JP7764012B2 - Raman spectrometer and method for measuring Raman spectrum - Google Patents

Description

The present invention relates to a method for measuring Raman spectra, and in particular to a method for correcting the baseline of measured spectral data.

When a substance is excited by light of a specific wavelength, it scatters light of several wavelengths different from the wavelength of the excitation light (Raman scattered light). The difference between the wavenumber of the excitation light and the wavenumber of the Raman peak is called the Raman shift, and is determined by factors such as the vibration and rotation of molecules in the substance. Therefore, a Raman spectrum, with the Raman shift on the horizontal axis and the Raman peak intensity on the vertical axis, can be used to identify a substance.

On the other hand, Raman scattered light is about 10 ⁻⁶ times weaker than Rayleigh scattered light having the same wavelength as the excitation light, and the spectrum measurement of Raman scattered light is significantly affected by fluorescence from the substance.

The effects of fluorescence in conventional Raman spectrum measurements can be explained in two ways. First, because the fluorescent components span a relatively wide band, the spectral baseline rises, obscuring weak Raman peaks. Second, the fluorescent components form interference fringes in optical elements such as films inside the analytical device, distorting the baseline into a wavy shape. These baseline rises and distortions are thought to be the cause of a lower success rate in spectral searches for qualitative analysis of samples.

Therefore, in Raman spectrum measurements, it is particularly important to remove or reduce the effects of fluorescence, i.e., to properly correct the baseline.

A common method for avoiding fluorescence is to change the wavelength of the excitation light to a longer wavelength. For example, with 532 nm excitation light, depending on the substance, a large fluorescent component may be generated, obscuring the Raman peak. By changing the wavelength of this excitation light to, for example, 1064 nm, the generation of fluorescent components can be suppressed, and the Raman peak may be visible.

There have also been various proposals for methods of correcting the baseline of a measured spectrum. For example, in Patent Document 1, shape data such as a semicircle or semiellipse is set for the measured spectrum data, and the baseline is estimated using software, thereby correcting the baseline so that it is flat.

Patent No. 4966337

However, even with conventional methods, there are cases where baseline elevation caused by fluorescent components cannot be fully eliminated, and there are also cases where baseline distortion caused by fluorescent interference fringes cannot be fully eliminated. The object of the present invention is to provide a Raman spectrometer and a method for measuring Raman spectra that can effectively eliminate or reduce the influence of fluorescence on measured spectra using a new method that differs from conventional methods.

That is, the Raman spectroscopic device according to the present invention:
A Raman spectroscopic device comprising: a light source for excitation light of a single wavelength; a stage for placing a sample; an irradiation area setting means for setting an irradiation area of the excitation light on the sample; a take-up area setting means for setting a take-up area of the light on the sample; a spectroscope for spectrally detecting light from the take-up area; and a data processing means for processing spectral data based on the detection value from the spectroscope,
At least one of the light source, the irradiation area setting means, and the capture area setting means sets, as a measurement condition for the main measurement, at least one of the capture area, the irradiation area, and the excitation light intensity so that the ratio of the Raman peak to the fluorescent component in the spectral data is larger than that in the base measurement ; and
as a measurement condition for the base measurement, at least one of the capture area, the irradiation area, and the excitation light intensity is set so that a ratio of a Raman peak to a fluorescent component in the spectral data is smaller than that in the main measurement;
The data processing means
The system is characterized by including a coefficient calculation unit that calculates the intensity change rate of the fluorescent component of the spectral data of the base measurement according to differences in measurement conditions and calculates a baseline coefficient based on the intensity change rate, and a difference spectrum calculation unit that calculates the difference spectrum between the spectral data of the base measurement to which the baseline coefficient has been applied and the spectral data of the main measurement.

The capture area setting means may be configured to narrow the area of the capture area as a measurement condition for the main measurement and to widen the area of the capture area as a measurement condition for the base measurement.

The capture area setting means may have an opening that limits the light captured by the spectroscope, and may be configured to change the shape or size of the opening.

