Uniform Radiation Heating Control Architecture

Abstract

[0005] Embodiments disclosed herein include a method for modeling a rapid thermal processing (RTF) tool. In one embodiment, the method includes developing a lamp model for the RTF tool, the lamp model including a plurality of lamp zones, calculating an irradiance graph for the plurality of lamp zones, multiplying irradiance values ​​of the plurality of lamp zones in the irradiance graph by an existing RTF tool power at a given time during a process recipe, summing the multiplied irradiance values ​​for the plurality of lamp zones to form an exposure graph of the lamp model, using the exposure graph as an input to a machine learning algorithm, and outputting a temperature across a virtual substrate from the machine learning algorithm.

Description

Technical Field

[0001] Cross - Reference to Related Applications This application claims the benefit of priority of U.S. Patent Application No. 17 / 695,619, filed Mar. 15, 2022, the entire content of which is incorporated herein by reference.

[0002] Embodiments relate to the field of semiconductor manufacturing, and more particularly, to a process for estimating thermal uniformity across a substrate in a modeled tool.

Background Art

[0003] In semiconductor processing environments, for example, rapid thermal processing (RTP) tools are used to perform heat treatments (e.g., annealing) and grow material layers (e.g., oxide growth). In an RTP tool, an array of lamps is used to heat a substrate placed under those lamps. Also, in some cases, a reflector may be provided under the substrate. Temperature control across the surface of the substrate is an important parameter of the RTP tool. Often, it is desired that the temperature be substantially uniform across the diameter of the substrate.

[0004] To control the temperature, RTP tools often include lamps grouped into multiple zones. Lamps within a single zone can be supplied with the same power, and different zones can have different power levels. For example, the power of a zone near the center of the substrate can be different from the power of a zone near the edge of the substrate.

[0005] Temperature control across the substrate is a complex engineering challenge. While placed over a region of the substrate, lamp irradiation can also heat adjacent regions of the substrate. Thermal modeling needs to take into account, among many other parameters, chamber wall temperature, edge ring temperature, reflector material, and substrate material.

[0006] Therefore, modeling RTP tools is extremely difficult. Furthermore, the models created are computationally intensive and require long time periods to generate thermal behavior within the system. The complexity of forming such models makes it difficult to model new RTP tool designs. For example, it may be desirable to reduce the number of lamps in an RTP tool (e.g., to reduce manufacturing costs, reduce power consumption). However, existing solutions limit the ability to test new designs before they are implemented in physical form. [Overview of the Initiative]

[0007] Embodiments disclosed herein include methods for modeling a rapid heat treatment (RTP) tool. In one embodiment, the method involves developing a lamp model of an RTP tool, wherein the lamp model includes a plurality of lamp zones; calculating an irradiance graph for the plurality of lamp zones; multiplying the irradiance values ​​of the plurality of lamp zones in the irradiance graph by the power of an existing RTP tool at a given time in the process policy; summing the multiplied irradiance values ​​for the plurality of lamp zones to form an irradiation graph of the lamp model; using the irradiation graph as input to a machine learning algorithm; and outputting a temperature across a hypothetical substrate from the machine learning algorithm.

[0008] Embodiments may also include a non-temporary computer-readable medium containing program instructions for causing a computer to carry out the method. In one embodiment, the method is to develop a lamp model for an RTP tool, the lamp model comprising a plurality of lamp zones; to calculate an irradiance graph for the plurality of lamp zones; to multiply the irradiance values ​​of the plurality of lamp zones in the irradiance graph by the power of an existing RTP tool at a given time in the process policy; to sum the multiplied irradiance values ​​for the plurality of lamp zones to form an irradiation graph of the lamp model; to use the irradiation graph as input to a machine learning algorithm; and to output a temperature across a virtual substrate from the machine learning algorithm.

