High-precision temperature control method and system based on industrial integrated circuit manufacturing

Abstract

The invention relates to the technical field of remote monitoring, and discloses a high-precision temperature control method and system based on industrial integrated circuit manufacturing, and the method comprises the steps: configuring a temperature sensor array of a photoetching machine in the industrial integrated circuit manufacturing, collecting the multi-source temperature data of the photoetching machine, building a cavity digital twin model of the photoetching machine, and obtaining a digital twin model of the cavity of the photoetching machine according to the multi-source temperature data; calculating the three-dimensional temperature gradient of the digital twin model of the cavity; analyzing a heat flow evolution path of the photoetching machine to determine a local hot spot area of the photoetching machine; analyzing a photoetching machine working condition of the photoetching machine to define a multi-mode control instruction of the photoetching machine; and decomposing the mixed control instruction into a heater group control instruction, a cooling system control instruction and a gas flow field modulation instruction so as to execute high-precision temperature control of industrial integrated circuit manufacturing. According to the invention, the temperature control precision of industrial integrated circuit manufacturing can be improved.

Description

TECHNICAL FIELD

[0001] The application relates to a high-precision temperature control method and system based on industrial integrated circuit manufacturing, and belongs to the technical field of remote monitoring. BACKGROUND

[0002] High-precision temperature control in industrial integrated circuit manufacturing refers to a series of advanced control strategies and technical means for ensuring that key components of a photolithography machine and a process environment are maintained in an extremely stable and uniform temperature range during the entire exposure process, thereby guaranteeing photolithography precision, overlay precision and final device yield, through multi-source sensing, real-time modeling, closed-loop feedback and intelligent regulation. The stability and consistency of the process can be significantly improved through high-precision temperature control in industrial integrated circuit manufacturing, ensuring that the temperature fluctuation of the wafer is controlled within a very small range in key links such as thin film deposition, photolithography, etching and heat treatment, thereby effectively improving the uniformity of the thin film, reducing the defect density, optimizing the electrical characteristics of the device, and greatly improving the overall yield and reliability of the chip, laying a solid technical foundation for the continuous iteration and mass production of advanced processes.

[0003] In the traditional high-precision temperature control method of industrial integrated circuit manufacturing, a single or limited number of temperature sensors are usually used to monitor the photolithography machine in a point-by-point manner, and open-loop or single-loop closed-loop control is performed based on a simplified thermodynamic model or empirical rules. This method is difficult to fully capture the complex three-dimensional temperature distribution and dynamic heat flow changes inside the chamber, resulting in limited temperature control accuracy. SUMMARY

[0004] The application provides a high-precision temperature control method and system based on industrial integrated circuit manufacturing, which mainly aims to improve the temperature control accuracy of industrial integrated circuit manufacturing.

[0005] To achieve the above-mentioned purpose, the application provides a high-precision temperature control method based on industrial integrated circuit manufacturing, which comprises: configuring a temperature sensor array of a photolithography machine in industrial integrated circuit manufacturing, wherein the temperature sensor array comprises a substrate level sensor group, a gas phase level sensor group and a chamber wall level sensor group; acquiring multi-source temperature data of the photolithography machine based on the temperature sensor array, establishing a chamber digital twin model of the photolithography machine, and calculating a three-dimensional temperature gradient of the chamber digital twin model according to the multi-source temperature data; According to the three-dimensional temperature gradient, the heat flow evolution path of the photolithography machine is analyzed to determine the local hot spot area of the photolithography machine; Based on the local hot spot area, the working condition of the photolithography machine is analyzed to define the multi-modal control instruction of the photolithography machine; The mixed control instruction is decomposed into a heater group control instruction, a cooling system control instruction, and a gas flow field modulation instruction to perform high-precision temperature control of the industrial integrated circuit manufacturing.

[0006] Optionally, the configuration of the temperature sensor array of the photolithography machine in the industrial integrated circuit manufacturing includes: Embedding a TFTC array on the wafer chuck surface of the photolithography machine; And configuring an FBG sensor on the edge of the chuck of the photolithography machine; Configuring a substrate-level sensor group of the photolithography machine through the TFTC array and the FBG sensor; Configuring a non-contact infrared laser spectrometer of the gas injection manifold, chamber center, and exhaust port of the photolithography machine; Establishing a gas-phase-level sensor group of the photolithography machine in combination with the non-contact infrared laser spectrometer; Installing a multi-spectral infrared thermal imager on the chamber sidewall, top cover, and pedestal of the photolithography machine; Identifying the key mechanical joints of the photolithography machine to configure a surface acoustic wave temperature sensor at the key mechanical joints; Combining the multi-spectral infrared thermal imager and the surface acoustic wave temperature sensor to construct a chamber wall-level sensor group of the photolithography machine; Based on the substrate-level sensor group, the gas-phase-level sensor group, and the chamber wall-level sensor group, a temperature sensor array of the photolithography machine in the industrial integrated circuit manufacturing is constructed.

[0007] Optionally, the acquisition of the multi-source temperature data of the photolithography machine based on the temperature sensor array includes: Establishing a time-sharing multiplexing strategy of the temperature sensor array to acquire initial multi-source temperature data of the photolithography machine; Constructing sensor coordinates of the temperature sensor array; And unifying the initial multi-source temperature data to the sensor coordinates to obtain coordinate-unified temperature data; Organizing a 5D tensor of the coordinate-unified temperature data, and fusing the 5D tensor to obtain the multi-source temperature data of the photolithography machine.

