KR100575894B1 - System and method for diagnosing plasma processing chamber - Google Patents

Abstract

The present invention relates to an optimized diagnosis system and diagnostic method of a plasma process chamber that enables efficient computer simulation through a theoretical model of the plasma process. The present invention includes a simulator for computer simulation of the plasma chamber through the numerical analysis of the basic theoretical model for the physicochemical changes occurring in the plasma chamber; A server for controlling driving of the simulator such as start or end of the computer simulation of the plasma chamber; VIP-SEPCAD including a client for generating / managing / controlling all input and output data for the computer simulation of the plasma chamber.

      According to the present invention, it is possible to efficiently control the properties of chemically active species that play an important role in plasma and surface chemical reactions to improve the quality and productivity of the semiconductor. In addition, it is possible to efficiently diagnose / measure the plasma and optimize the design and process conditions of the plasma chamber.

Plasma, Theoretical Model, Computer Simulation, Plasma Process Chamber, VIP-SEPCAD

Description

Optimal Diagnosis System and Diagnosis Method of Plasma Process Chamber {SYSTEM AND METHOD FOR DIAGNOSING PLASMA PROCESSING CHAMBER}             

1 is a view showing a correlation between phenomena occurring in a plasma process chamber and process variables;

2 is a block diagram showing the structure of an optimization diagnostic system of a plasma process chamber according to the present invention;

3 is a view showing the components of the theory module of the optimization diagnostic system of the plasma process chamber according to the present invention;

4 is a detailed configuration diagram illustrating a client of the present invention;

5 is a diagram showing an internal structure of a plasma chamber generated by input of data;

6 is a view generated for inputting execution condition data such as a reaction mechanism or simulation condition input;

7 is a diagram illustrating an example of result analysis using simulation result data;

8 is a flow chart showing a plasma analysis method of the present invention,

9 is a flow chart showing a gas flow analysis method of the present invention,

10 is a flowchart illustrating a method of simultaneously analyzing plasma and gas flows of the present invention;

11 is a flow chart showing a particulate contamination analysis method of the present invention;

12 is a flow chart showing a wafer surface analysis method of the present invention;

13 is a structural diagram showing an example of an argon plasma chamber having three external electrodes;

14 shows plasma characteristics in the plasma chamber shown in FIG. 13;

15 is a view showing the effect of the phase difference between the two electrodes when voltage is applied to the two electrodes of the chamber shown in FIG.

16 is a structural diagram showing a distribution of temperature and pressure and a flow of neutral gas as an example of an oxygen plasma chamber;

17 is a view showing the distribution of water concentration of cations and anions in the plasma chamber shown in FIG. 16;

18 is a view showing the distribution of water concentration of oxygen molecules in the plasma chamber shown in FIG. 16;

19 is a view showing the distribution of water concentration of oxygen atoms in the plasma chamber shown in FIG. 16;

20 is a view showing a distribution of contaminating fine particles in the plasma chamber shown in FIG. 16;

               <Description of Major Codes in Drawings>

1: Plasma Chamber Optimization Diagnosis System 10: Simulator

11: theory module 12: IPC library

20: Server 21: Simulation Communication Module

22: server communication module 30: client

31: GUI module 32: client communication module

40: first electrode 41: second electrode

42: first RF power supply 43: second RF power supply

44 shielding capacity 45 wafer

46: ground electrode

The present invention relates to a system and method for comprehensively diagnosing and optimizing a plasma process which is essential for etching and deposition of thin films in a semiconductor manufacturing process. More specifically, theoretically identify complex phenomena occurring inside the plasma chamber during the plasma process, and perform computer simulations through theoretical models based on the theoretical model to predict the performance from the design stage of the plasma equipment. The present invention relates to an optimized diagnosis system and diagnostic method of a plasma process chamber that can be used to maximize the performance of equipment as well as to reduce the time and cost required. "Computational simulation" refers to computer simulations that use mathematical models to clarify and predict the principles of the actual material manufacturing process.

Plasma is used in a wide range of fields such as the treatment of environmental pollutants, the surface treatment of flat panel displays (FPDs), and the display (Plasma Display Panels) using plasma characteristics. In particular, it is essential for plasma processes such as plasma enhanced chemical vapor deposition (PECVD) and plasma etching in the process of manufacturing integrated circuits (ICs). Due to the high integration of circuits and the miniaturization of line widths, it is impossible to manufacture next-generation semiconductors without using a plasma process. In addition, as the thin film processing of semiconductor materials newly developed for the next-generation semiconductor manufacturing and the large diameter of the processed wafer to increase the productivity, the modification of existing plasma equipment or the development of a new plasma apparatus that can satisfy the more stringent process conditions are required. This is required.

