JP6683318B2 - Odor evaluation device - Google Patents

Description

  The present invention relates to an odor evaluation device for evaluating the odors (odors and odors) of various substances.

  BACKGROUND ART An odor measuring device that captures a certain substance or odor in the atmosphere from the viewpoint of its quality and strength, and digitizes and expresses it has been conventionally known (for example, Patent Document 1). With this odor measuring device, the concentration of a plurality of reference odor species (eg, butyric acid, ester, etc.) is changed in multiple stages in a multidimensional space formed by the detection outputs of multiple odor sensors, and measurement is performed. Position the result as the reference axis. Then, the measurement result of the object to be measured is positioned in the same space as the measurement point and the positional relationship with the reference axis, for example, the shortest distance between the measurement point and the reference axis or the line drawn from the origin to the measurement point and the reference point The angle formed with the axis is calculated, and the similarity of the odor quality is judged from now on.

  By the way, the fragrance added to foods and drinks, cosmetics, detergents, etc. is mainly made to resemble the fragrance (natural fragrance) emitted by natural products such as flowers, herbs, and fruits. Most natural fragrances have a complex odor in which various components are mixed, and have various odors depending on the types of components and the ratio of each component. However, not all of the components that make up the natural fragrance contribute to the formation of the odor, and the ratio of contribution is very small, or components that do not contribute at all are included. By removing such components that do not contribute to the odor formation of natural flavors, it is possible to obtain a flavorless fragrance, at the production site of the fragrance, among the components contained in the natural fragrance, the components that contribute to odor formation. There is a desire to specify.

  Omission test is one of the methods to identify the components that contribute to the formation of odor of natural fragrance. In the omission test, the components that make up the natural fragrance are identified by using an analyzer such as a gas chromatograph (GC device), and then all the standards of each component are mixed to reproduce the natural fragrance. The odor of the one obtained by removing a certain component from the odor and the odor of the reproduced fragrance are measured by the odor measuring device described above, and the results are compared. Then, when the odors are different from each other, it is determined that the components removed from the reproduced fragrance contribute to the odor formation of the natural fragrance. By sequentially performing an omission test on all the components that constitute the reproduced flavor, it is possible to identify all the components that contribute to the odor formation of the natural flavor.

JP 2003-315298 JP

  Considering the original purpose of identifying the components that contribute to odor formation, it is not necessary to identify the components that do not contribute to odor formation among the components included in the natural flavor, but in the method described above, all components included in the natural flavor are included. The components have to be identified, which is time consuming. In addition, it is not always possible to prepare samples for all components contained in natural flavors. In this case, since the natural fragrance cannot be reproduced, there is a problem that the above-mentioned omission test cannot be performed.

  The problem to be solved by the present invention is to make it possible to identify a component that contributes to odor formation without identifying all the components contained in the gas to be analyzed having an odor.

The odor evaluation device according to the present invention made to solve the above problems,
a) a gas chromatograph having a separation column that separates a plurality of components contained in the analysis target gas having an odor in the time direction,
b) detection means for detecting the timing at which each of the plurality of components emerges from the separation column,
c) Based on the detection result by the detection means, by passing the analysis target gas through the separation column a plurality of times, a sample gas that is a gas containing all the components that come out from the separation column, and from the sample gas Recovery means for recovering each of the omission gas excluding predetermined components,
d) an odor measuring means including m (m is an integer of 2 or more) odor sensors having different response characteristics for measuring the odor of gas;
e) The measurement result of the sample gas and the measurement result of the omission gas by the odor measuring means are represented as one odor vector in an m-dimensional space formed by the detection outputs of the m odor sensors, respectively. Arithmetic processing means for calculating an index value representing the similarity between the sample gas and the omission gas based on the positional relationship between the odor vector of the sample gas and the odor vector of the omission gas.

  In the odor evaluation apparatus, by passing the gas to be analyzed through the separation column a plurality of times, the sample gas containing all the components coming out from the separation column and the omission gas obtained by removing the predetermined components from the sample gas are separated. It is collected by the collecting means. Then, based on the measurement results of the odor measuring means of the sample gas and the omission gas, an index value representing the similarity between the two is calculated. At this time, when the odor of the sample gas and the odor of the omission gas have a low similarity, it means that the odor of the sample gas is significantly changed by removing the predetermined component. It can be determined that the gas, that is, the gas to be analyzed contributes to the odor formation. On the contrary, if the two have a high similarity, it is determined that the predetermined component does not contribute to the odor formation of the gas to be analyzed. can do.

