Film formation simulation method, film formation simulation program, film formation simulator, and film formation device

Abstract

A film formation simulation method according to an embodiment of the present disclosure includes the steps of generating representative particles corresponding to an incident radical flux, and calculating adhesion, desorption, migration, and deposition of each representative particle on a film formation surface through probabilities. In the above steps, this film formation simulation method comprises calculating the deposition as voxels to which state information is imparted, the state information indicating the bonding or non-bonding between the representative particles and the film formation surface, and thereby expressing the quality of the film and the coverage of the film on the film formation surface.

Description

Technical Field

[0001] The present invention relates to a film formation simulation method for simulating the shape of a surface to be processed (i.e., a film formation surface, which is a film formation object) during a film formation simulation process, a film formation simulation program for implementing the film formation simulation method, and a film formation simulator. In addition, the present invention relates to a film formation device equipped with the film formation simulator. Background Art

[0002] The key process of semiconductor device manufacturing includes the film forming process of forming or embedding thin film (nanoscale) on the pattern. In recent years, with the increasing demand for high functionality of semiconductor devices, more semiconductor devices have complex stacked structures with a mixture of high aspect ratio and low aspect ratio, and the difficulty of embedding such patterns in the film forming process has increased. Therefore, it becomes more important to predict film coverage and film quality (e.g., density, defect density, permeability, adhesion, etc.) depending on process conditions. In this case, numerical simulation of film forming process is useful as one of the prediction techniques.

[0003] For example, Patent Document 1 uses a Monte Carlo method to calculate the attachment / de-attachment of gas particles to / from a pattern surface, migration dependent on substrate temperature, and deposition of particles taking into account the influence of surface roughness, and predicts the morphology of a film. In addition, Patent Document 1 also proposes a method for predicting film quality distribution using a film quality database related to morphology pre-constructed based on molecular dynamics (MD) calculations or first principle calculations.

[0004] Reference List

[0005] Patent Literature

[0006] Patent Document 1: Japanese Unexamined Patent Application Publication No. 2017-204757 Summary of the invention

[0007] However, in the method described in Patent Document 1, since the Monte Carlo method is used to calculate the irradiated gas, its accuracy and calculation time largely depend on the number of particles used in the Monte Carlo method. In addition, in order to calculate the film quality, it is necessary to prepare a database that depends on the morphology and gas flux, respectively, according to molecular dynamics (MD) calculations or first principle calculations. Therefore, it takes considerable time and effort to prepare before it actually becomes usable as a simulation tool. Therefore, it is desirable to provide a film formation simulation method, a film formation simulation program, a film formation simulator, and a film formation device including the film formation simulator that can shorten the calculation time and improve the calculation accuracy.

[0008] The film formation simulation method according to the embodiment of the present disclosure includes the following steps: generating representative particles according to the incident radical flux, and calculating the attachment, desorption from the film formation surface, migration and deposition of each representative particle to the film formation surface according to probability. The steps of the above-mentioned film formation simulation method include: expressing the film quality and film coverage on the film formation surface by calculating the deposition as voxels endowed with state information, and the state information indicates the bonding state or non-bonding state between the representative particle and the film formation surface.

[0009] The film formation simulation program of the embodiment of the present invention includes an input unit, a calculation unit and an output unit. The input unit obtains film formation conditions. The calculation unit generates representative particles according to the incident radical flux under the film formation conditions obtained by the input unit, and calculates the attachment, desorption from the film formation surface, migration and deposition of the representative particles to the film formation surface according to probability. The output unit outputs the calculation results obtained by the calculation unit. The calculation unit expresses the film quality and film coverage on the film formation surface by calculating the deposition as voxels given state information, and the state information indicates the bonding state or non-bonding state between the representative particles and the film formation surface.

[0010] A film forming simulator according to an embodiment of the present disclosure includes an input unit, a computing unit, and an output unit. The input unit acquires film forming conditions. The computing unit generates representative particles according to the incident radical flux under the film forming conditions acquired by the input unit, and calculates the attachment, desorption from the film forming surface, migration and deposition of the representative particles on the film forming surface according to probability. The output unit outputs the calculation result obtained by the computing unit. The computing unit expresses the film quality and film coverage on the film forming surface by calculating the deposition as voxels endowed with state information, and the state information indicates the bonding state or non-bonding state between the representative particles and the film forming surface.

[0011] According to an embodiment of the present disclosure, a film forming device includes a film forming chamber, a control unit, an optimization operation unit, and an output unit. The control unit controls the operation of the film forming chamber. The optimization operation unit generates representative particles according to the incident radical flux, and calculates the attachment, desorption, migration, and deposition of each representative particle on the film forming surface according to probability. The optimization operation unit retrieves the optimization conditions of the film forming process based on the calculation results. The output unit generates the data required to change the film forming conditions of the film forming chamber to the optimization conditions found by the optimization operation unit, and outputs the generated data to the control unit. The optimization operation unit expresses the film quality and film coverage on the film forming surface by calculating the deposition as voxels endowed with state information, and the state information indicates the bonding state or non-bonding state between the representative particles and the film forming surface. BRIEF DESCRIPTION OF THE DRAWINGS

[0012] Figure 1 It is a diagram showing an example of a processing sequence of a film formation simulation method according to an example of the present invention.

[0013] Figure 2 It shows Figure 1 An illustration of a specific example of the processing procedure shown in .

[0014] Figure 3 : is a diagram showing a configuration example of a film formation simulator according to an embodiment of the present invention.

[0015] Figure 4 is a diagram illustrating the concept of gas delivery.

[0016] Figure 5 This is a diagram showing the concept of an actual film-forming surface.

[0017] Figure 6 It shows that Figure 5 An illustration of the concepts shown in Figure 1 that are combined into a voxel model.

[0018] Figure 7 is a diagram used to describe the generation of representative particles in a single voxel.

[0019] Figure 8 This is a diagram conceptually showing an actual film-forming surface during migration.

[0020] Fig. 9 is a diagram illustrating the concept of a voxel model in migration.

[0021] Fig.10 is a diagram illustrating an example of a bracket determination method in a voxel space.

[0022] Fig.11 is a diagram illustrating an example of a bracket determination method in a voxel space.

[0023] Fig.12 is a diagram illustrating an example of a bracket determination method in a voxel space.

[0024] Fig.13 It is a diagram for explaining the calculation of film density and transmittance in voxel space.

[0025] Fig.14 is a diagram for describing the calculation of stickiness in voxel space.

[0026] Fig.15 is a diagram for describing the annealing process in voxel space.

[0027] Fig.16 is a diagram for explaining the annealing process in the voxel space.

[0028] Fig.17 is a diagram used to describe bubbles in voxel space.

[0029] Fig.18 is a diagram used to describe bubbles in voxel space.

[0030] Fig.19 is a diagram used to describe bubbles in voxel space.

[0031] Fig. 20 is a diagram used to describe bubbles in voxel space.

[0032] Fig.21 It is a diagram showing the calculation results of the coverage and film density distribution during the CVD process.

[0033] Fig. 22 is a diagram showing the results of voxel bonding / unbonding state distribution and gas flux distribution.

[0034] Fig.23 It is a diagram showing the calculation results of the permeability distribution.

[0035] Fig.24 is a diagram showing calculation results of adhesion distribution.

[0036] Fig.25 is a diagram showing the calculation results of the film density distribution after annealing.

[0037] Fig.26 It is a graph showing the calculation results of the adhesion distribution after annealing.

[0038] Fig. 27 It is a graph showing the calculation results of the transmittance distribution after annealing.

[0039] Fig.28 is a diagram illustrating the concept of bubble calculation.

[0040] Fig.29 is a diagram showing an example of a trench intersection structure.

[0041] Fig.30 It shows Figure 1 Schematic diagram of a variation of the process shown in .

