CN104572742B - It is a kind of to be used to improve the method and device that theoretical spectral database creates speed - Google Patents
It is a kind of to be used to improve the method and device that theoretical spectral database creates speed Download PDFInfo
- Publication number
- CN104572742B CN104572742B CN201310505175.3A CN201310505175A CN104572742B CN 104572742 B CN104572742 B CN 104572742B CN 201310505175 A CN201310505175 A CN 201310505175A CN 104572742 B CN104572742 B CN 104572742B
- Authority
- CN
- China
- Prior art keywords
- exponent number
- expansion exponent
- wavelength points
- spectroscopic data
- critical dimension
- 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.)
- Active
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
- G06F16/35—Clustering; Classification
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
- G06F16/33—Querying
- G06F16/3331—Query processing
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Databases & Information Systems (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Computational Linguistics (AREA)
- Spectrometry And Color Measurement (AREA)
Abstract
The invention discloses a kind of method and device for improving theoretical spectral database and creating speed, wherein, method includes：Obtain different Fourier expansion exponent numbers corresponding theoretical spectral data of the optical critical dimension model in each wavelength points；Convergence analysis is carried out to the theoretical spectral data acquired in the different expansion exponent numbers in each wavelength points, to determine the optimal expansion exponent number for meeting predetermined convergence condition of the wavelength points；According to the optical critical dimension model each wavelength points the corresponding theoretical spectral data of optimal expansion exponent number, to establish or update the theoretical spectral database of the optical critical dimension model.Carry out the calculating of theoretical spectral data using same expansion exponent number to all wavelengths point different from the prior art, the present invention not only ensure that the accuracy of measurement result, but also improve the establishment speed of theoretical spectral database.
Description
Technical field
The present invention relates to the process control technology of IC manufacturing, more particularly to a kind of raising optical critical dimension
【Optics Critical Dimension（OCD）】The theoretical spectral database of e measurement technology creates the method and device of speed.
Background technology
As semi-conductor industry is persistently promoted to deep sub-micron technique node, integrated circuit device size constantly reduces, device
Part structure design more sophisticated.The device of fully functional circuit and high speed operation could be only obtained by stringent technology controlling and process
Part.
Optical critical dimension e measurement technology is a kind of process control technology of mainstream, its base in current semiconductor manufacturing process
This operation principle can be described as：（1）Establish the corresponding theoretical spectral database of pattern with device example；（2）Pass through optics
Critical size measuring apparatus obtains the measure spectrum of sample；（3）Found and optimal of measure spectrum from theoretical spectral database
The characteristic spectrum matched somebody with somebody, so that it is determined that the structural parameters of the sample.
Spectral matching method based on storehouse needs substantial amounts of numerical computations, usually before measuring in dedicated computing server
Upper foundation.When matching measure spectrum, the light with measure spectrum best match is found out from the theoretical spectral database having built up
Spectrum, and index out parameters value.
However, with semiconductor technology development, it is necessary to measurement integrated circuit device sample structure and morphology increasingly
Complexity, thus the variable number of sample theoretical spectral database is also more；Also, technology controlling and process requires increasingly measurement accuracy
Height, the step-length of the variable of theoretical spectral database are finer；At the same time in order to avoid local matching, join in theoretical spectral database
Number scope is also increasing.In the case, the data volume of theoretical spectral database is increasing, how to improve theoretical spectral number
According to storehouse establishment speed it is particularly important that.
The content of the invention
The technical problems to be solved by the invention be to provide it is a kind of improve theoretical spectral database create speed method and
Device, not only to ensure the accuracy of optical critical dimension measurement result, but also can improve optical critical dimension theoretical spectral database
Establishment speed.
According to an aspect of the present invention, there is provided a kind of theoretical spectral database for improving optical critical dimension e measurement technology
The method for creating speed, wherein, including：
Obtain different Fourier expansion exponent numbers corresponding spectroscopic data of the optical critical dimension model in each wavelength points；
Convergence analysis is carried out to the spectroscopic data acquired in the different expansion exponent numbers in each wavelength points, to determine the wavelength
The optimal expansion exponent number for meeting predetermined convergence condition of point；
According to the optical critical dimension model each wavelength points the corresponding spectroscopic data of optimal expansion exponent number, to establish
Or the theoretical spectral database of the renewal optical critical dimension model.
According to another aspect of the present invention, a kind of theoretical spectral for improving optical critical dimension e measurement technology is additionally provided
Database creates the device of speed, wherein, including：
Theoretical spectral data acquisition facility, for obtaining different Fourier of the optical critical dimension model in each wavelength points
The corresponding theoretical spectral data of exponent number are unfolded；
Optimal expansion exponent number determining device, for the spectroscopic data acquired in the different expansion exponent numbers in each wavelength points
Convergence analysis is carried out, to determine the optimal expansion exponent number for meeting predetermined convergence condition of the wavelength points；
Theoretical spectral Database updating device, for according to the optical critical dimension model in each wavelength points most
It is excellent that the corresponding spectroscopic data of exponent number is unfolded, to establish or update the theoretical spectral database of the optical critical dimension model.
Compared with prior art, the present invention has the following advantages：The present invention is by determining OCD models in each wavelength points
Optimal expansion exponent number, to obtain the corresponding spectroscopic data of optimal expansion exponent number, for establishing the theoretical spectral data of OCD models
Storehouse, carries out all wavelengths point using same expansion exponent number different from the prior art calculating of spectroscopic data, so as to both ensure
The accuracy of measurement result, and improve the establishment speed of theoretical spectral database.
Brief description of the drawings
By reading the detailed description made to non-limiting example made with reference to the following drawings, of the invention is other
Feature, objects and advantages will become more apparent upon：
Fig. 1 is the method for being used for the theoretical spectral database that OCD is measured according to a kind of foundation of one embodiment of the invention
Flow chart；
Fig. 2 is the structure diagram of two dimension sample to be measured；
Fig. 3 is the device for being used for the theoretical spectral database that OCD is measured according to a kind of foundation of one embodiment of the invention
Schematic diagram.