[0008] Optionally, the establishment of the chamber digital twin model of the photolithography machine includes: Constructing a three-dimensional entity model of the photolithography machine; And giving the three-dimensional entity model a temperature-dependent material coefficient to obtain an attribute three-dimensional entity model; Establishing a control equation of the attribute three-dimensional entity model, wherein the control equation includes a heat conduction equation, a fluid dynamics equation, and an electromagnetic field equation; mapping multi-source temperature data of the lithography machine to a model boundary of the attribute three-dimensional entity model to obtain a model boundary condition; based on the attribute three-dimensional entity model, the control equation and the model boundary condition, establishing a chamber digital twin model of the lithography machine.

[0009] Optionally, the calculating a three-dimensional temperature gradient of the chamber digital twin model according to the multi-source temperature data comprises: preprocessing the multi-source temperature data to obtain processed multi-source temperature data; labeling a grid node of the chamber digital twin model; defining a temperature scalar field of the grid node; calculating a three-dimensional temperature gradient of the chamber digital twin model according to the temperature scalar field and the processed multi-source temperature data.

[0010] Optionally, the analyzing a heat flow evolution path of the lithography machine according to the three-dimensional temperature gradient comprises: calculating a heat flow vector field of the lithography machine according to the three-dimensional temperature gradient; analyzing a heat flow line set of the lithography machine based on the heat flow vector field; analyzing a feature position coordinate and a classification label of the lithography machine through the heat flow line set; determining a heat flow evolution path of the lithography machine based on the feature position coordinate and the classification label.

[0011] Optionally, the calculating a heat flow vector field of the lithography machine according to the three-dimensional temperature gradient comprises: calculating a heat flow density vector of the lithography machine according to the three-dimensional temperature gradient by using the following formula:

[0012] wherein, represents a heat flow density vector of the lithography machine, represents an anisotropic thermal conductivity coefficient tensor, represents a direction, represents a temperature, represents a three-dimensional temperature gradient, represents a plasma correction coefficient, represents an electron density, represents a Boltzmann constant, represents an electron temperature, represents a logarithmic gradient of the electron temperature, represents a fluid density, represents a specific heat capacity at constant pressure, represents a turbulent thermal diffusivity; construct a heat flow vector field of the lithography machine based on the heat flow density vector.

[0013] Optionally, the determining the local hotspot region of the lithography machine comprises: calculating a path energy flux of the corresponding heat flow evolution path of the lithography machine; analyzing an energy concentration index of the lithography machine according to the path energy flux to construct an energy concentration distribution map of the lithography machine; establishing a topological hotspot coordinate list of the lithography machine; generating a candidate hotspot region coordinate of the lithography machine according to the energy concentration distribution map and the topological hotspot coordinate list to determine a local hotspot region boundary of the lithography machine; determining the local hotspot region of the lithography machine through the local hotspot region boundary.

[0014] Optionally, the decomposing the mixed control instruction into the heater group control instruction, the cooling system control instruction and the gas flow field modulation instruction comprises: establishing a subsystem of the lithography machine corresponding to the mixed control instruction, wherein the subsystem comprises a heater group subsystem, a cooling subsystem and a gas flow field modulation subsystem; mapping the mixed control instruction to the subsystem to obtain a heating index, a cooling index and a gas flow field modulation index; generating the heater group control instruction, the cooling system control instruction and the gas flow field modulation instruction of the lithography machine according to the heating index, the cooling index and the gas flow field modulation index.

[0015] To solve the above problems, the application further provides a high-precision temperature control system based on industrial integrated circuit manufacturing, which comprises: a sensor array construction module configured to configure a temperature sensor array of a lithography machine in industrial integrated circuit manufacturing, wherein the temperature sensor array comprises a substrate level sensor group, a gas phase level sensor group and a cavity wall level sensor group; a temperature gradient calculation module configured to collect multi-source temperature data of the lithography machine based on the temperature sensor array, establish a chamber digital twin model of the lithography machine, and calculate a three-dimensional temperature gradient of the chamber digital twin model according to the multi-source temperature data; a hotspot region analysis module configured to analyze a heat flow evolution path of the lithography machine according to the three-dimensional temperature gradient to determine a local hotspot region of the lithography machine; a control instruction determination module configured to analyze a lithography machine working condition of the lithography machine based on the local hotspot region to define a multi-modal control instruction of the lithography machine; A temperature precision control module is configured to decompose the mixed control instruction into a heater group control instruction, a cooling system control instruction, and a gas flow field modulation instruction to perform high-precision temperature control of the industrial integrated circuit manufacturing.

[0016] The scheme realizes all-around and three-dimensional real-time monitoring of the temperature distribution inside the chamber by configuring a multi-level temperature sensor array including a substrate level, a gas phase level, and a chamber wall level in the industrial integrated circuit manufacturing photolithography machine, thereby significantly improving the comprehensiveness and accuracy of temperature data acquisition. The chamber digital twin model constructed based on these multi-source temperature data can accurately reproduce the thermodynamic state inside the chamber and reveal the evolution path of heat flow in space through three-dimensional temperature gradient calculation, effectively identifying local hot spot areas that are difficult to capture by traditional control methods. This hot spot identification mechanism based on digital twinning enables temperature control to go beyond single-point or empirical adjustment and possess global perception and prediction capabilities, significantly improving the targeting and foresight of control. Furthermore, in combination with local hot spot distribution and real-time working condition analysis, the scheme can intelligently define multi-modal control instructions and accurately decompose them into coordinated control strategies for the heater group, cooling system, and gas flow field. This multi-executive mechanism linkage control method not only enables rapid response to temperature changes but also realizes fine adjustment and dynamic balance of the thermal field, effectively suppressing temperature fluctuations and process drift, ensuring high uniformity and stability of the temperature during wafer processing. Ultimately, the scheme significantly improves the controllability and consistency of the process, reduces the defect rate caused by temperature abnormalities, improves the yield and reliability of chip manufacturing, provides solid technical support for the development and mass production of advanced process technology, and has the potential to be promoted to other complex industrial thermal control systems. Therefore, the present application can improve the temperature control precision of industrial integrated circuit manufacturing. BRIEF DESCRIPTION OF DRAWINGS

[0017] Figure 1 A flowchart of the high-precision temperature control method based on industrial integrated circuit manufacturing is provided for an embodiment of the present application. Figure 2 A module diagram for implementing the high-precision temperature control method based on industrial integrated circuit manufacturing is provided for an embodiment of the present application.