The irradiation area setting means may be configured to set the offset between the center of the irradiation area and the center of the capture area to zero or a small value as a measurement condition for the main measurement, and to set the offset between the center of the irradiation area and the center of the capture area to a large value as a measurement condition for the base measurement.

The light source may be configured to reduce the intensity of the excitation light as a measurement condition for the main measurement and to increase the intensity of the excitation light as a measurement condition for the base measurement.

Furthermore, an exposure time setting means may be provided for setting the exposure time of the sample to the excitation light, and the exposure time setting means may be configured to set the exposure time to be long as the measurement condition for the main measurement and to set the exposure time to be short as the measurement condition for the base measurement.

Further, the method for measuring a Raman spectrum according to the present invention comprises:
A method for measuring a Raman spectrum by capturing light from a sample excited with excitation light of a single wavelength into a spectrometer, comprising:
The method includes an irradiation area setting step of setting an irradiation area of the excitation light on the sample, a capture area setting step of setting a light capture area on the sample, an irradiation step of irradiating the excitation light, a step of acquiring spectral data based on detection values from a spectroscope, and a data processing step of processing the spectral data,
In at least one of the irradiation area setting step, the capture area setting step, and the irradiation step,
As a measurement condition for this measurement, at least one of the capture area, the irradiation area, and the excitation light intensity is set so that the ratio of the Raman peak to the fluorescent component in the spectrum data is larger than that in the base measurement; and
As the measurement conditions for the base measurement, at least one of the capture area, the irradiation area, and the excitation light intensity is set so that the ratio of the Raman peak to the fluorescent component in the spectral data is smaller than that in the main measurement;
The data processing procedure includes:
a coefficient calculation step of calculating a rate of change in intensity of a fluorescent component of spectral data of a base measurement according to differences in measurement conditions and calculating a baseline coefficient based on the rate of change in intensity; and a difference spectrum calculation step of calculating a difference spectrum between the spectral data of the base measurement to which the baseline coefficient has been assigned and the spectral data of the main measurement.
It is characterized by:

The coefficient calculation procedure may calculate the baseline coefficient based on spectral data from multiple base measurements under different measurement conditions.

The data processing procedure may include, after the difference spectrum calculation procedure, a baseline flattening procedure for flattening the baseline of the difference spectrum.

The measurement method of the present invention may be performed after changing the wavelength of the excitation light to a longer wavelength and measuring the Raman spectrum.

By performing Raman spectrum measurements as described above, it is possible to obtain Raman spectrum data that has been corrected for baseline rise (the influence of fluorescent components across a wide wavenumber range) and baseline distortion (the influence of interference fringe components due to fluorescence), thereby improving the accuracy rate of material analysis using such Raman spectrum data.

1 is a schematic diagram illustrating the configuration of a Raman spectroscopic device according to an embodiment. 10A and 10B are diagrams showing the difference in capture area between two types of openings in the device. FIG. 2 is a diagram showing a procedure for measuring Raman spectra using the device. FIG. 10 is a Raman spectrum actually measured under the first measurement condition. FIG. 10 is a diagram of a Raman spectrum actually measured under the second measurement condition. FIG. 10 is a diagram of a Raman spectrum actually measured under the third measurement condition. FIG. 10 is an explanatory diagram of verification using spatial offsets. FIG. 10 is a diagram showing the results of calculating the distribution of the fluorescence area and the Raman area in the verification. 1A and 1B show spectral data measured using a rectangular slit and a pinhole slit, respectively.

Preferred embodiments of the present invention will now be described with reference to the drawings. Figure 1 is a schematic diagram of a microscopic laser Raman spectrophotometer (hereinafter referred to as the Raman spectrophotometer), which is one embodiment of the Raman spectrophotometer of the present invention. The Raman spectrophotometer is a device for spectrally detecting Raman scattered light from a sample S when irradiated with excitation light (laser light) and acquiring a Raman spectrum. Measurement targets in this embodiment are a wide range, including solid samples (such as PEEK resin), powder samples (such as cellulose powder), and liquid samples (such as colored liquids such as whiskey).

