CN111895923A - Method for fitting and measuring thickness of thin film - Google Patents

Method for fitting and measuring thickness of thin film Download PDF

Info

Publication number
CN111895923A
CN111895923A CN202010645136.3A CN202010645136A CN111895923A CN 111895923 A CN111895923 A CN 111895923A CN 202010645136 A CN202010645136 A CN 202010645136A CN 111895923 A CN111895923 A CN 111895923A
Authority
CN
China
Prior art keywords
film
neural network
parameters
thickness
lstm
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Granted
Application number
CN202010645136.3A
Other languages
Chinese (zh)
Other versions
CN111895923B (en
Inventor
李相相
魏慎金
李晶
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Shanghai Fuwa Technology Co.,Ltd.
Original Assignee
Shanghai Chenhuiyuan Technology Development Co ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Shanghai Chenhuiyuan Technology Development Co ltd filed Critical Shanghai Chenhuiyuan Technology Development Co ltd
Priority to CN202010645136.3A priority Critical patent/CN111895923B/en
Publication of CN111895923A publication Critical patent/CN111895923A/en
Application granted granted Critical
Publication of CN111895923B publication Critical patent/CN111895923B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01B—MEASURING LENGTH, THICKNESS OR SIMILAR LINEAR DIMENSIONS; MEASURING ANGLES; MEASURING AREAS; MEASURING IRREGULARITIES OF SURFACES OR CONTOURS
    • G01B11/00—Measuring arrangements characterised by the use of optical techniques
    • G01B11/02—Measuring arrangements characterised by the use of optical techniques for measuring length, width or thickness
    • G01B11/06—Measuring arrangements characterised by the use of optical techniques for measuring length, width or thickness for measuring thickness ; e.g. of sheet material
    • G01B11/0616—Measuring arrangements characterised by the use of optical techniques for measuring length, width or thickness for measuring thickness ; e.g. of sheet material of coating
    • G01B11/0641—Measuring arrangements characterised by the use of optical techniques for measuring length, width or thickness for measuring thickness ; e.g. of sheet material of coating with measurement of polarization
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/02—Neural networks
    • G06N3/04—Architecture, e.g. interconnection topology
    • G06N3/045—Combinations of networks
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/02—Neural networks
    • G06N3/04—Architecture, e.g. interconnection topology
    • G06N3/049—Temporal neural networks, e.g. delay elements, oscillating neurons or pulsed inputs
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/02—Neural networks
    • G06N3/08—Learning methods

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Data Mining & Analysis (AREA)
  • Molecular Biology (AREA)
  • Biophysics (AREA)
  • Computational Linguistics (AREA)
  • Artificial Intelligence (AREA)
  • Evolutionary Computation (AREA)
  • General Health & Medical Sciences (AREA)
  • Biomedical Technology (AREA)
  • Computing Systems (AREA)
  • General Engineering & Computer Science (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Mathematical Physics (AREA)
  • Software Systems (AREA)
  • Health & Medical Sciences (AREA)
  • Length Measuring Devices By Optical Means (AREA)

Abstract

The invention provides a method for fitting and measuring the thickness of a film, which comprises the steps of adopting a method of combining an elliptical polarization measurement technology and a neural network, and fitting and measuring the thickness of the film through elliptical polarized light parameters of the film and a neural network model; the neural network is an LSTM recurrent neural network. The method for measuring the thickness of the film by fitting is suitable for Al2O3The thickness of the film was measured by fitting. The invention has the advantages that: 1) the result can be obtained more quickly and accurately; 2) the number of required measurement parameters is less, namely the measurement is easier; 3) repeated mathematical iteration is not needed, so that the result is more stable; 4) the invention is very suitable for the field of large-scale optical detection.