[0018] The object implementation, functional features, and advantages of the present application will be further described with reference to the embodiments and the accompanying drawings. DETAILED DESCRIPTION

[0019] It should be understood that the specific embodiments described herein are merely intended to explain the present application and are not intended to limit the present application.

[0020] The embodiment of the present application provides a high-precision temperature control method in industrial integrated circuit manufacturing. The execution subject of the high-precision temperature control method in industrial integrated circuit manufacturing includes but is not limited to at least one of electronic devices such as a server, a terminal and the like which can be configured to execute the method provided by the embodiment of the present application. In other words, the high-precision temperature control method in industrial integrated circuit manufacturing can be executed by software or hardware installed in a terminal device or a server device. The server includes but is not limited to a single server, a server cluster, a cloud server or a cloud server cluster and the like.

[0021] Referring to Figure 1 FIG. 1 is a flowchart of a high-precision temperature control method in industrial integrated circuit manufacturing provided by an embodiment of the present application. In the embodiment, the high-precision temperature control method in industrial integrated circuit manufacturing includes the following steps. S1, configuring a temperature sensor array of a lithography machine in industrial integrated circuit manufacturing, wherein the temperature sensor array includes a substrate-level sensor group, a gas-phase-level sensor group and a cavity-wall-level sensor group.

[0022] The present application configures a temperature sensor array of a lithography machine in industrial integrated circuit manufacturing to realize omnidirectional and stereoscopic real-time monitoring of temperature distribution inside a chamber, thereby significantly improving the comprehensiveness and accuracy of temperature data acquisition.

[0023] In detail, the configuration of the temperature sensor array of the lithography machine in industrial integrated circuit manufacturing includes the following steps. embedding a TFTC array on a wafer carrier disc surface of the lithography machine; and configuring an FBG sensor at an edge of the carrier disc of the lithography machine; configuring a substrate-level sensor group of the lithography machine through the TFTC array and the FBG sensor; configuring a non-contact infrared laser spectrometer of a gas injection manifold, a chamber center and an exhaust port of the lithography machine; establishing a gas-phase-level sensor group of the lithography machine in combination with the non-contact infrared laser spectrometer; installing a multispectral infrared thermal imager on a chamber sidewall, a top cover and a base of the lithography machine; identifying a key mechanical connection of the lithography machine to configure a surface acoustic wave temperature sensor at the key mechanical connection; establishing a cavity-wall-level sensor group of the lithography machine in combination with the multispectral infrared thermal imager and the surface acoustic wave temperature sensor; based on the substrate-level sensor group, the gas-phase-level sensor group and the cavity-wall-level sensor group, constructing a temperature sensor array of the lithography machine in industrial integrated circuit manufacturing.

[0024] The photolithography machine refers to a closed cavity structure used to perform thin film deposition, etching, heat treatment and other processes in the integrated circuit manufacturing process. The wafer carrier surface refers to the upper surface of the base (commonly called susceptor) that supports the wafer, which is in direct contact with the wafer and is a key area for heat conduction and temperature uniformity control. The TFTC array refers to a thin film thermocouple array (TFTC), which is a micro temperature sensor array directly integrated on the carrier surface through micro-nano processing technology. It has the characteristics of high sensitivity, fast response and high spatial resolution, and is used to monitor the temperature distribution on the wafer carrier surface in real time. The FBG sensor refers to a fiber Bragg grating sensor (FBG). GratingSensor is a sensor based on the principle that the reflected wavelength of optical fiber changes with temperature. It is often used in areas such as the edge of the carrier that require anti-electromagnetic interference, high temperature resistance, and high-precision temperature measurement. The substrate-level sensor group refers to a TFTC array and an FBG sensor, which are used to directly monitor the temperature state of the surface and edge of the wafer carrier. It is the temperature sensing layer closest to the core area of ​​the process. The gas injection manifold refers to the pipeline system in the lithography machine used to evenly introduce process gases (such as precursors, carrier gases, etc.) into the chamber. The chamber center refers to the geometric center area of ​​the lithography machine, which is usually the core reference point for gas flow and thermal field distribution. The exhaust port refers to the channel for exhausting process gases in the chamber. The non-contact infrared laser spectrometer refers to a non-contact temperature measurement device based on the principle of infrared laser absorption spectroscopy, which can obtain real-time temperature and composition information of gases at locations such as the gas injection manifold, chamber center and exhaust port. The gas phase-level sensor group refers to a non-contact infrared laser spectrometer and a miniature platinum resistance temperature sensor, which is used to monitor Measuring the temperature distribution and dynamic changes in the gas phase region of a lithography machine. The chamber sidewall refers to the lateral inner wall of the cylindrical main structure of the lithography machine. The top cover refers to the sealing structure above the lithography machine, which usually includes an observation window, a gas injection port, etc. The base refers to the bottom mechanical component that supports the entire lithography machine structure. The multispectral infrared thermal imager refers to a thermal imaging device that can simultaneously capture radiation information in multiple infrared bands. The key mechanical connection refers to the mechanical interface between different components in the lithography machine (such as the sidewall and the top cover, and the base and the cavity). The surface acoustic wave temperature sensor refers to a wireless sensor based on the principle that the frequency of surface acoustic waves (SAW) changes with temperature. The cavity wall-level sensor group refers to a sensor group composed of a multispectral infrared thermal imager and a surface acoustic wave temperature sensor, used to monitor the temperature distribution on the surface of the lithography machine structure and the thermal state of key connection points. The temperature sensor array refers to a multi-level, multi-dimensional temperature monitoring system composed of three groups of sensors at the substrate level, the gas phase level, and the cavity wall level, used to comprehensively perceive the thermal field state inside and on the surface of the lithography machine structure. Optionally, the air flow path of the lithography machine is marked to configure the micro platinum resistance temperature sensors of the lithography machine by arranging along the air flow path at equal intervals (10 cm).