Figure 1 shows the basic structure of a Raman spectrometer. Specifically, the Raman spectrometer comprises a confocal optical system consisting of a movable sample stage 11, an objective lens 12, a beam splitter (BS) 13 (or dichroic mirror DM) that reflects excitation light toward the sample and transmits light from the sample S, a rejection filter 14 for cutting Rayleigh light, an imaging lens 15, an aperture switching device 16, a spectroscope 17, and a CCD detector 18.

The Raman spectrometer also includes an excitation optical system consisting of a laser wavelength switching device 21, a beam spot position changing device 22, and a laser control device 23. The laser control device 23 includes an exposure time setting unit 24 and a laser intensity setting unit 25.

The Raman spectrometer also includes a spectral data processing device 30 and a storage device 40. The spectral data processing device 30 is configured as an arithmetic processing device such as a computer, and has the following functional blocks: a baseline correction unit 31, a baseline flattening processing unit 32, and a noise removal unit 33. The baseline correction unit 31 further includes a difference spectrum calculation unit 34 and a baseline coefficient calculation unit 35, which are characteristic of the present invention.

The storage device 40 stores a data processing program that causes the spectral data processing device 30 to realize these functional blocks, and the spectral data processing device 30 executes this data processing program as appropriate to perform these functions.

<Details of the excitation optical system>
The laser wavelength switching device 21 is equipped with a plurality of switchable light sources 26, 27 of different wavelengths. The device is configured so that excitation light from the light sources 26, 27 of the selected wavelength is output toward the beam spot position changing device 22. The beam spot position changing device 22 can be configured, for example, with a pair of alignment mirrors. The pair of alignment mirrors can adjust the optical axis of the excitation light from the light sources 26, 27, and can move the position of the beam spot on the sample without changing the positional relationship between the sample S and the objective lens 12.

The laser control device 23 controls the selected light sources 26, 27. The exposure time setting unit 24 turns the output of the light sources 26, 27 on and off according to the set exposure time. Alternatively, it may open and close shutters provided in the output windows of the light sources 26, 27. The laser intensity setting unit 25 adjusts the output intensity of the light sources 26, 27 according to the intensity conditions of the excitation light. Alternatively, it is configured to change the intensity of the excitation light by switching on and off a neutral density filter installed in the optical path of the excitation light.

The excitation light passes through the objective lens 12 on the sample to form a tiny beam spot, which is the target area for spectral measurement.

<Details of the confocal optical system>
The movable stage 11 is an XY two-axis stage having a mounting surface parallel to the XY plane, and can position the beam spot on the mounted sample at the focusing position of the objective lens 12 .

Objective lens 12 is an example of a focusing optical element of the present invention, such as a Cassegrain mirror. Objective lens 12 serves to form a beam spot of excitation light on the sample and to focus light from the beam spot and its surrounding area. Rayleigh scattered light contained in the light from sample S is removed by rejection filter 14. Imaging lens 15 focuses the focused light on an aperture, such as rectangular slit plate 16A or pinhole slit plate 16B, mounted in the downstream aperture switching device 16, to form an intermediate image of the beam spot. Note that while Figure 1 shows an example of a reflective optical system, a measurement device using a transmissive optical system, such as a Cassegrain mirror, may also be constructed.

The opening switching device 16 is configured to be able to switch between at least two types of openings. Figure 1 shows, as an example, a rectangular slit plate 16A with an elongated opening and a pinhole slit plate 16B with a circular opening held in a common sliding frame, allowing the selected opening to be installed online. Openings of various shapes other than elongated or circular shapes can be formed in the opening plate. It is also possible to form multiple openings of different sizes for each shape.

The spectrometer 17 is a dispersive spectrometer with a typical diffraction grating. Diffracted light from the diffraction grating is imaged on the downstream CCD detector 18, and the spectral values for each band are detected by a group of light-receiving elements arranged two-dimensionally in the CCD detector 18, and output to the spectral data processing device 30.