[0009] Embodiments may also include a method for modeling a rapid heat treatment (RTP) tool. In one embodiment, the method includes training a machine learning algorithm with training data including real temperature data from an existing RTP tool; developing a lamp model of the RTP tool, wherein the lamp model includes multiple lamp zones and the number of lamps in the lamp model is different from the number of lamps in the existing RTP tool; calculating an irradiance graph for the multiple lamp zones; multiplying the irradiance values ​​of the multiple lamp zones in the irradiance graph by the power of the existing RTP tool at a given time in the process policy; summing the multiplied irradiance values ​​for the multiple lamp zones to form an irradiation graph of the lamp model; using the irradiation graph as input to a machine learning algorithm; and outputting a temperature across a virtual substrate from the machine learning algorithm. [Brief explanation of the drawing]

[0010] [Figure 1A] This is a plan view of a lamp array for an existing rapid heat treatment (RTP) tool according to one embodiment. [Figure 1B]This is a plan view of a ramp array for an RTP tool being investigated by the processing method described herein, according to one embodiment. [Figure 2] This is a graph of the irradiance of a lamp array on a substrate, having multiple zones, according to one embodiment. [Figure 3A] This is a graph of the power applied to lamps in several groups over the duration of a treatment strategy, according to one embodiment. [Figure 3B] This is a graph of the substrate temperature at different locations over the duration of the processing method according to one embodiment. [Figure 3C] This is a graph of the temperature over the radius of a substrate used as a training dataset, according to one embodiment. [Figure 4] This is a graph of irradiation over the radius of a substrate used as input for a machine learning (ML) algorithm, according to one embodiment. [Figure 5] This is a schematic diagram of an ML algorithm used in one embodiment to convert input irradiation across a substrate into a temperature output across the substrate. [Figure 6] This is a process flow diagram illustrating a process for determining the temperature profile of a heated virtual substrate in a modeled RTP tool, according to one embodiment. [Figure 7] This is a block diagram of an exemplary computer system that may be used with a processing tool according to one embodiment. [Modes for carrying out the invention]

[0011] The system described herein includes a process for estimating thermal uniformity across a substrate in a modeled tool. Numerous specific details are provided in the following description to provide a complete understanding of the embodiments. It will be apparent to those skilled in the art that embodiments can be practiced without these specific details. In other cases, well-known aspects are not described in detail so as not to unnecessarily obscure the embodiments. Furthermore, it should be understood that the various embodiments shown in the accompanying drawings are illustrative and not necessarily drawn to a specific scale.

[0012] As mentioned above, modeling the rapidly heat-treating (RTP) tools currently under development is difficult. Therefore, it is not currently feasible to determine the substrate temperature profile in design without using overly complex models or actually building the RTP tool. This leads to excessive waste of resources and time. This becomes a problem, especially when redesigning the RTP tool is required. For example, it may be desirable to reduce the number of lamps in a lamp array to reduce costs and / or power consumption.

[0013] Accordingly, embodiments disclosed herein include methods for generating temperature profiles using machine learning (ML) algorithms. Generally, a new lamp design is created. The irradiance of the lamp on the substrate is calculated. This provides a graph of the irradiance supplied by multiple lamp zones. The irradiance can then be multiplied by the power in the policy supplied to each individual zone. The obtained values ​​for each zone can then be summed together to provide a graph of irradiation across the surface of the substrate. In one embodiment, the irradiation values ​​can then be input to an ML algorithm. The ML algorithm can output a temperature profile for the RTP tool being investigated. Thus, it is not necessary to model or construct the RTP tool extensively to determine the temperature profile.

[0014] Referring now to Figure 1A, a plan view of a lamp array 150 of an RTP tool according to one embodiment is shown. As shown, the lamp array 150 includes a plurality of lamps 155 configured in a given pattern. For example, the pattern may be a honeycomb pattern. Each of the lamps 155 in the lamp array 150 may be powered to heat a substrate (not shown) underneath. The substrate may be a semiconductor substrate, such as a silicon wafer. However, it should be noted that other substrates may also be used (for example, a glass substrate).