[0025] Optionally, the key mechanical connection of the lithography machine can be identified by X-ray computed tomography.

[0026] S2, based on the temperature sensor array, collecting multi-source temperature data of the lithography machine, establishing a chamber digital twin model of the lithography machine, and calculating a three-dimensional temperature gradient of the chamber digital twin model according to the multi-source temperature data.

[0027] The present application can be used as a basis for later temperature analysis based on the temperature sensor array collecting multi-source temperature data of the lithography machine.

[0028] In detail, the multi-source temperature data of the lithography machine based on the temperature sensor array includes: Establishing a time division multiplexing strategy of the temperature sensor array to collect initial multi-source temperature data of the lithography machine; Constructing a sensor coordinate of the temperature sensor array; And unify the initial multi-source temperature data to the sensor coordinate to get coordinate unified temperature data; Organizing a 5D tensor of the coordinate unified temperature data, and fusing the 5D tensor to get the multi-source temperature data of the lithography machine.

[0029] Wherein, the time division multiplexing strategy refers to activating different sensor groups for sampling in different time periods through time slice rotation method to optimize system resources (such as bandwidth, power consumption) and avoid signal conflict, the initial multi-source temperature data refers to the original temperature data set directly collected without processing, including voltage / current signal, optical signal, radiation intensity, etc., the sensor coordinate refers to the unique three-dimensional space identifier allocated to each temperature sensor, which is used to describe its physical position in the lithography machine, the coordinate unified temperature data refers to the temperature data set after converting the initial multi-source data to the same coordinate system, the 5D tensor refers to a multi-dimensional array structure for organizing coordinate unified data, including (X, Y, Z, time, temperature value) five-dimensional attributes, and the multi-source temperature data refers to the final temperature data set after fusing and optimizing the 5D tensor, reflecting the complete thermal state of the lithography machine Optionally, the time-sharing multiplexing strategy of the temperature sensor array dynamically adjusts the sensor sampling frequency according to the process stage, for example, the sampling frequency in the steady state stage: substrate level (10 Hz), gas phase level (5 Hz), cavity wall level (1 Hz); the sampling frequency in the transient stage (such as plasma ignition): forced full sensor switching to the highest frequency (substrate level 100 Hz, gas phase level 50 Hz, cavity wall level 10 Hz).

[0030] Optionally, the fusion of the 4D tensor to obtain the multi-source temperature data of the lithography machine can adopt a weighted Bayesian fusion algorithm to give dynamic weights to different sensor data for fusion.

[0031] The chamber digital twin model of the lithography machine established by the application can accurately reproduce the thermodynamic state in the chamber.

[0032] In detail, the establishment of the chamber digital twin model of the lithography machine comprises: constructing a three-dimensional entity model of the lithography machine; and giving the three-dimensional entity model a temperature-dependent material coefficient to obtain an attribute three-dimensional entity model; establishing a control equation of the attribute three-dimensional entity model, wherein the control equation comprises a heat conduction equation, a fluid dynamics equation and an electromagnetic field equation; mapping the multi-source temperature data of the lithography machine to the model boundary of the attribute three-dimensional entity model to obtain a model boundary condition; based on the attribute three-dimensional entity model, the control equation and the model boundary condition, establishing the chamber digital twin model of the lithography machine.

[0033] Wherein, the three-dimensional entity model refers to a digital expression of the geometric structure of the lithography machine based on computer-aided design (CAD), the temperature-dependent material coefficient refers to a functional relationship describing the change of material properties (such as thermal conductivity, specific heat capacity, electrical conductivity) with temperature, the attribute three-dimensional entity model refers to an enhanced model embedding physical parameters such as material properties, boundary type, initial condition on the basis of the three-dimensional entity model, the heat conduction equation refers to a partial differential equation describing the evolution of the temperature field inside a solid, the fluid dynamics equation refers to a Navier-Stokes equation set describing the flow of reactive gas and convective heat transfer, the electromagnetic field equation refers to a Maxwell equation set describing the electromagnetic field distribution in radio frequency heating, the model boundary condition refers to the conversion of multi-source temperature data into input constraints of the digital twin model, and the chamber digital twin model refers to a high-fidelity virtual lithography machine integrating geometry, material, equation and boundary condition.

[0034] Optionally, the temperature-dependent material coefficient of the three-dimensional entity model is given, and an attribute three-dimensional entity model is obtained, which can be bound through the "material library" of COMSOL or the "Engineering Data" module of ANSYS.

[0035] Optionally, the model boundary condition obtained by mapping the multi-source temperature data of the lithography machine to the model boundary of the attribute three-dimensional entity model includes a Dirichlet condition of directly specifying a boundary temperature (such as substrate temperature measurement data → platen surface T = measured value), a Robin condition of defining convective / radiative heat exchange, and a Neumann condition of specifying a heat flux density.

[0036] According to the multi-source temperature data, the three-dimensional temperature gradient of the chamber digital twin model is calculated, the evolution path of heat flow in space is revealed, and local hot spot areas that are difficult to capture by traditional control methods are effectively identified.

