1. Field of the Invention
The present invention relates to a method for analyzing a membrane structure and an apparatus therefor. More particularly, the invention relates to a method for analyzing a membrane structure by simultaneously fitting simulated operation data to measured data obtained with an optical system utilizing at least two sets of resolutions and dynamic ranges, and an apparatus therefor.
2. Description of the Related Art
In the X-ray reflectivity measurement, an interference phenomenon of X-rays reflected on interfaces among layers of a membrane is measured, and simulated operation data is fitted to the measurement results to analyze density, membrane thickness and roughness of the respective layers. The density of the outermost thin layer can be calculated from the total reflection critical angle, and the densities of the other layers can be calculated from the amplitudes of the interference fringes. The membrane thickness of the layers can be calculated from the frequencies of vibrations. The roughness can be calculated from the attenuation ratio of the total reflectivity measurement data and the attenuation of the amplitudes of the interference fringes on the higher angle side (as described for example in JP-A-2001-349849).
In the case where the X-ray reflectivity measurement is applied to a multi-layer thin membrane having a layer thickness of several hundred nanometers, it is necessary that the measurement is carried out with the divergence angle and the wavelength spread of the incident X-ray being set to sufficiently small values. For example, a perfect crystal monochromator of Si or Ge is used on the incident side. The perfect crystal monochromator suppresses the wavelength of the X-ray generated by a light source from being spread and abstracts only the parallel component from the X-ray, and therefore, the intensity of the X-ray thus irradiating the thin membrane is reduced by about from 1/10 to 1/100 in comparison to the case using no perfect crystal monochromator. In the case where a four-crystal optical system is used, in particular, the intensity of the incident X-ray is reduced to about 1/100, and there are some cases where a sufficient dynamic range cannot be obtained upon X-ray reflectivity measurement.
That is, in the case where the perfect crystal monochromator is used, the measurement cannot be attained to the higher angle side due to the small dynamic range, and there are some cases where the roughness cannot be sufficiently evaluated. In the case where a thin layer is present in the multi-layer membrane, furthermore, there are some cases where the fitting goes with neglecting the layer since the period of vibration corresponding to the layer thickness is small.
However, in the case where the fitting is again carried out with different initial values set to the parameters including density, thickness and roughness of the membrane, the fitting results in the profile shown in
It is understood from the comparison between the results shown in Table 2 and the results shown in Table 3 that the magnitude relation of the interface roughness is reversed.
Such a case that plural solutions are present is often encountered, and it is difficult to evaluate as to which is the correct solution or as to whether both are ascribed to insufficient analysis only by using the measurement results shown in
The invention has been made under the aforementioned circumstances, and an object thereof is to provide a method for analyzing a membrane structure by fitting simulated operation data to measured data obtained by X-ray reflectivity measurement to analyze the membrane structure. The analysis result obtained by the fitting can be prevented from falling into a local solution, so as to obtain an analysis result of the membrane structure with high accuracy, and to provide an apparatus therefor.
The invention relates to, as a first aspect, a method for analyzing a membrane structure for analyzing a structure of a membrane specimen having a single layer membrane or a multi-layer membrane by an X-ray reflectivity measurement. The method containing a step of determining the membrane structure by simultaneously analyzing a plurality of pieces of measured data obtained by measuring the membrane specimen under a plurality of sets of measuring conditions which are different from each other in at least one of a resolution and a dynamic range.
The invention also relates to, as a second aspect, a method for analyzing a structure of a membrane containing of the following steps. Fitting simulated operation data obtained through simulation operation carried out by differentiating at least one parameter showing a physical property of a membrane specimen having a single layer membrane or a multi-layer membrane, to measured data obtained with an X-ray incident on the specimen at an angle in a vicinity of a critical angle with respect to a surface of the specimen. Obtaining, as optimum values, values of parameters providing a minimum difference between the measured data and the simulated operation data, so as to determine the structure of the membrane specimen, the simulated operation data being fitted simultaneously to first measured data obtained by irradiating the membrane specimen with an X-ray having a high resolution and second measured data obtained by irradiating the membrane specimen with an X-ray having a low resolution.