Here, using Figure 2, we will show the difference in the capture area between the two types of apertures (rectangular slit plate 16A and pinhole slit plate 16B). Here, the beam spot is shown as the black circle area on the sample. Its periphery is shown as a white circle area larger than the black circle.

The light entering the spectrometer 17 is limited by the openings of the rectangular slit plate 16A and pinhole slit plate 16B. In other words, the light capture area on the sample is determined by the shape and size of these openings.

In Figure 2, the slit width of rectangular slit plate 16A is approximately the same as the size of the beam spot image, but with rectangular slit plate 16A, not only light from the beam spot but also much of the light from its periphery is taken into spectroscope 17. On the other hand, with pinhole slit plate 16B, especially when its opening is approximately the same size as the beam spot image, the confocal effect allows only light from the beam spot to be efficiently taken into spectroscope 17, allowing measurements to be made with higher spatial resolution than with rectangular slit plate 16A.

This embodiment is characterized by using the above-described Raman spectrometer to perform at least two types of spectral measurements (base measurement and main measurement) under different measurement conditions. Table 1 below lists example measurement conditions.

Under condition 1, the base measurement is performed with low spatial resolution, and the main measurement is performed with high spatial resolution. For example, a wide capture area and a narrow capture area are set for the same measurement position on the sample. Conditions 1A to 1C in Table 1 are used when a rectangular slit or pinhole slit opening is used to set the light capture area on the sample.

Condition 1A uses, for example, a rectangular slit with a slit width of 200 μm and a pinhole slit with a diameter of 100 μm. When using a rectangular slit, the proportion of the beam spot in the capture area on the sample should be small; when using a pinhole slit, conversely, that proportion should be larger than with a rectangular slit. Alternatively, the diameter of the pinhole slit should be the same as or smaller than the slit width of the rectangular slit.

Under conditions 1B and 1C, the size of the rectangular slit or pinhole slit is changed. For example, for the same measurement position on the sample, a slit with a larger opening is used for the base measurement and a slit with a smaller opening is used for the main measurement.

When using a pinhole slit in this measurement, the light capture area of the pinhole slit may be aligned with the beam spot, or the light capture area may be completely within the beam spot. In this embodiment, these conditions are set by the operation of the aperture switching device 16.

Under condition 2, the spatial offset is changed by changing the distance between the beam spot and the light capture area. In the base measurement, an offset state is established and the above distance is increased. In the main measurement, a coincidence state is established, meaning the above distance is zero, or the distance is decreased even in an offset state. Note that the beam spot may be fixed and the position of the capture area may be moved. Alternatively, the capture area may be fixed and the position of the beam spot may be moved by operating the beam spot position change device 22 in Figure 1.

Condition 3 increases the intensity of the excitation light in the base measurement and decreases the intensity of the excitation light in the main measurement. This condition setting is performed by the operation of the laser intensity setting unit 25 in Figure 1.

A common feature of each condition is that, using the exposure time setting unit 24 in Figure 1, the exposure time of the excitation light in this measurement should be set longer than the exposure time in the base measurement so that a Raman peak of appropriate height is obtained.

<Method for measuring Raman spectrum>
Fig. 3 shows the procedure for measuring Raman spectra by changing the spatial resolution (for example, condition 1B in Table 1). Figs. 4 to 6 show the results of measuring Raman spectra of powdered cellulose samples under three specific measurement conditions.

The measurement method according to this embodiment may be started immediately after optimizing the wavelength of the excitation light, a conventional method for avoiding fluorescence. The wavelength of the excitation light is switched using the laser wavelength switching device 21 shown in Figure 1. Even if the wavelength of the excitation light is changed to a longer wavelength and the Raman spectrum is measured, it may not be possible to fully eliminate the effects of fluorescence. However, by performing the measurement method according to this embodiment, it may be possible to eliminate or reduce the effects of such fluorescence.

First, a measurement position on the sample placed on the movable stage 11 is specified (Step S10). Next, the movable stage 11 is operated to align the beam spot position with the measurement position, thereby setting the beam spot (Step S12). Next, the aperture switching device 16, which is equipped with a wide rectangular slit plate and a narrow rectangular slit plate, is operated to set the wide rectangular slit on the optical path (Step S14).