[0037] In detail, the three-dimensional temperature gradient of the chamber digital twin model is calculated according to the multi-source temperature data, including: The multi-source temperature data is preprocessed to obtain processed multi-source temperature data; The grid nodes of the chamber digital twin model are marked; The temperature scalar field of the grid nodes is defined; According to the temperature scalar field and the processed multi-source temperature data, the three-dimensional temperature gradient of the chamber digital twin model is calculated.

[0038] The processed multi-source temperature data refers to cleaning, aligning, and normalizing temperature data from different sensors (such as TFTC arrays, FBG sensors, infrared laser spectrometers, and miniature platinum resistors) to form processed multi-source temperature data that can be used for subsequent modeling. The grid nodes refer to establishing structured or unstructured grids on the three-dimensional geometric model of the chamber for subsequent discrete expression of the temperature field. The temperature scalar field refers to assigning a temperature value to each grid node to form a discrete temperature scalar field. The three-dimensional temperature gradient refers to calculating the temperature gradient at each grid node based on the temperature scalar field T(x, y, z).

[0039] Optionally, the definition of the temperature scalar field of the grid node can be marked by radial basis function (RBF), Kriging interpolation for unknown node temperature, including first selecting Gaussian kernel function as RBF based on the temperature scalar field and optimizing its shape parameter by cross-validation method; then constructing a linear equation set with the Euclidean distance between nodes as the variable, and introducing Tikhonov regularization processing to solve the weight coefficient to enhance the numerical stability; further, the obtained weight is used for temperature interpolation calculation of the global unknown grid node, and the KD tree is used to accelerate the neighbor search to improve the calculation efficiency; finally, the Fourier law is introduced as a physical constraint to modify the interpolation result and quantify the prediction uncertainty, so as to generate a high-precision continuous temperature field meeting the thermodynamic law.

[0040] S3, according to the three-dimensional temperature gradient, analyzing the heat flow evolution path of the lithography machine to determine the local hot spot area of the lithography machine.

[0041] According to the three-dimensional temperature gradient, the application analyzes the heat flow evolution path of the lithography machine, which has global perception and prediction ability, and greatly improves the pertinence and foresight of control.

[0042] In detail, according to the three-dimensional temperature gradient, analyzing the heat flow evolution path of the lithography machine includes: According to the three-dimensional temperature gradient, calculating the heat flow vector field of the lithography machine; Based on the heat flow vector field, analyzing the heat flow line set of the lithography machine; Through the heat flow line set, analyzing the characteristic position coordinates and classification labels of the lithography machine; Based on the characteristic position coordinates and classification labels, determining the heat flow evolution path of the lithography machine.

[0043] The heat flow vector field refers to the three-dimensional vector distribution of the heat flow intensity and direction of each space point in the lithography machine, the heat flow line set refers to the curve family in the heat flow vector field that satisfies the differential equation, the characteristic position coordinates refer to the set of space points in the heat flow field that have special dynamic significance, the classification label refers to the discrete enumeration set of physical attribute labeling of the characteristic position, and the heat flow evolution path refers to the dynamic correlation of the characteristic position in the space-time domain.

[0044] Further, according to the three-dimensional temperature gradient, calculating the heat flow vector field of the lithography machine includes: According to the three-dimensional temperature gradient, calculating the heat flow density vector of the lithography machine: Based on the heat flow density vector, constructing the heat flow vector field of the lithography machine.

[0045] Further, as another embodiment of the present application, the heat flux vector is calculated by the following formula:

[0046] wherein, represents the heat flux vector of the lithography machine, represents the anisotropic thermal conductivity tensor, represents the direction, represents the temperature, represents the three-dimensional temperature gradient, represents the plasma correction coefficient, represents the electron density, represents the Boltzmann constant, represents the electron temperature, represents the logarithmic gradient of the electron temperature, represents the fluid density, represents the specific heat capacity at constant pressure, represents the turbulent thermal diffusivity.

[0047] wherein, the heat flux vector refers to the vector of heat transfer per unit time through a unit area, the direction is the heat transfer direction, the size is the heat flow intensity, the anisotropic thermal conductivity tensor refers to the second-order tensor describing the difference in thermal conductivity of the material in different directions, the direction refers to the coordinate axis direction (such as (x, y, z)) in the formula, which is used to represent the direction of the gradient, the temperature refers to the macroscopic measure of the intensity of internal thermal motion of a substance, the plasma correction coefficient refers to an empirical coefficient for correcting the heat transfer behavior in a plasma environment, the electron density refers to the number of free electrons per unit volume in the lithography machine, the Boltzmann constant refers to a physical constant representing the relationship between temperature and microscopic particle energy, with a value of (1.38 times 10^{-23} J / K), the electron temperature refers to the temperature corresponding to the average kinetic energy of electrons in the plasma in the lithography machine, the logarithmic gradient of the electron temperature refers to the gradient of the relative change of the electron temperature in space, the fluid density refers to the mass of the fluid per unit volume in the lithography machine, the specific heat capacity at constant pressure refers to the heat required to raise the temperature of the fluid per unit mass in the lithography machine under the condition that the pressure is constant, and the turbulent thermal diffusivity refers to a parameter of the heat diffusion capacity under turbulent conditions in the lithography machine.

[0048] The present application determines the local hot spot area of the lithography machine, which can provide a basis for later temperature control.

[0049] In detail, the determination of the local hot spot area of the lithography machine comprises: calculating the path energy flux of the corresponding heat flow evolution path of the lithography machine, According to the path energy flux, an energy accumulation index of the lithography machine is analyzed to construct an energy accumulation distribution map of the lithography machine; A topological hotspot coordinate list of the lithography machine is established; According to the energy accumulation distribution map and the topological hotspot coordinate list, candidate hotspot region coordinates of the lithography machine are generated to determine a local hotspot region boundary of the lithography machine; A local hotspot region of the lithography machine is determined through the local hotspot region boundary.