Description

Method for fitting and measuring thickness of thin film
Technical Field
The invention relates to a method for measuring the thickness of a thin film in a fitting manner, in particular to a method for obtaining the thickness of a thin film material through elliptical polarization measurement and neural network fitting, and belongs to the technical field of elliptical polarization measurement and neural networks.
Background
An elliptical polarization spectrometer is a traditional non-contact optical measurement device, has high surface sensitivity when approaching a pseudo-brewster angle and non-destructive depth analysis capability, is proved to be a very reliable measurement tool, and can be used for accurately measuring the optical properties of a given sample by combining a specific physical model and proper data spectrum analysis; meanwhile, it has been shown that the harmonic oscillator approximation method employed in the fitting of the measurement data is very effective for simulating the dielectric function spectrum of a given substrate, and as a result, not only the geometry including the sample surface and subsurface layers, but also the doping concentration dependence of the dielectric function spectrum can be determined and analyzed; the ellipsometry technology can analyze the property of a thin film with the thickness smaller than the wavelength by analyzing the phase and amplitude change of the reflected polarized light, the thickness of the tested thin film is generally between tens of nanometers and hundreds of nanometers, and the surface information of a monoatomic layer thin film sample and a body sample can be measured; for the information received by the test, the ellipsometry mainly analyzes the complex refractive index or dielectric function tensor of the material to different wavelengths, and other basic physical property parameters are calculated on the basis of the complex refractive index or dielectric function tensor; the physical property parameters are closely related to the properties of the material such as form, crystallization state, chemical component components and the like, on the basis of the property analysis, the thickness of the film layer can be fitted through a fitting method, different physical models are selected for different materials, so that the data measured through experiments are fitted to the parameters of the models, and the fitting value of the thickness of the sample is obtained; generally, the thickness measurement can be accurate to the nanometer level.
The application of artificial neural networks to various scientific problems is rapidly increasing, and particularly the trend of combining with the traditional physical science field is more obvious; their popularity and utility stems from their ability to mathematically mimic biological neural networks, and thus perform tasks that are generally classified as intelligent behavior; neural networks have been applied to many different problems, including classification and pattern recognition; the several scenarios described above all apply the theory of neural networks and the various techniques involved in "training" and using them; in many previous studies, scholars reported many applications of neural networks, mostly curve fitting to simulated experimental data, avoiding the need for lengthy non-linear least squares fitting procedures.
Disclosure of Invention
The invention provides a method for measuring the thickness of a thin film in a fitting manner, and aims to quickly and accurately obtain the thickness result of the thin film.
The technical solution of the invention is as follows: a method for fitting and measuring the thickness of a film comprises the steps of adopting a method of combining an elliptical polarization measurement technology and a neural network, and fitting and measuring the thickness of the film through elliptical polarized light parameters of the film and a neural network model; the neural network is an LSTM recurrent neural network.
The invention has the advantages that:
1) the result can be obtained more quickly and accurately;
2) the number of required measurement parameters is less, namely the measurement is easier;
3) repeated mathematical iteration is not needed, so that the result is more stable;
4) the invention is very suitable for the field of large-scale optical detection.
Drawings
FIG. 1 is a schematic diagram of the mapping process using the LSTM recurrent neural network.
Fig. 2 is a schematic diagram of the structure of the LSTM layer.
Detailed Description
A method for fitting and measuring the thickness of a film comprises the steps of adopting a method of combining an elliptical polarization measurement technology and a neural network, and fitting and measuring the thickness of the film through elliptical polarized light parameters of the film and a neural network model; the neural network is an LSTM (Long Short-Term Memory networks) cyclic neural network.
The method for measuring the thickness of the film through the elliptical polarized light parameters of the film and the neural network model specifically comprises the following steps:
(1) generating optical parameters of films with different thicknesses by using optical film design software, and then training an LSTM cyclic neural network by using the optical parameters to generate an LSTM cyclic neural network model, wherein the optical parameters comprise the thickness of the films and elliptical polarized light parameters under different angles; the LSTM recurrent neural network model takes the elliptical polarized light parameters as input and outputs the corresponding film thickness;
(2) measuring by an elliptical polarization spectrometer to obtain elliptical polarization parameters of the film;
(3) and inputting the elliptical polarization parameters of the film measured by the elliptical polarization spectrometer into the LSTM recurrent neural network model to obtain the thickness of the film.
The method for generating the LSTM recurrent neural network model by training the LSTM recurrent neural network with the optical parameters comprises the following specific steps:
(1) constructing LSTM recurrent neural network by using Keras deep learning framework and setting parameters of LSTM recurrent neural network
Figure 325642DEST_PATH_IMAGE001
Randomly initializing, wherein the constructed LSTM recurrent neural network comprises four layers, wherein the first three layers are LSTM layers, and the last layer is a full connection layer;
(2) using Psi parameters and Delta parameters of films with different thicknesses between 30nm and 200nm generated by Film Wizard optical Film design software as training sets to train parameters of an LSTM recurrent neural network;
(3) the LSTM recurrent neural network is trained in a way that Psi parameters and Delta parameters of a Film generated by Film Wizard optical Film design software are sent into the LSTM recurrent neural network, a network predicted thickness value h is obtained after network mapping of the LSTM recurrent neural network, and then a loss function L (h, h ') is calculated with the thickness h ' of the Film generated by the Film Wizard optical Film design software, wherein the loss function L is the mean square error loss of h and h ';
(4) iteratively calculating loss function L vs. LSTM recurrent neural network parameters
Figure 185013DEST_PATH_IMAGE001
And updating parameters of the LSTM recurrent neural network by Adam algorithm
Figure 793849DEST_PATH_IMAGE001
Thus, the training of the LSTM recurrent neural network is realized;
said parameter
Figure 967341DEST_PATH_IMAGE001
Refers to all parameters of all layers of the whole LSTM recurrent neural network model
Figure 541804DEST_PATH_IMAGE001
Including both LSTM layers
Figure 509760DEST_PATH_IMAGE002
、
Figure 668209DEST_PATH_IMAGE003
、
Figure 379813DEST_PATH_IMAGE004
Also comprising fully-connected layers
Figure 307318DEST_PATH_IMAGE002
And
Figure 446175DEST_PATH_IMAGE005
parameter of
Figure 29604DEST_PATH_IMAGE001
Is the set of all learnable parameters in the entire network structure.
The film is any one of a crystal film and a metal oxide film material, and is preferably Al2O3The film material is further preferably Al suitable for preparing a magnetron sputtering coating system2O3A film material.
The elliptical polarized light parameters comprise an amplitude reflectivity ratio and a phase difference; a beam of linearly polarized light with a known polarization state is incident on the surface of the sample to interact with the sample, the polarization state of the reflected light is changed into an elliptical polarization state, and the ratio of the reflectivity of p light and s light is measured, as shown in the following formula (3):
Figure 341636DEST_PATH_IMAGE006
formula (3)
Wherein r ispAnd rsComplex reflectivities for p-light and s-light; Ψ represents the ratio of the amplitude reflectivities of the p light and the s light, and Δ represents the phase difference between the p light and the s light, and is called an ellipsometric parameter, where Ψ is Psi and Δ is Delta.
The method for measuring the thickness of the film is suitable for Al2O3The thickness of the film is subjected to fitting measurement, specifically, Al is prepared by using a magnetron sputtering coating system2O3Film samples and Al measurement by ellipsometry2O3Elliptical polarized light parameters of the film sample; then using the trained LSTM recurrent neural network model and the prepared Al2O3Fitting and measuring Al by using elliptical polarized light parameters of film sample2O3The thickness of the film.
The elliptic polarization spectrometer is an advanced optical film nondestructive measuring instrument, and the obtained original measurement data must be subjected to spectrum analysis by a classical data fitting method (such as a Lorentz vibrator model, a Cauchy model and the like) to obtain the final optical parameters and thickness of a sample; as an advanced data fitting mode, the invention firstly proposes that an LSTM recurrent neural network model is used for fitting measurement in ellipsometric data fitting to replace a classical data fitting mode, and compared with a conventional fitting method, the method is higher in speed and accuracy.
Each LSTM layer comprises a plurality of cell structures, the different cell structures have transverse connection, and the mapping relation of each LSTM layer is represented by the following formula (1):
Figure 326910DEST_PATH_IMAGE007
formula (1)
Wherein:
Figure 203380DEST_PATH_IMAGE008
；
Figure 274104DEST_PATH_IMAGE009
is a time sequence, or step;
Figure 389827DEST_PATH_IMAGE010
is the state of the current step (or call timing) and therefore
Figure 229608DEST_PATH_IMAGE011
Is the state of the last time sequence;
Figure 772584DEST_PATH_IMAGE012
namely, the weighted summation of the current time sequence input and the last time sequence output can be understood as the information of the current input;
Figure 65025DEST_PATH_IMAGE013
as an input to the current timing sequence,
Figure 922123DEST_PATH_IMAGE014
the output of the last time sequence;
Figure 678726DEST_PATH_IMAGE002
is composed of
Figure 330288DEST_PATH_IMAGE015
The weight of (a) is determined,
Figure 375604DEST_PATH_IMAGE004
is composed of
Figure 334595DEST_PATH_IMAGE016
The weight of (a) is determined,
Figure 883388DEST_PATH_IMAGE003
is composed of
Figure 33747DEST_PATH_IMAGE017
The weight of (a) is determined,
Figure 300780DEST_PATH_IMAGE002
、
Figure 499680DEST_PATH_IMAGE004
、
Figure 965296DEST_PATH_IMAGE003
shared among all timings;
Figure 224239DEST_PATH_IMAGE018
and
Figure 40886DEST_PATH_IMAGE019
are all outputs of the current time sequence;
Figure 777898DEST_PATH_IMAGE020
function is as
Figure 301283DEST_PATH_IMAGE021
A function, or hyperbolic tangent function;
Figure 291979DEST_PATH_IMAGE022
in order to input the information into the gate,
Figure 268025DEST_PATH_IMAGE023
in order to forget to leave the door,
Figure 871045DEST_PATH_IMAGE024
is an output gate;
wherein sigmoid is an activation function, called sigmoid function.
In practical use, the size of the LSTM layer coefficient matrix is selected according to the size of the LSTM input and output of each layer, the input and output are vectors with different dimensions, and a matrix with a proper size needs to be selected to realize mapping from the input vector to the output vector.