The same or similar reference numeral represents the same or similar component in attached drawing.
Embodiment
The present invention is described in further detail below in conjunction with the accompanying drawings.
Fig. 1 is that a kind of theoretical spectral database of raising OCD e measurement technologies according to an embodiment of the invention creates speed
The method flow diagram of degree.
As shown in Figure 1, in step S101, different Fourier expansion exponent numbers pair of the OCD models in each wavelength points are obtained
The spectroscopic data answered.
Preferably, can the model parameter based on the OCD models that do not float, come calculate obtain the OCD models each
The corresponding spectroscopic data of different expansion exponent numbers of wavelength points.
Preferably, can the model parameter based on the OCD models of floating, obtain the OCD models in each ripple to calculate
The corresponding spectroscopic data of different expansion exponent numbers of long point.
The corresponding light of exponent numbers is unfolded in the different of each wavelength points to illustrate how to obtain OCD models with an example below
Modal data.
Example 1
By taking the sample to be measured of the two-dimensional periodic structure shown in Fig. 2 as an example, the sample to be measured is in normal section in week
Phase property changes, if the periodicity of periodic structure is enough, this two-dimensional periodic structure usually is known as optical grating.It is first
First, the sample message of sample to be measured, the overlying material of the sample to be measured are obtained（Such as it is air）With (n0,k0) describe, n is
The refractive index of material, k are material absorption co-efficient, and the first layer of sample to be measured is grating layer, are such as polysilicon, its material is
(n1,k1), cycle PITCH, shape is trapezoidal, is described with (TCD, BCD, HT).The second layer is film layer, is such as silica,
Its material is (n2,k2), thickness is described with Thickness.It is down substrate again, is such as silicon, its material is (n3,k3).Usual feelings
Condition, the information of material can be known by measured thin film technology.In this way, the model of sample can use parameter vector v=(TCD,
BCD,HT,Thickness)TDescription, if vague generalization describes, can be written as v=(V0,V1,...,VN-1)T, Vi, i=0 ..., N-1 is
The parameter of sandwich construction whole.Then, according to acquired sample message, to establish OCD models, specifically, OCD model building devices
Acquired sample message can be converted to OCD models, and the OCD models being converted to are saved as into configuration file.
After OCD models are established, OCD model building devices set the nominal value of each parameter of the OCD models, OCD measurements
The systematic parameter of equipment, and the wavelength parameter of OCD models.Wherein, the nominal value refers to the acquiescence of each parameter of the OCD models
Value；The systematic parameter of OCD measuring apparatus includes the incident angle of measurement light, orientation angles, numerical aperture etc.；OCD models
Wavelength parameter includes start wavelength, terminates wavelength, wavelength interval etc., the wavelength interval refer to start wavelength and terminate wavelength it
Between wavelength points.A usual spectrum has 50 to 150 wavelength points, and the step-length at wavelength points interval is 2 to 5 nanometers.
It is exemplified below, the parameters of the OCD models of the sample to be measured shown in Fig. 2 and OCD measuring apparatus are
Unite parameter, and the nominal value such as institute of table 1 of the systematic parameter of the parameters of the OCD models of sample to be measured and OCD measuring apparatus
Show.Wavelength parameter is：Wavelength initial value is 300nm, and wavelength stop value is 700nm, wavelength interval 5nm, and every wavelength has
81 wavelength points.
Table 1：
It should be noted that above example and parameter values are only for example used, not any limitation of the invention, this
Field technology personnel should be able to be as described above teaching, based on actual conditions, to determine or select suitable parameter values.
Then, OCD theoretical spectrals acquisition device is in the range of the pre-set expansion exponent number to be analyzed, based on ginseng
Number nominal value, to calculate the corresponding theoretical spectral data of different expansion exponent numbers for obtaining the OCD models in each wavelength points, its
In, the parametric nominal value is not floated.
Wherein, the scope of the pre-set expansion exponent number to be analyzed can rule of thumb, with sample to be measured
Based on, the requirement of measurement accuracy is considered to set, it is preferable that the scope of the expansion exponent number to be analyzed of two-dimentional sample generally may be used
To be arranged to 1 to 61, the scope of the expansion exponent number to be analyzed of three-dimensional sample generally could be provided as 1 to 21.
Below with RCWA（Rigorous Coupled-Wave Analysis, rigorous couple-wave analysis）Exemplified by method, come
Illustrate the calculation of spectroscopic data.It will be understood by those skilled in the art that following calculate spectroscopic data using RCWA methods
Mode be only for example, any other calculates the method for spectroscopic data, such as using method of the invention, should be included in this hair
Within bright protection domain, and it is incorporated herein by reference.
Light propagation law in material meets the universal theory of electromagnetic wave, can use RCWA numerical computation methods according to light
Bar phantom calculates the corresponding theoretical spectral of sample of model description.In RCWA methods, parameter vector v=(TCD, BCD, HT,
Thickness)TDetermine the periodic distribution rule of whole space material (n, k), the dielectric system of material is represented with e (x, y, z)
Several distributions.By taking two-dimensional grating as an example, it is located in (x, z) plane, the dielectric constant of material is in cyclically-varying, and in y directions
Uniformly.It is layered in z directions, every layer of material is expressed as εm(x,zm), m be z directions the number of plies, m=1,2 ..., M.Then εm
(x,zm) have periodically in x directions, therefore formula 1 can be used）Shown Fourier (Fourier) series description：
Wherein d is the cycle., can be such as following using Fourier expansion as (2N+1) order harmonics for TE mould electric fields E at m layers
Formula 2）It is shown：
Wherein, TE moulds refer in the waveguide, and the longitudinal component of electric field is zero, and the propagation that the longitudinal component in magnetic field is not zero
Pattern.knx,knzIn x, the component of z, i.e., each layer of electric field (2N+1) rank scattering optical electric field can be used to describe for wave vector.If exhibition
Open that exponent number (2N+1) is sufficiently large, then the precision of E can be guaranteed.Snm(z) the amplitude size of optical electric field is scattered for each rank.For
Each layer can obtain one group of Differential Equation with Constant Coefficients group using wave equation and the boundary condition of each layer.Finally by solution square
The problem of battle array characteristic value and feature vector, can solve the E of each layerm, and finally obtain scattering light at different levels.According to dissipating for solution
Light is penetrated, the description form of the measure spectrum of OCD measuring apparatus output can be converted to.