[0050] The path energy flux refers to the energy flux passing through per unit time along the heat flow evolution path, the energy accumulation index refers to a dimensionless index representing the degree of energy accumulation in a local region, the energy accumulation distribution map refers to a three-dimensional distribution map drawn after discretization of the energy accumulation index in space, the topological hotspot coordinate list refers to a spatial coordinate list extracted according to points in the energy accumulation distribution map whose energy accumulation index exceeds a threshold, the candidate hotspot region coordinates refer to a candidate region coordinate set based on the topological hotspot coordinate list, and the local hotspot region boundary refers to a closed boundary generated by geometric modeling of the candidate hotspot region coordinates, and the local hotspot region refers to a closed region in the lithography machine where the temperature is significantly higher than the surrounding environment due to energy accumulation, and the spatial range enclosed by the boundary.

[0051] Optionally, the generation of the candidate hotspot region coordinates of the lithography machine according to the energy accumulation distribution map and the topological hotspot coordinate list can adopt a density-based DBSCAN clustering algorithm to perform spatial clustering on high-heat-density regions with energy values as weights, and finally generate a continuous and physically meaningful candidate hotspot region boundary coordinate set.

[0052] Optionally, the determination of the local hotspot region boundary of the lithography machine can be achieved by calculating a minimum convex polygon to generate a lithography machine candidate hotspot region coordinate set containing all potential hotspots and having an explicit geometric boundary.

[0053] S4, based on the local hotspot region, analyzing the lithography machine working condition of the lithography machine to define the multi-modal control instruction of the lithography machine.

[0054] The application analyzes the lithography machine working condition of the lithography machine based on the local hotspot region to define the multi-modal control instruction of the lithography machine to achieve multi-dimensional control of the lithography machine temperature. In detail, the lithography machine working condition refers to a comprehensive state representation of multiple parameters of the chamber under a specific process stage, including steady state, transient state and chaos, and in detail, the lithography machine working condition is determined by thermodynamic state, plasma state, flow state and chemical state, and the multi-modal control instruction refers to a mixed control strategy set for different working conditions.

[0055] S5, decompose the mixed control instruction into heater group control instruction, cooling system control instruction and gas flow field modulation instruction to execute high-precision temperature control of the industrial integrated circuit manufacturing.

[0056] The application decomposes the mixed control instruction into the linkage control mode of the multi-executive mechanism of the heater group control instruction, the cooling system control instruction and the gas flow field modulation instruction, which can not only quickly respond to temperature change, but also realize fine adjustment and dynamic balance of the thermal field, effectively inhibit temperature fluctuation and process drift, and guarantee high uniformity and stability of temperature in wafer processing.

[0057] In detail, the decomposition of the mixed control instruction into the heater group control instruction, the cooling system control instruction and the gas flow field modulation instruction comprises: establishing a subsystem of the lithography machine corresponding to the mixed control instruction, wherein the subsystem comprises a heater group subsystem, a cooling subsystem and a gas flow field modulation subsystem; mapping the mixed control instruction to the subsystem to obtain heating index, cooling index and gas flow field modulation index; generating the heater group control instruction, the cooling system control instruction and the gas flow field modulation instruction of the lithography machine according to the heating index, the cooling index and the gas flow field modulation index.

[0058] The heater group subsystem refers to a combination of devices in the lithography machine for providing active heat sources, usually composed of multiple independently controllable heating units (such as resistive heaters, radio frequency heaters, infrared lamp tubes, etc.), distributed in the carrier disc, cavity wall or other key positions, for accurately adjusting the local or global temperature, the cooling subsystem refers to a combination of devices for removing excess heat in the lithography machine, usually including water cooling circuits, air cooling devices, heat exchangers, semiconductor refrigerators, etc., for suppressing local overheating or maintaining the overall thermal balance of the chamber, the gas flow field modulation subsystem refers to a combination of devices for affecting heat transfer and temperature distribution by adjusting the gas flow state (such as flow rate, direction, distribution) in the lithography machine, including gas spray heads, mass flow controllers, guide vanes, air exhaust systems, etc., the heating index refers to the quantitative control target of the heater group subsystem mapped by the mixed control instruction, including target temperature value, temperature change rate, power distribution ratio of each heater and hotspot compensation amount, the cooling index refers to the quantitative control target of the cooling subsystem mapped by the mixed control instruction, including target cooling power, cooling medium flow rate setting, temperature drop rate and hotspot suppression intensity, the gas flow field modulation index refers to the quantitative control target of the gas flow field modulation subsystem mapped by the mixed control instruction, including total gas flow rate, regional gas flow rate distribution, gas spray angle or speed and air exhaust rate or pressure setting, the heater group control instruction refers to the instruction set converted from the heating index for specific devices, the cooling system control instruction refers to the instruction set converted from the cooling index for cooling devices, controlling cooling valve opening degree, pump speed, fan speed, refrigeration power, etc., to achieve accurate heat removal, the gas flow field modulation instruction refers to the instruction set converted from the gas flow field modulation index for the gas flow field modulation subsystem, controlling MFC setting value, valve opening and closing, spray head angle adjustment, vacuum pump power, etc., to optimize gas distribution and convective heat transfer.

[0059] Finally, the high-precision temperature control for industrial integrated circuit manufacturing is realized by the execution of the industrial integrated circuit manufacturing.

