JPH05322807A - Method for identifying substance by x-ray diffraction - Google Patents

Abstract

PURPOSE:To easily identify a sample to be inspected as a whole without specifying individual pure substances contained in the sample. CONSTITUTION:A diffraction profile data base is prepared by measuring the X-ray diffraction patterns of soil samples collected from every place in Japan. After measuring the diffraction pattern of a soil sample, the standard profile which is most coincident with the measured diffraction pattern is decided by finding the degree of coincident between the diffraction file of the sample soil with the standard profile. In the first stage, the number of standard profiles to be compared with the measured diffraction pattern is reduced to about 500 from the numerous standard profiles stored in the data base by performing primary retrieval. Thereafter, the measured profile is compared with the selected standard profiles by performing secondary retrieval while the selected standard profiles are successively displayed in the order of the degree of coincidence. Then the degree of coincidence is further confirmed by shifting the standard profile selected by the secondary retrieval by a small angle. Thus the locality of the sample soil is specified.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a method for identifying substances by X-ray diffraction, and more particularly to a method for identifying substances in which a plurality of crystals and non-crystals are mixed in a state where they are compared with each other.

[0002]

2. Description of the Related Art Identification of a substance by X-ray diffraction is usually aimed at determining the crystal phase constituting the substance.
This is performed by a method of sequentially comparing a known single phase (pure substance) diffraction pattern and the diffraction pattern of an unknown sample.
This method is very effective when all the crystal phases of a substance can be identified, and it is widely used because it is easy to measure and operate. However, identification becomes difficult when the standard single-phase diffraction pattern contains an unknown component or when an amorphous substance is contained.

In order to identify a substance, in addition to the determination of the individual component phases as described above, it may be necessary to collate and compare as it is in the state where various crystalline phases and amorphous phases are mixed. is there. Even if the component phase of the substance and its mixing ratio are the same,
Depending on the crystallographic state (crystallinity, lattice defects, crystallite size, shape, etc.) of the component phase, the aggregated state of particles, the mixed state, and the mixed state of non-crystals, its physical properties and mechanical properties Different properties and chemical properties. Identification of such a mixed substance state as a whole substance is also practically useful. For example, the identification of the place of origin of the soil and the determination of the trade name of the paint are examples.

In the case of identifying such a mixed substance, conventionally, the mixed substance is measured by measuring an X-ray diffraction pattern of the mixed substance and then comparing this diffraction pattern with a standard diffraction pattern of a pure substance. A method was taken to determine what was included. Specifically, by comparing the diffraction pattern of the test substance with the diffraction peak position and relative intensity of the pure substance described in the JCPDS card, what pure substance is contained in the test substance Had decided.

[0005]

In the above-mentioned conventional identification method, although it is possible to determine each pure substance contained in the test substance, it is not possible to determine what the test substance is. , It is very difficult without considerable knowledge and experience. For example, regarding the identification of the soil, it is possible to determine what is contained in the soil of the test substance, but to identify whether the test substance is the soil at the point A or the soil at the point B. Is very difficult.

An object of the present invention is to provide a simple identification method for identifying what a test substance as a whole is, without determining individual pure substances for the test substance.

[0007]

The identification method of the first invention comprises a first step of preparing a database storing the profiles of X-ray diffraction patterns of a large number of standard substances, and a test substance to be identified. , A second step of measuring the profile of the X-ray diffraction pattern, and a third step of searching a standard substance that best matches the profile of the test substance from a large number of profiles in the database. .. Here, the "profile" of the X-ray diffraction pattern means the data of the curve itself of the diffraction pattern. With the conventional JCPDS card,
For each pure substance, only the angular position of the characteristic diffraction peak and the relative intensity are listed. On the other hand, in the present invention, all the curves of the diffraction pattern, that is, the data of the diffraction intensity for an arbitrary angular position in the predetermined angular range are stored as profile data. Then, the test substance is identified by comparing the degree of coincidence of the profile itself between the test substance and the standard substance.

In the present invention, naturally, it is necessary to create a database of standard substances corresponding to the test substances in advance. For example, in order to identify the soil, it is necessary to measure the X-ray diffraction pattern for each point based on the samples of the soil in various places in Japan and store it as standard profile data. Then, when identifying the soil of the test substance, by measuring the diffraction pattern of the test substance, by comparing the profile of the diffraction pattern and the standard profile data, by searching the standard data that best matches the profile. , The soil of the test substance is identified at which point.

A second invention is a more specific version of the third step of the first invention, wherein the third step is selected by a primary search that does not include profile comparison and by the primary search. The secondary search is performed to compare the profiles of the standard substances. Comparing the degree of coincidence of the profiles with each other is actually handled by a computer, but this process is more time consuming than just comparing the angular positions of the peaks. Therefore, it takes a lot of time to compare all the standard profiles contained in the database with the profile of the test substance to find the standard profile with the highest degree of agreement. The more standard profiles the database contains, the longer it will take. Therefore, it is more efficient to reduce the number of standard profiles in advance before comparing the profiles. In the above-mentioned primary search, such narrowing is performed. Factors used for narrowing down by the primary search include a method of comparing only the angular position of the diffraction peak, a method of comparing the diffraction peak and the diffraction intensity, and a method of narrowing down depending on the type of the test substance (for example,
Whether it is soil data, paper data, etc.),
A method of narrowing down by various keywords (a method of narrowing down by this search keyword by adding various search keywords to each of the standard profile data.
The color, hardness, etc. of the test substance are given as keywords. ) And so on. By such a primary search, standard profile data is narrowed down to, for example, about 500 items, and then a secondary search is performed by profile comparison.