In developed countries, including the United States, research on the improvement and development of plasma devices for next-generation semiconductor processing is being actively conducted. Currently, the size of processed wafers is shifting from 200mm to 300mm, and 450mm wafers are expected to be used by 2010, making it more important to develop a plasma device that can process a wider wafer surface more uniformly. From this point of view, the design and operation of the plasma apparatus and the optimization and control of the operating conditions are very important.However, the developed countries have been mainly experimentally designed and statistically optimized, which requires a lot of time and cost. Therefore, development costs are expected to skyrocket. Accordingly, in recent years, advanced semiconductor equipment companies have been using experimental tests and theoretical computer simulations to predict, analyze, and optimize equipment performance from the development stage in order to reduce development time and cost. Basic theoretical model development and computer simulation have been actively conducted.

On the other hand, in Korea, the technology for producing plasma equipment has been secured to some extent for the localization of semiconductor equipment, but it is still at the level of improving existing equipment performance based on experience. Therefore, in order to secure and maintain national competitiveness through rapid application of rapidly developing semiconductor technology manufacturing technology, a theoretical model must be developed to maximize the performance of equipment by rapidly developing, designing and modifying plasma devices at low cost. Based on this, a system capable of computer simulation is urgently needed.

In addition, the plasma etching and plasma enhanced chemical vapor deposition (PECVD) processes in the semiconductor manufacturing process have a problem that the surface of the wafer is contaminated by the fine particles. Particularly, since the flow of fine particles in the plasma is much different from that in the atmosphere, it is necessary to sufficiently study the flow characteristics of the fine particles in the plasma in order to effectively control the fine particles.

The present invention devised to solve the problems of the prior art as described above is to enable the computer simulation based on the theoretical model for the plasma optimization system and diagnosis of the plasma process chamber that can comprehensively diagnose the plasma process It is an object to provide a method.

Another object of the present invention is to optimize the diagnosis process of the plasma process chamber to improve the quality and productivity of semiconductors by predicting the properties of chemically active species that play an important role in plasma and surface chemical reactions and controlling them efficiently. And it provides a diagnostic method.

It is still another object of the present invention to provide an optimized diagnostic method of a plasma process chamber that enables continuous tracking and analysis of intermediate results during computer simulation using only necessary theoretical models.

In the present invention, VIP-SEPCAD (Virtual Integrated Prototyping-Simulation Environment for Plasma Chamber Analysis) efficiently identifies the basic theoretical model for such physical and chemical phenomena and implements an environment capable of performing computer simulation. and Design (hereinafter referred to as " VIP-SEPCAD &quot;) to provide an optimized diagnostic system for the plasma process chamber.

In the optimization diagnostic system of the plasma process chamber of the present invention for achieving the above object, a simulator for computer simulation of the plasma chamber through the numerical analysis of the basic theoretical model for the physical / chemical changes occurring inside the plasma chamber; A server for controlling driving of the simulator such as start or end of the computer simulation of the plasma chamber;
In the optimized diagnostic system of the conventional plasma process chamber, characterized in that the VIP-SEPCAD comprising a client for generating / managing / controlling all input / output data for the computer simulation of the plasma chamber,
The simulator includes a plurality of theoretical modules having basic theoretical models for various physical / chemical changes in the plasma chamber, and an IPC library for exchanging data with a server through IPC communication and comprehensively managing the simulator;
The server is a simulation communication module that controls the operation of the simulator for computer simulation by IPC communication with the IPC library, and exchanges data with the client through TCP / IP communication, and manages input / output data for the computer simulation in the client. A server communication module for enabling control;
The client includes a GUI module for performing data input, data file management, result data viewing, result data graph output, result data file management, and the like; And a client communication module for exchanging data with the server, outputting data for execution of the simulator, and receiving result data from the simulator.
The theoretical module includes a CSM which is a model for the physical and chemical phenomena of the charged components; EMM, which is a model for the electromagnetic field formed in the plasma; NSM, which is a model for reaction and flow of non-charged species; PFM, which is a model for the generation and growth of particulates in plasma; PTM which is a model for the movement of microparticles; It is characterized by consisting of SEM which is a model for the formation and etching of thin film by chemical species and ions on the wafer surface.
The theory module is preferably configured to be integrated into a user interface (GUI).