[0015] In one embodiment, the lamps 155 may be separated into multiple groups (also called zones). These zones may be substantially concentric. In a simple case, the first zone may be the central zone, and the second zone may be the group of lamps 155 outside the first zone. However, it should be noted that in more complex tools, the number of zones can be significantly higher. For example, there may be up to 15 (or more) zones in a given lamp array.

[0016] For convenience, the lamp array 150 may be considered a physical lamp array 150 in this specification; that is, the lamp array 150 may be an existing array that has already been designed and assembled. The lamp array 150 may be used for training purposes to instruct machine learning (ML) algorithms in order to help develop new RTP architectures.

[0017] Referring next to Figure 1B, a plan view of a lamp array 160 according to an additional embodiment is shown. As shown, the lamp array 160 may include a number of lamps 165. The configuration (and / or number) of lamps 165 in the lamp array 160 may differ from the number and / or configuration of lamps 155 in the lamp array 150 described above. For example, the lamp array 160 may have fewer lamps 165 than the number of lamps 155 in the lamp array 150. Furthermore, there may be empty spaces 166 where there are no lamps 165. Apart from the empty spaces 166, the lamps 165 may be configured in a honeycomb type configuration. However, it should be noted that other configurations (e.g., different packing methods, different spacing (or pitch), etc.) may be used.

[0018] As will be explained in more detail below, the lamp array 160 may be a theoretical or virtual lamp array 160; that is, the lamp array 160 may not be physically constructed. However, as a result of analysis methods, such as those explained in more detail below, the lamp array 160 may be analyzed to determine the temperature profile that will be implemented on the substrate. Thus, the output of the RTP tool can be characterized and compared to existing solutions to determine whether the design should be built into an actual product. This saves design time and cost because a faulty lamp array 160 can be excluded from consideration.

[0019] Next, referring to FIG. 2, a graph of the irradiance of the lamps 165 across the surface of the substrate according to one embodiment is shown. The irradiance shown in FIG. 2 is divided into a set of nine zones. Each zone may include a plurality of individual lamps 165. Although nine zones are shown in FIG. 2, it should be understood that the lamps 165 may be grouped into any number of zones (e.g., two or more zones). In certain embodiments, there may be 14 zones. The irradiance may be the calculated irradiance. That is, the irradiance may not necessarily be a measured quantity. Thus, the graph shown in FIG. 2 may be generated without actually fabricating the lamp array 160. In one embodiment, the irradiance values (Y-axis) are plotted against the radius of the substrate (X-axis). For example, the X-axis may range from 0 mm (i.e., the center of the substrate) to about 150 mm (i.e., the edge of the substrate). In such an embodiment, the substrate is a 300 mm substrate. However, it should be understood that substrates having other form factors according to other embodiments may also be used.

[0020] Next, referring to FIG. 3A, a graph of the power of different lamp groups as a function of time according to one embodiment is shown. In one embodiment, the graph in FIG. 3A may be a graph of a physical system. That is, the system shown in FIG. 3A may be actually constructed. For example, a lamp array similar to the lamp array 150 may be used to generate the graph in FIG. 3A. In FIG. 3A, seven groups are shown. However, it should be understood that any number of groups (e.g., two or more groups) may be used according to one embodiment. Each group may include two or more lamps, such as the lamps 155 described above.

[0021] The graph in FIG. 3A can be a graph of power during the duration of a process strategy. For example, the process strategy can have a duration of about 225 seconds. However, it should be understood that the process strategy can have any duration in order to provide the desired result regarding the substrate. As shown, the process strategy can include a thermal ramp up region around 70 seconds. The thermal ramp up region represents a rapid increase in the temperature of the substrate. After the ramp up region, a thermal soak is implemented. The thermal soak is a time period during which the substrate is held at a substantially constant elevated temperature. After the desired time of the thermal soak, a thermal ramp down region returns the temperature of the substrate to room temperature.