The first two LSTM layers of the three LSTM layers return a sequence with the same length as the input length; the returnThe sequence with the same length as the input sequence refers to the output sequence
Figure 514516DEST_PATH_IMAGE025
To
Figure 849682DEST_PATH_IMAGE026
Number of and input sequence
Figure 640921DEST_PATH_IMAGE027
To
Figure 719735DEST_PATH_IMAGE028
The number of the output terminals is the same, and the whole is selected and output in actual use
Figure 280030DEST_PATH_IMAGE016
Sequence or only the last sequence output i.e
Figure 786097DEST_PATH_IMAGE026
The neural network is a directed acyclic computational graph, each layer being a function that maps an input vector to an output vector.
The full connection layer is a matrix, and specifically represents the following formula (2):
Figure 2315DEST_PATH_IMAGE029
formula (2)
Wherein the content of the first and second substances,
Figure 448602DEST_PATH_IMAGE015
the sequence is the input sequence of the fully-connected layer,
Figure 801086DEST_PATH_IMAGE030
the sequence is an output sequence; by using
Figure 805951DEST_PATH_IMAGE002
Refers to the entire matrix:
Figure 243886DEST_PATH_IMAGE031
by using
Figure 664503DEST_PATH_IMAGE005
Referring to the entire bias vector:
Figure 199389DEST_PATH_IMAGE032
，
Figure 47260DEST_PATH_IMAGE002
is the coefficient of the number of the first and second,
Figure 34807DEST_PATH_IMAGE005
is an offset;
Figure 259115DEST_PATH_IMAGE002
、
Figure 586191DEST_PATH_IMAGE005
that is, parameters of the neural network, a proper size needs to be selected according to the dimension of the input vector and the dimension of the output vector, so that the input vector is mapped into the output vector.
The whole neural network is a vector mapper, and the input vector is mapped continuously through different layers to obtain the desired output, as shown in fig. 1, it can be seen that the size of the output of each layer is the same as the size of the input of the next layer.
Fitting Al2O3The method for the thickness of the film material comprises the following specific steps:
1) firstly, high-purity Al is taken as a target material, a reactive sputtering method is adopted, namely O is introduced during sputtering2Gas as reaction gas for producing Al2O3A film; different samples are prepared by changing the sputtering time through fixing the sputtering power, and then the samples are annealed in the nitrogen atmosphere to obtain Al with different thicknesses2O3A film; then, Al with different thicknesses is measured by using an elliptical polarization spectrometer2O3Elliptical polarization parameters of the film;
2) generation of Al of varying thickness using optical film design software2O3Optical parameters of the film, including the filmThe method comprises the following steps of (1) training and generating an LSTM (laser scanning) circulating neural network model by using optical parameters under the conditions of thickness and elliptical polarized light parameters under different angles, wherein the LSTM circulating neural network model takes the elliptical polarized light parameters as input and outputs corresponding film thickness;
3) using Al of different thicknesses measured in step 1)2O3Fitting the elliptical polarized light parameters of the film and the LSTM recurrent neural network model obtained by training in the step 2) to obtain corresponding Al2O3The thickness of the film.
The optical parameters used for training the LSTM recurrent neural network to generate the LSTM recurrent neural network model can use measured data; when the measured data is limited and insufficient, the data generated by Film Wizard optical Film design software can be further used for providing data for training an LSTM recurrent neural network; the FilmWizard optical film design software may not be used to generate data when the measured data is sufficient.
The Film Wizard optical Film design software comprises a plurality of physical models, such as Lorentz oscillator models, Cauchy models and the like, wherein one physical model can be understood as a constraint, so that the generated data can meet the constraint, and the data of the elliptical polarized light parameters and the Film thickness generated by the invention can meet the constraint of the Lorentz oscillator models.
Al of different thicknesses in the step 1)2O3The film is preferably 3-4 Al films of different thickness2O3A film.
In the step 3), an LSTM recurrent neural network model is used and an elliptical polarized light parameter is used as an input to fit Al2O3The thickness of the film material.
The method is simple and effective, does not need to select a specific physical model, uses a more universal LSTM (least squares) cyclic neural network model, and is favorable for realizing the aim of carrying out Al alignment through the elliptical polarized light parameters2O3The thickness of the film is fitted quickly, accurately and stably.
Example 1
Fitting measurement Al2O3Thickness of film materialThe method comprises the following specific steps:
Al2O3the film is prepared on a Si (100) substrate by using a Sunicoat 549L magnetron sputtering coating machine, the target material is an Al target with the purity of 99.99 percent, and the aluminum oxide (Al) is prepared by using a reactive sputtering method2O3) The film reactive sputtering is based on direct current magnetron sputtering, high-purity metal aluminum (Al) is taken as a target material, and high-purity oxygen (O) with the purity of 99.99 percent is introduced2) As a reaction gas, the metal aluminum can generate a chemical reaction with oxygen in the sputtering deposition process to generate an alumina film with higher purity;
in the sputtering preparation process, the sputtering power is set to 60W, the argon gas is introduced at a rate of 100sccm as a working gas, the oxygen partial pressure is controlled to be about 2%, and the working gas pressure is set to 3mTor (1Torr =1.33 × 10)2Pa), the distance between the Al target and the substrate is 10mm, the rotating speed of the substrate is 15r/s, the temperature in the cavity is kept consistent with the ambient temperature in the preparation process, the sputtering power is fixed during sputtering, and Al with consistent preparation conditions and different thicknesses can be obtained by using different sputtering time2O3The film is prepared under two sputtering times in actual experiments, wherein the sputtering times are respectively 100s and 200 s; one sample at 100s sputtering time was designated as A; three samples were taken at 200s sputtering time and designated B1, B2, and B3, respectively;
