JP3331696B2 - X-ray quantitative analysis method for granular samples - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an X-ray microanalyzer that focuses an electron beam on a point on a sample, disperses the X-rays emitted from the sample, and analyzes an electron beam irradiation point on the sample. The present invention relates to a method for performing elemental quantification of a granular sample using a device for performing the method.

[0002]

2. Description of the Related Art Conventionally, a method called ZAF method has been used as an elemental quantitative analysis of a massive sample by X-rays.
This method obtains the first approximation of the concentration of the target element in simple proportion from the ratio of the characteristic X-ray intensity of the pure sample and the characteristic X-ray intensity of the actual sample of the element to be quantified. Assuming that the concentration of the target element is reduced, the attenuation of the excitation ray in the sample, the absorption of the characteristic X-rays by the other elements, the fluorescent X-rays of the target element by the X-rays emitted by the other elements, and the like are performed. The second approximation value is obtained, the same calculation is repeated thereafter, and the quantitative value of the target element is obtained by the successive approximation method. For the X-ray spectroscopic quantitative analysis of the granular sample by electron beam irradiation, the above-described ZAF method has been used conventionally, but good analysis results have been difficult to obtain.

[0003]

SUMMARY OF THE INVENTION It is an object of the present invention to provide a method for quantitative analysis by X-ray which can obtain good results for a granular sample.

[0004]

[Means for Solving the Problems] Observing the size of a granular real sample, selecting particles to be quantitatively analyzed, considering a spherical sample model having substantially the same diameter as that, and assuming the element composition ratio of the sample model. Performing a simulation calculation of the trajectory of the electrons in the sample when one electron accelerated at a predetermined acceleration voltage is incident on the sample model, and is radiated during the movement of the electrons in the sample model. Calculate the intensity of the characteristic X-ray of the component element considering the absorption until it goes out of the sphere of the sample model, perform the above calculation for many electrons, and calculate the component from the sum of the characteristic X-ray intensity of the component element. Determine the mutual intensity ratio of characteristic X-rays of the element, and on the other hand, measure the mutual intensity ratio of characteristic X-rays of the component elements of the sample when the selected actual sample is irradiated with the electron beam accelerated at the predetermined acceleration voltage, Calculated by the above calculation The iteration mutual intensity ratio of the characteristic X-ray of each component by modifying the composition ratio of the initial to match the measured values, the quantitative analysis results of the samples with a hypothetical composition ratio when match.

[0005]

The sphere S in FIG. 1 is a spherical model of a sample considered for a granular sample, and the radius is r. Arrow e indicates electrons incident on this model. The electron accelerating voltage is the accelerating voltage of the electron beam when the actual sample is analyzed, and this is the predetermined voltage. Electrons that enter the sample repeatedly collide with atoms in the sample, gradually lose energy while drawing a zigzag trajectory, and stop in the sample (absorb by the sample).
Or go out of the particle before that. Such a zigzag locus can be calculated by a simulation calculation using the Monte Carlo method if the composition of the component elements constituting the sample is known. In addition, the situation where the electrons collide with atoms in the sample during the movement in the sample and emit characteristic X-rays can also be obtained by the calculation of the simulation. Regarding the characteristic X-ray of each element generated in the sample, the absorption by other atoms until the X-ray goes out of the sample in the detection direction can be calculated by assuming the element composition. These calculation methods are known. Therefore, the above calculation is repeated many times, and the ratio of the sum of the calculated intensities of the characteristic X-rays of each element emitted each time is the intensity ratio of the characteristic X-rays of each component element which will be observed in the spherical model considered. If this matches the ratio of the characteristic X-ray intensities of the respective component elements actually measured for the actual sample, it means that the initially assumed composition ratio was correct. In practice, it is not always possible to make a correct assumption about the composition ratio from the beginning, but if the above calculation is repeated while correcting the assumption, a correct result can be achieved.

[0006]

DESCRIPTION OF THE PREFERRED EMBODIMENTS The method of the present invention is most suitably applied when the diameter of a sample is about 10 μm. The diameter of the sample particles is determined directly by observing the sample with a scanning electron microscope image. That is, the particles to be quantitatively analyzed are selected by observing a scanning electron microscope image of the sample, the particles are moved on the optical axis of the electron optical system, the electron beam is intensively irradiated, and the emitted X-rays are spectrally analyzed. You do it. When the particle size is about 10 μm, the irradiated electron beam hardly reaches the substrate, so there is no need to consider the influence of the substrate, and the application of the present invention is simplified. In performing the above-described simulation calculation, it is necessary to first assume the element composition ratio of the sample.
The closer this assumption is to the truth, the less the number of repetitions of the calculation. In this example, the proportional X-ray intensity of the characteristic X-ray intensity of each component element included in the granular sample to be quantitatively analyzed and the characteristic X-ray intensity of the component element of the standard sample having a shape that is not necessarily granular is proportional to the ratio. Determine the first assumed concentration of each component in the calculation.
Thereafter, the above-described calculation is performed.