In addition, the diagnostic method of the plasma process chamber of the present invention using the diagnostic optimization system of the plasma process chamber, analyzing the characteristics of the plasma; Analyzing gas flow characteristics; Simultaneously analyzing the plasma and gas flow; Analyzing the contamination of the particulates; Analysis of the surface of the wafer, characterized in that the analysis by selecting any one of the steps or a combination of two or more steps.

Hereinafter, preferred embodiments of the present invention will be described in detail with reference to the drawings.

FIG. 1 is a diagram illustrating a correlation between phenomena occurring in a plasma process chamber and process variables, and shows that physical and chemical phenomena of plasma are complicated by various variables in a chamber during a plasma process. The final goal of the plasma process is to selectively and uniformly etch / deposit a wide wafer surface in a short process time to form trenches that meet the required specifications. The type of trench that is etched / deposited is induced by plasma because gas component introduced into the plasma chamber collides with electrons of high energy plasma and decomposes into reactive active species or ions and is involved in wafer surface chemical reaction. It is influenced by the distribution of active species and the energy of ions. In addition, since the ions induced by the collision between the neutral species and the electrons have electric charges, they affect the characteristics of the plasma again, and because of the characteristics of the plasma, the cations can always strike the solid surface, causing the fine particles to pop out when there is sufficient energy. Since active species are highly reactive, they may generate fine particles by vapor phase chemical reaction depending on the type of the component. Particularly, the particles present in the plasma are charged by the flow of electrons and ions, which is greatly affected by the flow of the neutral gas as well as the characteristics of the plasma and contaminates the wafer surface, which adversely affects the yield. Characteristics of the plasma, including water and energy concentrations of electrons and ions, which play an important role in achieving the final target value of the plasma process, reaction and flow characteristics of the neutral species including the distribution of active species derived from the plasma, The degree of particle generation and flow characteristics in the plasma are determined by process variables such as voltage and frequency applied to the electrode, pressure and temperature and type of chamber, type, composition and injection speed of source gas.

Figure 2 is a block diagram showing the structure of an optimization diagnostic system of the plasma process chamber according to the present invention. As shown in Figure 2, the optimization diagnostic system of the plasma process chamber according to the present invention is composed of a VIP-SEPCAD (1) consisting of a simulator 10, a server 20 and a client 30 of the known technology. At this time, the simulator 10 and the server 20 is to be run in the Unix (Unix) operating system, the client 30 is preferably to be run in the MS Windows operating system.

The simulator 10 includes a plurality of theoretical modules 11, which are theoretical models for identifying basic theories of various physical and chemical changes in the plasma chamber, and numerically analyze the theoretical models of the theoretical modules 11 to perform plasma analysis. Compute the chamber. Here, the theoretical module 11, which is a theoretical model, is a charged species module (hereinafter referred to as "CSM"), which is a model for identifying physical and chemical phenomena of charge-carrying components such as electrons and ions. Neutral, a model for the reaction and flow of non-charged species, the Electro-Magnetic Module (hereinafter referred to as "EMM"), which is a model for the field of electrons formed in plasma. Neutral Species Module (hereinafter referred to as "NSM"), Particle Formation Module (hereinafter referred to as "PFM"), which is a model for identifying the formation and growth of particulates in plasma, and the flow of particulates. Particle Transport Module (hereinafter referred to as "PTM"), which is a model to identify the model, and a model to investigate the formation and etching of thin film by chemical species and ions on the wafer surface Surface Evolution Module (hereinafter referred to as "SEM"). The function of the theoretical model of each module will be described in detail.

In this case, each theoretical module 11 is a graphical user interface that allows the user to easily input the structural design and operating conditions of the chamber and continuously track and analyze intermediate results while performing computer simulation using only necessary modules. Interface (hereinafter referred to as "GUI") is preferably configured. To this end, an IPC library 12 having a function of enabling IPC communication with the server 20 and comprehensively managing the simulator is used. At this time, the IPC library 12 (Inter-Process Communication library) receives the state data (state data), configuration data (configuration data) and result data type (result data type) for the simulation from the server and stores the information about this And set the result data type to be sent to the server 20. For example, the state information represents information of simulation start / pause / stop, such as "Play / Pause / Stop", the configuration data represents figure and condition information, and the result data type is pressure, temperature, gas / surface. Characteristic information important in the plasma process such as reaction rate, reactive active species, potential, electric field and magnetic field is stored in the form of string. Here, the simulator 10 collects simulation information through the IPC library 12, sets data, and transmits the result data to the server 20.

The server 20 is composed of a simulation communication module 21 is connected to the IPC library of the simulator and the IPC communication method, and a server communication module 22 connected to the client by the TCP / IP communication method. In this case, the simulation communication module 21 of the server 20 controls the driving of the simulator for computer simulation, and the server communication module 22 allows the client to manage and control the input / output data for the computer simulation. .