after preparing a sample, characterizing the thickness of the sample by using elliptical polarized light parameters and combining a Lorentz oscillator model; measuring Psi (amplitude reflectance ratio) and Delta (phase difference) parameters of the film sample at three different incident angles, namely 65 °, 70 ° and 75 °; the measuring range is 300nm-800nm, sampling points are taken every 10nm, and 51 sampling points are counted;
selecting five different initial thickness values, using Film Wizard optical Film design software, specifically selecting a Lorentz oscillator model, then fitting the thickness iteratively, wherein the five different initial thickness values are respectively 50nm, 75nm, 100nm, 125nm and 150nm, and then averaging the optimized results of the five different initial thickness values to serve as the final fitting result, wherein the fitting result is shown in Table 1; the initial thickness value is selected by estimating the approximate thickness, and then continuously and iteratively correcting and optimizing to make the thickness closer to the real thickness, wherein five initial thickness values are selected to reduce the influence of accidental errors.
A Keras deep learning framework is used for building an LSTM recurrent neural network, the network has four layers in total, the dimensionality of an input tensor is (6, 6) and is Psi (amplitude reflectivity ratio) and Delta (phase difference) values under 3 angles (65 degrees, 70 degrees, 75 degrees) and 6 wavelengths (300nm, 400nm, 500nm, 600nm, 700nm and 800nm), then final output is obtained through three LSTM layers and a full connection layer, each layer uses a rectification linear unit (ReLU) as an activation function, in the three LSTM layers, the first two layers still return a sequence with the same length as the input length to be used as the input of the next LSTM layer, the last LSTM layer only returns a single tensor at the last unit, the tensor obtains the thickness of the final output through the mapping of the full connection layer, and a model can be used for predicting Al within the thickness range of 30nm-200nm2O3And (3) randomly generating a large number of different thickness values in the range of 30-200 nm by using the Film Wizard optical Film design software, generating corresponding Psi and Delta parameter values by using the Film Wizard optical Film design software to construct a data set, and then training the LSTM recurrent neural network by using the constructed data set to obtain a final LSTM recurrent neural network model.
Applying the well-trained LSTM recurrent neural network model to Al2O3Film samples A, B1, B2, and B3 were fitted to their thickness using Psi and Delta parameters, and the results are shown in Table 2.
From the fitting results, it can be seen that the LSTM neural network is used for Al versus the thickness fitted by the Film Wizard optical Film design software in the classical fitting method2O3The fitting of the film thickness is accurate, and the method has the following advantages:
1. the result can be directly output according to the trained model after only one training without repeated iterative optimization process, so that the calculation is simpler and more stable;
2. the required data volume is less, and a more accurate result can be obtained only by using 1/10 of the parameters required by the Lorentz oscillator model, so that the measurement workload is reduced.
TABLE 1 thickness optimization results for different samples at different initial values of thickness
Figure DEST_PATH_IMAGE033
TABLE 2 comparison of the mean thickness values corresponding to each sample obtained using classical fitting methods for different samples with the thickness values obtained by LSTM neural network fitting
Sample (I) A B1 B2 B3
Thickness (nm) by classical fitting method 37.26 72.77 66.33 66.76
Neural network prediction thickness (nm) 39.20 73.55 68.11 66.20
From table 2, it can be seen that the thickness values obtained by fitting the LSTM neural network only need to be fitted once, and the results consistent with the thickness mean obtained by using the classical fitting method can be obtained; if a classical fitting method is adopted, an initial value of the thickness must be defined firstly, and as shown in table 1, the final fitting values of the thickness of the same sample are different by adopting the classical fitting method due to the difference of the initial values, so that in order to reduce accidental errors, more initial values must be defined by adopting the classical fitting method, and then the optimization results obtained by corresponding to each initial value are averaged, so that the fitting and measuring process of the thickness of the whole film is relatively complicated; according to the technical scheme, after the LSTM recurrent neural network is trained, stable thickness values can be obtained by once fitting for different film samples, and the whole fitting test process is rapid and stable; the fitting by the classical fitting method has a problem that the optimization process from defining an initial value to obtaining a final value needs repeated iteration, and the fitting can be completed at one time by adopting an LSTM neural network.
In this embodiment, different groups of Al are studied by an ellipsometric technique in combination with a neural network method2O3The film is subjected to thickness fitting, and the analysis of the principle of the method and the fitting result shows that compared with the classical fitting method and the fitting method of a corresponding physical model, the method has the advantages of rapidness, accuracy, stability and the like, so that the method is very suitable for the field of large-scale optical detection.

Claims (10)

1. A method for fitting and measuring the thickness of a film is characterized in that the method comprises the steps of adopting a method of combining an elliptical polarization measurement technology and a neural network, and fitting and measuring the thickness of the film through elliptical polarized light parameters of the film and a neural network model; the neural network is an LSTM recurrent neural network.
2. The method of claim 1, wherein the step of fitting the thickness of the thin film by the ellipsometric parameters of the thin film and the neural network model comprises the steps of:
(1) generating optical parameters of films with different thicknesses by using optical film design software, and then training an LSTM cyclic neural network by using the optical parameters to generate an LSTM cyclic neural network model, wherein the optical parameters comprise the thickness of the films and elliptical polarized light parameters under different angles; the LSTM recurrent neural network model takes the elliptical polarized light parameters as input and outputs the corresponding film thickness;
(2) measuring by an elliptical polarization spectrometer to obtain elliptical polarization parameters of the film;
(3) and inputting the elliptical polarization parameters of the film measured by the elliptical polarization spectrometer into the LSTM recurrent neural network model to obtain the thickness of the film.
3. The method of claim 2, wherein the step of using the optical parameters to train the LSTM recurrent neural network to generate the model of the LSTM recurrent neural network comprises the steps of:
(1) constructing LSTM recurrent neural network by using Keras deep learning framework and setting parameters of LSTM recurrent neural network
Figure 919735DEST_PATH_IMAGE001
Randomly initializing, wherein the constructed LSTM recurrent neural network comprises four layers, wherein the first three layers are LSTM layers, and the last layer is a full connection layer;
(2) using Psi parameters and Delta parameters of films with different thicknesses between 30nm and 200nm generated by Film Wizard optical Film design software as training sets to train parameters of an LSTM recurrent neural network;
(3) the LSTM recurrent neural network is trained in a way that Psi parameters and Delta parameters of a Film generated by Film Wizard optical Film design software are sent into the LSTM recurrent neural network, the predicted thickness value h of the neural network is obtained after network mapping of the LSTM recurrent neural network, and then a loss function L (h, h ') is calculated with the thickness h ' of the Film generated by the Film Wizard optical Film design software, wherein the loss function L is the mean square error loss of h and h ';
(4) iteratively calculating loss function L vs. LSTM recurrent neural network parameters
Figure 899192DEST_PATH_IMAGE001
And updating parameters of the LSTM recurrent neural network by Adam algorithm
Figure 678930DEST_PATH_IMAGE001
Thus, the training of the LSTM recurrent neural network is realized;
said parameter
Figure 637921DEST_PATH_IMAGE001
Refers to all parameters of all layers of the entire LSTM recurrent neural network model.
4. A method of fitting measurement of film thickness according to any of claims 1-3, wherein said film is any of a crystalline film, a metal oxide film.
5. The method of claim 1, wherein said elliptically polarized parameters include amplitude reflectance ratio and phase difference.
6. A method for fitting measurement of film thickness according to claim 3, wherein each LSTM layer comprises a plurality of cell structures, different cell structures have transverse connections therebetween, and the mapping relationship of each LSTM layer is expressed by the following formula (1):
Figure 452293DEST_PATH_IMAGE002
formula (1)
Wherein:
Figure 337072DEST_PATH_IMAGE003
；
Figure 604106DEST_PATH_IMAGE004
is a time sequence;
Figure 865323DEST_PATH_IMAGE005
is the state of the current time sequence, therefore
Figure 534201DEST_PATH_IMAGE006
Is the state of the last time sequence;
Figure 589882DEST_PATH_IMAGE007
namely, the weighted summation of the current time sequence input and the last time sequence output is understood as the current input information;
Figure 344211DEST_PATH_IMAGE008
as an input to the current timing sequence,
Figure 81223DEST_PATH_IMAGE009
the output of the last time sequence;
Figure 183039DEST_PATH_IMAGE010
is composed of
Figure 612883DEST_PATH_IMAGE011
The weight of (a) is determined,
Figure 916825DEST_PATH_IMAGE012
is composed of
Figure 191949DEST_PATH_IMAGE013
The weight of (a) is determined,
Figure 897737DEST_PATH_IMAGE014
is composed of
Figure 232903DEST_PATH_IMAGE015
The weight of (a) is determined,
Figure 961825DEST_PATH_IMAGE010
、
Figure 102956DEST_PATH_IMAGE012
、
Figure 600933DEST_PATH_IMAGE014
shared among all timings;
Figure 372580DEST_PATH_IMAGE016
and
Figure 887001DEST_PATH_IMAGE017
are all outputs of the current time sequence;
the f function is
Figure 769506DEST_PATH_IMAGE018
A function, or hyperbolic tangent function;
Figure 121990DEST_PATH_IMAGE019
in order to input the information into the gate,
Figure 126855DEST_PATH_IMAGE020
in order to forget to leave the door,
Figure 564790DEST_PATH_IMAGE021
is an output gate;
wherein sigmoid is an activation function, called sigmoid function.
7. The method of claim 6, wherein the first two of the three LSTM layers return a sequence of the same length as the input length; the returning of a sequence with the same length as the input is inputGo out of sequence
Figure 985407DEST_PATH_IMAGE022
To
Figure 520293DEST_PATH_IMAGE023
Number of and input sequence
Figure 633743DEST_PATH_IMAGE024
To
Figure 621290DEST_PATH_IMAGE025
The number of the output terminals is the same, and the whole is selected and output in actual use
Figure 845598DEST_PATH_IMAGE013
Sequence or only the last sequence output i.e
Figure 907095DEST_PATH_IMAGE023
The neural network is a directed acyclic computational graph, each layer being a function that maps an input vector to an output vector.
8. A method of fitting measurements of film thickness as claimed in claim 3 wherein said fully-connected layer is a matrix, and is expressed in more detail by the following equation (2):
Figure 752298DEST_PATH_IMAGE026
formula (2)
Wherein the content of the first and second substances,
Figure 164825DEST_PATH_IMAGE011
the sequence is the input sequence of the fully-connected layer,
Figure 192824DEST_PATH_IMAGE027
the sequence is an output sequence; by using
Figure 436723DEST_PATH_IMAGE010
Refers to the entire matrix:
Figure 688713DEST_PATH_IMAGE028
by using
Figure 588536DEST_PATH_IMAGE029
Referring to the entire bias vector:
Figure 216963DEST_PATH_IMAGE030
，
Figure 518632DEST_PATH_IMAGE010
is the coefficient of the number of the first and second,
Figure 442988DEST_PATH_IMAGE029
is an offset;
Figure 830107DEST_PATH_IMAGE010
、
Figure 262225DEST_PATH_IMAGE029
the input vector needs to be mapped to the output vector by selecting a proper size according to the dimension of the input vector and the dimension of the output vector.
9. Fitting measurement Al2O3The method for the thickness of the thin film material is characterized by comprising the following specific steps of:
1) firstly, high-purity Al is taken as a target material, a reactive sputtering method is adopted, namely O is introduced during sputtering2Gas as reaction gas for producing Al2O3A film; different samples are prepared by changing the sputtering time through fixing the sputtering power, and then the samples are annealed in the nitrogen atmosphere to obtain Al with different thicknesses2O3A film; then, Al with different thicknesses is measured by using an elliptical polarization spectrometer2O3Elliptical polarization parameters of the film;
2) generation of disks using optical film design softwareAl of the same thickness2O3Optical parameters of the film comprise film thickness and elliptical polarized light parameters under different angles, and then the optical parameters are used for training to generate an LSTM circulating neural network model, wherein the LSTM circulating neural network model takes the elliptical polarized light parameters as input and outputs the corresponding film thickness;
3) using Al of different thicknesses measured in step 1)2O3Fitting the elliptical polarized light parameters of the film and the LSTM recurrent neural network model obtained by training in the step 2) to obtain corresponding Al2O3The thickness of the film.
10. A fitted Al according to claim 92O3Method for the thickness of thin film materials, characterized in that Al of different thickness in step 1) is added2O3The film is 3-4 Al with different thicknesses2O3A film; in the step 3), an LSTM recurrent neural network model is used and an elliptical polarized light parameter is used as an input to fit Al2O3The thickness of the film material.
CN202010645136.3A 2020-07-07 2020-07-07 Method for fitting and measuring thickness of thin film Active CN111895923B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202010645136.3A CN111895923B (en) 2020-07-07 2020-07-07 Method for fitting and measuring thickness of thin film

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202010645136.3A CN111895923B (en) 2020-07-07 2020-07-07 Method for fitting and measuring thickness of thin film

Publications (2)

Publication Number Publication Date
CN111895923A true CN111895923A (en) 2020-11-06
CN111895923B CN111895923B (en) 2021-09-21

Family

ID=73191839

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202010645136.3A Active CN111895923B (en) 2020-07-07 2020-07-07 Method for fitting and measuring thickness of thin film

Country Status (1)

Country Link
CN (1) CN111895923B (en)

Cited By (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112504141A (en) * 2020-11-24 2021-03-16 水利部南京水利水文自动化研究所 Intelligent monitoring method and system for ice thickness of whole scene
CN114107935A (en) * 2021-11-29 2022-03-01 重庆忽米网络科技有限公司 Automatic PVD (physical vapor deposition) coating thickness adjusting method based on AI (Artificial Intelligence) algorithm
CN116086330A (en) * 2023-02-17 2023-05-09 无锡星微科技有限公司 Thickness detection platform for large-size wafer
CN117436355A (en) * 2023-12-21 2024-01-23 东莞市钜欣电子有限公司 Method for establishing optical film thickness model and related equipment

Citations (11)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101666926A (en) * 2008-09-02 2010-03-10 株式会社茉莉特斯 Determination method and determination device for optical anisotropic parameters
US20120158138A1 (en) * 2010-12-20 2012-06-21 Restoration Robotics, Inc. Adjustable Hair Transplantation Chair
US20150258304A1 (en) * 2009-06-03 2015-09-17 Medline Industries, Inc. Catheter Tray, Packaging System, and Associated Methods
CN107574415A (en) * 2017-08-16 2018-01-12 电子科技大学 A kind of metal-base film sensor gradual transition layer and preparation method
CN108492302A (en) * 2018-03-26 2018-09-04 北京市商汤科技开发有限公司 Nervous layer dividing method and device, electronic equipment, storage medium, program
US20190120610A1 (en) * 2017-10-20 2019-04-25 Institute of Nuclear Energy Research, Atomic Energy Council, Executive Yuan, R.O.C. Method and system for measuring thickness of thin film
CN110070713A (en) * 2019-04-15 2019-07-30 浙江工业大学 A kind of traffic flow forecasting method based on two-way nested-grid ocean LSTM neural network
CN110411952A (en) * 2019-07-15 2019-11-05 复旦大学 The spectroscopic ellipsometry of more channel of polarization faces array detection obtains system and method
CN111292525A (en) * 2019-12-30 2020-06-16 南京艾特斯科技有限公司 Traffic flow prediction method based on neural network
CN111324990A (en) * 2020-03-19 2020-06-23 长江大学 Porosity prediction method based on multilayer long-short term memory neural network model
CN111336935A (en) * 2020-03-31 2020-06-26 上海市计量测试技术研究院 Nano-film parameter inversion calculation method based on improved hybrid optimization algorithm