The base measurement is performed (step S16). In the base measurement, first, excitation light from the light sources 26 and 27 is irradiated onto the beam spot position, and only the light from the sample that passes through the opening of the wide rectangular slit plate is captured by the spectroscope 17. The spectroscope 17 detects the spectral values for each band, and these detected values are read into the spectral data processing device 30. Note that the exposure time in the base measurement is short.

After the base measurement is completed, the aperture switching device 16 is operated to set a narrow rectangular slit plate on the optical path (step S18), and the main measurement is performed in the same manner as the base measurement (step S20). In the main measurement, the exposure time is set longer than in the base measurement.

After the spectral data processing device 30 has completed reading the detected values from the base measurement and the main measurement using the above procedure, data processing is performed by the spectral data processing device 30.

The upper left of Figure 4 shows the spectral data SP10 and SP12 from the base measurement and this measurement. For the base measurement, a rectangular slit 200 μm wide was used, with an exposure time of 1 second. For this measurement, a rectangular slit 100 μm wide was used, with an exposure time of 10 seconds.

In the data processing shown in Figure 3 (steps S22 to S40), the baseline coefficient A is first calculated (step S22) using the function of the baseline coefficient calculation unit 35. Here, a wavenumber range in which no Raman scattered light occurs in either of the two measurement spectra is selected, and the intensity change rate of the fluorescent component occurring in that wavenumber range (e.g., the spectral area corresponding to the fluorescent component) is calculated. The baseline coefficient A is then calculated based on this intensity change rate.

Next, the baseline coefficient A is used to calculate the difference spectrum between the spectral data of the main measurement and the spectral data of the base measurement (step S24). Here, the spectral data of the base measurement is multiplied by the baseline coefficient A to obtain the difference from the spectral data of the main measurement. The above processing is the baseline correction according to this embodiment.

The processing up to this point removes the effects of fluorescence (such as interference fringes), which were difficult to remove using conventional methods, but some fluorescent components may remain. Any fluorescent components remaining at this stage can be easily removed using conventional software-based baseline flattening processing, such as that described in Patent Document 1 (step S30).

In the measurement example shown in Figure 4, no Raman peaks appear in the wavenumber range of 2500 to 2000 cm ^-1 for either of the two spectral data sets SP10 and SP12. The fluorescence areas A10 and A12 in this wavenumber range were calculated, and a baseline coefficient A was calculated so that the difference between the two fluorescence areas would be zero (step S22). The difference spectrum SP14, using this coefficient A, is shown in the upper right corner of Figure 4 (step S24). This difference spectrum SP14 was subjected to baseline flattening processing using existing software (step S30) after calculation of the difference spectrum. For comparison, the upper right corner of Figure 4 also shows comparison data Ref14, which was obtained by performing only baseline flattening on the spectral data from this measurement. It can be seen that the difference spectrum SP14 has less noise and interference fringes.

In the data processing of Figure 3, existing software-based noise removal processing (step S40) may be performed at the end, if necessary. For example, spectral data may be subjected to an inverse Fourier transform, and the resulting power spectrum may be subjected to a low-pass filter to remove high-frequency spectral fluctuations occurring on the baseline of the spectral data as noise.

In the measurement example of Figure 4, noise removal processing (step S40) was performed on the difference spectrum SP14 of this embodiment after baseline flattening processing. The difference spectrum SP16, with an improved S/N ratio due to noise removal, is shown at the bottom of Figure 4. Furthermore, both existing noise removal processing and interference fringe removal processing were performed on the comparison data Ref14. The results are also shown together as comparison data Ref16. It can be seen that while the comparison data Ref16 contains interference fringes (wavy deformation of the baseline) that are not removed by existing data processing, such interference fringes have been significantly reduced in the difference spectrum SP16 of this embodiment.