The invention further relates to, as a third aspect, an apparatus for analyzing a membrane structure for analyzing a structure of a membrane specimen having a single layer membrane or a multi-layer membrane by an X-ray reflectivity measurement, the apparatus containing the following. A means for measuring a plurality of pieces of data by measuring the membrane specimen under a plurality of sets of measuring conditions different from each other in at least one of a resolution and a dynamic range. A means for simultaneously analyzing the plurality of pieces of data measured by the measuring means, so as to determine the structure of the membrane specimen.
The invention further relates to, as a fourth aspect, an apparatus for analyzing a structure of a membrane containing the following. A means for fitting simulated operation data obtained through simulation operation carried out by differentiating at least one parameter showing a physical property of a membrane specimen having a single layer membrane or a multi-layer membrane, to measured data obtained with an X-ray incident on the specimen at an angle in a vicinity of a critical angle with respect to a surface of the specimen. A means for obtaining, as optimum values, values of parameters providing a minimum difference between the measured data and the simulated operation data, so as to determine a structure of the membrane specimen, the simulated operation data being fitted simultaneously to first measured data obtained by irradiating the membrane specimen with an X-ray having a high resolution and second measured data obtained by irradiating the membrane specimen with an X-ray having a low resolution.
According to the method for analyzing a membrane structure and the apparatus of the invention, the simulated operation data is fitted to the measured data obtained by X-ray reflectivity measurement to analyze the membrane structure, in which the analysis result obtained by the fitting can be prevented from falling into a local solution, so as to obtain an analysis result of the membrane structure with high accuracy.
a is a schematic diagram of the X-ray reflectivity measuring device in which no perfect crystal monochromator is provided.
b is a schematic diagram of the X-ray reflectivity measuring device in which a perfect crystal monochromator is provided.
The method for analyzing a membrane structure according to the invention is a method for analyzing a membrane structure for analyzing a structure of a membrane specimen having a single layer membrane or a multi-layer membrane by an X-ray reflectivity measurement, and a plurality of pieces of measured data, which are obtained by measuring the membrane specimen under a plurality of sets of measuring conditions which are different from each other in at least one of a resolution and a dynamic range, are simultaneously analyzed. Specifically, simulated operation data, which is obtained through simulation operation carried out by differentiating at least one parameter showing a physical property of a membrane specimen having a single layer membrane or a multi-layer membrane, is fitted to measured data, which is obtained with an X-ray incident on the specimen at an angle in a vicinity of a critical angle with respect to a surface of the specimen, so as to determine the structure of the membrane specimen.
As a procedure for the fitting, for example, analysis by the method of least squares may be employed, in which an X-ray reflectivity is calculated according to a parameter, which is a factor for obtaining the reflectivity, being incrementally adjusted, for minimizing the residual sum of squares to the actual reflectivity data. Accordingly, one set of parameters that is most fitting to the measured data can be obtained.
In the analysis by the method of least squares, a problem arises because the residual sum of squares is simply minimized based on the order and method of fitting or adequacy and inadequacy of the initial value. Accordingly, the fitting often fails to complete due to divergence. It has been known that in the ordinary analysis, the solution is converged by carrying out the fitting of simulated operation data to measured data of X-ray reflectivity in the following order.
1. A layer structure model of thin layers including a substrate is formed.
2. With respect to the measured data and the simulated operation data, the maximum intensities are conformed at an angle lower than the critical angle.
3. The background intensity of the model is conformed to the intensity of the measured data.
4. A parameter to be fixed is selected from the parameters of the model. In many cases, the substrate is known, and therefore, the density thereof is fixed. In the case where other parameters, such as thickness and density, are known, they are determined as constants.
5. The parameters are manually determined in such a manner that the measured data is visually conformed to the simulated operation data.
6. Automatic fitting is carried out.
7. A value of the parameter providing a minimum difference between the measured data and the simulated operation data is obtained as an optimum value. At this time, the residual of the minimum sum of squares is in the second decimal place or less (in an order of 10−2 or less), it is determined that the fitting is completed with substantial accuracy.
In the method for analyzing a membrane structure according to the invention, upon measurement, the simulated operation data is fitted simultaneously to the first measured data obtained by irradiating the membrane specimen with an X-ray having a high resolution and the second measured data obtained by irradiating the membrane specimen with an X-ray having a low resolution, so as to optimize the parameters.