[0060] The scheme realizes all-around and stereoscopic real-time monitoring of the temperature distribution inside the chamber by configuring a multi-level temperature sensor array including a substrate level, a gas phase level and a cavity wall level in an industrial integrated circuit manufacturing photoetching machine, thereby significantly improving the comprehensiveness and accuracy of temperature data collection, and the chamber digital twin model constructed based on the multi-source temperature data can accurately reproduce the thermodynamic state inside the chamber and reveal the evolution path of the heat flow in space through three-dimensional temperature gradient calculation, effectively identifying the local hot spot area that is difficult to capture by traditional control methods. This hot spot identification mechanism based on digital twin enables temperature control to be no longer limited to single-point or empirical adjustment, but has global perception and prediction capabilities, significantly improving the pertinence and foresight of control. Furthermore, in combination with local hot spot distribution and real-time working condition analysis, the scheme can intelligently define multi-modal control instructions and accurately decompose them into coordinated control strategies of the heater group, the cooling system and the gas flow field. This multi-executive mechanism linkage control mode can not only quickly respond to temperature changes, but also realize fine adjustment and dynamic balance of the thermal field, effectively suppress temperature fluctuations and process drift, and ensure the high uniformity and stability of the temperature in the wafer processing process. Ultimately, the scheme significantly improves the controllability and consistency of the process, reduces the defect rate caused by temperature abnormalities, improves the yield and reliability of chip manufacturing, provides solid technical support for the development and mass production of advanced process technology, and has the potential to be popularized to other complex industrial thermal control systems. Therefore, the present application can improve the temperature control precision of industrial integrated circuit manufacturing.

[0061] As Figure 2 shown, it is a functional module diagram of a high-precision temperature control system based on industrial integrated circuit manufacturing.

[0062] The high-precision temperature control system based on industrial integrated circuit manufacturing 200 can be installed in an electronic device. According to the functions implemented, the high-precision temperature control system based on industrial integrated circuit manufacturing can include a sensor array construction module 201, a temperature gradient calculation module 202, a hot spot area analysis module 203, a control instruction determination module 204, and a temperature precision control module 205. The modules of the present application can also be referred to as units, which refer to a series of computer program segments that can be executed by an electronic device processor and can complete fixed functions, stored in the memory of the electronic device.

[0063] In the embodiments of the present application, the functions of each module / unit are as follows: The sensor array construction module 201 is configured to configure a temperature sensor array of a photoetching machine in industrial integrated circuit manufacturing, wherein the temperature sensor array includes a substrate level sensor group, a gas phase level sensor group and a cavity wall level sensor group. The temperature gradient calculation module 202 is configured to collect multi-source temperature data of the lithography machine based on the temperature sensor array, establish a chamber digital twin model of the lithography machine, and calculate a three-dimensional temperature gradient of the chamber digital twin model according to the multi-source temperature data. The hotspot area analysis module 203 is configured to analyze a heat flow evolution path of the lithography machine according to the three-dimensional temperature gradient, so as to determine a local hotspot area of the lithography machine. The control instruction determination module 204 is configured to analyze a lithography machine working condition of the lithography machine based on the local hotspot area, so as to define a multi-modal control instruction of the lithography machine. The temperature precision control module 205 is configured to decompose the hybrid control instruction into a heater group control instruction, a cooling system control instruction, and a gas flow field modulation instruction, so as to perform high-precision temperature control of the industrial integrated circuit manufacturing.

[0064] In detail, the modules in the high-precision temperature control system 200 for industrial integrated circuit manufacturing in the embodiments of the present application adopt the same technical means as the high-precision temperature control method for industrial integrated circuit manufacturing in the above-mentioned Figure 1 application, and can produce the same technical effects, which will not be described here.

[0065] It is obvious for those skilled in the art that the present application is not limited to the details of the above exemplary embodiments, and the present application can be implemented in other specific forms without departing from the spirit or essential characteristics of the present application.

[0066] Finally, it should be noted that in the above embodiments, each embodiment can be combined or independent, and deleting any one of them does not affect the technical implementation of the other embodiments. The above embodiments are only used to illustrate the technical solutions of the present application and are not limiting. Although the present application has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present application can be modified or replaced by equivalents without departing from the spirit and scope of the present application.

Claims

1. A high-precision temperature control method based on industrial integrated circuit manufacturing, characterized in that: The method comprises: Configuring a temperature sensor array for a photolithography machine in industrial integrated circuit manufacturing, wherein the temperature sensor array includes a substrate-level sensor group, a gas-phase-level sensor group, and a cavity-wall-level sensor group; Collecting multi-source temperature data of the lithography machine based on the temperature sensor array, establishing a chamber digital twin model of the lithography machine, and calculating a three-dimensional temperature gradient of the chamber digital twin model according to the multi-source temperature data; Analyzing the heat flow evolution path of the lithography machine according to the three-dimensional temperature gradient to determine the local hot spot area of ​​the lithography machine; Analyzing the working condition of the lithography machine based on the local hot spot area to define a multimodal control instruction of the lithography machine; The hybrid control instruction is decomposed into a heater group control instruction, a cooling system control instruction, and a gas flow field modulation instruction to perform high-precision temperature control for the industrial integrated circuit manufacturing.

2. The high-precision temperature control method based on industrial integrated circuit manufacturing according to claim 1, characterized in that: The temperature sensor array configured for a photolithography machine in industrial integrated circuit manufacturing includes: Embedding a TFTC array on the surface of a wafer carrier of the lithography machine; and configuring an FBG sensor at the edge of the carrier plate of the lithography machine; Configuring a substrate-level sensor group of the lithography machine through the TFTC array and the FBG sensor; A non-contact infrared laser spectrometer configured with the gas injection manifold, chamber center, and exhaust port of the lithography machine; In combination with the non-contact infrared laser spectrometer, a gas-phase level sensor group of the photolithography machine is established; Installing a multispectral infrared thermal imager on the chamber sidewall, top cover and base of the lithography machine; Identifying key mechanical connections of the lithography machine to configure surface acoustic wave temperature sensors at the key mechanical connections; Combining the multispectral infrared thermal imager and the surface acoustic wave temperature sensor to construct a cavity wall level sensor group of the lithography machine; Based on the substrate-level sensor group, the gas-phase-level sensor group and the cavity-wall-level sensor group, a temperature sensor array of the photolithography machine in the industrial integrated circuit manufacturing is constructed.

