibit different temperature responses under the same base voltage. Therefore, the present invention maintains a constant heating power, using an initial voltage of 5% outer ring power and 20% inner ring power. Following process requirements, heating is performed for 5 seconds during the pin-up phase and then for another 5 seconds after the pin-down phase. For multiple experiments, the temperature difference between the pin-down phase and the temperature 2 seconds after the pin-down phase is obtained, as well as the temperature data at 5 seconds after the pin-down phase. Because these two sets of data have higher resolution than other time points, they are selected along with the wafer emissivity parameters to form the data set, resulting in the data shown in Table 1.

[0040] Table 1

[0041]

[0042] (2) The experimental data are median filtered and preprocessed. The target label is the type of wafer, and the data labels are the emissivity of the wafer, the 7-way temperature difference, and the 7-way temperature value. All the data are divided into a training set and a test set with a ratio of 3:1.

[0043] (3) Use support vector machine (SVM), neural network (NN), RF (random forest), and KNN (K nearest neighbors) algorithms to train the data.

[0044] (4) Deploy the trained model, and the results are as follows Figure 5 As shown in Table 2.

[0045] Table 2

[0046] algorithm 5-fold cross validation results Neural Network (NN) 20.67％(+ / -6.46％) Random Forest (RF) 93.33％(+ / -13.33％) Support Vector Machine Linear (SVM) 96.67％(+ / -6.67％) K-nearest neighbors (KNN) 79.33％(+ / -16.25％)

[0047] Finally, the support vector machine was selected as the wafer type recognition model. The wafers that needed to be classified were heated for 10 seconds, and the obtained data labels were input into the trained support vector machine to obtain the wafer category. The rapid heating process was implemented according to the obtained wafer category.

[0048] Corresponding to the aforementioned embodiment of a wafer classification method for a wafer rapid heating process, the present invention also provides an embodiment of a wafer classification device for a wafer rapid heating process.

[0049] See also Figure 6 An embodiment of the present invention provides a wafer sorting device for a wafer rapid heating process, comprising a memory and one or more processors, wherein the memory stores executable code, and when the processor executes the executable code, it is used to implement a wafer sorting method for a wafer rapid heating process in the above embodiment.

[0050] The embodiment of a wafer sorting device for a wafer rapid heating process provided by the present invention can be applied to any device with data processing capabilities, and the device with data processing capabilities can be a device or apparatus such as a computer. The device embodiment can be implemented through software, or through hardware or a combination of software and hardware. Taking software implementation as an example, as a device in a logical sense, it is formed by the processor of any device with data processing capabilities in which it is located reading the corresponding computer program instructions in the non-volatile memory into the memory for execution. From the hardware level, if Figure 6 As shown, it is a hardware structure diagram of a wafer sorting device for wafer rapid heating process provided by the present invention, in which any device with data processing capability is located, except Figure 6 In addition to the processor, memory, network interface, and non-volatile memory shown, any device with data processing capabilities in which the apparatus in the embodiment is located may also include other hardware, generally based on the actual functions of the device with data processing capabilities, which will not be described in detail.

[0051] The implementation process of the functions and effects of each unit in the above-mentioned device is specifically described in the implementation process of the corresponding steps in the above-mentioned method, and will not be repeated here.

[0052] For the device embodiments, since they basically correspond to the method embodiments, the relevant parts can be referred to the partial description of the method embodiments. The device embodiments described above are merely illustrative, wherein the units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed across multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the present invention. A person of ordinary skill in the art can understand and implement the present invention without inventive work.

[0053] An embodiment of the present invention further provides a computer-readable storage medium having a program stored thereon. When the program is executed by a processor, a wafer classification method for a wafer rapid heating process in the above embodiment is implemented.

[0054] The computer-readable storage medium may be an internal storage unit of any device with data processing capabilities described in any of the aforementioned embodiments, such as a hard disk or memory. The computer-readable storage medium may also be an external storage device of any device with data processing capabilities, such as a plug-in hard disk, a smart media card (SMC), an SD card, a flash card, etc. equipped on the device. Furthermore, the computer-readable storage medium may also include both an internal storage unit and an external storage device of any device with data processing capabilities. The computer-readable storage medium is used to store the computer program and other programs and data required by any device with data processing capabilities, and may also be used to temporarily store data that has been output or is to be output.

[0055] The present invention also provides a computer program product, comprising a computer program, which, when executed by a processor, implements the wafer classification method for a wafer rapid heating process.

[0056] The above embodiments are used to illustrate the present invention rather than to limit the present invention. Any modifications and changes made to the present invention within the spirit of the present invention and the protection scope of the claims shall fall within the protection scope of the present invention.

Claims

1. A wafer sorting method for a wafer rapid heating process, characterized in that: The method comprises the following steps: (1) Obtaining several types of wafers, heating them for a period of time to obtain temperature change data to construct a data set. The temperature change data includes emissivity, temperature value after heating, and temperature difference with resolution during heating. A support vector machine is trained based on the data set. (2) The wafer to be classified is heated for the same time as in step (1), and the obtained temperature change data is input into the trained support vector machine to obtain the wafer category, and a rapid heating process is implemented according to the obtained wafer category.

2. The wafer sorting method for a wafer rapid heating process according to claim 1, characterized in that: In step (1), the heating power is controlled to be constant during heating, and the heating time is 10 seconds, wherein the heating is performed for 5 seconds during the pin-up stage and then for 5 seconds after the pin-down stage.

3. The wafer sorting method for a wafer rapid heating process according to claim 2, characterized in that: The temperature difference with resolution is the temperature difference between the pin-down moment of the multi-channel experiment and the temperature 2 seconds after the pin-down; the temperature value after heating is the temperature value data 5 seconds after the pin-down.

4. The wafer sorting method for a wafer rapid heating process according to claim 1, characterized in that: The data set includes the emissivity of the wafer, 7-way temperature difference and 7-way temperature value. All data are divided into training set and test set with a ratio of 3:

1.

5. The wafer sorting method for a wafer rapid heating process according to claim 1, characterized in that: After median filtering and preprocessing, the temperature change data in the dataset is input into the support vector machine and trained with the corresponding wafer categories as labels.

6. The wafer sorting method for a wafer rapid heating process according to claim 1, characterized in that: Confirm the wafer rapid heating process based on the thermal response characteristics of different wafer types.

7. A wafer sorting device for a wafer rapid heating process, comprising a memory and one or more processors, wherein the memory stores executable code, characterized in that: When the processor executes the executable code, a wafer classification method for a wafer rapid heating process according to any one of claims 1 to 6 is implemented.

8. A computer-readable storage medium having a program stored thereon, characterized in that: When the program is executed by a processor, a wafer classification method for a wafer rapid heating process according to any one of claims 1 to 6 is implemented.

9. A computer program product comprising a computer program, characterized in that When the computer program is executed by a processor, a wafer classification method for a wafer rapid heating process as described in any one of claims 1 to 6 is implemented.

Images