For example, the first measured data is obtained by making an X-ray incident on the membrane specimen through a perfect crystal monochromator, and the second measured data is obtained by making an X-ray incident on the membrane specimen through no perfect crystal monochromator. The simulated operation data is then fitted simultaneously to both the first and second measured data, so as to optimize the parameters. In the case where the measurement is carried out with a perfect crystal monochromator provided on the incident side, measured data with a high resolution and a narrow dynamic range is obtained, and in the case where the measurement is carried out with no perfect crystal monochromator provided on the incident side, measured data with a low resolution and a broad dynamic range is obtained.
It is understood from the aforementioned analysis that an SiO2 oxide layer is formed, which has not appeared in the case where the fitting is carried out by using only the measured data obtained with a perfect crystal monochromator provided on the incident side. Furthermore, the accurate solution is obtained for the roughness without falling into a local solution.
As having been described, the roughness and the thickness of the membrane can be simultaneously analyzed by fitting two pieces of data with conditions different from each other in resolution and dynamic range. The number of pieces of data with conditions different from each other in resolution and dynamic range may be two or more. The similar effect as differentiating the resolution and the dynamic range can be expected, for example, by selecting plural wavelengths for simultaneous analysis of data.
The apparatus for analyzing a membrane structure according to the invention has a constitution for practicing the aforementioned method for analyzing a membrane structure. Specifically, the apparatus for analyzing a membrane structure according to the invention has, as a basic constitution, an X-ray reflectivity measuring device for practicing X-ray reflectivity measurement, and contains a means for measuring a plurality of pieces of data by measuring one membrane specimen under a plurality of sets of measuring conditions different from each other in at least one of a resolution and a dynamic range, and a means for simultaneously analyzing the plurality of pieces of data measured by the measuring means, so as to determine a structure of the membrane specimen.
More specifically, the apparatus for analyzing a membrane structure according to the invention may have such a constitution that contains an ordinary apparatus for measuring X-ray reflectivity equipped, in which a perfect crystal monochromator is movably provided on the X-ray incident side. Upon measurement, an X-ray is incident on a membrane specimen through the perfect crystal monochromator to obtain the first measured data, and then after moving the perfect crystal monochromator, an X-ray is incident on a membrane specimen through no perfect crystal monochromator to obtain the second measured data. The first measured data and the second measured data thus obtained are loaded to the means for determining the membrane structure, such as a computer, and subjected to the method for analyzing a membrane structure having been described in detail, so as to calculate optimum parameters showing the membrane structure.
The invention has been described with reference to the aforementioned examples, but the invention is not construed as being limited to them, and various changes and modifications in detail may be made therein without departing from the spirit and scope thereof.
According to the invention, in the method for analyzing a membrane structure by fitting simulated operation data to measured data obtained by X-ray reflectivity measurement to analyze the membrane structure, the analysis result obtained by the fitting can be prevented from falling into a local solution, so as to obtain an analysis result of the membrane structure with high accuracy.
There are some cases in the conventional techniques that the dynamic range is narrowed as a result of improvement in resolution to fail to obtain information sufficient for analyzing a membrane structure, and the parameters used in fitting are converged to a local solution. In the invention, measured data with a high resolution and a narrow dynamic range and measured data with a low resolution and a broad dynamic range are used to enable fitting to such data that reflects the overall structure of the membrane specimen, whereby the aforementioned problem of local solution is resolved. In other words, according to the invention, parameters showing a structure can be obtained with high accuracy even in the case of a membrane specimen having a complex layer structure.
The X-ray reflectivity measurement is being widely used for evaluation of functions and characteristics of various functional thin membrane materials, and it is considered that the method for analyzing a membrane structure and the apparatus therefor according to the invention realizing structure analysis with high accuracy are considerably useful, and practical applications thereof are strongly expected.
Number | Date | Country | Kind |
---|---|---|---|
2003-347224 | Oct 2003 | JP | national |
Number | Name | Date | Kind |
---|---|---|---|
6025596 | Shirai et al. | Feb 2000 | A |
6907107 | Wallis et al. | Jun 2005 | B1 |
20020150208 | Yokhin et al. | Oct 2002 | A1 |
Number | Date | Country |
---|---|---|
9-105726 | Apr 1997 | JP |
Number | Date | Country | |
---|---|---|---|
20050102110 A1 | May 2005 | US |