The client 30 generates / manages / controls all input / output data for the computer simulation of the plasma chamber, and provides a GUI for graphic processing for analyzing the result data of the computer simulation through the simulator 10. And a client communication module 32 that enables data exchange between the module 31 and the server.

At this time, the GUI module 31 performs functions such as data input, data file management, result data viewing, result data graph output, and result data file management. As shown in FIG. 4, the initial condition module defining initial conditions such as process pressure, raw material gas, applied voltage and frequency, and a chamber drawing capable of drawing the size and structure of the plasma chamber using blocks. Module and block and area designation module to specify the characteristics of the block, extract the selected area after block formation and specify the shape of the area, and set boundary conditions and group the boundary according to the line selection of each area after the area extraction. Boundary and grid definition modules that define symmetric lines of area lines and define parameters after separation, volume condition modules that define volume conditions such as pressure, temperature, coil current, chemical species and gas phase reactions, and plasma chamber A simulation selection and execution condition module that selects a theory module for computer simulation of a computer and defines the execution condition of a simulator, and a plasma chamber An analysis condition module for selecting an analysis condition for diagnosis / analysis of the object, a three-dimensional graph generation / storage module for displaying a three-dimensional graph based on the result data and storing the displayed three-dimensional graph as an image file; It consists of a geometry file module that can save all the information entered into a file and load the saved file. Herein, the shape of the similar group has been integrated for convenience, but the present invention is not limited thereto. As shown in FIG. 4, a separate configuration of each group may be preferable.

With such a configuration as described above, the optimization diagnostic system of the plasma process chamber is provided with a client menu such as File, Module, Geometry, Condition, Execution and Analysis. Implementation and computer simulation. 5 to 7 are exemplified drawings for clarity of understanding. FIG. 5 is a diagram illustrating an internal structure of a plasma chamber generated by inputting graphic data. FIG. 6 is a plasma processing condition and a reaction mechanism to be considered. FIG. 7 is a screen generated for inputting execution condition data such as simulation condition input. FIG. 7 is a result analysis screen for outputting numerical or three-dimensional graphics by simulation result data.

Meanwhile, a method for comprehensively diagnosing / analyzing the plasma process chamber using the optimized diagnosis system of the plasma process chamber according to the present invention will be described with reference to FIGS. 8 to 12.

The diagnostic method of the plasma process chamber of the present invention includes a method of analyzing only plasma, a method of analyzing only gas flow, a method of simultaneously analyzing plasma and gas flow, a method of analyzing particulate contamination, and a method of analyzing the surface of a wafer. . Any one of the above methods can be selected and analyzed, or a combination of two or more methods can be analyzed.

FIG. 8 is a flowchart illustrating a method of analyzing plasma only in the diagnostic method of the present invention. After drawing a figure of a plasma chamber, a module to be used, namely, a CSM for plasma characteristic analysis, is selected from theoretical modules, and parameters are inputted. At this time, process conditions such as process pressure, raw material gas, applied voltage and frequency are input.

Once the parameters are entered and computer simulations are run, the computer simulations are performed using the EMM simultaneously with the CSM. The CSMs / EMMs characterize the plasma and continuously display intermediate results during the analysis.

After checking whether the CSM converges to a steady state during the continuous simulation, if the CSM does not converge, it is fed back to the initial stage using the CSM and the EMM, and the CSM converges. If so, terminate the operation.

9 is a flowchart illustrating a method of analyzing only gas flow in the diagnostic method of the present invention. After drawing a plasma chamber figure, a neutral module NSM is selected from the theoretical modules, and parameters are input. At this time, process conditions such as process pressure, raw material gas composition and flow rate are input.

Computer simulations are performed using the NSM, and the characteristics of the neutral species that do not charge through the NSM during the execution of the computer simulation are analyzed and the intermediate results are continuously displayed.

Check whether the NSM converges to a steady state during the execution of the computer simulation. If the NSM did not converge, feed back to the initial stage using the NSM, and if the NSM converges, terminate the work. do.

FIG. 10 is a flowchart illustrating a method of simultaneously analyzing plasma and gas flow in a diagnostic method of the present invention. After drawing a diagram of a plasma chamber, a module to be used, such as a CSM and an NSM, is selected, and parameters are input. At this time, process conditions such as process pressure, raw material gas composition, flow rate, applied voltage and frequency are input.

The computer simulation is executed by using CSM / EMM and NSM separately, and the characteristics of each module are separately analyzed during the execution of the computer simulation, and the intermediate results are continuously displayed.