Figure 5 shows the spectral data SP22 from this measurement using a 50 μm-wide rectangular slit. The data from the base measurement is the same as Figure 4. Comparing the difference spectra SP24 and SP26 calculated from this spectral data with the comparison data Ref24 and Ref26, it is clear that the difference spectrum SP24 from this embodiment has less noise and interference fringes than the comparison data Ref24, and that the difference spectrum SP26 after noise removal has significantly reduced interference fringes compared to the comparison data Ref26.

Furthermore, Figure 6 shows difference spectra SP34 and SP36 based on the spectral data SP32 from this measurement using a pinhole slit with a diameter of 100 μm, along with comparison data Ref34 and Ref36. As with Figure 5, it can be seen that the difference spectrum from this embodiment is superior to the comparison data Ref34 and Ref36 in terms of noise and interference fringes.

Note that while the procedure in Figure 3 describes a case where the base measurement (step S16) is performed only once, it is also possible to perform the base measurement multiple times by setting different measurement conditions and calculate the baseline coefficient A based on multiple pieces of spectral data (step S22).

<Verification by spatial offset>
The first feature of the present invention is that the Raman spectrum measurement method of the present invention performs two types of spectrum measurements (main measurement and base measurement) on the same sample under different measurement conditions. In the main measurement, measurement conditions are set so that "the ratio of the Raman peak to the fluorescent component in the spectral data is relatively large," and in the base measurement, measurement conditions are set so that "the ratio is relatively small."

Specific examples include: measurement condition 1 is changing the area of the light capture region (measurement field) on the sample; measurement condition 2 is changing the positional relationship between the excitation light irradiation region (beam spot) on the sample and the capture region; and measurement condition 3 is changing the intensity of the excitation light.

The second feature is that the difference between the spectral data obtained under two different measurement conditions is taken to cancel out the fluorescent components contained in the spectral data from this measurement. However, because this is spectral data from the same sample, you might be wondering whether it would be difficult to cancel out just the fluorescent components even if the measurement conditions were changed (wouldn't the desired Raman peak also be canceled out?).

In response to this, the inventors conducted a "spatial offset verification" as shown in Figure 7. Here, PEEK resin was used as the sample. An objective lens 12 with a magnification of 100 was used so that the 532 nm wavelength excitation light would form a beam spot with a diameter of approximately 1 μm on the sample. The spatial resolution on the sample was also set to a 1 μm equivalent (capture area). The offset between the center of the beam spot on the sample and the center of the light capture area (both circular areas with a diameter of 1 μm) was gradually increased from 0 μm to 5 μm, and the respective spectra were measured. Differences were observed between the change in the peak height of the Raman peak and the change in the height of the fluorescent component. The offset was achieved by shifting the optical axis of the excitation light (black area) by 0 to 5 μm relative to the optical axis of the measurement light, without changing the relative positions of the sample S and the objective lens 12, as shown in Figure 7.

Even with an offset of 3 μm or 5 μm, fluorescent components can be observed over a wide wavenumber range (a wide rise in the baseline), but Raman peaks are barely discernible, even with an offset of 3 μm. Raman peaks can be observed with an offset of up to 1 μm.

Therefore, we investigated the change in peak area for each offset for the Raman peak near 1600 cm ^-1 and the change in the spectral area of the fluorescent component in the band where no Raman peak occurs (2500 to 2000 cm ^-1 ), as shown in Figure 8. We confirmed that the fluorescent component (fluorescence area) gradually decreases from its peak at the beam spot, while the Raman peak (Raman area) rapidly decreases from its peak at the beam spot. Note that the graph showing the area for each offset in Figure 8 has been normalized by setting the fluorescent area and Raman area at 0 μm to 100.

The inventors then estimated that while the intensity distribution of fluorescence is a relatively gentle mountain-shaped distribution with the beam spot at its peak, the intensity distribution of Raman scattered light is a Gaussian distribution with the beam spot at its peak. This is because Raman scattered light decreases rapidly as it moves away from the beam spot, and becomes almost zero after a short distance.

Raman scattered light is inelastically scattered light, and is proportional to the intensity of the excitation light; therefore, the distribution of Raman light can be said to be Gaussian, just like excitation light (laser light). In contrast, fluorescence is light emitted as a result of the absorption of light by a substance, and tends to saturate above a certain level of excitation light. For this reason, the distribution of fluorescence is thought to be mountain-shaped rather than Gaussian. In other words, when the same excitation light is irradiated onto a sample, the volume of the sample from which Raman scattered light is detected is smaller than the volume of the sample from which fluorescence is detected.

The inventors discovered that, due to these differences in the distribution of fluorescence and Raman light on a sample, by performing two different spectral measurements on the same sample under different conditions and taking the difference between the two spectra, they could cancel out the effects of fluorescence and obtain a good Raman spectrum that retains the desired Raman peaks.

Several measurement conditions can be set. What they all have in common is that some measurement conditions are set so that the ratio of Raman peaks to fluorescent components in the spectral data is relatively large, and other measurement conditions are set so that the ratio is relatively small.

As shown in the light distribution in Figure 8, the ratio of Raman light to fluorescence is relatively high near the beam spot, and this can be called the "Raman light-rich region." Furthermore, away from the beam spot, the ratio of Raman light to fluorescence is almost zero, and this can be called the "fluorescence-rich region."

For example, under measurement condition 1, in this measurement, a relatively narrow region that mainly includes the "Raman light-rich region" is set as the light capture region. Then, in the base measurement, a relatively wide region that mainly includes the "fluorescence-rich region" is set as the light capture region. In other words, by changing the area of the light capture region, the spatial resolution can be varied.

Furthermore, for example, under measurement condition 2, as shown in Figure 7, the "Raman light-rich region" is set as the light capture region in this measurement, and the "fluorescence-rich region" is set as the light capture region in the base measurement. In other words, the positional relationship between the beam spot and the capture region is changed.

Also, for example, measurement condition 3 involves changing the intensity of the excitation light. In this measurement, the light capture region is set to a "Raman light-rich region" to obtain spectral data in which the ratio of Raman peaks to fluorescent components is relatively large. Then, in the base measurement, the spectrum of the same capture region is measured, but the intensity of the excitation light is increased to saturate the fluorescence and obtain spectral data in which the ratio of Raman peaks to fluorescent components is relatively small.

In addition, as shown in Figure 7, the spectral data measured under multiple offset conditions shows that the fluorescent components have similar shapes (similar shapes along the vertical axis of the spectral data). Furthermore, as shown in Figure 9, the fluorescent components also have similar shapes in the spectral data measured under different spatial resolution conditions (rectangular slit, pinhole slit). Therefore, when calculating the difference spectrum, fluorescent components can be neatly canceled out by applying an appropriate baseline coefficient to the spectral data from the base measurement and taking the difference with the spectral data from the main measurement. In this invention, the intensity change rate of fluorescent components due to changes in measurement conditions is calculated from the fluorescent area of each component, and a baseline coefficient is calculated based on this intensity change rate.

For this reason, the spectral measurement method of this embodiment makes it possible to obtain Raman spectral data in which the baseline rise and distortion caused by the influence of fluorescence have been corrected.

11 Movable stage (irradiation area setting means), 12 Objective lens, 13 Beam splitter, 14 Rejection filter, 15 Imaging lens, 16 Aperture switching device (capture area setting means), 16A Rectangular slit plate, 16B Pinhole slit plate, 17 Spectrometer, 18 CCD detector, 21 Laser wavelength switching device, 22 Beam spot position changing device (irradiation area setting means), 23 Laser control device, 24 Exposure time setting unit, 25 Laser intensity setting unit, 26, 27 Light source, 30 Spectral data processing device, 31 Baseline correction unit, 32 Baseline flattening processing unit, 33 Noise removal unit, 34 Difference spectrum calculation unit, 35 Baseline coefficient calculation unit, 40 Storage device.