Wherein, exponent number (2N+1) is unfolded influences the computational accuracy of theoretical spectral.N is bigger, and the error that it is calculated is smaller, but counts
Calculation amount also increases.Under certain precision conditions, the size of required N is related with the structure and material of grating.What RCWA methods calculated
Complexity is O (N3), it is directly proportional to number of plies M, it is proportional to the cube of exponent number N.If three-dimensional structure, complexity higher, is O
((Nx*Ny)3), with x directions exponent number NxWith y directions exponent number NyProduct cube it is proportional.Therefore, usually by high-performance calculation
The calculating for the theoretical spectral that the calculation server of processor composition specialty or evaluation work station complete to need in OCD measurements is appointed
Business.
In step s 102, OCD theoretical spectrals acquisition device is to acquired in the different expansion exponent numbers in each wavelength points
Spectroscopic data carries out Convergence analysis, to determine the optimal expansion exponent number for meeting predetermined convergence condition of the wavelength points.
Specifically, OCD theoretical spectrals acquisition device can be to the light acquired in the different expansion exponent numbers in each wavelength points
Modal data carries out Convergence analysis, when the corresponding spectroscopic data of the expansion exponent number benchmark corresponding with benchmark exponent number of continuous predetermined number
The difference of spectroscopic data is less than predetermined accuracy requirement, it is determined that an expansion rank in the expansion exponent number of the continuous predetermined number
Number is the optimal expansion exponent number of the wavelength points.
Wherein, the benchmark exponent number can rule of thumb be set, it is preferable that the benchmark exponent number of two dimension sample to be measured can
To be arranged to 99 to 961, the benchmark exponent number of three-dimensional sample to be measured could be provided as 21 to 31.The reference spectra data are base
The corresponding spectroscopic data of quasi- exponent number.
Wherein, the predetermined accuracy requires related with OCD measuring apparatus, can pass through the noise analysis of OCD measuring apparatus
Middle acquisition.
It should be noted that above example and benchmark exponent number numerical value be only for example it is used, not to any limit of the present invention
System, the teaching that those skilled in the art should be able to be as described above, based on actual conditions, to determine or select suitable benchmark rank
Number numerical value.
Illustrate to carry out the spectroscopic datas acquired in the different expansion exponent numbers in each wavelength points with an example below
Convergence analysis, when the corresponding spectroscopic data of the expansion exponent number reference spectra data corresponding with benchmark exponent number of continuous predetermined number
Difference is less than predetermined accuracy requirement, it is determined that an expansion exponent number in the expansion exponent number of the continuous predetermined number is the wavelength
The mode of the optimal expansion exponent number of point.
Example 2
To the expansion exponent number N of two dimension sample to be measured, only analysis x-axis directionx=N.To three-dimensional sample to be measured, x-axis is set
The expansion exponent number in direction is equal with the expansion exponent number in y-axis direction, i.e. the expansion exponent number N in x-axis directionx=Ny=N.It is to be measured with two dimension
Exemplified by sample, the spectroscopic data acquired in different expansion exponent numbers in each wavelength points carries out Convergence analysis, compares Data Convergence
Situation, if the corresponding spectra values of continuous three expansion exponent numbers N, N+2, N+4 benchmark oval thickness spectrum corresponding with benchmark exponent number
Data αNAnd βNDifference is both less than predetermined accuracy and requires ε, such as equation 3 below）It is shown：
Then this expansion exponent number N is the minimum value for meeting the requirement of Data Convergence predetermined accuracy in this wavelength points, i.e., most
Excellent expansion exponent number.
Or by taking sample to be measured shown in Fig. 2 as an example, the scope for setting the expansion exponent number to be analyzed is 1 to 41, expansion
Exponent number is at intervals of 2, and it is 0.005 that predetermined accuracy, which requires ε, and benchmark exponent number is 99, wavelength points 500nm, using above-mentioned Convergence analysis
Method calculates all wavelengths point, obtains one group of expansion exponent number list corresponding with each wavelength points, as shown in table 2.
Table 2：
Wherein, when it is 9 that exponent number, which is unfolded,
Then wavelength points are in 500nm, and expansion exponent number is the minimum value that meets the requirement of Data Convergence predetermined accuracy for 9, i.e.,
Optimal expansion exponent number.
Specifically, OCD theoretical spectrals acquisition device can also be to acquired in the different expansion exponent numbers in each wavelength points
Spectroscopic data carries out Convergence analysis, and the corresponding spectroscopic data of exponent number and the first predetermined number before the expansion exponent number are unfolded when one
Expansion exponent number and/or the expansion exponent number after the second predetermined number expansion exponent number corresponding to spectroscopic data in each it
Between difference be respectively less than predetermined accuracy requirement, it is determined that the expansion exponent number be the wavelength points optimal expansion exponent number.
Illustrate to carry out the spectroscopic datas acquired in the different expansion exponent numbers in each wavelength points with an example below
Convergence analysis, the expansion exponent number of the corresponding spectroscopic data of exponent number and the first predetermined number before the expansion exponent number is unfolded when one
And/or the difference between each in the spectroscopic data after the expansion exponent number corresponding to the expansion exponent number of the second predetermined number is equal
Less than predetermined accuracy requirement, it is determined that the expansion exponent number is the mode of the optimal expansion exponent number of the wavelength points.