After checking whether the intermediate results continuously analyzed during the computer simulation are converged to the normal state, the CSM and the NSM converge, and if the CSM and the NSM do not converge, they are fed back to the initial computer simulation execution step to perform repetitive work.

After the CSM and NSM converge, check if they continue to affect each other.If the CSM and NSM continue to influence each other, exchange data with each other and feed back to the initial computer simulation execution step to perform repetitive tasks. End work if there is no mutual impact.

11 is a flowchart illustrating a method for analyzing particulate contamination in a diagnostic method of the present invention. After drawing a diagram of a plasma chamber, a module to be used of CSM / NSM / PTM / PFM is selected and parameters are input. At this time, process conditions such as process pressure, raw material gas composition, flow rate, applied voltage and frequency are input.

The computer simulation is executed by using CSM / EMM and NSM separately, and the characteristics of each module are separately analyzed during the execution of the computer simulation, and the intermediate results are continuously displayed.

After verifying that the CSM and NSM converge to the steady state, the intermediate results continuously analyzed during the computer simulation run, and if the CSM and NSM did not converge, feed back to the initial computer simulation step to perform the repetitive work.

After the CSM and NSM converge, check if they continue to affect each other. If the CSM and NSM continue to influence each other, exchange data with each other and feed back to the initial computer simulation step to perform repetitive work.

If CSM and NSM do not affect each other, perform computer simulation using PTM / PFM, a module for particle contamination analysis, and analyze particle size and distribution change over time through PTM / PFM with computer simulation. The results are shown (S51).

After confirming that PTM / PFM converges to a steady state in the intermediate result, if the PTM / PFM does not converge, it feeds back to the computer simulation execution step by the PTM / PFM to perform the repetitive work, and the PTM / PFM If it has converged, terminate the operation.

12 is a flowchart illustrating a method of analyzing a wafer surface among diagnostic methods of the present invention. After drawing a plasma chamber figure, a module to be used for CSM / NSM / SEM is selected and parameters are input. At this time, process conditions such as process pressure, raw material gas composition, flow rate, applied voltage, frequency, and processing time are input.

Computational simulation is performed using CSM / EMM simultaneously and NSM separately, and the characteristics of each module are separately analyzed during computational simulation and the intermediate results are continuously displayed.

After confirming that the CSM and the NSM converge to the steady state, the intermediate results are analyzed, and if the CSM and the NSM do not converge, they are fed back to the initial computer simulation execution step to perform repetitive tasks.

After the CSM and NSM converge, check if they continue to affect each other. If the CSM and NSM continue to influence each other, exchange data with each other and feed back to the initial computer simulation execution step to perform repetitive tasks.

If CSM and NSM do not affect each other, computer simulation is performed using SEM, a module for wafer surface analysis, and the SEM analyzes the surface change of the wafer over time and continuously displays the intermediate results.

After checking whether the machining time entered in the parameter input step of inputting the process condition has been reached during the computer simulation, if the specified machining time has not been reached, it is fed back to the computer simulation execution step by the SEM to perform the repetitive work. If the specified machining time is reached, the operation is terminated.

11 and 12 illustrate a step of drawing a diagram of the plasma chamber and selecting a module to be used.

(Example 1)

In Example 1 of the present invention, a simulation using CSM and EMM was performed on an argon plasma chamber having the following structure and process conditions.

FIG. 13 is a view showing the structure of a capacitive argon plasma chamber applied with two separate external power sources 43 and 44 and process conditions. The argon plasma chamber is provided with electrodes formed at the same electrode size at the top and the bottom. The electrode is separated from the grounded electrode 46 with an insulator interposed therebetween to form a predetermined electrode gap (eg, 4 cm). A high voltage and frequency are applied to the first electrode 40 on the top where the wafer 45 is not placed in order to increase the plasma density, and ion bombardment on the second electrode 41 on the bottom where the wafer 45 is placed. Apply relatively low voltage and frequency to control energy.

At this time, the operating pressure is 70mTorr, the applied voltage and frequency of the first electrode is 200V and 30MHz, respectively, the shielding capacitance is 500pF.