Claims (10)

A Raman spectroscopic device comprising: a light source for excitation light of a single wavelength; a stage for placing a sample; an irradiation area setting means for setting an irradiation area of the excitation light on the sample; a take-up area setting means for setting a take-up area of the light on the sample; a spectroscope for spectrally detecting light from the take-up area; and a data processing means for processing spectral data based on the detection value from the spectroscope,
At least one of the light source, the irradiation area setting means, and the capture area setting means sets, as a measurement condition for the main measurement, at least one of the capture area, the irradiation area, and the excitation light intensity so that the ratio of the Raman peak to the fluorescent component in the spectral data is larger than that in the base measurement ; and
as a measurement condition for the base measurement, at least one of the capture area, the irradiation area, and the excitation light intensity is set so that a ratio of a Raman peak to a fluorescent component in the spectral data is smaller than that in the main measurement;
The data processing means
a coefficient calculation unit that calculates a rate of change in intensity of a fluorescent component of spectral data of a base measurement according to differences in measurement conditions and calculates a baseline coefficient based on the rate of change in intensity, and a difference spectrum calculation unit that calculates a difference spectrum between the spectral data of the base measurement to which the baseline coefficient has been assigned and the spectral data of the main measurement,
A Raman spectroscopic device characterized by: A Raman spectroscopic device according to claim 1, wherein the capture area setting means is configured to narrow the area of the capture area as a measurement condition for the main measurement and to widen the area of the capture area as a measurement condition for the base measurement. In the device described in claim 2, the capture area setting means has an opening that limits the light captured by the spectrometer, and is configured to change the shape or size of the opening. A Raman spectroscopic device according to any one of claims 1 to 3, wherein the irradiation area setting means is configured to set the offset between the center of the irradiation area and the center of the capture area to zero or a small value as a measurement condition for the main measurement, and to set the offset between the center of the irradiation area and the center of the capture area to a large value as a measurement condition for the base measurement. A Raman spectroscopic device according to any one of claims 1 to 4, wherein the light source is configured to reduce the intensity of the excitation light as a measurement condition for the main measurement and to increase the intensity of the excitation light as a measurement condition for the base measurement. A Raman spectroscopic apparatus according to any one of claims 1 to 5, further comprising an exposure time setting means for setting the exposure time of the sample to excitation light, the exposure time setting means being configured to set the exposure time to a long value as a measurement condition for the main measurement and to set the exposure time to a short value as a measurement condition for the base measurement. A method for measuring a Raman spectrum by capturing light from a sample excited with excitation light of a single wavelength into a spectrometer, comprising:
The method includes an irradiation area setting step of setting an irradiation area of the excitation light on the sample, a capture area setting step of setting a light capture area on the sample, an irradiation step of irradiating the excitation light, a step of acquiring spectral data based on detection values from a spectroscope, and a data processing step of processing the spectral data,
In at least one of the irradiation area setting step, the capture area setting step, and the irradiation step,
As a measurement condition for this measurement, at least one of the capture area, the irradiation area, and the excitation light intensity is set so that the ratio of the Raman peak to the fluorescent component in the spectrum data is larger than that in the base measurement; and
As the measurement conditions for the base measurement, at least one of the capture area, the irradiation area, and the excitation light intensity is set so that the ratio of the Raman peak to the fluorescent component in the spectral data is smaller than that in the main measurement;
The data processing procedure includes:
a coefficient calculation step of calculating a rate of change in intensity of a fluorescent component of spectral data of a base measurement according to differences in measurement conditions and calculating a baseline coefficient based on the rate of change in intensity; and a difference spectrum calculation step of calculating a difference spectrum between the spectral data of the base measurement to which the baseline coefficient has been assigned and the spectral data of the main measurement.
A method for measuring a Raman spectrum. A method for measuring Raman spectra according to claim 7, wherein the coefficient calculation step calculates the baseline coefficient based on spectral data obtained by multiple base measurements under different measurement conditions. 9. The method according to claim 7, wherein the data processing step comprises:
A method for measuring a Raman spectrum, comprising, after the difference spectrum calculation step, a baseline flattening step of flattening the baseline of the difference spectrum. A method for measuring a Raman spectrum according to any one of claims 7 to 9, characterized in that the method is carried out after changing the wavelength of the excitation light to a longer wavelength and measuring the Raman spectrum.