[0042] Fig.31 : is a diagram showing a configuration example of a film formation simulation program according to an embodiment of the present invention.

[0043] Fig.32 is a diagram showing a configuration example of a film forming apparatus according to an embodiment of the present disclosure. DETAILED DESCRIPTION

[0044] Hereinafter, embodiments for implementing the present disclosure will be described in detail with reference to the accompanying drawings.

[0045] <Implementation Method>

[0046] <Overview>

[0047] First, the outline of the film formation simulation method according to the embodiment of the present invention will be described. The film formation simulation method according to the present embodiment is a film formation method for making a film containing material particles protrude toward a processed surface (film formation surface).

[0048] That is, in the film formation simulation method of this embodiment, various vapor deposition methods for predicting the film quality and coverage of the film are described. For example, the film formation simulation method according to this embodiment can handle film formation methods including physical vapor deposition (PVD) (such as resistance heating deposition, electron beam evaporation, molecular beam epitaxy, ion plating or sputtering) and chemical vapor deposition (CVD) (such as thermal CVD, plasma enhanced CVD, atomic layer deposition (ALD) or metal organic CVD).

[0049] Examples of material particles include atoms, molecules, and ions obtained by ionization thereof. Material particles may be formed by decomposing or ionizing a material gas introduced into a film forming chamber using heat, plasma, or the like, or may be formed by collision of a metal target with rare gas atoms, or the like. In addition, the material particles may be one or more of them. In other words, the film may be a film formed of a single type of material, or may be a film formed by reacting multiple types of materials.

[0050] As the film forming target, a metal substrate, a semiconductor substrate, a glass substrate, a quartz substrate, a resin substrate, etc. can be cited. The shape and material of the surface of the film forming target are not particularly limited. For example, a thin film or a fine structure can be formed on the surface of the film forming target. For example, the film formed on the film forming target can be a thin film with a film thickness of about several microns. In addition, in the film forming simulation method of this embodiment, for example, an area with a side length of about several microns can be processed.

[0051] According to the film formation simulation method of the present embodiment, the film quality or coverage of a film formed by the above-mentioned film formation method can be predicted within a range of several tens of nanometers.

[0052] <Flow of film formation simulation method>

[0053] Next, see Figure 1 and Figure 2 An outline of the flow of the film formation simulation method according to the present embodiment will be described. Figure 1 This is a flowchart schematically showing the flow of the film formation simulation method according to the present embodiment. Figure 2 Yes means Figure 1 Detailed flow chart of the flow of the film formation simulation method shown. Figure 3 : is a diagram showing a configuration example of an information processing device (film formation simulator) for realizing the film formation simulation method.

[0054] (Structure of film formation simulator)

[0055] Figure 3 The film forming simulator 1 shown in FIG. 1 includes an input unit 11, a calculation unit 12, and an output unit 13. The input unit 11 acquires film forming conditions for performing a predetermined film forming process on a film forming surface, and inputs the acquired conditions to the calculation unit 12. The input unit 11 is implemented by a character-based user interface (CUI) or a graphical user interface (GUI) to set the film forming conditions. The calculation unit 12 generates an output signal based on the film forming conditions input through the input unit 11. Figure 1 and Figure 2 The simulation method shown in (to be described later) calculates the film quality and shape development of the film-forming surface.

[0056] It should be noted that, in the present embodiment, the operation unit 13 may be configured by hardware to implement the calculation process described later. However, a predetermined simulation program (software) may be used to perform the calculation process. In this case, the operation unit 13 may be configured by an operation device such as a CPU (central processing unit), and the operation unit 13 may read the simulation program from the outside and may execute the program, thereby performing the calculation.

[0057] The simulation program can be stored in, for example, a database not shown in the figure, a storage unit such as a ROM (read-only memory) provided separately, or the like. In this case, for example, the simulation program can be pre-implemented in components such as a database and a separately provided storage unit. Alternatively, the simulation program can be implemented from the outside to components such as a database and a separately provided storage unit. In addition, when the simulation program is obtained from the outside, the simulation program can be distributed from a medium such as an optical disk or a semiconductor memory, or can be downloaded via a transmission mechanism such as the Internet.

[0058] The output unit 13 outputs the operation result of the operation unit 12 (the simulation result of the predetermined film forming process calculated by the operation unit 12). The output unit 13 is implemented by a GUI, and is used to visualize the operation result obtained by the operation unit 12 (the simulation result of the predetermined film forming process calculated by the operation unit 12). In addition, at this time, in addition to the result of the film forming simulation process, the output unit 13 can also output information such as parameters used for the operation, film forming conditions, etc. The output unit 13 can be configured by, for example, any device, such as a display device unit that displays the simulation result, a printing device that prints out the simulation result, and a recording device that records the simulation result, or a suitable combination thereof. It should be noted that although an example in which the simulator includes the output unit 13 is described in the present embodiment, the present technology is not limited to this example and the output unit 13 can be provided outside the simulator.

[0059] The simulator may further include a database section that stores various parameters required for the calculation process performed by the operation section 12. Alternatively, such a database section may be provided outside the simulator. It should be noted that, in the case where various parameters required for the calculation process are appropriately input from the outside, it is not necessary to provide a database section.

[0060] (Details of film formation simulation method)

[0061] In this simulation, a voxel model based on the flux method is used as a prediction technology for the film forming process. Generally, in the voxel model, the voxels arranged in the calculation area are cubes. In this embodiment, each voxel contains not only information indicating the presence or absence of a film, but also information such as film coverage, film quality (such as density, defect density, permeability, adhesion, etc.). In addition, in this embodiment, not only the gas component that directly enters (direct incident component) but also the gas component that expands through the surrounding pattern (gas component that responds to the expansion of the surrounding structure) is considered to calculate the gas component (incident component) flowing into the voxel. In addition, the existence, coverage and film quality of the film are predicted by using the concept of representative particles that depend on the incident radical flux. This makes it possible to predict the existence of films, coverage and film quality within a range of tens of nm with higher accuracy and lower cost compared to fluid calculations using the Monte Carlo method.

[0062] Next, explain Figure 1 and Figure 2 The film formation simulation method shown.

[0063] First, the calculation unit 12 sets the initial conditions for film formation ( Figure 1 and Figure 2 Specifically, the initial conditions for film formation include information related to film formation conditions, information related to the base layer, and other information. For example, in the case where the film formation method is a vapor deposition method using gas as a material, in step S101, model parameters related to surface reactions, incident gas flux, ion incident energy / angle distribution, etc. may be set as information related to film formation conditions. In addition, the shape and material of the base layer may be set as information related to the base layer.

[0064] Next, the operation unit 12 selects the surface voxel ( Figure 2 The voxels in the air region adjacent to the voxels of the deposited film that have been deposited in the last time step (air voxels) are defined as surface voxels. The surface voxels are used as targets for incident radical flux calculations and surface reaction calculations (attachment, desorption, migration, and deposition).

[0065] Next, the calculation unit 12 calculates the incident radical flux ( Figure 1 and Figure 2If ordinary fluid calculations are performed continuously, a large amount of calculation cost is required, and it is not realistic to apply this to nanometer-scale patterns used in semiconductor processes. Therefore, the operation unit 12 according to the present embodiment performs calculations while classifying the gas into two types: gas components that directly enter the surface voxels (direct incident components) and gas components that diffuse through the surrounding patterns (diffusion components).

[0066] Here, Figure 4 A conceptual diagram showing gas delivery. Pattern A has a wider inlet than pattern B. B Represents the direct incident component (density). C B depends on the solid angle obtained when viewing the pattern entrance from a surface voxel. However, this does not include the solid angle contribution of neighboring patterns. A represents the expansion component (density). The calculation unit 12 calculates C using the following equations (1) to (3): A By using the continuity equation (Expression (1)) and Bernoulli's principle (Expressions (2) and (3)), the amount of gas flowing from mode A (with a large inlet) to mode B (with a small inlet) at the junction surface can be found.

[0067] [Mathematical formula 1]

[0068]

[0069] [Mathematical formula 2]

[0070]

[0071]

[0072] P A =C A k B T r P B =C B k B T…(3)

[0073] Here, S A Indicates the opening area of ​​pattern A. S B Represents the area of ​​the connection surface between pattern A and pattern B. V A Represents the gas heating rate of pattern A. V B represents the gas heating rate in mode B. V is V A and V B A generic term for K B represents the Boltzmann constant. T N represents the gas temperature. M represents the mass of the gas. A P represents the pressure of the direct incident component.B Indicates the pressure of the spread component.

[0074] In addition, the operation unit 12 calculates C according to the following expression (4): A The contribution component C(z) to the corresponding voxel can be B While the area between the black circles is approximately treated as a partial groove (the area shown by the dot pattern), C(z) is approximately found according to the following expression (4). Here, W represents the width of the groove, and L represents the distance between the connecting surface and the relative boundary of pattern B. The total flux F in the corresponding voxel can be calculated according to the following expression (5). The total flux F corresponds to the incident radical flux in the corresponding voxel. In the case where there are multiple patterns around pattern B (hereinafter referred to as "surrounding patterns"), C(z) can be weighted by multiplying C(z) of each surrounding pattern by 1 / L, adding the obtained weighted values ​​(C(z)×1 / L), and thus finding the total flux F in the corresponding voxel.

[0075] [Formula 4]

[0076]

[0077] Kundsen diffusion coefficient (Groove shape)

[0078] Surface reaction rate constant (Through hole shape)

[0079] [Formula 5]

[0080] F=C B V B +C(z)V A …(5)

[0081] Next, the calculation unit 12 generates representative particles ( Figure 1 and Figure 2 . , step S103 in the above process. Figure 4 A conceptual diagram showing an actual film formation surface (e.g., an O2 step after a BDEAS step of ALD-SiO2). Figure 6 Shows the Figure 5 The conceptual map shown in FIG. 1 is incorporated into the conceptual map of the voxel model.

[0082] On the actual surface, a film is formed by repeated adhesion, migration, and deposition while each incident particle is bonded to the surface particle, and this causes a potential change. In order to significantly reduce the calculation load, the algorithm according to the present embodiment treats multiple incident particles as a single representative particle. This makes it easy to process 10 18 [ / cm 2 / S], and treats the attachment, desorption, migration and deposition of representative particles according to probability. This is the application of the so-called statistical integration method.

[0083] The calculation unit 12 calculates the number N of representative particles to be generated for each voxel. By setting the provisional density ρ [particles / cm 3 ] and will enter a volume L 3 ×10 -21 [cm 3 ] of a single voxel (e.g., Figure 7 The number of free radical particles (F×L 2 ×10 -14 ×dt[particles]) divided by the number of particles in a single voxel of the deposited film (ρ×L 3 ×10 -21 [particles]) to calculate the number N of representative particles to be generated (see expression (6)). The number N of representative particles to be generated depends on the incident radical flux F [flux / cm 2 / s] varies.

[0084] [Mathematical formula 6]

[0085] N=FxL 2 x10 -14 dt / (ρxL 3 x10 -21 )…(6)

[0086] For example, when F = 10 18 [Flux / cm 2 / s], dt = 0.1 [s], L = 5 [nm] and ρ = 10 22 [Particles / cm 3 ], the number N of representative particles to be generated becomes 20. In the case where there are multiple types of free radical particles, the operation unit 12 performs this operation for each type of free radical particle. In the case of forming a film by CVD or PVD, the operation unit 12 calculates the number N of representative particles to be generated for each type of gas in the same time step. In the case of film formation using ALD, the operation unit 12 calculates the number N of representative particles generated for each gas at different time stages.

[0087] Next, the calculation unit 12 determines the attachment / detachment of each representative particle ( Figure 1 and Figure 2Step S104 in the process). Specifically, the operation unit 12 uses random numbers to determine whether each representative particle is attached with an attachment probability Y (0≤Y≤1), or whether it is desorbed with an attachment probability Y (1-Y). In the case where the representative particle is in contact with the base layer (a very early stage of film formation), the influence of the change in the attachment probability Ys depending on the base damage Da caused by processing is considered (Expression (7)). On the other hand, the attachment probability Yd remains constant on the deposited film (Expression (8)). It should be noted that a, b, and c represent constants set by the user. Da is assigned an experimental value or result calculated by other film formation simulators. In the case of determination of desorption, no further surface reaction is calculated.

[0088] [Formula 7]

[0089] Y S =a+b×D a …(7)

[0090] [Mathematical formula 8]

[0091] Y d =c…(8)

[0092] Next, the calculation unit 12 determines the migration range ( Figure 1 and Figure 2 Step S105). Figure 8 A conceptual diagram showing the actual film-forming surface during migration and deposition. Fig. 9 Conceptual diagram showing voxel model in migration and deposition.

[0093] The attached radical particles migrate on the surface using the energy of the substrate temperature. In the area where the surface potential is stable, the radical particles are bonded to the deposited film (film-forming surface) to be deposited as a film. At this time, some radical particles may have dangling bonds (see Figure 8 ). When simulating this phenomenon into a voxel model, consider the following.

[0094] First, determine the travel length L calculated from the substrate temperature T and the activation energy Ed D (Formula (9))2L D Stereo Range(Movement Range)( Figure 1 and Figure 2 Step S105). Next, in 2L D Search for brackets (depressions with small radius of curvature) within a cubic range ( Fig. 9 ). Here, D0 represents the diffusion constant, τ represents the time constant, K B Denotes the Boltzmann constant. √D0τ and Ed on the right side of Expression (9) are parameters whose values ​​are determined based on the correlation (slope: Ed, intercept: √D0τ) between the substrate temperature calculated by first principles or actually measured and the film density actually measured.

[0095] [Mathematical formula 9]

[0096]

[0097] The following expression (10) represents the surface potential μ by using the radius of curvature 1 / R. The radical particles migrate in such a way that the surface potential μ is minimized, that is, the structure (bracket) having a small radius of curvature 1 / R disappears. Here, μ0 represents the chemical potential on the flat film, γ represents the surface tension, Ω represents the atomic volume, and 1 / R represents the radius of curvature.

[0098] [Formula 10]

[0099]

[0100] Next, the computing unit 12 recognizes the bracket in the voxel space ( Figure 2 Specifically, when focusing on the Fig.10 The migration range surrounded by the thick frame shown in (2L D ), the materials of the top, bottom, left, and right voxels adjacent to the focused air voxel are determined. In the case where the voxel adjacent to the focused air voxel corresponds to the deposited film, the determination index Nv(i, j, k) of the focused air voxel is increased by 1. Based on the value of the determination index Nv, the focused air voxel is determined as follows.

[0101] - In the case of Nv ≥ 3, the focused air voxel is determined to be the bracket.

[0102] - In case Nv=1 or 2, it is determined that the focused air voxel is not a bracket.

[0103] Assume flag: F(i, j, k)=1.

[0104] - In the case of Nv=0, the air voxels that are determined to be in focus remain air.

[0105] For example, assuming that one of the two air voxels has Nv=2 and the other has Nv=3, Fig.10 Each of them is surrounded by a thick frame. At this time, it is determined that the air voxel with Nv=2 is not a bracket, and it is determined that the air voxel with Nv=3 is a bracket.