14 is a view showing a simulation result when different frequencies and voltages are simultaneously applied to the first electrode and the second electrode by using the CSM and the EMM of the present invention. It is relatively smaller than the surface area of the wall of the plasma chamber that is grounded, and there is a self-DC bias, although there is a difference in degree depending on the applied voltage, frequency, and process pressure. In the plasma potential distribution shown in FIG. 14A, a self-DC bias of about -160V was applied to the first electrode that applied 200V at a frequency of 30MHz, and a relatively weak voltage was applied to the second electrode where the wafer is placed. It shows a self-DC bias of about -80V. If the surface area of the power electrode is smaller than the grounded electrode, the same amount must be exited into a narrow area as the flow of electrons from a large area. Therefore, the current density increases in the narrow area, so as to push out the excessively outgoing electron per unit area. negative self-DC bias. However, since the plasma potential is kept higher than the wall of the grounded plasma chamber, the potential drops sharply in the plasma sheath region where negative self-DC bias is applied. Therefore, a stronger electric field is generated, and as shown in Fig. 14B, the electron temperature increases rapidly. Also, due to the high electron temperature, the ionization reaction also rapidly increases as shown in Fig. 14 (c). As a result, as shown in Fig. 14 (d), the asymmetric plasma is formed at the point where the plasma density is maximized, not toward the middle between the two electrodes, but toward the more self-DC bias.

Here, the plasma process characteristics obtained for the case where a phase lag exists between waves when different voltage waves are applied to the upper and lower electrodes are described. In this case, the voltage and frequency applied to the first electrode were fixed, 100 V was applied at various frequencies to the second electrode, and the phase lag with the voltage wave applied to the first electrode was changed. As can be seen in FIG. 15, the self-DC bias applied to the second electrode is more affected than the other characteristics, whereas the plasma potential or the initial self-DC bias is hardly changed and thus the plasma density is not affected much. . The potential change between the plasma potential and the self-DC bias can be thought of as the ion bombardment energy that strikes the surface of the wafer approximately.

As can be seen, the present invention can be usefully used to find and analyze the optimal process conditions that can more effectively control ion bombardment energy and plasma density.

(Example 2)

In Example 2 of the present invention, in addition to the electron impact reaction with the CSM and the EMM, the plasma analysis module for the oxygen plasma chamber used for etching the photoresist or depositing the oxide thin film in the semiconductor process, as well as neutral ( Computational simulations were performed in conjunction with NSMs that characterize neutral species that are accompanied by chemical reactions between neutrals and no flowing charges.

16 is a diagram showing the structure of the oxygen plasma chamber with the flow mode of neutral gas as well as the temperature and pressure distribution, the process conditions were the absolute temperature 323K, pressure 500mTorr, oxygen (O ₂ ) supply flow rate 100sccm and operating voltage 150V.

In the second embodiment, since the pure oxygen molecules are continuously introduced into the small holes at the top of the chamber shown in FIG. 16 and the ionization amount is very small compared to the incoming oxygen to generate the oxygen plasma, the energy of the electrons is prevented from colliding with the oxygen molecules. The electron energy distribution function was calculated on the assumption that it was determined by the equation, and the reaction rate constant for the electron collision reaction was predicted and used. In addition to the electron collision reaction, the reaction between the ions and the neutral species is also considered, and the computer simulation is carried out by using the reaction between 24 electron impact collisions and the reaction between 24 ions and the neutral species. It was.

17 to 19 illustrate the results. In this case, when the plasma characteristics are examined through FIG. 17, unlike an argon plasma, which is a case of a positive plasma, a component to match an electric neutrality with a cation O ₂ ⁺ is anion O. ^- a show that has the characteristics of a sound-conductive plasma. In addition, when looking at the density distribution of the neutral species through FIGS. 18 to 19, metastables of oxygen atoms are generated in the bulk plasma region by electron collision reaction from oxygen molecules introduced from the upper showerhead electrode to reach the lower electrode. It shows that it is being converted to another ingredient before. However, the metastables of oxygen molecules can play an important role as they show a large amount reaching the bottom electrode where the wafer is placed due to the long life with oxygen atoms in the ground state.

(Example 3)

In Embodiment 3 of the present invention, the flow characteristics according to the particle size are simulated by using the CSM, EMM, NSM, and PTM in the VIP-SEPCAD, an optimized diagnostic system of the plasma process chamber, to simulate the presence of the particles in the oxygen plasma chamber. Was investigated.

It is assumed that the microparticles exist inside the plasma chamber through various paths, and there may be a myriad of microparticles of various sizes, but the size of the microparticles is limited from 1 nanometer to 1 micrometer and d _p = 0.001 ~ 0.01㎛, d _p The simulation was carried out on the assumption that 10 ⁴ particles were freely distributed in each case, divided into three groups of = 0.01 to 0.1 μm and d _p = 0.1 to 1 μm.

The results are shown in FIG. 20, and FIG. 20 is a view showing that the particles accumulate at a specific position according to the size in the oxygen plasma chamber. The particles are distributed inside the plasma chamber when the particles reach steady state. . Here, the relatively large particles are affected by the neutral force and follow where the process gas escapes. They are staying at the edge of the electrode where the wafer is placed without being completely escaped to the pumping port by the strong electric field. As a result, very small particles are not significantly affected by neutral drag. The smaller the particle size, the greater the action of thermophoretic and brown forces, which remain together with the electrostatic force in the bulk plasma.

Thus, it provides an advantage that the problem of particulates that may be inside the plasma chamber, that is, contamination of the wafer surface by the particulates, can be analyzed and easily controlled.

The optimized diagnostic system and diagnostic method of the plasma process chamber according to the present invention are not particularly limited by the specific embodiments and drawings as described above, and the design change and the shape change made within the scope of the claims of the present invention are not limited. All are considered to be within the scope of the present invention.

As described above, according to the optimized diagnosis system and diagnostic method of the plasma process chamber according to the present invention, the theoretical model for the plasma and computer simulations linked thereto facilitate the characteristics of various physical / chemical phenomena occurring in the plasma chamber. Can be identified / predicted.

In addition, the user can easily input the structural design and operating conditions of the plasma chamber, perform computer simulation by selecting only the required theoretical model, and accordingly, perform selective characteristic analysis as well as intermediate results during computer simulation. There is an advantage that can be tracked and analyzed continuously.

In addition, it is possible to efficiently control the properties of chemically active species that play an important role in plasma and surface chemical reactions to improve the quality and productivity of the semiconductor.

In addition, the present invention can efficiently comprehensively diagnose / measure the plasma, and optimize the design and process conditions of the plasma chamber. Therefore, it is possible to reduce the time and cost required for the development of plasma equipment, and to maximize the performance of the equipment.

Claims (14)