FIG. 2 is a flowchart showing a procedure for executing the above-described calculation. Before starting the calculation, the following parameters are input to the computer. Elements constituting electrons, electron accelerating voltage (acceleration voltage used for actual measurement of the above-mentioned sample to be measured and standard sample), particle diameter (value obtained from a scanning electron microscope image), composition of component elements of particles Ratio (assumed value obtained from actual measurement of the sample to be measured and the standard sample described above), and others. In the computer, the cross-sectional area of elastic collision with electrons of various energies of atoms of each element, the cross-sectional area of inelastic collision (collision that emits characteristic X-rays), the absorptivity of each element for characteristic X-rays, Data such as the energy loss rate per unit traveling distance of electrons of various energies traveling inside the sample are stored. In order to perform the simulation, in this embodiment, electrons are made incident on a point P on the top of the sphere in FIG. 1 (a in the flowchart). The simulation is performed as follows. The electron path after scattering is determined by Monte Carlo method with respect to a position point A (this point is point P at the start of the calculation) of the sample sphere (FIG. 1) S before the electron scattering. That is, the scattering direction θ is determined by a random number, and since the traveling distance l is assumed to be the composition of the sample, the mean free path is calculated from the energy of the electron and is set to l (c). Since point B is thus determined (d), it is determined whether point B is inside or outside the sample sphere (e).
I do. When the determination is YES, that is, when the point B is inside the sphere, the amount of X-rays scattered at the point A and emitted in the process of colliding with other atoms at the point B is obtained (f). There are several simulation methods for calculating the amount of X-ray generation.
For example, there is a method of stochastically determining one as the element type of the atom at the point B, whether it is an elastic collision or an inelastic collision, from the element composition and the cross-sectional area of each collision of the atoms of each element. In this embodiment, assuming that continuous X-rays and characteristic X-rays of the respective elements are emitted during the process between electrons A and B, the continuous X-rays and characteristic X-rays of the respective constituent elements are proportionally distributed according to the above-mentioned probability. As a result, the X-ray intensity due to one collision is determined. The point at which this X-ray is generated is defined as the midpoint between A and B (g), and the absorption of the X-ray from that point until the X-ray traveling in the detection direction leaves the sample sphere is calculated (h). The energy consumption of the electrons in the collision at point B is calculated (R), and it is determined whether the remaining energy of the electrons is sufficient to generate characteristic X-rays (N). If the energy is sufficient (YES), the operation is (B). And the above operation is repeated with the point B considered as the point A. If this determination is NO, one simulation calculation is completed, and it is determined whether the simulation has been repeated a predetermined number of times (NO). If NO, the operation returns to step A. If YES, the simulation ends, and each component obtained by the previous calculation is calculated. The sum of the characteristic X-ray intensities of the elements is determined, and the mutual ratio of the intensities is calculated.

If the determination in step (e) is NO, that is, if the electrons scattered at point A go out of the sample sphere, the intersection R of the straight line AB with the sphere S is determined (e). As in the case of (f), the intensity of each X-ray generated between the distances AR is calculated (W), and the X-ray absorption in the observation direction is calculated using the midpoint R 'of the AR as the X-ray generation point (F). And the operation is (Le)
go to.

The intensity ratio of characteristic X-rays of each component element obtained by the simulation calculation of a predetermined number of times as described above is compared with the actually measured intensity ratio of the sample, and the component composition ratio initially assumed so that the two coincide. Is corrected, and the above-described simulation is performed again. Such operations are repeated, and the calculation is completed when the difference between the calculated X-ray intensity ratio and the actually measured intensity ratio falls within an allowable range, and the finally assumed composition ratio is used as a quantitative value.

[0010]

As described above, it is difficult to obtain a granular standard sample for a granular sample, and the method of making a calibration curve with the granular standard sample and quantifying it cannot be applied. The calibration curve differs depending on the type and composition ratio of other coexisting elements. In the case of a lump sample, the effect of the difference in the composition ratio of other coexisting elements is corrected by the ZAF method. However, in the case of a granular sample, the ZAF method does not give a good result. According to the present invention, a granular standard sample that is difficult to obtain is unnecessary, and a standard sample having a general shape that is not granular is sufficient as the standard sample. Can be obtained.

[Brief description of the drawings]

FIG. 1 is a diagram of an electron trajectory in a spherical sample model.

FIG. 2 is a flowchart showing a procedure for executing the method of the present invention.

[Explanation of symbols]

 S spherical sample model e incident electron

────────────────────────────────────────────────── ─── Continuation of front page (56) References JP-A-1-219550 (JP, A) JP-A-2-35345 (JP, A) JP-A-4-43944 (JP, A) JP-A-4-43 42037 (JP, A) JP-A-4-249741 (JP, A) JP-A-3-172744 (JP, A) JP-A-5-26826 (JP, A) (58) Fields investigated (Int. Cl. ⁷ , DB name) G01N 23/225

Claims (1)

(57) [Claims] 1. Observing the size of a granular actual sample, selecting particles to be quantitatively analyzed, considering a spherical sample model having substantially the same diameter as the sample, assuming the element composition ratio of the sample model, and performing a predetermined acceleration. By performing a simulation calculation of the trajectory of the electrons in the sample when one electron accelerated by voltage is incident on the sample model, the component elements emitted during the movement of the electrons in the sample model are calculated. Calculate the intensity of characteristic X-rays in consideration of the absorption until the sample model goes out of the sphere, perform the above calculation for many electrons, and calculate the characteristic X of the component element from the sum of the characteristic X-ray intensities of the component elements. Find the mutual intensity ratio of the lines,
On the other hand, when the selected actual sample is irradiated with the electron beam accelerated at the predetermined acceleration voltage, the mutual intensity ratio of the characteristic X-rays of the component elements of the sample is actually measured, and the mutual intensity ratio of the characteristic X-rays of each component obtained by the above calculation is calculated. Quantitative analysis by X-ray of a granular sample, wherein the initial composition ratio is corrected so that the intensity ratio matches the actual measurement value, and the above calculation is repeated. Method.