Patent Citations (11)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101666926A (en) * 2008-09-02 2010-03-10 株式会社茉莉特斯 Determination method and determination device for optical anisotropic parameters
US20150258304A1 (en) * 2009-06-03 2015-09-17 Medline Industries, Inc. Catheter Tray, Packaging System, and Associated Methods
US20120158138A1 (en) * 2010-12-20 2012-06-21 Restoration Robotics, Inc. Adjustable Hair Transplantation Chair
CN107574415A (en) * 2017-08-16 2018-01-12 电子科技大学 A kind of metal-base film sensor gradual transition layer and preparation method
US20190120610A1 (en) * 2017-10-20 2019-04-25 Institute of Nuclear Energy Research, Atomic Energy Council, Executive Yuan, R.O.C. Method and system for measuring thickness of thin film
CN108492302A (en) * 2018-03-26 2018-09-04 北京市商汤科技开发有限公司 Nervous layer dividing method and device, electronic equipment, storage medium, program
CN110070713A (en) * 2019-04-15 2019-07-30 浙江工业大学 A kind of traffic flow forecasting method based on two-way nested-grid ocean LSTM neural network
CN110411952A (en) * 2019-07-15 2019-11-05 复旦大学 The spectroscopic ellipsometry of more channel of polarization faces array detection obtains system and method
CN111292525A (en) * 2019-12-30 2020-06-16 南京艾特斯科技有限公司 Traffic flow prediction method based on neural network
CN111324990A (en) * 2020-03-19 2020-06-23 长江大学 Porosity prediction method based on multilayer long-short term memory neural network model
CN111336935A (en) * 2020-03-31 2020-06-26 上海市计量测试技术研究院 Nano-film parameter inversion calculation method based on improved hybrid optimization algorithm

Cited By (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112504141A (en) * 2020-11-24 2021-03-16 水利部南京水利水文自动化研究所 Intelligent monitoring method and system for ice thickness of whole scene
CN112504141B (en) * 2020-11-24 2022-03-29 水利部南京水利水文自动化研究所 Intelligent monitoring method and system for ice thickness of whole scene
CN114107935A (en) * 2021-11-29 2022-03-01 重庆忽米网络科技有限公司 Automatic PVD (physical vapor deposition) coating thickness adjusting method based on AI (Artificial Intelligence) algorithm
CN116086330A (en) * 2023-02-17 2023-05-09 无锡星微科技有限公司 Thickness detection platform for large-size wafer
CN116086330B (en) * 2023-02-17 2024-01-12 无锡星微科技有限公司 Thickness detection platform for large-size wafer
CN117436355A (en) * 2023-12-21 2024-01-23 东莞市钜欣电子有限公司 Method for establishing optical film thickness model and related equipment
CN117436355B (en) * 2023-12-21 2024-04-16 东莞市钜欣电子有限公司 Method for establishing optical film thickness model and related equipment

Also Published As

Publication number Publication date
CN111895923B (en) 2021-09-21

Similar Documents

Publication Publication Date Title
CN111895923B (en) Method for fitting and measuring thickness of thin film
CN107092958B (en) Accurate and fast neural network training for library-based critical dimension CD metrology
CN111336935A (en) Nano-film parameter inversion calculation method based on improved hybrid optimization algorithm
Liu et al. Machine learning aided solution to the inverse problem in optical scatterometry
US5999267A (en) Nondestructive optical techniques for simultaneously measuring optical constants and thicknesses of single and multilayer films
JP7072924B2 (en) Method for determining material optical constants, method for expanding material database and equipment
CN111122459A (en) Method and device for correcting depolarization effect caused by uneven thickness in spectroscopic ellipsometry
Kwak et al. Semiconductor multilayer nanometrology with machine learning
KR100769566B1 (en) Method and apparatus for measuring thin film thickness using neural network and record media therefor
Lihua et al. Characterization of nanofilm parameters based on hybrid optimization algorithm
Zhaoxian et al. Generalized simulated annealing algorithm applied in the ellipsometric inversion problem
CN115219435B (en) Polarization detection method combining broad spectrum ellipsometry and simulation
CN111781148B (en) Method, device, terminal and system for detecting longitudinal non-uniformity of film
Tian et al. Rapid and accurate thin film thickness extraction via UV-Vis and machine learning
CN115289988A (en) Method for measuring thickness and density nonuniformity of nano film material
CN113267454A (en) Thin film quality detection method, thin film quality detection device, electronic device, and storage medium
Patel et al. Determination of refractive index and thickness of thin-film from reflectivity spectrum using genetic algorithm
CN108344698B (en) Ellipsometry method for inverting rough surface optical constant based on electromagnetic first-order principle
WO2021170995A1 (en) Method of and system for processing spectroscopic ellipsometry data
Park et al. Neural-net computing for interpretation of semiconductor film optical ellipsometry parameters
CN113483677A (en) In-situ film property parameter real-time characterization method based on ellipsometer
Gereige et al. Rapid control of submicrometer periodic structures by a neural inversion from ellipsometric measurement
Zabrocki et al. Adaptive Online Time-Series Prediction for Virtual Metrology in Semiconductor Manufacturing
Zhu et al. A Lightweight Neural Network for Spectroscopic Ellipsometry Analysis
Kanclíř et al. Precision of silicon oxynitride refractive-index profile retrieval using optical characterization

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant
TR01 Transfer of patent right

Effective date of registration: 20220323

Address after: 201800 room j4641, building 2, No. 4268, Zhennan Road, Jiading District, Shanghai

Patentee after: Shanghai Fuwa Technology Co.,Ltd.

Address before: Room j4504, building 2, No. 4268, Zhennan Road, Jiading District, Shanghai 201800

Patentee before: Shanghai chenhuiyuan Technology Development Co.,Ltd.

TR01 Transfer of patent right