delete In the optimized diagnostic system of the plasma process chamber, a simulator for computer simulation of the plasma chamber through the numerical analysis of the basic theoretical model for the physical and chemical changes occurring in the plasma chamber; A server for controlling driving of the simulator such as start or end of the computer simulation of the plasma chamber; In the optimized diagnostic system of the conventional plasma process chamber, characterized in that the VIP-SEPCAD comprising a client for generating / managing / controlling all input / output data for the computer simulation of the plasma chamber, The simulator includes a plurality of theoretical modules having basic theoretical models for various physical / chemical changes in the plasma chamber, and an IPC library for exchanging data with a server through IPC communication and comprehensively managing the simulator; The server exchanges data with the client through TCP / IP communication with a simulation communication module 21 for controlling the operation of the simulator for computer simulation by IPC communication with the IPC library, and input / output data for the computer simulation from the client. A server communication module 22 for enabling management / control of the server; The client includes a GUI module 31 for performing data input, data file management, result data viewing, result data graph output, result data file management, and the like; And a client communication module (32) for exchanging data with the server (20), outputting data for execution of the simulator (10), and receiving result data from the simulator (10). Optimization diagnostic system. 3. The apparatus of claim 2, wherein the theory module comprises: a CSM which is a model for the physical and chemical phenomena of the charged components; EMM, which is a model for the electromagnetic field formed in the plasma; NSM, which is a model for reaction and flow of non-charged species; PFM, which is a model for the generation and growth of particulates in plasma; PTM which is a model for the movement of microparticles; An optimization diagnostic system for a plasma process chamber, comprising SEM, a model for the formation and etching of thin films by chemical species and ions on the wafer surface. 4. The system of claim 3, wherein the theory module is integrated into a user interface (GUI). delete delete 3. The GUI module of claim 2, further comprising: an initial condition module for defining initial conditions such as process pressure, source gas, applied voltage and frequency; A Geometry Drawing module which can define / change the size of the cylinder and add / delete / move / size blocks; A block and area designation module for designating characteristics of the block, and extracting a selected area after block formation and designating a shape of the area; A boundary and grid definition module for setting boundary conditions according to line selection of each region after region extraction and grouping the boundaries, and defining parameters after finding and separating symmetric lines of region lines; A volume condition module for defining volume conditions such as pressure, temperature, coil current, chemical species and gas phase reactions; A simulation selection and execution condition module for selecting a theoretical module for computer simulation of the plasma chamber and defining execution conditions of the simulator; An analysis condition module for selecting an analysis condition for diagnosis / analysis of the plasma chamber; A 3D graph generation / storage module for displaying a 3D graph based on the result data and storing the displayed 3D graph as an image file; And a figure file module for storing the chamber figure information and all the inputted information as a file and retrieving the stored file. delete A diagnostic method of a plasma process chamber, comprising: analyzing a characteristic of a plasma using a diagnostic system of a plasma process chamber according to claim 1; Analyzing gas flow characteristics; Simultaneously analyzing the plasma and gas flow; Analyzing the contamination of the particulates; A method of diagnosing a plasma processing chamber, comprising: analyzing a surface of a wafer, and selecting any one of the above-described steps, or analyzing two or more steps in combination. The method of claim 9, wherein analyzing the plasma comprises: drawing a shape of a plasma chamber; Selecting a CSM as a module to use; Inputting parameters of process conditions; Performing computer simulation using CSM and EMM simultaneously; Analyzing the characteristics of the plasma according to each module during the computer simulation and displaying intermediate results; Checking whether the CSM converges to an intermediate state in a steady state; If the CSM did not converge, it performs a feedback by repeating the execution of the computer simulation process, and if the CSM converged, the process comprising the step of terminating the work. 10. The method of claim 9, wherein analyzing the gas flow characteristics comprises: drawing a plasma chamber figure; Selecting an NSM as a module to use; Inputting parameters of process conditions; Performing computer simulation using NSM; Analyzing and characterizing the neutral species of non-charged neutral species during execution of computer simulation; Confirming whether the NSM converges to an intermediate result in a steady state; If the NSM did not converge, it performs a feedback by repeatedly performing the computer simulation execution process, and if the NSM converged, the process of ending the operation characterized in that the plasma process chamber comprising a. 10. The method of claim 9, wherein simultaneously analyzing the plasma and gas flows comprises: drawing a shape of a plasma chamber; Selecting CSM and NSM as modules to use; Inputting parameters of process conditions; Performing computer simulation using both CSM / EMM and NSM separately; Analyzing the characteristics of each module during the simulation and displaying intermediate results; Checking whether the CSM and the NSM converge with the intermediate result in a normal state; If the CSM and NSM have not converged, performing a repetitive operation by feeding back each of the computer simulation execution processes; Checking if the CSM and NSM have converged and continue to have mutual influence; If CSM and NSM continue to influence each other, exchange data with each other, and then feed back to the computer simulation execution process to perform repetitive tasks, and if the CSM and NSM do not affect each other, the process is completed. Diagnostic method of the plasma process chamber. 10. The method of claim 9, wherein analyzing the contamination of the particulates comprises: drawing a shape of a plasma chamber; Selecting CSM, NSM, PTM and PFM as modules to use; Inputting parameters of process conditions; Performing computer simulation using both CSM / EMM and NSM separately; Analyzing the characteristics of each module during the simulation and displaying intermediate results; Checking whether the CSM and the NSM converge with the intermediate result in a normal state; If the CSM and NSM have not converged, performing a repetitive operation by feeding back each of the computer simulation execution processes; Checking if the CSM and NSM have converged and continue to have mutual influence; If the CSM and the NSM continue to influence each other, exchanging data with each other, and then feeding back to the computer simulation process to perform repetitive tasks; Performing computer simulation using PTM / PFM, a particle contamination analysis module, if CSM and NSM do not affect each other; Analyzing the change in particle size and distribution over time through PTM / PFM with computer simulation and presenting intermediate results; Checking whether the intermediate result converges to a steady state; If the PTM / PFM did not converge, the PTM / PFM feeds back to the computer simulation execution process and repeats the operation, and if the PTM / PFM converges, the method comprises diagnosing the plasma process chamber. . The method of claim 9, wherein analyzing the surface of the wafer comprises: drawing a shape of a plasma chamber; Selecting a module to use (CSM, NSM, SEM); Inputting parameters of process conditions; Performing computer simulation using both CSM / EMM and NSM separately; Analyzing the characteristics of each module during the simulation and presenting the intermediate results; Checking whether the CSM and the NSM converge with the intermediate result in a normal state; If the CSM and NSM have not converged, performing a repetitive operation by feeding back each of the computer simulation execution processes; Checking if the CSM and NSM have converged and continue to have mutual influence; If the CSM and the NSM continue to influence each other, exchanging data with each other, and then feeding back to the computer simulation process to perform repetitive tasks; Performing computer simulation using SEM, a wafer surface analysis module, if the CSM and NSM do not affect each other; Analyzing the surface change of the wafer with time through SEM and showing the intermediate result through computer simulation; Checking whether the processing time entered during the parameter input process has been reached; If the specified processing time has not been reached, performing a repetitive operation by feeding back to the computer simulation execution process by the SEM, and ending the operation if the specified processing time has been reached.

Images