JPH04127039A - Identification of material by fluorescent spectrum - Google Patents

Abstract

PURPOSE:To enable identification of materials at a high efficiency by comparing fluorescent spectrums for unknown material and respective reference materials in measurements with more than one excitation wavelengths. CONSTITUTION:Combination of a band pass filter 2, a dichroic mirror 3 and an absorption filter 6 is prepared in plurality so as to select a plurality of excitation wavelengths and the combinations are switched with an excitation wavelength switching mechanism 20 which is operated by control of a microcomputer 13. Fluorescence transmitted through the absorption filter 6 is analyzed with a diffraction grating 9 and converted to a digital signal with a photomultiplier 11 and an A/D converter 12 to be taken into the microcomputer 13. Comparison using fluorescent spectrums in measurements with a plurality of excitation wavelengths enables identification of a plurality of materials thereby yielding an effect for improving an identifying capacity of unknown material.

Description

DETAILED DESCRIPTION OF THE INVENTION [Industrial Application Field] The present invention relates to an identification method for identifying an unknown substance, which is mainly an organic substance, from among a plurality of standard substances using a fluorescence spectrum.

[Conventional technology]

A method of identifying an unknown substance by comparing the shape of the fluorescence spectrum of the unknown substance with that of a known standard substance has been attempted in the past. However, conventionally, when performing these analyses, comparisons have been made using one fluorescence spectrum measured with the excitation wavelength fixed at one specific wavelength. Related devices of this type include, for example,
Proceedings of the 5th Applied Spectrometry Tokyo Conference (1999)
Discussed in 2010).

[Problem to be solved by the invention]

The above conventional technology does not take into account the case where multiple standard substances exhibit very similar fluorescence spectra at the excitation wavelength used, and in such cases, the amounts of these substances cannot be distinguished, making it difficult to identify the unknown substance. There was a problem that was difficult.

An object of the present invention is to provide an identification method that enables discrimination between these similar standard substances even in such cases.

[Means to solve the problem]

In order to achieve the above objective, we measure not only the fluorescence spectrum at one excitation wavelength but also the fluorescence spectra at two or more predetermined excitation wavelengths for the unknown substance and each standard substance, and analyze these multiple fluorescence spectra. It is designed to be used for comparison.

[Effect]

When excited with excitation light in the range from near ultraviolet light around 350 nm to ultraviolet light around 550 nm, the fluorescence spectrum of organic substances generally has a small number of peaks, usually one, and a wide half-width of 1100 nm to 200 nm. It often takes the form of As an example, FIG. 3 shows typical substances that pose a problem as foreign matter in a clean room where semiconductors are manufactured. The fluorescence spectrum of human skin flakes at an excitation wavelength of 405 nm is shown. Because of these characteristics, two or more substances often have similar fluorescence spectrum shapes at a certain excitation wavelength. However, even in such a case, fluorescence spectra with significant differences in shape may be obtained at different excitation wavelengths. For this reason, it is beneficial to compare and identify fluorescence spectra trained at two or more excitation wavelengths. Furthermore, even if a similar fluorescence spectrum is obtained again with this new excitation wavelength, it is unlikely that the intensity relationship will be the same. That is, now the substance A is F at the excitation wavelength λl
At excitation wavelengths λ1 and λ2, substance B exhibits a fluorescence spectrum of F^2(λ) (where λ represents the wavelength on the fluorescence side) at excitation wavelengths λ1 and λ2.
λ), (however, kl and 2 are constants), that is, F^1(λ) and Fal(λ), F to 2(
Assuming that each pair of λ) and FBI(λ) has a similar relationship,
Furthermore, although k s = k 2 may hold true in some cases, in many cases 1≠2. Each of kl and ni2 itself does not have much meaning. This is because the intensity of a fluorescence spectrum is influenced by 1) the size (amount) or concentration of the substance, 2) the shape and surface condition of the substance, and in general, in qualitative analysis, unless special pretreatment is performed, certain conditions are not met. This is because it is difficult to make assumptions. However, these effects are independent of the excitation wavelength or do not change even if the excitation wavelength is slightly changed, and the ratio of fluorescence spectral intensities due to two excitation wavelengths at a certain wavelength λ0 is
and Fat(λo)/FB2(λ0) can be considered to be approximately constant if the substances are the same. kt earlier,
In such a case, kz is 1≠2 in many cases. In this case, even if the shapes of the fluorescence spectra at each excitation wavelength are similar and there is no significant difference, discrimination is possible if the intensity relationship is satisfied.

In this way, fluorescence spectra measured at two or more excitation wavelengths can be compared and each substance can be identified with high efficiency.

〔Example〕

An embodiment of the present invention will be described below with reference to FIG. The excitation light source 1 is one that has emission lines of multiple wavelengths in the near-ultraviolet to visible range, such as an ultra-high-pressure mercury lamp, or one that has a continuous wavelength distribution, such as a Cannon lamp, to excite the sample and emit fluorescence. Use a material with sufficient luminous energy. A bandpass filter 2 is used to extract only light in a wavelength band necessary as excitation light from the light emitted from this excitation light g1. The excitation light that has passed through the bandpass filter 2 is guided to the objective lens 4 via the dichroic ink mirror 3, and is irradiated onto the sample 5. This dichroic ink mirror 3 has a characteristic of reflecting light in a wavelength range selected by the bandpass filter 2 with high reflectance, and transmitting light in a longer wavelength range with high transmittance. Fluorescence and scattered light of the excitation light are generated from the sample 5 by irradiation with the excitation light, but these are collected again by the objective lens 4, and most of the fluorescence is transmitted through the dichroic ink mirror 3.

Most of the scattered light of the excitation light is reflected, but some of this scattered light is transmitted, so in order to completely eliminate this, we must completely absorb the excitation wavelength and use light in a wavelength range longer than the excitation wavelength. uses an absorption filter 6 that is made to transmit light with a wavelength as close as possible to the excitation wavelength with high transmittance. The wavelength characteristics of the dichroic ink mirror 3 and the absorption filter 6 need to correspond to the transmission wavelength of the bandpass filter 2.
These three optical components work as a set. In this embodiment, a bandpass filter 2.
A plurality of combinations of dichroic mirrors 3 and absorption filters 6 are provided, and these can be switched by an excitation wavelength switching mechanism 2o operated under control from a microcomputer 13. A portion of the fluorescence that has passed through the absorption filter 6 is reflected by the beam splinter 18 and guided to the eyepiece 19, where a fluorescence image of the sample 5 is observed with the naked eye.

The fluorescence transmitted through the beam splitter 18 is guided to the spectrometer 8. The spectrometer 8 has a diffraction grating 91 and a diffraction grating scanning motor 0. It consists of a photomultiplier tube 11, the fluorescence is separated by a diffraction grating 9, the light intensity of the wavelength component is converted into an electric signal in the photomultiplier tube 11, and converted into a digital signal by an A/D converter 12 and sent to a microcomputer. Incorporated into 13. At this time, the diffraction grating scanning motor 10 is controlled by a microcomputer, and in synchronization with this scanning, the A/D
By reading the output signal of the converter 12, the fluorescence spectrum is captured into the microcomputer 13. The microcomputer 13 has a magnetic disk storage device 14, and can store data of measured fluorescence spectra. Also, the microcomputer 13 has a keyboard 15.
It also has a CRT display device 16 and a printer 17, and is capable of inputting information from an operator, displaying results, etc. on a CRT, and printing on a printer.

The operation of this embodiment will be explained below with reference to FIG. Second
The figure is a flowchart of software running on the microcomputer 13. Here two excitation wavelengths λ^ and λ
The operation when performing identification based on fluorescence spectra captured at each sampling wavelength interval Δλ in the range of wavelengths λ1 to λ2 measured in B will be described. Assume that there are N standard substances to be identified, and the first one (i=1.2...,
The fluorescence spectrum data sequence when the standard material of N) is excited at λ8 is (Aha) F=λ1. λ engineering + Δλ,...λ
2) The fluorescence spectrum data sequence when excited with λB is (
BIJ) (j=λ1.λ1+Δλ, . . . , λ2), and each data string for N standard substances and the name of the substance are stored in the magnetic disk storage device 14 in advance. When this software starts operating, it first controls the excitation wavelength switching mechanism 20 to switch the excitation wavelength to λ^. Next, the diffraction grating scanning motor 1o is controlled to scan the wavelength of the spectrometer 8 from λ1 to λ2, and the A/D
The output signal of the converter 12 is taken in and the fluorescence spectrum data string (a,) (j=λ) of the unknown sample 5 at the excitation wavelength λ
1. λ1+Δλ, ..., λ2). Next, the excitation wavelength is switched to λB, and in the same way, the fluorescence spectrum data string (b,) at the excitation wavelength λB (j=λ1.λ1+Δ
λ, ..., λ2). When the measurement of these two spectra is completed, comparison with the fluorescence spectra of each standard substance measured in advance begins. For this comparison, the fluorescence spectrum data strings (a,) and (tz) of the unknown sample
and the fluorescence spectrum data string of the first standard material (A, ,
) and (BIJ) are calculated using the correlation coefficient between these data as shown in the equation below.

Σ(xk-ma)(yk-y) However, the data string (xk) is the data string (a,) followed by the data string (b,), and similarly the data string (y,
) is a combination of data string (Atd followed by data string (B, d). Also, X represents the average value of (X, ), and y represents the average value of (yk).

From 1:1 to i=N, determine the degree of agreement α between the fluorescence spectrum of each standard substance, and display and print the name of the standard substance and the degree of agreement on the CRT and printer in order from the standard substance that gave the largest α1. This software will terminate its operation.

According to this example, when there are multiple standard substances whose shape matches the fluorescence spectrum of an unknown sample at either the excitation wavelength λ^ or λB, a difference is observed in the shape of the fluorescence spectrum at the other excitation wavelength. Since there are differences in the degree of agreement with these multiple standard materials, identification becomes possible.

Furthermore, even if there are multiple standard substances whose shapes match the fluorescence spectra of the unknown spectrum at both excitation wavelengths, if the ratio of the fluorescence spectrum intensities between the two excitation wavelengths differs for each of the standard substances, Since the difference is reflected in the value of the degree of agreement, it becomes possible to identify the amounts of these multiple substances. In this way, this example has the effect of improving the ability to identify amounts of substances that exhibit shapes in similar fluorescence spectra, compared to the case where comparison is made using fluorescence spectra at only one excitation wavelength. Another effect is that when comparing fluorescence spectra at only one excitation wavelength, it is necessary to determine the excitation wavelength at which the unknown substance is efficiently excited by determining the excitation wavelength in advance. If one of these excitation wavelengths is used, it can be expected that the unknown substance will be efficiently excited by one of these excitation wavelengths, and there is also the effect of saving the effort of determining the excitation wavelength for each unknown sample.

The excitation wavelength switching mechanism 20 in the above embodiment may be a mechanism in which switching is performed manually in synchronization with the operation of the software. Furthermore, a spectrometer may be used instead of the bandpass filter 2 as a means for selecting the excitation wavelength. Furthermore, instead of using the diffraction grating 9, the spectroscope 8 that measures the fluorescence spectrum may use a variable wavelength filter. Further, the diffraction grating 9 may be fixed, and a one-dimensional or two-dimensional multichannel photodetector may be used instead of the photomultiplier tube.

[Effects of the Invention] According to the present invention, when there are multiple standard substances that exhibit similar fluorescence spectrum shapes at one excitation wavelength, more information can be obtained by comparing fluorescence spectra measured at multiple excitation wavelengths. In this case, it becomes possible to identify the amounts of these multiple substances, which has the effect of improving the ability to identify unknown substances.

[Brief explanation of the drawing]

FIG. 1 is a block diagram showing the configuration of an embodiment of the present invention, FIG. 2 is a flowchart showing the operation of the embodiment of the invention, and FIG. 3 is a diagram showing a fluorescence spectrum. 1...Excitation light source, 2...Band pass filter, 3...
・Dichroic mirror, 4...Objective lens, 5...
・Sample, 6... Absorption filter, 7... Eyepiece,
8... Spectrometer, 9... Diffraction grating, 10... Diffraction grating scanning motor, 11... Photomultiplier tube, 12... A/D
Converter, 13...Microcomputer, 14...Magnetic disk storage device, 15...Keyboard, 16...
- CRT display device, 17... printer, 20... excitation wavelength switching mechanism. Figure 2 Figure 2

Claims (1)

[Claims] 1. In an identification method that compares the fluorescence spectrum of an unknown substance with the fluorescence spectra of multiple pre-measured standard substances and searches for the best match, the unknown substance and each standard substance are compared at two or more excitation wavelengths. A method for identifying substances based on fluorescence spectra, the method comprising comparing a plurality of measured fluorescence spectra.