Example 3
To carry out Convergence analysis to the spectroscopic data acquired in the different expansion exponent numbers in each wavelength points, when an expansion
Between each in the spectroscopic data corresponding to two expansion exponent numbers after the corresponding spectroscopic data of exponent number and the expansion exponent number
Difference be respectively less than predetermined accuracy requirement, it is determined that the expansion exponent number for the wavelength points optimal expansion exponent number mode exemplified by,
Convergence analysis method is as follows：
|αN-αN+2|<ε, | αN-αN+4|<ε,
|βN-βN+2|<ε, | βN-βN+4|<ε。
Exponent number N is then unfolded to meet the minimum value of Data Convergence predetermined accuracy requirement, i.e., optimal exhibition in this wavelength points
Open exponent number.
In step s 103, OCD theoretical spectrals acquisition device according to the optical critical dimension model in each wavelength points
It is optimal that the corresponding spectroscopic data of exponent number is unfolded, to establish or update the theoretical spectral database of the optical critical dimension model.
Specifically, OCD theoretical spectrals acquisition device obtains the OCD models and is corresponded in the optimal expansion exponent number of each wavelength points
Spectroscopic data, then according to the OCD models each wavelength points the corresponding spectroscopic data of optimal expansion exponent number, to establish
Or the theoretical spectral database of the renewal optical critical dimension model.
Preferably, can the model parameter based on the optical critical dimension model of floating, obtain the optics to calculate and close
Optimal expansion exponent number corresponding spectroscopic data of the key dimension model in each wavelength points.Wherein, domain of walker can be according to be measured
Possibility offset of the sample in relation to technique is measured to be configured.
Preferably, can the model parameter based on the optical critical dimension model that do not float, come calculate obtain the optics
Optimal expansion exponent number corresponding spectroscopic data of the critical size model in each wavelength points.
The theoretical spectral database of the OCD models of the foundation or renewal is used for OCD measuring apparatus to sample to be measured
When measuring, when measurement compared with acquired measure spectrum, if according to predetermined matching standard, measure spectrum with
A parameter vector in the theoretical spectral database of the OCD modelsSpectroscopic data match, then
The sample to be measured of the measurement can use the parameter vectorTo describe.Preferably, predetermined
With standard can use GOF (Goodness of Fit, the goodness of fit) or RMSE (Root Mean Square Error, just
Root error) etc..
Illustrate difference with the prior art of the present invention below by an example：
As shown in the following Table 3, the method for establishing the theoretical spectral database for being used for OCD measurements according to the present invention, wherein,
In the optimal expansion exponent number of each wavelength points obtained, maximum expansion exponent number is 29, and minimum expansion exponent number is 5.And
In the prior art, the spectroscopic data of each wavelength points is calculated using same expansion exponent number to all wavelengths point, i.e., each wavelength
Point is all based on expansion exponent number as 29 to calculate the spectroscopic data of each wavelength points.
If carrying out parallel computation using the server of five calculating cores, it is used for OCD using foundation according to the present invention
The method of the theoretical spectral database of measurement uses the foundation of the prior art to be used for the reason of OCD measurements, it is necessary to time-consuming 71.5 minutes
By the method for spectra database, it is necessary to time-consuming 178.2 minutes.As it can be seen that using according to the present invention for improving theoretical spectral number
The optimization method of speed is created according to storehouse, not only ensure that the accuracy of measurement result, but also improves and creates theoretical spectral database
Speed..
Table 3：
Fig. 3 is that a kind of device for being used to improve theoretical spectral database establishment speed according to an embodiment of the invention shows
It is intended to.
As shown in figure 3, a kind of foundation of one embodiment of the invention is used for the device of the theoretical spectral database of OCD measurements
Including theoretical spectral data acquisition facility 301, optimal expansion exponent number determining device 302 and theoretical spectral Database more new clothes
Put 303.
Wherein, theoretical spectral data acquisition facility 301, for obtaining different expansion ranks of the OCD models in each wavelength points
The corresponding theoretical spectral data of number.
Preferably, the theoretical spectral data acquisition facility 301 can include the first spectroscopic data computing module（In figure not
Show）, for the model parameter based on the OCD models that do not float, the OCD models are obtained in each wavelength points not to calculate
The corresponding spectroscopic data with expansion exponent number.
Preferably, the theoretical spectral data acquisition facility 301 can also include the second spectroscopic data computing module（In figure
It is not shown）, for the model parameter based on the OCD models of floating, the OCD models are obtained in each wavelength points not to calculate
The corresponding spectroscopic data with expansion exponent number.
The corresponding reason of exponent numbers is unfolded in the different of each wavelength points to illustrate how to obtain OCD models with an example below
By spectroscopic data.
Example 4
Or by taking the sample to be measured of the two-dimensional periodic structure shown in Fig. 2 as an example, sample to be measured shown in Fig. 2
The parameters of OCD models and the systematic parameter of OCD measuring apparatus, and the parameters and OCD of the OCD models of sample to be measured
The nominal value of the systematic parameter of measuring apparatus is as shown in table 4.Wavelength parameter is：Wavelength initial value is 300nm, and wavelength stop value is
700nm, wavelength interval 5nm, every wavelength have 81 wavelength points.
Table 4：
It should be noted that above example and parameter values are only for example used, not any limitation of the invention, this
Field technology personnel should be able to be as described above teaching, based on actual conditions, to determine or select suitable parameter values.
Then, theoretical spectral data acquisition facility 301 is in the range of the pre-set expansion exponent number to be analyzed, base
In parametric nominal value, the corresponding theoretical spectral data of exponent numbers are unfolded in the different of each wavelength points to calculate the acquisition OCD models,
Wherein, the parametric nominal value is not floated.
Wherein, the scope of the pre-set expansion exponent number to be analyzed can rule of thumb, with sample to be measured
Based on, the requirement of measurement accuracy is considered to set, it is preferable that the scope of the expansion exponent number to be analyzed of two-dimentional sample generally may be used
To be arranged to 1 to 61, the scope of the expansion exponent number to be analyzed of three-dimensional sample generally could be provided as 1 to 21.
Below by taking RCWA methods as an example, for Mingguang City's modal data calculation.Under it will be understood by those skilled in the art that
State and be only for example using RCWA to calculate by the way of spectroscopic data, any other calculates the method for spectroscopic data, such as using
The method of invention, should be included within protection scope of the present invention, and be incorporated herein by reference.
Light propagation law in material meets the universal theory of electromagnetic wave, can use RCWA numerical computation methods according to light
Bar phantom calculates the corresponding theoretical spectral of sample of model description.In RCWA methods, parameter vector v=(TCD, BCD, HT,
Thickness)TDetermine the periodic distribution rule of whole space material (n, k), the dielectric system of material is represented with e (x, y, z)
Several distributions.By taking two-dimensional grating as an example, it is located in (x, z) plane, the dielectric constant of material is in cyclically-varying, and in y directions
Uniformly.It is layered in z directions, every layer of material is expressed as εm(x,zm), m be z directions the number of plies, m=1,2 ..., M.Then εm
(x,zm) have periodically in x directions, therefore formula 4 can be used）Shown Fourier (Fourier) series description：
Wherein d is the cycle.At m layers, (2N+1) order harmonics, such as equation 5 below can be expanded into for TE mould electric fields E）
It is shown：
Wherein, TE moulds refer in the waveguide, and the longitudinal component of electric field is zero, and the propagation that the longitudinal component in magnetic field is not zero
Pattern.knx,knzIn x, the component of z, i.e., each layer of electric field (2N+1) rank scattering optical electric field can be used to describe for wave vector.If exhibition
Open that exponent number (2N+1) is sufficiently large, then the precision of E can be guaranteed.Snm(z) the amplitude size of optical electric field is scattered for each rank.For
Each layer can obtain one group of Differential Equation with Constant Coefficients group using wave equation and the boundary condition of each layer.Finally by solution square
The problem of battle array characteristic value and feature vector, can solve the E of each layerm, and finally obtain scattering light at different levels.According to dissipating for solution
Light is penetrated, the description form of the measure spectrum of OCD measuring apparatus output can be converted to.
Wherein, exponent number (2N+1) is unfolded influences the computational accuracy of theoretical spectral.N is bigger, and the error that it is calculated is smaller, but counts
Calculation amount also increases.Under certain precision conditions, the size of required N is related with the structure and material of grating.What RCWA methods calculated
Complexity is O (N3), it is directly proportional to number of plies M, it is proportional to the cube of exponent number N.If three-dimensional structure, complexity higher, is O
((Nx*Ny)3), with x directions exponent number NxWith y directions exponent number NyProduct cube it is proportional.Therefore, usually by high-performance calculation
The calculating for the theoretical spectral that the calculation server of processor composition specialty or evaluation work station complete to need in OCD measurements is appointed
Business.
Optimal expansion exponent number determining device 302, for the spectrum acquired in the different expansion exponent numbers in each wavelength points
Data carry out Convergence analysis, to determine the optimal expansion exponent number for meeting predetermined convergence condition of the wavelength points.
Specifically, optimal expansion exponent number determining device 302 can include the first optimal expansion exponent number determining module（In figure not
Show）, for carrying out Convergence analysis to the spectroscopic data acquired in the different expansion exponent numbers in each wavelength points, when continuous predetermined
The difference of the corresponding spectroscopic data of the expansion exponent number reference spectra data corresponding with benchmark exponent number of number will less than predetermined accuracy
Ask, it is determined that an expansion exponent number in the expansion exponent number of the continuous predetermined number is the optimal expansion exponent number of the wavelength points.
Wherein, the benchmark exponent number can rule of thumb be set, it is preferable that the benchmark exponent number of two dimension sample to be measured can
To be arranged to 99 to 961, the benchmark exponent number of three-dimensional sample to be measured could be provided as 21 to 31.The reference spectra data are base
The corresponding spectroscopic data of quasi- exponent number.
Wherein, the predetermined accuracy requires related with OCD measuring apparatus, can pass through the noise analysis of OCD measuring apparatus
Middle acquisition.
It should be noted that above example and benchmark exponent number numerical value be only for example it is used, not to any limit of the present invention
System, the teaching that those skilled in the art should be able to be as described above, based on actual conditions, to determine or select suitable benchmark rank
Number numerical value.
Illustrate to carry out the spectroscopic datas acquired in the different expansion exponent numbers in each wavelength points with an example below
Convergence analysis, when the corresponding spectroscopic data of the expansion exponent number reference spectra data corresponding with benchmark exponent number of continuous predetermined number
Difference is less than predetermined accuracy requirement, it is determined that an expansion exponent number in the expansion exponent number of the continuous predetermined number is the wavelength
The mode of the optimal expansion exponent number of point.
Example 5
To the expansion exponent number N of two dimension sample to be measured, only analysis x-axis directionx=N.To three-dimensional sample to be measured, x-axis is set
The expansion exponent number in direction is equal with the expansion exponent number in y-axis direction, i.e. the expansion exponent number N in x-axis directionx=Ny=N.It is to be measured with two dimension
Exemplified by sample, the spectroscopic data acquired in different expansion exponent numbers in each wavelength points carries out Convergence analysis, compares Data Convergence
Situation, if the corresponding spectra values of continuous three expansion exponent numbers N, N+2, N+4 benchmark oval thickness spectrum corresponding with benchmark exponent number
Data αNAnd βNDifference is both less than predetermined accuracy and requires ε, such as equation 6 below）It is shown：
Then this expansion exponent number N is the minimum value for meeting the requirement of Data Convergence predetermined accuracy in this wavelength points, i.e., most
Excellent expansion exponent number.
Or by taking sample to be measured shown in Fig. 2 as an example, the scope for setting the expansion exponent number to be analyzed is 1 to 41, expansion
Exponent number is at intervals of 2, and it is 0.005 that predetermined accuracy, which requires ε, and benchmark exponent number is 99, wavelength points 500nm, using above-mentioned Convergence analysis
Method calculates all wavelengths point, obtains one group of expansion exponent number list corresponding with each wavelength points, as shown in table 5.
Table 5：
Wherein, when it is 9 that exponent number, which is unfolded,
Then wavelength points are in 500nm, and expansion exponent number is the minimum value that meets the requirement of Data Convergence predetermined accuracy for 9, i.e.,
Optimal expansion exponent number.
Specifically, optimal expansion exponent number determining device 302 can also include the second optimal expansion exponent number determining module（In figure
It is not shown）, for carrying out Convergence analysis to the spectroscopic data acquired in the different expansion exponent numbers in each wavelength points, when an exhibition
Open before the corresponding spectroscopic data of exponent number and the expansion exponent number after the expansion exponent number and/or the expansion exponent number of the first predetermined number
Difference between each in spectroscopic data corresponding to the expansion exponent number of second predetermined number is respectively less than predetermined accuracy requirement, then
Determine the optimal expansion exponent number that the expansion exponent number is the wavelength points.
Illustrate to carry out the spectroscopic datas acquired in the different expansion exponent numbers in each wavelength points with an example below
Convergence analysis, the expansion exponent number of the corresponding spectroscopic data of exponent number and the first predetermined number before the expansion exponent number is unfolded when one
And/or the difference between each in the spectroscopic data after the expansion exponent number corresponding to the expansion exponent number of the second predetermined number is equal
Less than predetermined accuracy requirement, it is determined that the expansion exponent number is the mode of the optimal expansion exponent number of the wavelength points.
Example 6
To carry out Convergence analysis to the spectroscopic data acquired in the different expansion exponent numbers in each wavelength points, when an expansion
Between each in the spectroscopic data corresponding to two expansion exponent numbers after the corresponding spectroscopic data of exponent number and the expansion exponent number
Difference be respectively less than predetermined accuracy requirement, it is determined that the expansion exponent number for the wavelength points optimal expansion exponent number mode exemplified by,
Convergence analysis method is as follows：
|αN-αN+2|<ε, | αN-αN+4|<ε,
|βN-βN+2|<ε, | βN-βN+4|<ε。
Exponent number N is then unfolded to meet the minimum value of Data Convergence predetermined accuracy requirement, i.e., optimal exhibition in this wavelength points
Open exponent number.
Theoretical spectral Database updating device 303, for according to the optical critical dimension model in each wavelength points
The corresponding spectroscopic data of optimal expansion exponent number, to establish or update the theoretical spectral data of the optical critical dimension model
Storehouse.
Specifically, theoretical spectral Database updating device 303 can include：First acquisition module（Do not show in figure
Go out）, for obtaining optimal expansion exponent number corresponding spectroscopic data of the OCD models in each wavelength points；Theoretical spectral database
Establish update module（Not shown in figure）, for corresponding in the optimal expansion exponent number of each wavelength points according to the OCD models
Spectroscopic data, to establish or update the theoretical spectral database of the optical critical dimension model.
Preferably, the first acquisition module can include the first acquisition submodule（Not shown in figure）, for based on floating institute
The model parameter of optical critical dimension model is stated, to calculate the optimal exhibition for obtaining the optical critical dimension model in each wavelength points
Open the corresponding spectroscopic data of exponent number.Wherein, domain of walker can according to possibility offset of the sample to be measured in relation to technique come into
Row is set.
Preferably, the first acquisition module can also include the second acquisition submodule（Not shown in figure）, for based on not floating
Move the optical critical dimension model model parameter, come calculate obtain the optical critical dimension model in each wavelength points most
It is excellent that the corresponding spectroscopic data of exponent number is unfolded.
It should be noted that the present invention can be carried out in the assembly of software and/or software and hardware, for example, this hair
Bright each device can use application-specific integrated circuit（ASIC）Or any other is realized similar to hardware device.In one embodiment
In, software program of the invention can be performed by processor to realize steps described above or function.Similarly, it is of the invention
Software program（Including relevant data structure）It can be stored in computer readable recording medium storing program for performing, for example, RAM memory,
Magnetically or optically driver or floppy disc and similar devices.In addition, some steps or function of the present invention can employ hardware to realize, example
Such as, as coordinating with processor so as to performing the circuit of each step or function.
It is obvious to a person skilled in the art that the invention is not restricted to the details of above-mentioned one exemplary embodiment, Er Qie
In the case of without departing substantially from spirit or essential attributes of the invention, the present invention can be realized in other specific forms.Therefore, no matter
From the point of view of which point, the present embodiments are to be considered as illustrative and not restrictive, and the scope of the present invention is by appended power
Profit requires rather than described above limits, it is intended that all in the implication and scope of the equivalency of claim by falling
Change is included in the present invention.Any reference numeral in claim should not be considered as to the involved claim of limitation.This
Outside, it is clear that one word of " comprising " is not excluded for other units or step, and odd number is not excluded for plural number.That is stated in system claims is multiple
Unit or device can also be realized by a unit or device by software or hardware.The first, the second grade word is used for table
Show title, and be not offered as any specific order.
Claims (12)
1. a kind of be used to improve the method that theoretical spectral database creates speed, wherein, including：
- obtain different Fourier expansion exponent numbers corresponding spectroscopic data of the optical critical dimension model in each wavelength points；
- Convergence analysis is carried out to the spectroscopic data acquired in the different expansion exponent numbers in each wavelength points, to determine the wavelength points
The optimal expansion exponent number for meeting predetermined convergence condition；
- according to the optical critical dimension model each wavelength points the corresponding spectroscopic data of optimal expansion exponent number, establishing or
Update the theoretical spectral database of the optical critical dimension model.
2. according to the method described in claim 1, wherein, the difference for obtaining optical critical dimension model in each wavelength points
The step of expansion exponent number corresponding spectroscopic data, includes：
- the model parameter based on the optical critical dimension model that do not float, come calculate obtain the optical critical dimension model exist
The corresponding spectroscopic data of different expansion exponent numbers of each wavelength points.
3. method according to claim 1 or 2, wherein, acquired in the different expansion exponent numbers in each wavelength points
Spectroscopic data carry out Convergence analysis and include the step of determining the optimal expansion exponent number of the wavelength points：
- Convergence analysis is carried out to the spectroscopic data acquired in the different expansion exponent numbers in each wavelength points, when continuous predetermined number
The differences of the corresponding spectroscopic data of expansion exponent number reference spectra data corresponding with benchmark exponent number required less than predetermined accuracy, then
Determine the optimal expansion exponent number that an expansion exponent number in the expansion exponent number of the continuous predetermined number is the wavelength points.
4. method according to claim 1 or 2, wherein, acquired in the different expansion exponent numbers in each wavelength points
Spectroscopic data carry out Convergence analysis and include the step of determining the optimal expansion exponent number of the wavelength points：
- Convergence analysis is carried out to the spectroscopic data acquired in the different expansion exponent numbers in each wavelength points, exponent number is unfolded when one
Corresponding spectroscopic data and after the expansion exponent number and/or the expansion exponent number of the first predetermined number before the expansion exponent number second pre-
Difference between each in spectroscopic data corresponding to fixed number purpose expansion exponent number is respectively less than predetermined accuracy requirement, it is determined that should
The optimal expansion exponent number that exponent number is the wavelength points is unfolded.
5. method according to claim 1 or 2, wherein, it is described according to the optical critical dimension model in each wavelength points
The corresponding spectroscopic data of optimal expansion exponent number establish or the step of updating optimal expansion exponent number corresponding spectroscopic data includes：
- obtain optimal expansion exponent number corresponding spectroscopic data of the optical critical dimension model in each wavelength points；
- according to described optical critical dimension model the optimal expansion exponent number of each wavelength points spectroscopic data, to establish
Or the theoretical spectral database of the renewal optical critical dimension model.
6. according to the method described in claim 5, wherein, the acquisition optical critical dimension model is in the optimal of each wavelength points
The step of expansion exponent number corresponding spectroscopic data, includes：
- the model parameter based on the optical critical dimension model that floats, the optical critical dimension model is obtained every to calculate
The corresponding spectroscopic data of optimal expansion exponent number of a wavelength points.
7. a kind of be used to improve the device that theoretical spectral database creates speed, wherein, including：
Theoretical spectral data acquisition facility, for obtaining different Fourier expansions of the optical critical dimension model in each wavelength points
The corresponding theoretical spectral data of exponent number；
Optimal expansion exponent number determining device, for being carried out to the spectroscopic data acquired in the different expansion exponent numbers in each wavelength points
Convergence analysis, to determine the optimal expansion exponent number for meeting predetermined convergence condition of the wavelength points；
Theoretical spectral Database updating device, for according to the optical critical dimension model each wavelength points optimal exhibition
The corresponding spectroscopic data of exponent number is opened, to establish or update the theoretical spectral database of the optical critical dimension model.
8. device according to claim 7, wherein, the spectroscopic data acquisition device includes：
First spectroscopic data computing module, for the model parameter based on the optical critical dimension model that do not float, to calculate
Obtain different expansion exponent numbers corresponding spectroscopic data of the optical critical dimension model in each wavelength points.
9. the device according to claim 7 or 8, wherein, the optimal expansion exponent number determining device includes：
First optimal expansion exponent number determining module, for the spectroscopic data acquired in the different expansion exponent numbers in each wavelength points
Convergence analysis is carried out, when the corresponding spectroscopic data of the expansion exponent number reference spectra number corresponding with benchmark exponent number of continuous predetermined number
According to difference less than predetermined accuracy requirement, it is determined that an expansion exponent number in the expansion exponent number of the continuous predetermined number is should
The optimal expansion exponent number of wavelength points.
10. the device according to claim 7 or 8, wherein, the optimal exponent number determining device includes：
Second optimal expansion exponent number determining module, for the spectroscopic data acquired in the different expansion exponent numbers in each wavelength points
Convergence analysis is carried out, the expansion rank of the corresponding spectroscopic data of exponent number and the first predetermined number before the expansion exponent number is unfolded when one
Difference between each in spectroscopic data after number and/or the expansion exponent number corresponding to the expansion exponent number of the second predetermined number
Respectively less than predetermined accuracy requirement, it is determined that the expansion exponent number is the optimal expansion exponent number of the wavelength points.
11. the device according to claim 7 or 8, wherein, the theoretical spectral Database updating device includes：
First acquisition module, for obtaining optimal expansion exponent number corresponding light of the optical critical dimension model in each wavelength points
Modal data；
Theoretical spectral Database update module, for according to described optical critical dimension model in each wavelength points
The spectroscopic data of optimal expansion exponent number, to establish or update the theoretical spectral database of the optical critical dimension model.
12. according to the devices described in claim 11, wherein, first acquisition module includes：
First acquisition submodule, for the model parameter based on the optical critical dimension model that floats, the light is obtained to calculate
Learn optimal expansion exponent number corresponding spectroscopic data of the critical size model in each wavelength points.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201310505175.3A CN104572742B (en) | 2013-10-23 | 2013-10-23 | It is a kind of to be used to improve the method and device that theoretical spectral database creates speed |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201310505175.3A CN104572742B (en) | 2013-10-23 | 2013-10-23 | It is a kind of to be used to improve the method and device that theoretical spectral database creates speed |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN104572742A CN104572742A (en) | 2015-04-29 |
| CN104572742B true CN104572742B (en) | 2018-05-01 |
Family
ID=53088825
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201310505175.3A Active CN104572742B (en) | 2013-10-23 | 2013-10-23 | It is a kind of to be used to improve the method and device that theoretical spectral database creates speed |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN104572742B (en) |
Families Citing this family (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112729108B (en) * | 2020-12-18 | 2022-12-06 | 长江存储科技有限责任公司 | Calibration method of optical critical dimension OCD measuring equipment |
| CN113343182B (en) * | 2021-06-30 | 2024-04-02 | 上海精测半导体技术有限公司 | Optimization method and system of theoretical spectrum data, electronic equipment and measurement method |
| CN113343492B (en) * | 2021-06-30 | 2023-07-18 | 上海精测半导体技术有限公司 | Optimization method, system and optical measurement method of theoretical spectrum data |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101271268A (en) * | 2007-03-21 | 2008-09-24 | 台湾积体电路制造股份有限公司 | Method and system for optimizing lithography focus and/or energy using a specially-designed optical critical dimension pattern |
| CN102183212A (en) * | 2010-12-28 | 2011-09-14 | 睿励科学仪器（上海）有限公司 | Method and device for rapidly determining topography parameters of microperiodic structure |
| CN102750333A (en) * | 2012-05-31 | 2012-10-24 | 华中科技大学 | Method for extracting semiconductor nano-structure feature size |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US7580129B2 (en) * | 2007-02-23 | 2009-08-25 | Taiwan Semiconductor Manufacturing Company | Method and system for improving accuracy of critical dimension metrology |
-
2013
- 2013-10-23 CN CN201310505175.3A patent/CN104572742B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101271268A (en) * | 2007-03-21 | 2008-09-24 | 台湾积体电路制造股份有限公司 | Method and system for optimizing lithography focus and/or energy using a specially-designed optical critical dimension pattern |
| CN102183212A (en) * | 2010-12-28 | 2011-09-14 | 睿励科学仪器（上海）有限公司 | Method and device for rapidly determining topography parameters of microperiodic structure |
| CN102750333A (en) * | 2012-05-31 | 2012-10-24 | 华中科技大学 | Method for extracting semiconductor nano-structure feature size |
Also Published As
| Publication number | Publication date |
|---|---|
| CN104572742A (en) | 2015-04-29 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Lu et al. | Performance prediction for silicon photonics integrated circuits with layout-dependent correlated manufacturing variability | |
| KR101342847B1 (en) | 2 optimization of diffraction order selection for two-dimensional structures | |
| CN101413791B (en) | Determining profile parameters of a structure using approximation and fine diffraction models in optical metrology | |
| CN102183212B (en) | Method and device for rapidly determining topography parameters of microperiodic structure | |
| US11086288B2 (en) | Optimizing computational efficiency by multiple truncation of spatial harmonics | |
| JP2018527282A5 (en) | ||
| CN101331378B (en) | Selecting unit cell configuration for repeating structures in optical metrology | |
| CN103154664A (en) | Method of determining an asymmetric property of a structure | |
| CN104572742B (en) | It is a kind of to be used to improve the method and device that theoretical spectral database creates speed | |
| CN104778181B (en) | A kind of method and its equipment measuring spectrum and library Spectral matching | |
| US20090063076A1 (en) | Determining profile parameters of a structure using approximation and fine diffraction models in optical metrology | |
| TWI401549B (en) | Method for designing two dimensional overlay array target and method and system for measureing overlay error using the same | |
| CN104750748B (en) | A kind of method and device creating speed for improving theoretical spectral database | |
| CN103559329B (en) | The measuring method of coarse nanostructured characterisitic parameter in optical scattering measurement | |
| Nishidate et al. | Ray-tracing method for isotropic inhomogeneous refractive-index media from arbitrary discrete input | |
| CN115406377B (en) | Method and device for establishing theoretical spectrum library, and method and device for measuring morphological parameters | |
| Gross et al. | Sensitivity analysis for indirect measurement in scatterometry and the reconstruction of periodic grating structures | |
| Madsen et al. | Replacing libraries in scatterometry | |
| Zhang et al. | Fitting-determined formulation of effective medium approximation for 3D trench structures in model-based infrared reflectrometry | |
| CN104679770B (en) | A kind of method and apparatus for the spectra database and the information that gets parms for generating sample | |
| Hammerschmidt et al. | Reduced basis method for Maxwell's equations with resonance phenomena | |
| Gross et al. | Optimal sets of measurement data for profile reconstruction in scatterometry | |
| Kleemann et al. | Fast online inverse scattering with Reduced Basis Method (RBM) for a 3D phase grating with specific line roughness | |
| Li et al. | Predicting the Critical Dimensions of Micron and Sub-micron Structures Using Joint Training Models and Electromagnetic Simulation Tools | |
| JP4040269B2 (en) | Two-dimensional electromagnetic field simulation method |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| PE01 | Entry into force of the registration of the contract for pledge of patent right | ||
| PE01 | Entry into force of the registration of the contract for pledge of patent right |
Denomination of invention: Method and device for increasing speed of creating theoretical spectral database Effective date of registration: 20190121 Granted publication date: 20180501 Pledgee: Shanghai Xingcheng Investment Management Co., Ltd. Pledgor: Raintree Scientific Instruments (Shanghai) Corporation Registration number: 2019310000002 |
|
| PC01 | Cancellation of the registration of the contract for pledge of patent right | ||
| PC01 | Cancellation of the registration of the contract for pledge of patent right |
Date of cancellation: 20191029 Granted publication date: 20180501 Pledgee: Shanghai Xingcheng Investment Management Co., Ltd. Pledgor: Ruili Scientific Instruments (Shanghai) Co., Ltd. Registration number: 2019310000002 |