In a third aspect of the present invention, after the secondary search of the second aspect, the profile of the standard substance selected in the secondary search and the profile of the test substance are relatively shifted to determine the degree of coincidence of the profiles. A shift search for comparison is carried out. Even if the X-ray diffraction pattern is measured for exactly the same sample, the angular position of the diffraction peak may slightly shift for each measurement. For example, the state when the sample is attached to the sample table may be slightly different for each measurement, which may shift the diffraction pattern as the measurement result. Therefore, by correcting such a shift, the degree of coincidence of the profiles may be increased, which enables more accurate identification.

[0011]

In the present invention, the substance is identified by directly comparing the profiles of the diffraction patterns. Therefore, not only the angular position and intensity of the diffraction peak but also the shape including all the sharpness and asymmetry of the peak are included. Then, the test substance and the standard substance can be compared. In this way, by comparing the degrees of coincidence of all the profiles with each other, it is possible to identify the entire mixed substance without determining individual pure substances.

[0012]

FIG. 1 is a flow chart of an embodiment of the present invention, and an example of soil identification will be described with reference to this flow chart. First, soil samples from all over Japan are collected, their X-ray diffraction patterns are measured, and a database of soil diffraction profiles is created. That is, the standard profile is registered for each soil sample. The method of registering the standard profile will be described below. First, set the measurement conditions. The measurement conditions include the type of X-ray tube used, the radius of the goniometer of the X-ray diffractometer, the slit width of the divergence slit, the sampling width of the diffraction angle (2θ), and the measurement range of the diffraction angle. In order to compare the degree of coincidence of profiles, it is desirable to make these measurement conditions as uniform as possible. Also, at this point, various search keywords can be input. This search keyword is used in the primary search described later, and the color and hardness of soil can be input.

Next, the X-ray diffraction pattern of the sample to be registered is measured by the X-ray diffractometer. The obtained diffraction pattern is subjected to predetermined data processing before registering the profile. In this embodiment, as data processing, smoothing, background removal, and peak search are performed. In the peak search, the diffraction angle position and the diffraction intensity of 50 peaks at the maximum are obtained. The data obtained by this peak search is used in the primary search described later. After performing the above data processing, profile data (that is, diffraction intensity data at each sampling angle position) and peak search data are registered in the database as standard data. The process from measurement to registration is automatically performed by computer control.

As described above, the diffraction profile is successively registered for each soil sample, for example, about 1
Create a million soil database.

Next, a method for identifying a test substance using this database will be described. First, the measurement conditions are set. It is desirable that the measurement conditions be the same as when the standard profile was measured. Next, the X-ray diffraction pattern of the test substance is measured, and predetermined data processing is performed. The content of the data processing at this time is the same as that in the case of registration of the standard profile.

Next, a primary search is carried out to narrow down the standard profiles for which profile comparison should be performed with the test substance. Comparing profiles requires a considerable amount of processing on the computer, so if the number of standard profiles included in the database is large,
It is efficient to narrow down the number of standard profiles in advance. In this embodiment, the standard search narrows down to about 500 standard profiles.

Keys for the primary search are roughly classified as follows.
There are those related to measurement conditions, those related to peak search data, and those related to search keywords. As a search key regarding the measurement conditions, the goniometer radius, the slit width of the divergence slit, the sampling width of the diffraction angle (2θ),
There is a diffraction angle measurement range. Only these matching standard data can be compared. Also,
As a search key for peak search data, there is a method of narrowing down only by the degree of coincidence of the angular position of the peak or a method of narrowing down by degree of coincidence of both the angular position of the peak and the intensity. The narrowing down by the search keyword is to use the search keyword input when the standard profile is registered. These various search keys are used alone or in combination to narrow down the number of standard profiles to about 500.

Next, a secondary search is carried out using the standard profile narrowed down by the above primary search. In this secondary search, the degree of coincidence between the diffraction profile of the test substance and the standard profile is compared, and the standard profile with the highest degree of coincidence is selected. In this embodiment, the index shown in the following formula 1 is used as the index showing the degree of coincidence of the profiles.

[0019]

[Equation 1]

Here, X (n _i ) represents the diffraction intensity at the angular position n _i of the profile X (n) of the test substance as shown in FIG. 2, and A (n _i ) is the standard profile A. It represents the diffraction intensity in the angular position n _i of (n).