[0106] Next, the calculation unit 12 moves the representative particle ( Figure 2Specifically, if an air voxel with Nv≥3 exists within the migration range, the representative particle (gas 1, for example, BDEAS in ALD-SiO2 processing) migrates to the air voxel with Nv≥3, and the flag F of the voxel used to determine its bonding state is B (i, j, k) is set to 1. In the case where there are multiple scaffolds within the migration range, the F of the scaffold closest to the representative particle is B (i, j, k) is set to 1 ( Fig.11 ).

[0107] Meanwhile, in the case where there are no air voxels with Nv≥3 within the migration range, the migration target voxel of the representative particle is determined from the voxels with F(i, j, k)=1 by using a random number (according to probability). The flag F of the migration target voxel for determining its bonding state B (i, j, k) is set to 1 ( Fig.12 As described above, the calculation unit 12 selects (determines) the deposition position of the representative particle ( Figure 1 and Figure 2 , or step S106 in the above procedure.

[0108] The calculation unit 12 executes the above-described steps S104 to S106 for all generated representative particles (No in step S204 ) until the deposition position is fixed.

[0109] It should be noted that in the case of forming a film by using a plurality of types of gases (e.g., gas 1 and gas 2), the operation unit 12 selects (decides) the deposition position of the representative particles (e.g., BDEAS) of gas 1, and then performs attachment / desorption to move, for example, the deposition position selection (determination) of the representative particles (e.g., O2) of gas 2 (steps S205 and S206). In other words, the film is formed by ALD. At this time, for the F that has been deposited by gas 1, B (i, j, k) is set to 1 (Yes in step S207), the operation unit 12 uses a random number to determine whether the state obtained during deposition is a bonding state (with probability Y B ) or non-bonded state (dangling bond: probability (1-Y B )). At the same time, for F B For a voxel with (i, j, k) = 0, the operation unit 12 determines that the state obtained during deposition is an unbonded state (step S208). It should be noted that if ions enter the voxel (i.e., there is an incident ion flux Γ) when determining the state of the voxel, i ), the operation unit 12 determines the effect of the bonding enhancement of these ions (determines whether the state is a bonding state or an unbonded state) based on the flux and energy of these incident ions.

[0110] Here, the detailed calculation method will be described. In other words, the following expression (11) represents the number of ions entering the migration target voxel within the time step dt (voxel size: L). By using expression (11) and the ion energy distribution function (IEDF), the total incident energy E entering the voxel can be calculated. tot At the same time, by using the temporary density ρ of the deposited film, the number of bonds in a single voxel becomes ρ×L 3 Therefore, energy expressed by the following formula (12) is added to each bond.

[0111] [Mathematical formula 11]

[0112] Γ i (I,J,K)×L 2 ×dt…(11)

[0113] [Mathematical formula 12]

[0114] Γ i (I,J,K)×L 2 ×dt / (ρ×L 3 -1)…(12)

[0115] When the energy obtained as above is greater than the bond energy E between atoms in the film B (value set by the user) is large, the representative particle will be B On the other hand, when the energy obtained as above is less than the bond energy E between atoms in the film formation, B (user set value), with probability Y B '(user set value) is deposited with a representative particle in a bonding state. As described above, the calculation unit 12 determines the film state (bonded state or non-bonded state) of the voxel (step S209). In other words, the deposition is calculated as a voxel that gives a bonding state or a non-bonded state between the representative particle and the film-forming surface.

[0116] The operation unit 12 performs the above-mentioned steps S205 to S209 (No in step S210) until the film state (bonded state or unbonded state) of all voxels in the migration range is determined. The operation unit 12 performs shape development (Yes in step S210) while determining the film state (bonded state or unbonded state) of all voxels in the migration range. Specifically, the operation unit 12 identifies the voxels whose film state is the bonded state as deposited films, and identifies the voxels whose film state is the unbonded state as unbonded films (defective films). In this way, the films are stacked and their shapes change in each time step.

[0117] Next, the calculation unit 12 calculates the film quality ( Figure 1 and Figure 2As the membrane quality, the membrane density, permeability and adhesion distribution are calculated. Fig.12 The film morphology shown, by using the information assigned to each voxel, the film density, permeability and adhesion distribution are calculated as physical quantities averaged within the set range indicated by the thick frame. The information indicates the bonded state (B), the unbonded state (UB) or the blank area. In other words, the film quality and film coverage on the film-forming surface are represented by calculating the deposition as voxels, which are assigned state information indicating the bonded state or unbonded state between the representative particles and the film-forming surface. The setting range is a parameter set by the user. For example, the setting range is 100nm.

[0118] (Film density)

[0119] exist Fig.13 In the setting range shown, the operation unit 12 gives a density weight W to each voxel, and defines the total average value thereof as the local film density DL(1, m, n) (Expression (13)). The density weight W of the bonded voxels is W B , the density weight W of the unbonded voxels is W UB , the density weight W of the voxels in the empty region is W V It should be noted that (l, m, n) represents the barycentric coordinates of the set region, and Q represents the number of voxels in the set region. L0 Found through actual measurement under typical process conditions, or through MD calculation, etc. L0 It is the parameter entered by the user.

[0120] [Mathematical formula 13]

[0121]

[0122] (Permeability)

[0123] The calculation unit 12 converts Fig.13 The volume V of the void area in the setting area shown v and Fig.13 The volume V of the setting area shown tot The ratio of φ is calculated as the exponent of the magnetic permeability WP (expression (14)). Note that (1, m, n) represents the centroid coordinates of the set area.

[0124] [Formula 14]

[0125] WP(l,m,n)=V v / V tot …(14)

[0126] (Adhesion)

[0127] like Fig.14As shown, the operation unit 12 processes the voxels that are in contact with the base layer as targets and calculates Fig.14 The volume V of the unbonded voxels in the set region shown UB (UB) and Fig.14 The volume V of the setting area shown tot The ratio between them is taken as the index of adhesion Ad (expression (15)). (l, m, n) represents the centroid coordinates of the set area.

[0128] [Mathematical formula 15]

[0129] Ad(l,m,n)=V UB / V tot …(15)

[0130] After the above film quality calculation, the operation unit 12 advances the clock in the calculation by the time step dt (step S211). Next, in the case where the calculation time does not reach the predetermined processing time (No in step S212), the operation unit 12 continues to execute the above step S201 and subsequent steps. On the other hand, when the calculation time reaches the prescribed processing time (Yes in step S212), the operation unit 12 ends the film forming process and calculates the film quality of the annealing treatment (step S111).

[0131] (Annealing calculation)

[0132] Here, the calculation section 12 calculates the thermal denaturation of the film caused by annealing, that is, the film density to be obtained after annealing. Fig.15 A conceptual diagram showing changes in density and film morphology caused by annealing is shown. First, the calculation unit 12 calculates the diffusion length La depending on the substrate temperature T and the Fig.15 The equal segmentation density ρ(I, J, K) of Na voxels within the range 2La surrounded by the thick frame in FIG. The voxels include air voxels. Fig.16 This allows representative particles to be separated from Fig.15 The voxels indicated by the black circles in Fig.15 The other voxels in the range 2La surrounded by the thick frame in FIG. In addition, regarding the calculation of the diffusion length La, the same formula as the above formula (8) is used. However, the parameter values ​​in the formula (8) are different from those used to calculate the migration length L. D Parameter value.

[0133] In addition, considering the representative particles that flow in from the surrounding voxels (I', J', K') at the same time, the density ρ after annealing a The following instructions may be given.

[0134] [Formula 16]

[0135]

[0136] [Mathematical formula 17]

[0137]

[0138] In order to reflect the improvement of the crystallinity of the deposited film due to heat, ρ in the following expression (18) is a Multiply by the correction term β which depends on the substrate temperature T. Here, A represents the coefficient, E a represents the activation energy, K B represents the Boltzmann constant.

[0139] [Mathematical formula 18]

[0140]

[0141] After the above calculation, in the case where voxel UB has a higher density than that of voxel B (bonded state) before annealing, the state of voxel UB (unbonded state) is changed to voxel B (bonded state). In addition, in the case where the air voxel has a higher density than a set density threshold (set by the user), the state of the air voxel is changed to a deposited film voxel and similar processing is applied thereto. This calculation is performed on each voxel except the base voxel.

[0142] (Bubble calculation)

[0143] As calculation of bubbles (film peeling) caused immediately after completion of the film forming process or immediately after (at the time of) annealing, the operation section 12 performs a voxel moving process (to be described below) on voxels UB (unbonded state) in the deposited film.

[0144] First, calculate the gas density n as the bubbling factor in UB B In addition, when calculating the flux of gas 1 and gas 2 in the film formation calculation, the flux F of the gas coefficient (hydrogen (H) during SiO2 film formation) is also derived. B , the derived flux F B Stored in voxels. The following expression (19) is obtained by using F B , voxel size L, film formation time t and the number of super particles NS to represent the gas density n B .

[0145] [Mathematical formula 19]

[0146] n B =F B / L×t / N s …(19)

[0147] The following expression (20) represents the gas density n obtained by using expression (19) assuming that BPart of (R B ≤1 and depends on the incident ion energy) as the energy of the gas pressure obtained in the case of degassing away from the voxel.

[0148] [Mathematical formula 20]

[0149] P B (L 3 )=R B n B ×k B ×T×L 3 …(20)

[0150] Here, T represents the substrate temperature during film formation. B (L 3 ) and by cross Fig.17 The thick solid line in the thick box indicates the sum of the bond energies of the adjacent faces of the voxel E Btot The comparison is made and the result indicates P B (L 3 ) is greater than the sum E Btot In the case of Fig.17 Each voxel in the thick box in moves up one voxel, and the voxel at the initial position becomes an air voxel ( Fig.18 ). In addition, when P B (L 3 ) and by Fig.19 The thick frame in Fig.17 The sum of the bond energies E of the voxels on the adjacent faces indicated by the thick solid lines on the thick boxes outside the thick boxes in Btot The comparison is made and the result indicates P B (L 3 ) is greater than the sum E Btot In the case of Fig.19 Each voxel in the thick box in moves up one voxel, and the voxel at the initial position becomes an air voxel ( Fig. 20 ). Repeat this process until the sum E Btot More than P B (L 3 ). In the case where a plurality of air regions caused by UB overlap with each other, which region has a larger size is determined in a manner similar to the above-described manner regarding the region including the plurality of air regions overlapping with each other.

[0151] It should be noted that the following expression is obtained by using the area density σ of the neighboring voxels and the average bond energy E B Approximately represents the sum E Btot In addition, in order to reflect the influence of ions on bond formation during film formation, the value multiplied by the correction term (exp(-E0 / E)×2.718) that depends on the ion energy E is used as RB Here, E0 represents the standard ion energy.

[0152] [Mathematical formula 21]

[0153]

[0154] <Effect>

[0155] According to the film formation simulation method and film formation simulator 1 of the present embodiment, representative particles are generated according to the incident radical flux, and the attachment, desorption, migration, and deposition of each representative particle on the film formation surface are calculated according to probability. This makes the accuracy and calculation time independent of the number of particles. In addition, according to the present embodiment, through calculation, the film quality and film coverage on the film formation surface can be expressed by calculating the deposition as voxels, and the voxels are given state information representing the bonding state or unbonded state between the representative particles and the film formation surface. Therefore, it is not necessary to separately prepare a database that depends on the morphology and gas flux to calculate the film quality. This makes it possible to shorten the calculation time and improve the calculation accuracy. In addition, it is also expected to shorten the turnaround time (TAT) before using the algorithm according to this example.

[0156] According to this embodiment, the incident radical flux can be calculated based on the gas component and the diffused gas component. The gas component directly enters the voxel of the film-forming surface, and the diffused gas component enters the voxel of the film-forming surface in response to the surrounding structure. This makes it possible to improve the calculation accuracy that reflects the real structure.

[0157] According to this embodiment, whether a representative particle is bonded or not bonded to the film-forming surface can be determined based on the energy and ion flux of the voxel entering the film-forming surface. Thus, a film-forming process that takes into account the effect of ions on adhesion enhancement can be realized.

[0158] According to this embodiment, by using state information indicating a bonding state or a non-bonding state, film density, permeability, and adhesion to the film formation surface can be calculated, thereby achieving a more accurate film formation process.

[0159] According to this embodiment, when a film is formed using multiple gases, the number of representative particles generated by the type of gas in the same time step can be calculated. This means a film forming process using CVD or PVD.

[0160] According to this embodiment, when a film is formed using a plurality of gases, the number of representative particles generated by the types of gases at different time steps can be calculated. This represents a film forming process using ALD.

[0161] According to this embodiment, blistering, film density, or both can be calculated during annealing after film formation. As described above, according to this embodiment, not only the film formation process but also the film change caused by annealing can be expressed.

[0162] The film formation simulator 1 according to the present embodiment includes an input section 11 implemented by a CUI or a GUI to set film formation conditions, and an output section 143 implemented by a GUI to visualize the calculation results obtained by the calculation section 12. This enables the user to operate the film formation simulator in a relatively simple manner, and to recognize the arithmetic results obtained by the film formation simulator 1 in a relatively simple manner.

[0163] <Example 1>

[0164] Basic CVD calculations (coverage + film density)

[0165] Fig.21 Calculation of coverage and film density distribution according to the CVD method in this example is shown. In this example, the substrate temperature is set to 800°C, the gas pressure is set to 400 Pa, and a thin film is formed in the base Si trench by using BDEAS BDEAS+O2 gas. The base Si trench has a structure including two intersecting trenches with a line width of 150nm, a depth of 2μm, and a length of 450nm. Fig.21 The simulation results are shown, which take into account the adhesion probability (0.5+0.5×Da) reflecting the processing damage caused during trench etching and the voxel bonding state probability during deposition caused by incident ions (energy up to 50 eV) (0.2 in the case of no ion entry, but 1 in the case of ion entry). The activation energy (determination of the bonding state by ion incidence and migration) is uniformly set to 0.4 eV. The voxel size is set to 5 nm. The set range for the average value used to calculate the film quality is set to 25 nm. In addition, the incident ion flux into the pattern is set to 4×10 16 / cm 2 / s, and the radical flux (BDEAS, O2) was set to 1×10 18 / cm 2 / s.

[0166] As the film formation time increases, the slit area of ​​the groove is sealed, SiO2 is thinly stacked in the groove, and a gap is formed. In addition, a high film density is obtained in the flat part where the ion irradiation has a greater impact (the possibility that the voxel is determined to be in a bonding state is high), and a low film density is obtained in the groove where the ion irradiation has a smaller impact. Fig. 22 Results showing voxel bonded / unbonded state distribution and gas flux distribution.

[0167] <Example 2>

[0168] Basic CVD calculations (transmittance)

[0169] Fig.23 The calculation result of the permeability distribution in this example is shown. This is the calculation result obtained using the basic structure and film formation conditions according to Example 1. Low permeability is obtained in the flat portion, and high permeability is obtained in the groove.

[0170] <Example 3>

[0171] Basic CVD calculations (adhesion)

[0172] Fig.24 The calculation results of the adhesion distribution in this example are shown. This is the calculation result obtained using the base structure and film forming conditions according to Example 1. Strong adhesion is obtained in the flat portion, and weak adhesion is obtained in the groove.

[0173] <Example 4>

[0174] Basic CVD calculations (annealing)

[0175] Fig.25 , Fig.26 and Fig. 27 The results of annealing calculations in this example are shown. These calculation results show the film density distribution ( Fig.25 ), Adhesion Distribution ( Fig.26 ), and permeability distribution ( Fig. 27 ) changes, and these calculation results show that the film quality improves towards the inner part of the groove.

[0176] <Example 5>

[0177] Basic CVD calculations (bubbling)

[0178] Fig.28 A conceptual diagram showing the bubble calculation in this example. Bubble calculation was performed after film formation according to Example 1. Assume that the substrate temperature is set to 800°C and the gas factor flux is set to 1×10 18 / cm 2 / s,R B The value was set to 0.01, the bonding energy was set to 5 eV, and a bubble having a size corresponding to the two-layer voxel was formed on the flat portion (film peeling).

[0179] <Example 6>

[0180] Apply CVD calculation (illustrative pattern with wide line width + exemplary pattern with narrow line width)

[0181] In Example 1, calculation is performed under the assumption that there is a structure in which two grooves having the same line width intersect each other. Meanwhile, calculation using the algorithm according to Example 6 is not limited thereto. The algorithm according to Example 6 can be applied to any structure including a mixture of a pattern having a wide line width and a pattern having a narrow line width.

[0182] For example, Fig.29 As shown in (A), regarding the illustrated pattern including a circular pattern B and four rectangular patterns A coupled to the periphery of the pattern B, calculation can be performed using the algorithm according to Example 6. In addition, for example, Fig.29 As shown in (B), regarding the illustrated pattern including a rectangular pattern B and four rectangular patterns A coupled to the periphery of pattern B, calculation can be performed using the algorithm according to Example 6.

[0183] In addition, for example, Fig.29 As shown in (C), regarding the illustrated pattern including a circular pattern B and two rectangular patterns A, calculation can be performed using the algorithm according to Example 6, and the two rectangular patterns A are connected to the periphery of the pattern B at relative positions across the circular pattern B. In addition, for example, Fig.29 As shown in (D), regarding the illustrated pattern including a rectangular pattern B and two rectangular patterns A, calculation can be performed using the algorithm according to Example 6, and the two rectangular patterns A are connected to the periphery of the pattern B at relative positions across the rectangular pattern B.

[0184] In addition, for example, Fig.29 As shown in (E), regarding the illustrated pattern including a circular pattern B and two rectangular patterns A, the calculation can be performed using the algorithm according to Example 6, wherein the two rectangular patterns A are coupled to the periphery of the pattern B at a position where their extension directions intersect at an angle of 90 degrees. In addition, for example, Fig.29 As shown in (F), regarding the illustrated pattern including a rectangular pattern B and two rectangular patterns A, calculation can be performed using the algorithm according to Example 6, and the two rectangular patterns A are coupled to the periphery of pattern B at a position where their extension directions intersect at an angle of 90 degrees.

[0185] <Example 7>

[0186] ALD calculation

[0187] The above examples 1 to 6 describe CVD calculation. Meanwhile, according to example 7, by alternately repeating the calculation of Figure 2 By calculating the gas 1 and gas 2 in the ALD process until the end of film formation, the coverage and film quality can be calculated.

[0188] <Example 8>

[0189] PVD calculation

[0190] Examples 1 to 6 above describe CVD calculations. Meanwhile, according to Example 8, by only calculating Figure 2 Gas 1 in the calculation can perform PVD calculations on a metal film target including a single element. Alternatively, by calculating Gas 3 and more gases in addition to Gas 1 and Gas 2, PVD calculations can be performed on a composite target including multiple elements.

[0191] <Example 9>

[0192] Collaboration with Gas Phase Computing

[0193] Fig.30 Example 9 is shown. By executing Figure 1 The gas phase calculation of the incident flux calculation in , can simulate the density distribution of gas 1 and gas 2 in the chamber plane. By inputting these density distributions and multiplying by the gas heating rate, the flux is obtained. In the gas phase calculation, the power to be applied to the upper electrode, the frequency, the gas type, the gas flow rate, the gas pressure, the chamber wall state (gas attachment probability), the chamber configuration, and the exhaust speed are input. Alternatively, interpolation calculations based on an actual measurement database or calculations using machine learning (e.g., Gaussian process regression, deep learning, etc., but the method is not limited) can be used.

[0194] <Example 10>

[0195] Fig.31 It is used to implement Figure 1 and Figure 2 A diagram showing a configuration example of information processing software (film formation simulation program) for the film formation simulation method shown.

[0196] (Structure of film formation simulation program 2)

[0197] Fig.31 The film formation simulation program 2 shown includes an input section 21, an arithmetic engine section 22 and a display section 23 to visualize simulation results. The arithmetic engine section 22 includes an airflow calculation section 22a, a coverage calculation section 22b, a film quality calculation section 22c and an output section 22d.

[0198] The input unit 21 is implemented by a character-based user interface (CUI) or a graphical user interface (GUI) to set the initial condition. The input unit 21 outputs the value of the set initial condition to the calculation engine unit 22. The display unit 23 is implemented by a GUI for visualizing the data obtained from the output unit 22d.

[0199] The gas flux calculation unit 22a calculates the incident radical flux incident surface voxel by executing the above-mentioned step S102 using the value of the initial condition set by the input unit 21. The coverage calculation unit 22b calculates the shape and state of the film by executing the above-mentioned steps S103 to S109 using the value of the incident radical flux calculated by the gas flux calculation unit 22a. The film quality calculation unit 22c calculates the film quality by executing the above-mentioned steps S110 and S111 using the shape and state of the film obtained by the coverage calculation unit 22b. The output unit 22d outputs the simulation results of the prescribed film forming process calculated by the coverage calculation unit 22b and the film quality calculation unit 22c.

[0200] The execution platform of the film formation simulation program 2 is, for example, any of Windows (registered trademark), Linux (registered trademark), Unix (registered trademark), and Mac (registered trademark). In addition, the GUI used as the input unit 21 and the display unit 23 can use any configuration language, such as OpenGL, Motif, and tcl / tk. The programming language of the calculation engine unit 22 can be any type of programming language, such as C, C++, Fortran, and JAVA (registered trademark). In addition, the film formation simulation program 2 may also have a function of importing the calculation results of other simulators into the calculation engine unit 22.

[0201] As initial conditions, for example, recipe information, operation parameters, pattern construction information are input. After the calculation is completed by the coverage calculation unit 22b and the film quality calculation unit 22c, the output unit 22d outputs the material / coordinate of each voxel, the state of each voxel (bonded state or unbonded state) / film quality (film density, permeability and adhesion) and surface flux to a file or a temporary storage area. Further, these results can also be visualized through the GUI. Data can be output or visualized in real time during calculation.

[0202] The film-forming simulation program 2 of this example generates representative particles based on the incident radical flow, and calculates the attachment, desorption, migration, and deposition of each representative particle on the film-forming surface based on probability. This makes the accuracy and calculation time independent of the number of particles. In addition, according to this example, through the above calculations, the film mass and film coverage on the film-forming surface can be expressed by calculating the deposition as voxels, and the voxels are assigned state information representing the bonding state or unbonded state between the representative particles and the film-forming surface. Therefore, it is not necessary to separately prepare a database that depends on morphology and gas flux to calculate the film mass. This makes it possible to shorten the calculation time and improve the calculation accuracy. In addition, it is also expected to shorten the turnaround time (TAT) before using the algorithm according to this example.

[0203] In this example, the input section 21 is implemented by a CUI or GUI to set film forming conditions, and the output section 22d is implemented by a GUI to visualize the arithmetic results obtained by the coverage calculation section 22d and the film quality calculation section 22c. Thus, the user can relatively easily execute the film forming simulation program 2 and can relatively easily recognize the calculation results obtained by the film forming simulation program 2.

[0204] <Example 11>

[0205] Fig.32 It means that it is applied Figure 1 and Figure 2 A diagram showing an example of the configuration of the film forming apparatus 3 of the film forming simulation method (film forming simulation program 2) shown.

[0206] (Structure of Film Forming Device 3)

[0207] Fig.32 A conceptual diagram showing a film forming apparatus 3 . The film forming apparatus 3 includes a film forming chamber 31 , a film forming simulation system 32 , a control system 33 , and a fault detection and classification / equipment engineering system (FDC / EES) system 34 .

[0208] The film forming chamber 31 includes a monitoring device 31A for monitoring the state in the chamber. For example, the monitoring device 31A includes an optical emission spectrometer (OES). The OES is a measuring device for monitoring the light emitted from the plasma in the chamber. When a film is formed, light of a wavelength specific to each gas type present in the plasma is emitted. The OES measures the light. The OES measures the light emission intensity of each wavelength. The OES specifies the gas type based on the measured wavelength and outputs information about the specified gas type as monitoring data. For example, the monitoring device 31A includes a system for monitoring the state in the chamber. The system measures the time changes of gas pressure, flow rate, temperature, power, bias, matching capacity, vacuum pump opening, etc., and outputs the data obtained by the measurement as monitoring data.

[0209] The film forming chamber 31 sends the data (monitoring data) obtained by the monitoring device 31A to the film forming simulation system 32. The film forming simulation system 32 includes a film forming simulator 1 that executes the film forming simulation program 2. The film forming simulation system 32 includes a gas flux calculation unit 32A, an optimization calculation unit 32B, and a correction condition output unit 32C. The gas flux calculation unit 32A, the optimization calculation unit 32B, and the correction condition output unit 32C can be implemented by an integrated circuit or by a calculation device loaded with the film forming simulation program 2.

[0210] The gas flux operation section 32A calculates the incident radical flux using the monitoring data input from the monitoring device 31A. By performing the above steps S103 to S109 using the value of the incident radical flux obtained by the gas flux operation section 32A, the optimization operation section 32B generates representative particles according to the value of the incident radical flux, and calculates the attachment, desorption, migration and deposition of each representative particle to the film formation surface according to probability. This allows the optimization operation section 32B to predict the shape and state of the film. In addition, the optimization operation section 32B calculates the film quality by performing the above steps S110 and S111 using the obtained shape and state of the film.

[0211] The optimization operation unit 32B compares the calculated value (predicted value) obtained in this way with the desired specification (for example, the set film thickness information D2). In the case where the calculated value (predicted value) exceeds the allowable range, the optimization operation unit 32B searches for the optimization conditions of the film forming process and finds the calculated value (predicted value) within the allowable range. Specifically, the optimization operation unit 32B changes the gas flow rate, gas pressure, substrate temperature and processing time to find a solution from a predetermined algorithm. The optimization operation unit 32B searches for the optimization conditions for each wafer or each batch.

[0212] In the case where the optimization condition is not found as a result, the optimization arithmetic section 32B sends an abnormality signal to the FDC / EES system 34. When receiving the abnormality signal from the optimization arithmetic section 32B, the FDC / EES system 34 outputs a signal for stopping the operation of the film forming chamber 31 to the control system 33. When receiving the signal for stopping the operation of the film forming chamber 31, the control system 33 stops the operation of the film forming chamber 31.

[0213] Alternatively, in the case where the optimized condition is found, the optimization operation section 32B outputs the found optimized condition to the correction condition output section 32C. When the optimized condition is input, the correction condition output section 32C generates data required for the film forming condition of the film forming chamber 31 to become the optimized condition and outputs the generated data to the control system 33. When receiving the data required for the optimized condition, the control system 33 controls the operation of the film forming chamber 31 based on the received data. The control system 33 changes the recipe information D1 based on the received data.

[0214] Regarding the optimization operation unit 32B, when the calculation time is equal to or greater than the actual processing time, as described above, the optimal solution is obtained not online but by a method (offline method) of preparing a database for simulation using the algorithm according to Example 11 with respect to various processing conditions in advance, and the above steps S103 to S109 are performed by using the value of the incident radical flux predicted using the database, and representative particles are generated based on the value of the incident radical flux. Alternatively, a machine learning model can also be constructed by using the database and used as the online optimization operation unit 32B.

[0215] The film-forming device 3 of this example generates representative particles based on the incident free radical flow, and calculates the attachment, desorption, migration, and deposition of each representative particle on the film-forming surface based on probability. This makes the accuracy and calculation time independent of the number of particles. In addition, according to this example, through the above calculation, the film quality and film coverage on the film-forming surface can be expressed by calculating the deposition as voxels, and the voxels are given state information representing the bonding state or unbonded state between the representative particles and the film-forming surface. Therefore, it is not necessary to separately prepare a database that depends on morphology and gas flux to calculate the film quality. This makes it possible to shorten the calculation time and improve the calculation accuracy. In addition, it is also expected to shorten the turnaround time (TAT) before using the algorithm according to this example.

[0216] Furthermore, in the film forming apparatus 3 according to the present example, the optimal conditions for the film forming process are searched for based on the calculation results obtained by the above-mentioned calculations, data required for the film forming conditions of the film forming chamber to become the found optimal conditions are generated, and the generated data are output to the control section 33. This feedback makes it possible to manufacture semiconductor devices with higher quality.

[0217] In this example, the monitoring device 31A monitors the state in the film forming chamber, and the incident radical flux is calculated by using the monitoring data obtained by the monitoring device 31 A. Such calculation using actual measured values ​​makes it possible to manufacture semiconductor devices with higher quality.

[0218] In this example, representative particles are generated based on the value of the incident radical flux predicted by using the database. Thus, even when the calculation time is larger than the actual processing time, it is possible to manufacture a semiconductor device with high quality.

[0219] In this example, the optimal conditions are searched for each wafer or each lot, thereby making it possible to manufacture semiconductor devices with higher quality for each wafer or each lot.

[0220] The present disclosure has been described above with reference to the embodiments and their variations. However, the present disclosure is not limited thereto, and various variations may be made thereto. It should be noted that the effects described in this specification are merely examples. The effects of the present disclosure are not limited to those described in this specification. The present disclosure may include effects other than those described in this specification.

[0221] For example, the present disclosure may be configured as follows.

[0222] (1) A film formation simulation method comprising:

[0223] In the first step, representative particles are generated according to the incident free radical flux, and the attachment, desorption, migration and deposition of each representative particle on the film-forming surface are calculated according to probability.

[0224] The first step includes expressing the film quality and film coverage on the film formation surface by calculating the deposition as voxels endowed with state information, the state information indicating the bonding state or non-bonding state between the representative particles and the film formation surface.

[0225] (2) The film formation simulation method according to (1), further comprising:

[0226] In the second step, the incident radical flux is calculated based on the gas component that directly enters the voxels of the air region adjacent to the film-forming surface and the diffuse gas component that enters the voxels of the air region adjacent to the film-forming surface in response to the surrounding structure.

[0227] (3) A film formation simulation method according to (1) or (2), wherein the first step further includes: determining whether the representative particle is bonded or not bonded to the film formation surface based on the energy and ion flux of voxels entering the air region adjacent to the film formation surface.

[0228] (4) The film formation simulation method according to any one of (1) to (3), wherein the first step further includes: calculating film density, permeability, and adhesion to a film formation surface by using the state information.

[0229] (5) The film formation simulation method according to any one of (1) to (3), wherein, when a plurality of gases are used to form a film, the first step further includes: calculating the number of representative particles generated by the type of gas in the same time step.

[0230] (6) The film formation simulation method according to any one of (1) to (3), wherein, when a plurality of gases are used to form the film, the first step further includes: calculating the number of representative particles generated by the types of gases in different time steps.