3. The high-precision temperature control method based on industrial integrated circuit manufacturing according to claim 2, characterized in that: The collecting multi-source temperature data of the lithography machine based on the temperature sensor array includes: Establishing a time-division multiplexing strategy for the temperature sensor array to collect initial multi-source temperature data of the lithography machine; constructing sensor coordinates of the temperature sensor array; and unifying the initial multi-source temperature data to the sensor coordinates to obtain coordinate-unified temperature data; A 5D tensor of the coordinate unified temperature data is organized and the 5D tensor is fused to obtain multi-source temperature data of the lithography machine.

4. The high-precision temperature control method based on industrial integrated circuit manufacturing according to claim 3, characterized in that: The establishing of the chamber digital twin model of the lithography machine includes: Constructing a three-dimensional solid model of the lithography machine; and assigning a temperature-dependent material coefficient to the three-dimensional solid model to obtain a property three-dimensional solid model; Establishing control equations of the attribute three-dimensional solid model, wherein the control equations include heat conduction equations, fluid dynamics equations, and electromagnetic field equations; Mapping the multi-source temperature data of the lithography machine to the model boundary of the attribute three-dimensional solid model to obtain the model boundary condition; Based on the attribute three-dimensional solid model, the control equations and the model boundary conditions, a digital twin model of the chamber of the lithography machine is established.

5. The high-precision temperature control method based on industrial integrated circuit manufacturing according to claim 4, characterized in that: Calculating the three-dimensional temperature gradient of the chamber digital twin model based on the multi-source temperature data includes: Preprocessing the multi-source temperature data to obtain processed multi-source temperature data; Marking the grid nodes of the chamber digital twin model; defining a temperature scalar field for the grid nodes; A three-dimensional temperature gradient of the chamber digital twin model is calculated based on the temperature scalar field and the processed multi-source temperature data.

6. The high-precision temperature control method based on industrial integrated circuit manufacturing according to claim 5, characterized in that: Analyzing the heat flow evolution path of the lithography machine according to the three-dimensional temperature gradient includes: Calculating a heat flux vector field of the lithography machine according to the three-dimensional temperature gradient; Analyzing a set of heat flow lines of the lithography machine based on the heat flow vector field; Analyzing the characteristic position coordinates and classification labels of the lithography machine through the heat flow line set; Based on the feature position coordinates and classification labels, a heat flow evolution path of the lithography machine is determined.

7. The high-precision temperature control method in industrial integrated circuit manufacturing according to claim 6, characterized in that: Calculating the heat flux vector field of the lithography machine according to the three-dimensional temperature gradient includes: According to the three-dimensional temperature gradient, the heat flux density vector of the lithography machine is calculated using the following formula: in, represents the heat flux density vector of the lithography machine, represents the anisotropic thermal conductivity tensor, Indicates direction, Indicates temperature, represents the three-dimensional temperature gradient, represents the plasma correction factor, represents the electron density, represents the Boltzmann constant, represents the electron temperature, represents the logarithmic gradient of the electron temperature, represents the fluid density, represents the specific heat capacity at constant pressure, represents the turbulent thermal diffusivity; Based on the heat flux density vector, a heat flux vector field of the lithography machine is constructed.

8. The high-precision temperature control method in industrial integrated circuit manufacturing according to claim 7, characterized in that: Determining the local hot spot area of ​​the lithography machine includes: Calculating the path energy flux of the heat flow evolution path corresponding to the lithography machine; Analyzing the energy concentration index of the lithography machine according to the path energy flux to construct an energy concentration distribution map of the lithography machine; Establishing a topological hotspot coordinate list of the lithography machine; Generating candidate hotspot region coordinates of the lithography machine according to the energy concentration distribution map and the topological hotspot coordinate list to determine the local hotspot region boundary of the lithography machine; The local hot spot area of ​​the lithography machine is determined by the local hot spot area boundary.

9. The high-precision temperature control method in industrial integrated circuit manufacturing according to claim 8, characterized in that: Decomposing the mixed control instruction into a heater group control instruction, a cooling system control instruction, and a gas flow field modulation instruction includes: Establishing a subsystem of a lithography machine corresponding to the hybrid control instruction, wherein the subsystem includes a heater group subsystem, a cooling subsystem, and a gas flow field modulation subsystem; Mapping the hybrid control instruction to the subsystem to obtain a heating index, a cooling index, and a gas flow field modulation index; According to the heating index, cooling index and gas flow field modulation index, a heater group control instruction, a cooling system control instruction and a gas flow field modulation instruction of the lithography machine are generated.

10. A high-precision temperature control system based on industrial integrated circuit manufacturing, characterized in that: The system comprises: A sensor array building module is used to configure a temperature sensor array of a lithography machine in industrial integrated circuit manufacturing, wherein the temperature sensor array includes a substrate-level sensor group, a gas-phase-level sensor group, and a cavity-level sensor group; a temperature gradient calculation module, configured to collect multi-source temperature data of the lithography machine based on the temperature sensor array, establish a chamber digital twin model of the lithography machine, and calculate a three-dimensional temperature gradient of the chamber digital twin model based on the multi-source temperature data; a hot spot analysis module, configured to analyze a heat flow evolution path of the lithography machine according to the three-dimensional temperature gradient to determine a local hot spot of the lithography machine; a control instruction determination module, configured to analyze the lithography machine operating conditions of the lithography machine based on the local hotspot area to define a multimodal control instruction of the lithography machine; A temperature precision control module is used to decompose the mixed control instructions into heater group control instructions, cooling system control instructions and gas flow field modulation instructions to perform high-precision temperature control for the industrial integrated circuit manufacturing.

Images