The correlation coefficient C1 is an index representing the degree of similarity between the profiles of X and A. Therefore, even when the entire profiles are different from each other by a uniform intensity ratio, they are completely matched, and C1 = 1. The closer C1 is to 1, the higher the similarity of the profiles. As another calculation method of the correlation coefficient, the square root may be used instead of X (n _i ) and the square root may be used instead of A (n _i ) to calculate the equation (1). When the correlation coefficient calculated by this method is C2, this is a correlation coefficient using square root data. The degree of coincidence of C2 near the base of the profile has a greater effect than C1.

The R factor R1 is an index for observing the degree of coincidence in consideration of absolute strength, and if R1 is used, even if the profiles are completely similar, the result is that the profiles do not coincide. is there. Therefore, the degree of coincidence can be quantitatively determined. The smaller R1 is, the higher the degree of coincidence is. As another method for calculating the R factor, the square root instead of _{X (n i), A (} n i)
Equation (2) can be calculated using the square root instead of If this is R2, this will be an R factor using square root data. Also in R2, the degree of coincidence near the base of the profile has a greater effect than in R1. The secondary search can be executed by selecting any of these indexes C1, C2, R1, and R2.

In this embodiment, as a result of the above secondary search,
Up to 30 standard profiles can be displayed in descending order of the degree of agreement with the profile of the test substance. For example,
When the secondary search is executed using C1 as an index indicating the degree of coincidence, standard profile numbers and the like are displayed in descending order of C1 (in order of being closer to 1). Further, the profile of the test substance and the standard profile can be displayed in an overlapping manner on the screen. FIG. 3 shows an example of this superimposed profile. The solid line is the profile of the test substance, and the broken line is the standard profile with the highest degree of agreement.
The curve 10 above the superimposed profile represents the difference between the two profiles.

When performing the secondary search, it is possible to specify the entire range of effective diffraction angles, or it is possible to specify only a specific angle range.

Next, a shift search is carried out for the standard profile selected in the secondary search. This shift search is for the standard profile selected in the secondary search.
The degree of coincidence is evaluated in consideration of the angle shift.
In this shift search, first, the specific standard profile selected in the secondary search and one step and two steps before and after this are added.
A total of five profiles, which are profiles that are angle-shifted by steps, are used. One step is a sampling width (for example, 1/100 degree) when the profile is measured. For each of the above five profiles, the degree of agreement with the profile of the test substance is examined.
For example, the value of the above correlation coefficient C1 is obtained for the above five profiles. Then, as shown in FIG. 4, the horizontal axis represents the shift amount of the standard profile, and the vertical axis represents C1 at that time.
The values of are plotted and a parabolic approximation of these 5 points is performed. Then, the shift amount S corresponding to the maximum value of this parabola is obtained. That is, when the standard profile is angularly shifted by this shift amount S, the profile of the test substance and the standard profile should be more consistent. Therefore, the standard profile is set to this shift amount S
Only after the angle is shifted, the degree of agreement with the profile of the test substance is re-evaluated. In reality, the above-mentioned shift amount S is smaller than the sampling width, so that the diffraction intensity when the standard profile is shifted by the shift amount S is obtained by interpolating the diffraction intensity at the sampling positions before and after that.

In the shift search described above, the shift search may be performed uniformly over the entire angle range of the standard profile, or the shift search may be performed separately for each designated angle range. In general, the optimum shift amount on the low angle side and the optimum shift amount on the wide angle side are different, and therefore the degree of coincidence increases when a separate shift search is performed for each angle range.

When the shift search as described above is completed,
The standard profile numbers are displayed in descending order of the degree of agreement with the profile of the test substance. It should be noted that the identification can be finished at the stage of the secondary search.

[0028]

INDUSTRIAL APPLICABILITY According to the present invention, the profile of the test substance and the standard profile in the database are compared in the profile state to determine the degree of coincidence, so that each pure substance contained in the test substance is determined. It is possible to identify the entire mixed substance without specifying it. According to this method, the test substance can be easily identified without special experience or knowledge.

[Brief description of drawings]

FIG. 1 is a flowchart showing an embodiment of the present invention.

FIG. 2 is a graph illustrating a sampling position of a diffraction angle and a diffraction intensity corresponding to the sampling position.

FIG. 3 is a diffraction pattern in which the standard profile selected in the secondary search and the profile of the test substance are superimposed.

FIG. 4 is a graph illustrating an optimum shift amount in shift search.

Claims (3)

[Claims] 1. A first step of preparing a database that stores profiles of X-ray diffraction patterns of many standard substances, and a second step of measuring profiles of X-ray diffraction patterns of test substances to be identified. A method for identifying a substance by X-ray diffraction, comprising: a step; and a third step of searching for a standard substance that most closely matches the profile of the test substance from a large number of profiles in the database. 2. The third step comprises a primary search that does not include profile comparison, and a secondary search that performs profile comparison on the standard substance selected in the primary search. The identification method according to claim 1. 3. After the secondary search, a shift search is performed in which the profile of the standard substance selected in the secondary search and the profile of the test substance are relatively shifted to compare the degree of coincidence of the profiles. The identification method according to claim 2, which is characterized in that: