Inspection method and inspection device of brewage

Abstract

To provide an inspection method and an inspection device for highly accurately analyzing brewage components according to Raman spectroscopy using a wavelength range by which an intensity ratio of a Raman signal and fluorescence noise falls within a prescribed range, as an excitation light wavelength.SOLUTION: An inspection method of conducting simultaneous determination of ethanol and glucose of brewage includes the steps of: extracting a sample from a brewing vessel; obtaining a liquid component by performing solid-liquid separation, and in applying Raman spectroscopy, a wavelength of excitation light is set to a wavelength by which an S / N ratio is 90% or more of a maximal value. S / N is a signal intensity ratio Isignal / Inoise.SELECTED DRAWING: Figure 1

Description

[Technical Field]

[0001] The present invention relates to a method for testing brewed alcoholic beverages, and more particularly to a testing method and device for simply and accurately analyzing ethanol and glucose in the mash components of sake. [Background technology]

[0002] In the brewing process of sake and other beverages, it is necessary to constantly measure alcohol content (ethanol concentration) and sugar content (glucose concentration) for process control and product quality control. Conventional methods for quantifying alcohol content include gas chromatography and near-infrared spectroscopy, but these methods are time-consuming and cost-intensive. While sugar content can sometimes be measured using a simple saccharimeter, there are issues with measurement accuracy, and high-precision saccharimeters have high consumable costs. In particular, the coexistence of alcohol, sugar, and other soluble and insoluble solids adversely affects measurement accuracy, making simple measurements difficult.

[0003] In light of this situation, as shown in Patent Document 1, a quantitative method and filtration device are known that simultaneously quantify the ethanol and glucose concentrations in the filtrate obtained from sake mash using a Raman spectrometer. Here, moromi refers to the fermented mixture of moto (starter), steamed rice, koji, and water. Sake moromi (fermented soybean paste) is defined in the Liquor Tax Act, and the solid matter remaining after squeezing the liquid sake from moromi becomes sake lees, which are used as ingredients in pickles, amazake (sweet sake), and vinegar. Hereinafter, in this invention, it will be simply referred to as moromi.

[0004] Patent Document 1 shows that when using laser light in the near-infrared region shorter than 650 nm, measurement sensitivity is reduced due to noise caused by fluorescence emitted by components of sake, but it is suggested to use excitation light of a longer wavelength, such as 785 nm, which does not reduce measurement sensitivity. [Prior art documents] [Patent documents]

[0005] [Patent Document 1] Patent No. 6606352 (Claims 1 and 5, paragraphs 0020-0021) Summary of the Invention [Problem to be solved by the invention]

[0006] However, although the intensity of the fluorescence noise decreases toward longer wavelengths such as 785 nm as described in Patent Document 1, the Raman intensity also decreases, resulting in a problem in that the effect (relative intensity) of the fluorescence noise on the Raman signal cannot be ignored.

[0007] In light of this background, the inventors have conducted extensive research and have found that a region where the effect (relative intensity) of the fluorescence noise on the Raman signal is small (i.e., a region with a high S / N ratio) exists in a specific region between the short wavelength side, where the absolute intensity of the fluorescence noise is high, and the long wavelength side, where the absolute intensity is low, and further that the size and region of this region with a good S / N ratio change depending on the magnitude of the noise in the sensor (spectroscope).

[0008] In particular, when using an inexpensive (high-noise) spectrometer for sake, it was found that the high S / N range was around 650nm to 700nm, making it ideal for mash analysis using an inexpensive sensor.

[0009] Therefore, in order to solve the above problems, the present invention aims to provide a method and device for inspecting brewed alcoholic beverages that analyzes the components of brewed alcoholic beverages with high accuracy using Raman spectroscopy, where the wavelength of the excitation light is within a wavelength range in which the ratio of the Raman signal intensity (S, signal) to the sum (N, noise) of the fluorescence noise intensity and the spectrometer-derived noise intensity falls within a predetermined range. [Means for solving the problem]

[0010] The invention described in claim 1 is a testing method for simultaneously quantifying ethanol and glucose in brewed alcoholic beverages during brewing, which involves extracting a sample from a brewing vessel, performing solid-liquid separation of the sample to obtain liquid components, and performing Raman spectroscopic analysis of the liquid components, selecting a wavelength range for the excitation light such that the S / N ratio is 90% or more of the maximum value. Here, S / N is the signal intensity ratio I signal / I noise , A is a proportionality constant calculated from the laser intensity, polarizability, and component concentration, B i is the proportionality constant calculated from the fluorescence intensity, C is the speed of light, ν s is the frequency shift due to Raman scattering, λ L is the laser wavelength, λ s is the wavelength shift due to Raman scattering, σ i is the half-width of the fluorescence spectrum, and I signal and I noise is defined by the following equations (1) and (2): I sensor noise is the specific noise signal intensity value measured for each spectrometer.

[0011]

number

[0012]

number

[0013] The invention of claim 2 is a brewed alcohol inspection device that implements the brewed alcohol inspection method of claim 1, comprising a filtration unit that filters the sample of the brewed alcohol, an objective optical unit that receives the liquid components of the sample separated from the filtration unit, an analysis unit that performs Raman spectroscopic analysis of the liquid components, and a transmission unit that optically couples the objective optical unit and the analysis unit, wherein the wavelength of the excitation light in the Raman spectroscopic analysis of the liquid components is selected to be in a wavelength range in which the S / N ratio is 90% or more of the maximum value. Here, S / N is the signal intensity ratio I signal / I noise, A is a proportionality constant calculated from the laser intensity, polarizability, and component concentration, B i is the proportionality constant calculated from the fluorescence intensity, C is the speed of light, ν s is the frequency shift due to Raman scattering, λ L is the laser wavelength, λ s is the wavelength shift due to Raman scattering, σ i is the half-width of the fluorescence spectrum, and I signal and I noise is defined by the following equations (1) and (2): I sensor noise is the specific noise signal intensity value measured for each spectrometer.

[0014]

number

[0015]

number

[0016] The excitation light used in the present invention is on the shorter wavelength side than conventional light, which is prone to producing noise due to fluorescence and noise derived from the spectrometer. However, the wavelength range is such that the Raman signal intensity to noise intensity ratio S / N is 90% or more of the maximum value. Therefore, measurement sensitivity is not reduced by fluorescence noise or noise derived from the spectrometer, and moromi components can be analyzed optimally. [Brief explanation of the drawings]

[0017] [Figure 1] This is a graph showing the intensity of the Raman signal, noise (the sum of fluorescence and noise from the spectrometer), and S / N for each wavelength when a spectrometer with three types of noise is used in Raman spectroscopy of sake. [Figure 2] 1 is a schematic diagram of a brewed alcohol inspection device according to the present invention. [Figure 3] 1 is a graph showing the transition of the ethanol and glucose concentrations measured by Raman spectroscopy. [Figure 4]1 is a graph showing the transition of ethanol concentration determined by Raman spectroscopy and the distillation-density method. [Figure 5] 1 is a graph showing the error in ethanol measurement by Raman spectroscopy and the distillation-density method. [Figure 6] 1 is a graph showing the change in glucose concentration as determined by Raman spectroscopy. DETAILED DESCRIPTION OF THE INVENTION

[0018] (1. Selection of excitation wavelength range for Raman spectroscopic analysis of sake) Because sake is brewed through a complex process known as multiple parallel fermentation, in which starch saccharification and sugar alcoholic fermentation occur in the same system, feedback to the brewing process based on the amount of components in the mash is essential to producing high-quality sake. In particular, ethanol and glucose, which are components directly related to the fermentation process, which is the key to brewing, are important monitoring indicators.

[0019] However, component monitoring requires a large number of samples to be measured and each measurement takes a long time, so current measurement methods require workers to start work early in the morning, placing a significant burden on workers. Furthermore, even measuring a single sample requires traveling to an analysis room where the equipment is installed, which is time-inefficient. Therefore, if the components in the mash (especially ethanol and glucose) could be measured in a short time (targeting less than five minutes) right in the brewing tank, a significant reduction in the workload would be expected.

[0020] Currently, the most common method used to measure ethanol in mash is the distillation-density method. This requires a pretreatment process of filtering and distilling the mash to remove impurities before measuring density, and the entire process takes approximately one hour. The number of tanks in a brewery ranges from a few to several hundred, depending on the brewery's size, and each tank is typically analyzed once every two days. Meanwhile, glucose is measured using an enzyme-absorbance method, which also takes approximately 30 minutes. Thus, analyzing both ethanol and glucose is a lengthy process that takes more than an hour. Furthermore, because the analytical methods for each component are completely different, it is impossible for one person to measure them simultaneously. Most analytical methods designated by the National Tax Agency other than those mentioned above also involve chemical procedures, so it is difficult to say that any of these methods are currently time-efficient.

[0021] Therefore, in this invention, we aim to establish a method for measuring both ethanol and glucose in mash on-site in a short time by using Raman spectroscopy, which can directly measure molecules, with the aim of shortening daily work time and reducing the burden. This specification discloses the results of an analysis of ethanol in actual mash at a sake brewery, as well as the current progress on glucose analysis as an additional analysis.

[0022] Ethanol and glucose in the mash were measured by capturing the backscattered component of the linear spontaneous Raman spectrum generated from the position where the excitation light was irradiated using the same objective lens as that used for the excitation light irradiation, and then observing only the Stokes component as a Raman spectrum using a spectrometer.

[0023] Raman spectroscopy is a type of vibrational spectroscopy that utilizes the Raman effect. The Raman effect occurs when a material is exposed to an angular frequency ω i When monochromatic light of s is the same as the incident light (ω s =ω i ) component (Rayleigh scattering n), each frequency is shifted by Ω, s =ω i This is the phenomenon (Raman scattering) in which ±Ω components are generated. Therefore, the basic equation (3) holds true for Raman scattering.

[0024]

number

[0025] In addition, since the measurement environment in this experiment can be considered to be in thermal equilibrium, the energy levels of the molecules follow the Boltzmann distribution. Therefore, En = Em + hΩ occurs relatively predominantly, so in this measurement, ω s =ω i We decided to observe only -Ω.

[0026] The wavelength of the excitation light is also an important factor in the generation of Raman scattered light. In this measurement, the wavelength of the excitation light was selected taking into consideration the amount of Raman scattered light generated and the fluorescent substances in sake. Furthermore, quantitative values ​​were obtained by analyzing the highly independent peaks for the ethanol and glucose components in the measured Raman spectrum.

[0027] In Raman spectroscopy, the Raman scattered light intensity I s is the scattering angular frequency ω s , excitation light angular frequency ω i , speed of light c, KHD dispersion formula α ρσ , incident light intensity I i is expressed by the following equation (4) using the molecular weight A in the sample.

[0028]

number

[0029] Among these, the scattering angular frequency ω s is calculated from the excitation wavelength so that the equation is satisfied. i Therefore, in the measurement of vibrationally excited states targeted by Raman spectroscopy, s Since the shift amount of ω is small, s 3 ×ω i ≒ω i 4 Therefore, the Raman scattered light intensity I sis proportional to the fourth power of the frequency of the excitation light. Therefore, in order to increase the Raman scattering intensity (signal), it is better to select an excitation light wavelength on the short wavelength side.

[0030] However, samples containing organic matter generally generate fluorescence (noise), and as shown in Figure 1, the fluorescence excitation wavelength is often on the short wavelength side. noise Of these, this is the first term. Like most other products, sake contains fluorescent substances and has an excitation wavelength range on the short wavelength side. In other words, in order to obtain an appropriate Raman signal from sake, it is best to avoid the excitation wavelength range of the fluorescent substances in sake and select an excitation light wavelength on the short wavelength side.

[0031] Table 1 shows the fluorescent substances found in sake. It shows that the maximum excitation wavelength of the fluorescent substances found in sake is 450 nm, which is the wavelength of flavin compounds. The center wavelength of the fluorescent wavelength is 530 nm, and the range of wavelengths broadens to 650 nm. Therefore, to avoid the fluorescent band, it is best to select an excitation light wavelength of 650 nm or higher.

[0032] [Table 1]

[0033] As mentioned above, the excitation wavelength is selected to avoid fluorescence, but there is also noise from the spectrometer used in Raman spectroscopy. noise Of these, item 2, I sensor noiseThis is the case. Generally, cheaper spectrometers have higher noise, while more expensive ones are adjusted to minimize noise. Furthermore, when measuring over a short period of time, noise reduction through long exposure times cannot be expected, so spectrometer noise significantly affects the S / N ratio. Figure 1 shows a schematic diagram of the S / N ratio. Tests were conducted on spectrometers in three price ranges: a high-priced spectrometer S (low noise), a medium-priced spectrometer M (medium noise), and a low-priced spectrometer L (high noise). The vertical axis represents relative intensity and the horizontal axis represents wavelength. The more noise a spectrometer has, the more its noise is combined with the fluorescence noise, shifting the wavelength at which the S / N ratio is maximized toward shorter wavelengths (where fluorescence is present). Taking this into consideration, selecting an excitation wavelength at which the S / N ratio is 90% or greater of its peak value allows for the most accurate analysis of mash. The graph in Figure 1 shows the shaded area when the S / N ratio is 95% of its peak value.

[0034] Considering these factors, the optimal excitation wavelength range is approximately 730 to 860 nm when spectrometer noise can be largely ignored and only fluorescence noise can be considered. If a certain degree of spectrometer noise is acceptable, approximately 680 to 780 nm is preferred. If spectrometer noise is significant, approximately 640 to 720 nm is preferred. Thus, when using a spectrometer with significant noise, the excitation wavelength range shifts as shown in Figure 1. However, even in this case, by taking spectrometer noise into account and selecting a wavelength range that provides an S / N ratio of 90% or higher, it is possible to eliminate the effects of any spectrometer, allowing for the selection of a low-cost spectrometer, which accounts for a large portion of the cost of an inspection system.

[0035] (2. Selection of excitation wavelength range for general Raman spectroscopic analysis of brewed alcohol) While we explained above that the excitation wavelength range of approximately 730–860 nm is preferable for Raman spectroscopy of sake without considering spectrometer noise, this method can also be generalized to the brewing of other alcoholic beverages, such as shochu, awamori, beer, wine, and makgeolli. The schematic diagram in Figure 1, which uses sake as an example, can be obtained for each type of brewed alcohol. Depending on the dissolved components and solids present in the brewing system, different graphs of fluorescence noise and Raman signal intensities are drawn, and the resulting S / N curves are also obtained for each type of brewed alcohol. Using these S / N curves, the optimal Raman excitation wavelength range can be determined for each brewed alcohol by selecting the wavelength range where the intensity relationship defined by Equation (1) and Equation (2) is 90% or greater.

[0036] (3. Configuration of Brewed Alcohol Inspection Device) Figure 2 shows a schematic diagram of the configuration of a brewed alcohol inspection device 1 of the present invention. The inspection device 1 is equipped with an analysis unit 10 that has a laser and a spectrometer that perform the Raman spectroscopic analysis of the present invention. One end of a transmission unit 11, such as an optical fiber, is optically connected to the analysis unit 10, and the other end of the transmission unit 11 is optically connected to an objective optical unit 13, which is a sensor probe that contains the sample to be analyzed. The upper end of the objective optical unit 13 is equipped with a filtration unit 12 that separates the solid components of the sample and introduces only the liquid components into the objective optical unit 13.

[0037] Reference numeral 20 denotes a brewing vessel installed in a sake brewery or the like, and the brewing vessel 20 is filled with an alcoholic beverage being brewed, and in the case of sake brewing, reference numeral 21 denotes sake mash.

[0038] The filtration unit 12 and the objective optical unit 13 are installed in close proximity to the brewing vessel 20, allowing an operator to collect mash 21 from the brewing vessel 20 and immediately drop it into the filtration unit 12 for analysis. By configuring the transmission unit 11 to be short, the analysis unit 10 can be installed integrally with or in close proximity to the objective optical unit 13. Alternatively, by configuring the transmission unit 11 to be long, the analysis unit 10 can be installed in a remote location outside the sake brewery.

[0039] Although the interior of the analysis unit 10 is not shown, a known device configuration for Raman spectroscopy can be adopted. In one embodiment, a semiconductor laser provided in the analysis unit 10 is used to irradiate excitation light onto an objective lens incorporated in the objective optical unit 13 via the transmission unit 11. Scattered light is generated from the mash sample 21 and transmitted to the analysis unit 10 via the transmission unit 11. Of the scattered light transmitted to the analysis unit 10, Rayleigh scattered light is removed by a dichroic mirror, a bandpass filter, or the like. On the other hand, Raman scattered light passes through these filters and enters a spectrometer, where it is measured as a Raman spectrum.

[0040] Known filtering means can be used to filter the sample of mash 21. For example, the sample may be dropped onto a single-stage filter placed directly above the objective optical unit 13, or a two-stage filter in which large solid particles are removed using a relatively coarse-mesh filter cloth, followed by a filter with a desired pore size, e.g., 1 μm, to remove small solid particles, or even a three- or more-stage filter using a coarse-mesh filter cloth, a filter with a 5 μm pore size, and a filter with a 1 μm pore size, which are preferred because they significantly reduce the time and effort required for filtration.

[0041] The pore size of the final filter is preferably 1 μm or less, and more preferably 0.5 μm or less, to remove fine solid particles that affect Raman spectroscopy. Filtration can be performed by natural filtration at normal pressure, or by vacuum filtration or pressure filtration. For example, a two-stage filtration can be performed by connecting a filter unit containing filter paper to the tip of a syringe, drawing the filtrate into the syringe by pulling the plunger, and then replacing the filter unit with one having a pore size of 1 μm or less and pushing the plunger. [Example]

[0042] The present invention will be described in more detail below with reference to examples. (A. Measurement of mash during the brewing process using Raman spectroscopy) A semiconductor laser (Lambda mini FIBER, RGB Lasersystems) with a center wavelength of 662 nm and a maximum output of 50 mW was used as the excitation light for Raman scattering. The excitation light was irradiated at 5 mW (at the sample surface) through a 10x magnification, 0.3 NA, and 6.4 mm WD objective lens (Sigma Koki Co., Ltd.). Rayleigh scattering from the scattered light generated by the sample was removed using a dihedral mirror (FF677-Di, Semrock) and a bandpass filter (BLP01-664R, Semrock). Meanwhile, Raman scattering passed through these filters and was transmitted via an optical fiber to a spectrometer (HR-1 (custom range model), ASEQ Instruments) where it was measured as a Raman spectrum. One Raman spectrum was obtained by integrating six 10-s exposure times, and the accuracy was calculated from 10 measurements.

[0043] A total of 19 samples (from the 5th to 38th brewing days) of mash were collected from actual brewing tanks at Haginishiki Sake Brewery Co., Ltd. For Raman spectroscopy measurements, all samples were filtered after collection. The mash was filtered using 1 μm filter paper to remove most of the scattering particles, such as insoluble rice with large particle size and yeast and koji with small particle size. The samples were then sealed in a 3.5 ml quartz cell and measured.

[0044] The obtained Raman spectrum shows a baseline reflecting the fluorescence of the objective lens and the spectral transmittance of each element, as well as the Raman spectra of water, ethanol, and glucose. Table 2 shows the Raman peaks of ethanol and glucose.

[0045] [Table 2]

[0046] This time, we will compare the CCO asymmetric stretching vibration of ethanol and the (-CH2) of glucose. βTo quantify the out-of-plane vibrations, the baseline was removed using a fourth-order polynomial fitting based on five arbitrary measurement points, and peak fitting was performed on the target spectrum using the Breit-Wigner-Fano (BWF) function, after which the peak height was used as the quantitative value.

[0047] As a reference for quantifying the ethanol content in the mash, the quantitative value obtained by the distillation-density method currently used for analysis in sake breweries was used as an indicator. Measurements by the distillation-density method were conducted in accordance with the analytical method prescribed by the National Tax Agency, and the distillate extracted by distillation from 150 ml of mash taken from the brewing tank was used to obtain the quantitative value by reading the scale on a hydrometer.

[0048] Figure 3 shows the transition of component concentrations over the number of brewing days measured by Raman spectroscopy. These results show that the simultaneous parallel fermentation process of sake brewing, characterized by an increasing trend in ethanol and a decreasing trend in glucose, was successfully measured. In particular, the ethanol and glucose concentrations fluctuate due to the simultaneous parallel fermentation of koji and yeast in the early stages of brewing, and the changes in concentration due to the addition of water depending on the fermentation status from the middle to the end of the brewing process appear to be captured.

[0049] (B. Cost, time, and comparison with other methods) For this measurement, impurities in the collected mash were removed by filtration before measurement with a Raman spectrometer, and the entire analysis series, including the filtration time of 35 seconds (average) and measurement time of 1 minute, was completed within the target time of 5 minutes. Since the distillation-density method, which is currently the de facto method for ethanol analysis as shown in Table 3, takes 1 hour, we have succeeded in reducing the working time for ethanol analysis to approximately 1 / 12.

[0050] [Table 3]

[0051] (C. Quantitative Accuracy of Ethanol by Raman Spectroscopy) The results of ethanol measurement using Raman spectroscopy and the reference distillation-density method are shown in Figure 4. We confirmed that the accuracy of ethanol quantification in mash using Raman spectroscopy was extremely high, at 0.20% (σ).

[0052] Furthermore, the maximum error in the ethanol quantification values ​​obtained by comparing Raman spectroscopy and the distillation-density method, as shown in Figure 5, was 0.29%, which indicates that the ethanol in the mash can be quantified with sufficient accuracy, considering that the accuracy of the distillation-density method is estimated to be around 0.4%.