[0231] (7) The film formation simulation method according to any one of (1) to (6), wherein the first step further includes calculating bubbles, film density, or both during annealing after film formation.

[0232] (8) A film forming simulation program comprising:

[0233] An input part for obtaining film forming conditions;

[0234] a calculation unit that generates representative particles based on the incident radical flux under the film forming conditions obtained by the input unit, and calculates the attachment, desorption from the film forming surface, migration and deposition of each representative particle on the film forming surface based on probability; and

[0235] The output part outputs the calculation result of the calculation part.

[0236] The calculation unit expresses the film quality and film coverage on the film formation surface by calculating the deposition as voxels to which the state information is assigned, and the state information indicates the bonding state or non-bonding state between the representative particles and the film formation surface.

[0237] (9) The film formation simulation program according to (8), wherein:

[0238] The input unit is implemented by a character-based user interface (CUI) or a graphical user interface (GUI) to set film forming conditions, and

[0239] The output unit is implemented by a GUI to visualize the calculation result obtained by the calculation unit.

[0240] (10) A film forming simulator comprising:

[0241] An input part for obtaining film forming conditions;

[0242] a calculation unit that generates representative particles based on the incident radical flux under the film forming conditions obtained by the input unit, and calculates the attachment, desorption from the film forming surface, migration and deposition of the representative particles on the film forming surface based on probability; and

[0243] The output part outputs the calculation result of the calculation part.

[0244] The calculation unit expresses the film quality and film coverage on the film formation surface by calculating the deposition as voxels to which the state information is assigned, and the state information indicates the bonding state or non-bonding state between the representative particles and the film formation surface.

[0245] (11) The film formation simulator according to (10), wherein

[0246] The input unit is implemented by a character-based user interface (CUI) or a graphical user interface (GUI) to set film forming conditions, and

[0247] The output unit is implemented by a GUI to visualize the calculation result obtained by the calculation unit.

[0248] (12) A film forming device comprising:

[0249] Film forming room;

[0250] A control unit, which controls the operation of the film forming chamber;

[0251] An optimization operation unit generates representative particles according to the value of the incident radical flux, calculates the attachment, desorption, migration, and deposition of each representative particle on the film-forming surface according to probability, and retrieves the optimization conditions of the film-forming process based on the calculation results; and

[0252] an output unit that generates data required to change the film forming conditions of the film forming chamber to the optimal conditions found by the optimization calculation unit, and outputs the generated data to the control unit,

[0253] The optimization operation unit expresses the film quality and film coverage on the film formation surface by calculating the deposition as voxels to which state information is assigned, the state information indicating the bonding state or non-bonding state between the representative particles and the film formation surface.

[0254] (13) The film forming apparatus according to (12), further comprising:

[0255] A monitoring device for monitoring the status in the film forming chamber; and

[0256] The gas flux calculation unit calculates the incident radical flux using the monitoring data obtained by the monitoring device.

[0257] (14) The film forming apparatus according to (12), wherein the optimization calculation unit generates the representative particles based on a value of the incident radical flux predicted by using a database.

[0258] (15) The film forming apparatus according to any one of (12) to (14), wherein the optimization calculation unit searches for the optimization condition for each wafer or each lot.

[0259] This application claims the benefit of Japanese Priority Patent Application JP2022-094465 filed with the Japan Patent Office on Jun. 10, 2022, the entire contents of which are hereby incorporated by reference.

[0260] It should be understood by those skilled in the art that various modifications, combinations, sub-combinations and substitutions may occur depending on design requirements and other factors insofar as they are within the scope of the appended claims or the equivalents thereof.

Claims

1. A film formation simulation method, comprising: In the first step, representative particles are generated according to the incident free radical flux, and the attachment, desorption, migration and deposition of each representative particle on the film-forming surface are calculated according to probability. The first step includes expressing the film quality and film coverage on the film-forming surface by calculating the deposition as voxels endowed with state information, wherein the state information indicates a bonding state or a non-bonding state between the representative particles and the film-forming surface.

2. The film formation simulation method according to claim 1, further comprising: In the second step, the incident radical flux is calculated based on a gas component that directly enters a voxel in the air region adjacent to the film-forming surface and a diffuse gas component that enters a voxel in the air region adjacent to the film-forming surface in response to a surrounding structure.

3. The film formation simulation method according to claim 1, wherein: The first step further includes determining whether the representative particle is bonded or unbonded to the film-forming surface based on energy and ion flux of voxels entering an air region adjacent to the film-forming surface.

4. The film formation simulation method according to claim 1, wherein: The first step further includes calculating film density, permeability, and adhesion to the film forming surface by using the state information.

5. The film formation simulation method according to claim 1, wherein: In the case where a plurality of gases are used to form the film, the first step further includes: calculating the number of representative particles generated by the types of gases in the same time step.

6. The film formation simulation method according to claim 1, wherein: In the case of using a plurality of gases to form the film, the first step further comprises: calculating the number of representative particles generated by the types of gases in different time steps.

7. The film formation simulation method according to claim 1, wherein: The first step further includes calculating bubbles, film density, or both during annealing after film formation.

8. A film forming simulation program, comprising: An input part, for obtaining film forming conditions; a calculation unit that generates representative particles according to the incident radical flux under the film forming conditions obtained by the input unit, and calculates the attachment, desorption from the film forming surface, migration and deposition of each representative particle on the film forming surface according to probability; as well as an output unit, outputting the calculation result of the calculation unit, The calculation unit expresses the film quality and film coverage on the film formation surface by calculating the deposition as voxels to which state information is assigned, the state information indicating a bonding state or a non-bonding state between the representative particles and the film formation surface.

9. The film formation simulation program according to claim 8, wherein: The input unit is implemented by a character-based user interface (CUI) or a graphical user interface (GUI) to set the film forming conditions, and The output unit is implemented by a GUI to visualize the calculation result obtained by the calculation unit.

10. A film forming simulator, comprising: An input part, for obtaining film forming conditions; a calculation unit that generates representative particles according to the incident radical flux under the film forming conditions obtained by the input unit, and calculates the attachment, desorption from the film forming surface, migration and deposition of the representative particles on the film forming surface according to probability; as well as an output unit, outputting the calculation result of the calculation unit, The calculation unit expresses the film quality and film coverage on the film formation surface by calculating the deposition as voxels to which state information is assigned, the state information indicating a bonding state or a non-bonding state between the representative particles and the film formation surface.

11. The film formation simulator according to claim 10, wherein: The input unit is implemented by a character-based user interface (CUI) or a graphical user interface (GUI) to set the film forming conditions, and The output unit is implemented by a GUI to visualize the calculation result obtained by the calculation unit.

12. A film forming device, comprising: Film forming room; A control unit, controlling the operation of the film forming chamber; An optimization operation unit generates representative particles according to the value of the incident radical flux, calculates the attachment, desorption, migration and deposition of each representative particle to the film-forming surface according to probability, and retrieves the optimization conditions of the film-forming process based on the calculation results; as well as an output unit that generates data required to change the film forming conditions of the film forming chamber to the optimal conditions found by the optimization calculation unit, and outputs the generated data to the control unit, The optimization operation unit expresses the film quality and film coverage on the film formation surface by calculating the deposition as voxels to which state information is assigned, the state information indicating a bonding state or a non-bonding state between the representative particles and the film formation surface.

13. The film forming device according to claim 12, further comprising: A monitoring device for monitoring the state in the film forming chamber; as well as The gas flux calculation unit calculates the incident radical flux by using the monitoring data obtained by the monitoring device.

14. The film forming device according to claim 12, wherein: The optimization operation section generates representative particles according to a value of incident radical flux predicted by using a database.

15. The film forming apparatus according to claim 12, wherein: The optimization operation unit searches for optimization conditions for each wafer or each batch.

Images