[0022] In the illustrated embodiment, Group 1 (G1) can be at the center of the substrate and Group 7 (G7) can be at the edge of the substrate. As shown, Group 7 can have the highest power during the thermal soak and Group 1 can have the lowest power during the thermal soak. The remaining groups (G2 - G6) can have power between Group 1 and Group 7.

[0023] Next, referring to FIG. 3B, a graph of the substrate temperature during the duration of a strategy according to one embodiment is shown. In one embodiment, the temperature graph can include a plurality of groups (e.g., G1 - G7). Groups G1 - G7 can be substantially similar to groups G1 - G7 described above with respect to FIG. 3A. Thus, although seven groups are shown, it should be understood that any number of groups (e.g., two or more groups) can be used according to one embodiment. As shown, the temperatures of G1 - G7 are substantially uniform with respect to each other, despite having significantly different power levels (as shown in FIG. 3A). As shown, there is a thermal ramp at about 100 seconds and after the thermal ramp, a thermal soak continues from about 110 seconds to about 160 seconds.

[0024] Next, referring to Figure 3C, a graph of the substrate temperature at a given time according to one embodiment is shown. The Y-axis may represent temperature, and the X-axis may represent distance from the center of the substrate. As shown, six positions 3711 to 3716 are shown. Each of the positions 371 may be a pyrometer location, which measures the substrate temperature at that location. In one embodiment, the portion of the curve between the six positions 371 may be a fitted line; that is, there may be no actual temperature measurements between positions 371.

[0025] In certain embodiments, the time moment represented by the graph in Figure 3C may be the point indicated by the dashed line 372 in Figures 3A and 3B. For example, the temperature snapshot in Figure 3C may be around 150 seconds into the process policy. That is, in some embodiments, this time may be around the end of the heat soak. However, it should be understood that the temperature snapshot may be provided at any time in the process policy.

[0026] In one embodiment, the temperature snapshot in Figure 3C may be used as a training dataset. For example, a machine learning (ML) algorithm, described in more detail below, may use the temperature snapshot and other data in Figures 3A and 3B as a set of training data. The snapshot in Figure 3C may be the output value, and other data from the graphs in Figures 3A and 3B may be used as input data.

[0027] Next, referring to Figure 4, a graph of irradiation pair positions across a substrate is shown according to one embodiment. In one embodiment, the graph in Figure 4 is generated from the data in the irradiance graph in Figure 2. In detail, the irradiance values ​​in Figure 2 are multiplied by the power values ​​in Figure 3A at time 372 of the dashed line. After the irradiance values ​​are multiplied, each group is summed together to provide an irradiation value. This irradiation value shown in Figure 4 can then be used as input to an ML algorithm to output a temperature snapshot, similar to the embodiment shown in Figure 3C. In this way, temperature uniformity can be predicted without actually having to build and test the lamp array. A more detailed description of the process for generating the temperature uniformity plot is given below in more detail with respect to Figure 6.

[0028] Next, referring to Figure 5, a schematic diagram of an ML algorithm 580 according to one embodiment is shown. In one embodiment, the ML algorithm may include an input side 581 and an output side 582. Multiple hidden layers 583 may be provided between the input side 581 and the output side 582. Each hidden layer 583 may include multiple nodes 584 that are connected to each other in a communicative manner (as shown by lines between nodes). In one embodiment, two hidden layers 583 are shown. However, please understand that any number of hidden layers may be used depending on the complexity of the ML algorithm.

[0029] In one embodiment, the ML algorithm takes irradiation values ​​(similar to the graph shown in Figure 4, for example) as input and outputs a temperature uniformity plot (similar to the graph shown in Figure 3C, for example). The structure of the ML algorithm can be any type of ML algorithm. For example, the ML algorithm can be a supervised ML algorithm, a semi-supervised ML algorithm, an unsupervised ML algorithm, an enhanced ML algorithm, etc.

[0030] Referring next to Figure 6, a process flow diagram is shown illustrating a process 690 for modeling an RTP tool according to one embodiment. In one embodiment, process 690 may begin with operation 691, which includes training an ML algorithm with training data, including real temperature data from an existing RTP tool. For example, the training data may be obtained from an RTP tool having a lamp array, similar to the lamp array 150 described in more detail above. The training data may include information about the power supplied to various lamp zones in the lamp array, the temperature of the various zones over time, and snapshots of the temperature across the substrate at a given time. For example, the snapshot of the temperature across the substrate may be similar to the snapshot graph shown in Figure 3C. The snapshot of the temperature across the substrate may be the output value of the ML algorithm, and other data may be fed as input data to the ML algorithm. In some embodiments, there may be two or more training datasets used. For example, there may be up to 25 or more training datasets to properly train the ML algorithm.

[0031] In one embodiment, process 690 may continue with operation 692, which includes developing a ramp model for an RTP tool. In one embodiment, the ramp model for an RTP tool may have a different configuration from the ramp array of an existing RTP tool used in the ML algorithm training process. For example, the ramp model may have a ramp array in which the individual ramps have different layouts and / or different numbers of ramps. In certain embodiments, it is desirable that the RTP tool being investigated has the same or similar performance as an existing RTP tool, but includes fewer ramps to allow for cost and power reductions.

[0032] In one embodiment, process 690 may continue with operation 693, which includes calculating irradiance graphs for multiple zones of a lamp model. In one embodiment, the irradiance graphs may be similar to the graph shown in Figure 2 above, that is, several different zones across the surface of a substrate and the irradiance of those zones are provided. The irradiance can be calculated. Since the value is calculated, there is no need to physically construct the lamp model.

[0033] In one embodiment, process 690 may follow operation 694, which includes multiplying the irradiance values ​​of several lamp zones in an irradiance graph by the power of an existing RTP tool at a given time in the process policy. For example, the power value may be provided by a graph, such as the graph shown in Figure 3A. The given time may be indicated by the dashed line 372. For example, the power level may be during a thermal soak. In another embodiment, the power level used may be during a thermal lamp. In yet another embodiment, the power at several different times is multiplied by the irradiance value.

[0034] In one embodiment, process 690 may continue with operation 695, which includes summing multiplied irradiance values ​​for multiple lamp zones to form an irradiation graph of the lamp model. The irradiation graph of the lamp model may be similar to the graph shown in Figure 4, i.e., irradiation across the surface of the substrate may be provided. In embodiments, there may be multiple irradiation graphs provided, where power at multiple different times is multiplied by the irradiance value.

[0035] In one embodiment, process 690 may continue with operation 696, which includes using an irradiation graph (or multiple graphs) as input to an ML algorithm. The irradiation graph (or multiple graphs) may be input to the ML algorithm trained in operation 691. In one embodiment, process 690 may continue with operation 697, which includes outputting the temperature across a virtual substrate from the machine learning algorithm. Thus, the performance of the RTP tool can be determined without the need to build a model of the RTP tool. Therefore, many different models can be easily investigated using a similar process to select the best candidate for further consideration regarding minimum cost and development time.

[0036] Referring now to Figure 7, a block diagram of an exemplary computer system 700 of a processing tool according to one embodiment is shown. In one embodiment, the computer system 700 is coupled to the processing tool and controls the processing in the processing tool. The computer system 700 may be connected to other machines (e.g., networked) in a local area network (LAN), intranet, extranet, or internet. The computer system 700 may operate as a server machine or a client machine in a client-server network environment, or as a peer machine in a peer-to-peer (or distributed) network environment. The computer system 700 may be a personal computer (PC), tablet PC, set-top box (STB), personal digital assistant (PDA), cellular telephone, web appliance, server, network router, switch or bridge, or any machine capable of executing a set of instructions (sequential or otherwise) specifying the actions to be taken by that machine. Furthermore, although only a single machine is shown for computer system 700, the term “machine” shall also be interpreted to include any set of machines (e.g., computers) that individually or collectively execute a set (or set) of instructions to implement one or more of the methodologies described herein.

[0037] The computer system 700 may include a computer program product or software 722 having a non-temporary machine-readable medium storing instructions, which may be used to program the computer system 700 (or other electronic devices) to perform a process, according to the embodiment. The machine-readable medium includes any mechanism for storing or transmitting information in a form readable by a machine (e.g., a computer). For example, machine-readable (e.g., computer-readable) media include machine (e.g., computer)-readable storage media (e.g., read-only memory ("ROM"), random-access memory ("RAM"), magnetic disk storage media, optical storage media, flash memory devices, etc.), machine (e.g., computer)-readable transmission media (e.g., electrical, optical, acoustic or other forms of propagating signals (e.g., infrared signals, digital signals, etc.)), etc.

[0038] In one embodiment, the computer system 700 includes a system processor 702 that communicates with each other via a bus 730, a main memory 704 (for example, read-only memory (ROM), flash memory, dynamic random access memory (DRAM) such as synchronous DRAM (SDRAM) or Rambus DRAM (RDRAM), a static memory 706 (for example, flash memory, static random access memory (SRAM), etc.), and a secondary memory 718 (for example, a data storage device).

[0039] The system processor 702 represents one or more general-purpose processing devices, such as a microsystem processor or a central processing unit. More specifically, the system processor may be a composite instruction set computing (CISC) microsystem processor, a reduced instruction set computing (RISC) microsystem processor, a very long instruction word (VLIW) microsystem processor, a system processor implementing another instruction set, or a system processor implementing a combination of instruction sets. The system processor 702 may also be one or more dedicated processing devices, such as an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA), a digital signal system processor (DSP), or a network system processor. The system processor 702 is configured to execute processing logic 726 for performing the operations described herein.

[0040] The computer system 700 may further include a system network interface device 708 for communicating with other devices or machines. The computer system 700 may also include a video display unit 710 (e.g., a liquid crystal display (LCD), a light-emitting diode display (LED), or a cathode ray tube (CRT)), an alphanumeric input device 712 (e.g., a keyboard), a cursor control device 714 (e.g., a mouse), and a signal generation device 716 (e.g., a speaker).

[0041] The secondary memory 718 may include a machine-accessible storage medium 732 (or more specifically, a computer-readable storage medium) storing one or more sets of instructions (e.g., software 722) that embody any one or more of the methodologies or functions described herein. The software 722 may also reside entirely or at least partially in the main memory 704 and / or the system processor 702 while the computer system 700 is executing the software 722, the main memory 704 and the system processor 702 also constitute the machine-readable storage medium. The software 722 may further be transmitted or received over the network 720 via a system network interface device 708. In one embodiment, the network interface device 708 may operate using RF coupling, optical coupling, acoustic coupling, or inductive coupling.

[0042] Although the machine-accessible storage medium 732 is shown to be a single medium in exemplary embodiments, the term “machine-readable storage medium” shall be interpreted to include a single or multiple mediums (e.g., a centralized or distributed database, and / or associated caches and servers) that store one or more sets of instructions. The term “machine-readable storage medium” shall also be interpreted to include any medium capable of storing or encoding a set of instructions for machine execution, causing a machine to implement one or more methodologies. The term “machine-readable storage medium” shall therefore be interpreted to include, but are not limited to, solid memory and optical and magnetic media.

[0043] The above specification describes specific exemplary embodiments. It will be apparent that various modifications can be made thereto without departing from the scope of the following claims. Therefore, this specification and the drawings should be considered illustrative rather than restrictive.

Claims

1. A method for modeling rapid heat treatment (RTP) tools, Developing a ramp model for an RTP tool, wherein the ramp model includes multiple ramp zones. Calculating irradiance graphs for the aforementioned multiple lamp zones, The irradiance values ​​of the multiple lamp zones in the irradiance graph are multiplied by the power of the existing RTP tool at a given time in the process strategy. To form the irradiation graph of the lamp model, the multiplied irradiance values ​​for the plurality of lamp zones are summed, The aforementioned irradiation graph is used as input to a machine learning algorithm, The machine learning algorithm outputs the temperature across the virtual board. Methods that include...

2. Training the machine learning algorithm with training data including real temperature data from the aforementioned existing RTP tool. The method according to claim 1, further comprising:

3. The method according to claim 2, wherein the training includes at least 25 sets of different training data.

4. The method according to claim 1, wherein the plurality of ramp zones include up to 15 ramp zones.

5. The method according to claim 1, wherein the lamp configuration of the lamp model is different from the lamp configuration of the existing RTP tool.

6. The method according to claim 5, wherein the number of lamps in the lamp configuration of the lamp model is different from the number of lamps in the lamp configuration of the existing RTP tool.

7. The method according to claim 1, wherein the machine learning algorithm includes two or more hidden layers.

8. The method according to claim 1, wherein the irradiation graph includes data points for at least 15 different locations on the virtual substrate.

9. The method according to claim 1, wherein the given time in the process strategy is during a thermal soak.

10. The method according to claim 1, wherein the given time in the process strategy is in the heat lamp.

11. The method according to claim 1, wherein the temperature across the virtual board matches the set of training data.

12. On the computer, Developing a ramp model for an RTP tool, wherein the ramp model includes multiple ramp zones. Calculating irradiance graphs for the aforementioned multiple lamp zones, The irradiance values ​​of the multiple lamp zones in the irradiance graph are multiplied by the power of the existing RTP tool at a given time in the process strategy. To form the irradiation graph of the lamp model, the multiplied irradiance values ​​for the plurality of lamp zones are summed, The aforementioned irradiation graph is used as input to a machine learning algorithm, The machine learning algorithm outputs the temperature across the virtual board. A non-temporary computer-readable medium containing program instructions for performing a method including [a specific method].

13. Training the machine learning algorithm with training data including real temperature data from the aforementioned existing RTP tool. A non-temporary computer-readable medium according to claim 12, further comprising:

14. The non-temporary computer-readable medium according to claim 13, wherein the training comprises at least 25 sets of different training data.

15. The non-temporary computer-readable medium according to claim 12, wherein the plurality of lamp zones include up to 15 lamp zones.

16. The non-temporary computer-readable medium according to claim 12, wherein the lamp configuration of the lamp model differs from the lamp configuration of the existing RTP tool.

17. The non-temporary computer-readable medium according to claim 16, wherein the number of lamps in the lamp configuration of the lamp model is different from the number of lamps in the lamp configuration of the existing RTP tool.

18. The method according to claim 1, wherein the given time in the process strategy is during a thermal soak and / or during a thermal lamp.

19. A method for modeling rapid heat treatment (RTP) tools, Training a machine learning algorithm with training data that includes real temperature data from existing RTP tools, To develop a ramp model for an RTP tool, wherein the ramp model includes multiple ramp zones, and the number of ramps in the ramp model is different from the number of ramps in the existing RTP tool. Calculating irradiance graphs for the aforementioned multiple lamp zones, Multiplying the irradiance values ​​of the multiple lamp zones in the irradiance graph by the power of the existing RTP tool at a given time in the process strategy, To form the irradiation graph of the lamp model, the multiplied irradiance values ​​for the plurality of lamp zones are summed, The irradiation graph is used as input to the machine learning algorithm, The machine learning algorithm outputs the temperature across the virtual board. Methods that include...

20. The method according to claim 19, wherein the given time in the process strategy is during a thermal soak and / or thermal lamp.

Images