[0053] (D. Quantitative results of glucose by Raman spectroscopy) We also report the results of simultaneous measurements of glucose in the mash. Because no reference measurements were performed for glucose this time, we report only the quantitative accuracy of Raman spectroscopy. The measurement results are shown in Figure 6. These results demonstrate that although we were successful in capturing concentration changes due to the addition of water and fermentation from the middle to the end of brewing, the quantitative accuracy of glucose in the mash by Raman spectroscopy was 0.34% (σ), worsening for ethanol. Possible causes of this error include fluctuations in the excitation light output and insufficient optimization of the measurement point for baseline estimation. Furthermore, the glucose peak used for quantification was relatively smaller than the peak used for ethanol quantification, and the influence of fluctuations in the two adjacent peaks also contributed to the large error.

[0054] (E. Summary) In this study, we used a 662 nm excitation Raman spectroscopy system to successfully quantify the ethanol concentration of mash during the sake brewing process to an accuracy of 0.20% (σ). This measurement was achieved in a much shorter time, just 1 / 12 of the current ethanol analysis time. Furthermore, we confirmed that glucose analysis could also be simultaneously quantified with an accuracy of 0.34% (σ). Thus, we demonstrated that Raman spectroscopy analysis of mash can simultaneously quantify the concentrations of each component very quickly and with great accuracy. [Industrial Applicability]

[0055] This technology can be packaged into analytical equipment that can be used by sake brewers in actual breweries that do not have conventional ethanol and glucose analyzers installed. Furthermore, because of its ability to directly measure molecules, this Raman spectroscopy technology is not only useful for quantifying components in beverages other than brewed alcohol, but is also expected to contribute in the future to quantifying molecules in other samples, such as agricultural sensors that measure components in soil. [Explanation of symbols]

[0056] 1: Brewed sake inspection device, 10: Analysis section, 11: Transmission section, 12: Filtration section, 13: Objective optical section, 20: Brewing vessel, 21: Sake mash

Claims

1. A testing method for simultaneously quantifying ethanol and glucose in brewed alcoholic beverages during brewing, comprising: A sample was taken from the brewing vessel. performing solid-liquid separation of the sample to obtain a liquid component; When performing Raman spectroscopic analysis of the liquid component, A method for inspecting brewed alcoholic beverages, characterized in that the wavelength of the excitation light is selected from a wavelength range in which the S / N ratio is 90% or more of the maximum value. Here, S / N is the signal intensity ratio I ｓｉｇｎａｌ / I ｎｏｉｓｅ , A is a proportionality constant calculated from the laser intensity, polarizability, and component concentration, B ｉ is a proportionality constant calculated from the fluorescence intensity, C is the speed of light, ν s is the frequency shift due to Raman scattering, λ Ｌ is the laser wavelength, λ s is the wavelength shift due to Raman scattering, σ ｉ is the half-width of the fluorescence spectrum, and I ｓｉｇｎａｌ and I ｎｏｉｓｅ is defined by the following equations (1) and (2): ｓｅｎｓｏｒ ｎｏｉｓｅ is the specific noise signal intensity value measured for each spectrometer. [Equation 1] [Equation 2]

2. A brewed alcohol inspection device that implements the brewed alcohol inspection method according to claim 1, a filtration unit that filters the sample of brewed alcohol; an objective optical unit that receives the liquid component of the sample separated from the filtration unit; an analysis unit that performs Raman spectroscopic analysis of the liquid components; A brewed alcohol inspection device comprising a transmission unit that optically couples the objective optical unit and the analysis unit, and wherein, when performing Raman spectroscopic analysis of the liquid components, the wavelength of the excitation light is selected from a wavelength range in which the S / N ratio is 90% or more of the maximum value. Here, S / N is the signal intensity ratio I ｓｉｇｎａｌ / I ｎｏｉｓｅ , A is a proportionality constant calculated from the laser intensity, polarizability, and component concentration, B ｉ is a proportionality constant calculated from the fluorescence intensity, C is the speed of light, ν s is the frequency shift due to Raman scattering, λ Ｌ is the laser wavelength, λ s is the wavelength shift due to Raman scattering, σ ｉ is the half-width of the fluorescence spectrum, and I ｓｉｇｎａｌ and I ｎｏｉｓｅ is defined by the following equations (1) and (2): ｓｅｎｓｏｒ ｎｏｉｓｅ is the specific noise signal intensity value measured for each spectrometer. [Equation 1] [Equation 2]

Images