In this case, when the sample gas contains two types of components, the sample gas containing both of the two types of components and the omission gas excluding the two types of components are collected. The same applies when the sample gas contains three or more components, but a gas obtained by removing two or more components from the sample gas may be recovered as the omission gas. For example, if it is found that one component does not contribute to the odor formation of the gas to be analyzed based on the similarity between the sample gas and the omission gas, that component and another component (components whose contribution to odor formation is unknown) Another omission gas is recovered by removing both of them from the sample gas.

  In the above-mentioned odor evaluation device, the recovery means comprises a plurality of recovery containers, a recovery flow path switching means for switching the recovery flow path for recovering the sample gas and the omission gas into the recovery container, and recovery in the recovery container. It is preferable to provide an introduction flow path switching means for switching the introduction flow path for introducing the sample gas and the emission gas thus obtained into the odor measuring means.

  With this configuration, the sample gas coming out of the separation column and one or a plurality of omission gases obtained by removing a predetermined component from the sample gas can be collected in a plurality of recovery containers, respectively. Then, it is possible to collectively judge which of the components contained in the sample gas contributes to the odor formation of the analysis target gas by sequentially introducing the odor gas collected in the plurality of collection containers into the odor measurement means. it can.

  In the above-mentioned odor evaluation device, the detection means only needs to be able to specify the timing at which the components contained in the gas to be analyzed come out of the separation column, so the flame flame ionization detection which is a standard detector used in gas chromatographs. Although an instrument (FID) can be used, the mass spectrometric means can identify each component together with the detection of the timing when each of the plurality of components emerges from the separation column.

  According to the present invention, it is not necessary to identify all the components contained in the gas to be analyzed, and even if there is no sample of all the components, the components contributing to the odor formation of the gas to be analyzed having an odor can be identified. can do.

The schematic block diagram of the odor evaluation apparatus which concerns on one Example of this invention. The schematic block diagram of an odor measuring part. The figure which shows the analysis result of the sample gas and omission gas which were introduce | transduced into the mass spectrometry part. The table which shows the output value of 10 odor sensors of sample gas and two types of omission gas. The figure which shows the odor vector showing the detection result of sample gas and two types of omission gas.

An odor evaluation device according to an embodiment of the present invention will be described below with reference to the drawings. FIG. 1 is a schematic configuration diagram of the odor evaluation device according to the present embodiment.
The odor evaluation device of the present embodiment is roughly divided into a gas chromatograph section (GC section) 1, a mass spectrometric section (MS section) 2, an odor measurement section 3, an interface section 4, and a gas recovery section 5. Consists of.

  The GC unit 1 includes a separation column 10, a column oven 11 that incorporates the column 10, a sample injection unit 12 provided at the inlet of the column 10, and a heart-cut flow path 13 provided at the outlet of the column 10. And a GC control unit 14 that controls these units. Although not shown, the heart-cut flow path 13 includes a valve and a drive motor for switching the valve.

  The MS unit 2 includes a vacuum container 20, an ion source 21 that ionizes sample component molecules in the introduced sample gas, an ion optical system 22 that transports the generated ions, and a mass that separates the ions according to their mass numbers. It includes a quadrupole mass filter 23 as a separation unit, an ion detector 24 that detects the mass-separated ions, and an MS control unit 25 that controls these units.

  The interface unit 4 is provided between the GC unit 1 and the MS unit 2 and includes a heater 41 that maintains the temperature of the conduit at a high temperature to facilitate the flow of the sample gas.

The gas recovery unit 5 is provided between the GC unit 1 and the odor measuring unit 3, and has a mounting opening 51 for mounting a plurality of sample bags SB (12 sample bags SB are shown in FIG. 1), and mounting. An auto sampler 54 including a lead-out inlet 52 for taking a sample gas in and out of a sample bag SB attached to the mouth 51, and a first flow passage switching unit 53 for switching a flow passage between the lead-out inlet 52 and the mounting opening 51. , A second flow path switching unit 55 that switches the flow path between the outlet port 52 and the odor measuring unit 3 and the GC unit 1, and a flow path control unit that controls the first flow path switching unit 53 and the second flow path switching unit 55. 56 and. The first channel switching unit 53 corresponds to the recovery channel switching unit of the present invention, and the second channel switching unit 55 corresponds to the introduction channel switching unit.

  As shown in FIGS. 1 and 2, the odor measuring unit 3 includes an inlet 31 for sucking the sample gas, a diluting unit 32 for diluting the sucked sample gas, a concentrating unit 33 for concentrating the sucked sample gas, and various types. A sensor cell 34 for measuring a sample gas containing an odorous component and having a plurality of (10 in this example) odor sensors 341a to 341j having different response characteristics, and a pump for drawing the sample gas into the sensor cell 34 35, an A / D conversion unit 36 that converts the detection signals from the odor sensors 341a to 341j into digital signals, a signal processing unit 37 that analyzes and processes the digitized detection signals, and an odor measurement that controls the overall operation of the odor measurement unit 3. The control unit 39 and the like are included.

The odor sensors 341a to 341j are generally metal oxide semiconductor sensors whose resistance value changes depending on the odor component, but in addition to this, a conductive polymer sensor, a quartz oscillator, or a gas on the surface of a SAW device. A sensor having another detection mechanism such as a sensor having an adsorption film may be used.
The signal processing unit 37 and the odor measurement control unit 39 are mainly configured by the personal computer 6, and by executing a predetermined program on the computer 6, a vector calculation unit 371, a dilution / concentration ratio calculation unit 372, and a storage unit described later. The function as the 373 and the similarity rate calculation unit 374 is achieved.

  In addition to the above, the personal computer 6 centrally controls the data processing unit 61 for analyzing the signal acquired by the ion detector 24 of the MS unit 2 and the control units 14, 25, 39, 56. A central control unit 62 for controlling is included as a function, and an input unit 63 such as a keyboard and a mouse and a display unit 64 are connected (see FIG. 1). In this embodiment, the MS unit 2 corresponds to the detecting means of the present invention. Further, the gas recovery unit 5 corresponds to a recovery means, and the sample bag SB corresponds to a recovery container. Further, the signal processing unit 37 corresponds to the arithmetic processing means.

  The diluting unit 32 includes a syringe 321 and a first valve (4 port 3 position) that selectively connects the suction / discharge port of the syringe 321 to any of the suction port 31, the nitrogen gas supply port 324, and the second valve 333. Valve) 323. The syringe 321 includes a cylinder 321a having a predetermined volume and a plunger 321b that is reciprocally fitted in the cylinder 321a. The plunger 321b is driven by a plunger drive unit 322 including a drive source such as a motor. In the diluting unit 32, the sample gas flowing from the suction port 31 to the sensor cell 34 can be diluted with nitrogen gas by performing the suction / discharge operation by the syringe 321 while switching the first valve 323. The dilution rate of the sample gas is determined by the amount of the sample gas sucked by the syringe 321 and the amount of nitrogen gas. The amount of sample gas and the amount of nitrogen gas sucked into the syringe 321 are determined by the driving amount of the plunger 321b by the plunger driving unit 322. Therefore, the odor measurement control unit 39 controls the drive of the plunger drive unit 322, so that the dilution rate of the sample gas introduced into the sensor cell 34 can be adjusted.

  The concentrating unit 33 includes a collection pipe 331 provided with a heater 332 for heating, a gas passage connected to one end of the collection pipe 331, the first valve 323, and the sensor cell 34. A second valve (3 port, 3 position valve) 333 for selectively connecting and a gas flow path connected to the other end of the collecting pipe 331 are alternatively connected to the nitrogen gas supply port 335 or the pump 336. And a third valve (3 port, 2 position valve) 334 for The collection tube 331 is filled with, for example, a carbon-based adsorbent or another appropriate adsorbent according to the sample component to be measured.

  When concentrating the sample gas in the concentrating unit 33, first, the second valve 333 and the third valve 334 are switched to the state shown by the solid line in FIG. 2, and the pump 336 is operated. Then, the gas supplied through the first valve 323 is discharged after passing through the collection pipe 331, and at this time, various components contained in the gas are adsorbed by the adsorbent in the collection pipe 331. After that, both the second valve 333 and the third valve 334 are switched to the state shown by the dotted line in FIG. Then, the dry nitrogen gas (carrier gas), which is supplied to the nitrogen gas supply port 335 at a high gas pressure (at least higher than the atmospheric pressure), flows to the sensor cell 34 via the collection pipe 331. In this state, when the temperature of the collection tube 331 is rapidly increased by energizing the heater 332, the odorous component adsorbed by the adsorbent is released from the adsorbent, and is carried on the dry nitrogen gas flow to the sensor cell 34. be introduced. In the concentrating section 33, the sample gas is determined by the total flow rate of the sample gas passing through the collection tube 331 at the time of adsorbing the component and the total flow rate of the carrier gas passing through the collection tube 331 at the time of desorption of the component in the range where the adsorption action is not saturated. The concentration rate of is determined.

  The odor measuring unit 3 measures the components of the sample gas as follows. That is, when the sample gas to be measured is introduced into the sensor cell 34, the components in the sample gas contact the odor sensors 341a to 341j, and different detection signals are output in parallel from the odor sensors 341a to 341j. The detection signal is sampled by the A / D conversion unit 36, digitized, and input to the signal processing unit 37. Since the signal processing unit 37 obtains one detection data for each odor sensor for one sample gas, a total of ten detection data DS1 to DS10 can be obtained by measuring a certain sample gas. Since the 10 odor sensors 341a to 341j have different response characteristics, it is possible to consider a 10-dimensional odor space in which the outputs of the 10 odor sensors 341a to 341j are axes in different directions. The state where the outputs of all the odor sensors 341a to 341j are zero is the origin of this odor space.

  In the odor space, the 10 detection data can be positioned as one measurement point (DS1, DS2, DS3, DS4, DS5, DS6, DS7, DS8, DS9, DS10). Here, when an odor vector having the origin of the odor space as the starting point and the measurement point as the end point is considered, the length of the odor vector is the “strength of odor” of the sample gas (that is, the odor component of the sample gas). The direction of the odor vector corresponds to the "quality of odor". That is, if the odor vector obtained by the measurement of a certain sample gas is oriented in a direction close to the odor vector obtained by the measurement of another sample gas, they can be considered as a similar type of odor, and vice versa. If the directions of the vectors are greatly different, it can be considered that the odor is a distant type. Therefore, as an index for determining the closeness of orientation of the two vectors, the angle θ formed by the two vectors is used, and the similarity of “smell quality” can be determined based on this angle θ. For example, when two odor vectors overlap (have the same direction) (that is, when θ = 0), the similarity rate is set to 100%, and when θ is equal to or larger than a predetermined value α, the similarity rate is 0%. Establish. Then, when θ is in the range of 0 to α, the similarity rate is defined according to the θ.

  Here, when the detection output level of the odor sensor is almost linear with respect to the concentration of the sample gas (concentration of the odor component), if the odor is of the same type, the direction of the odor vector is constant regardless of the concentration. Become. Therefore, the angle θ formed by the two odor vectors is constant regardless of the concentration, so that the difference in odor quality between a plurality of sample gases can be accurately determined. However, when the sensor output with respect to the concentration of the odor component is non-linear like the metal oxide semiconductor sensor generally used for the odor identifying device, even if the odor is of the same type, the direction of the odor vector varies depending on the concentration. Therefore, it becomes difficult to accurately discriminate the difference in odor quality between a plurality of sample gases.

  Therefore, in the odor evaluation apparatus according to the present embodiment, when the sample gas is measured, the dilution unit 32 and the concentrating unit 33 are feedback-controlled based on the output values of the odor sensors 341a to 341j and introduced into the sensor cell 34. Make sure that the concentration of the sample gas is always appropriate. Specifically, the detection signals DS1 to DS10 obtained from the 10 odor sensors 341a to 341j are positioned by the vector calculation unit 371 as one measurement point in the odor space described above, and the measurement point with the origin as the starting point. An odor vector whose end point is is created and the length of the vector is obtained. Then, the dilution rate in the dilution section 32 or the concentration rate in the concentration section 33 is controlled so that the length becomes a predetermined value.

Next, the operation of identifying the component that contributes to the odor formation of the gas to be analyzed using the above odor evaluation device will be described.
The analysis target gas such as a natural fragrance to be evaluated is stored in a liquid state (analysis target liquid), for example, in a sample bag and attached to the sample injection unit 12. Then, under the control of the GC control unit 14, when a small amount of the analysis target liquid is injected from the sample injection unit 12 into the sample vaporization chamber 12a at a predetermined timing, the analysis target liquid is vaporized in a short time. It becomes the gas to be analyzed, is pushed by the carrier gas, and is introduced into the column 10 from the sample vaporization chamber 12a. Each component contained in the gas to be analyzed is separated while passing through the column 10, and comes out of the column 10 with a time lag. Each component coming out of the column 10 is branched into two channels through the heart cut channel 13, one of which is introduced into the MS section 2 through the interface section 4 and the other of which is introduced into the gas recovery section 5.

  First, in order to check the timing when the components contained in the gas to be analyzed come out of the column 10, a gas (sample gas) containing all the components coming out of the column 10 is introduced into the MS section 2. That is, at this time, the heart cut flow path 13 is in a state in which the GC section 1 and the MS section 2 are in communication with each other, and the sample gas discharged from the column 10 is not recovered by the gas recovery section 5. As a result, in the MS unit 2, the odor components contained in the introduced sample gas are sequentially ionized by the ion source 21 and only the ions having the specific mass number selected by the quadrupole mass filter 23 are detected by the ion detector. Reach 24. Under the control of the MS controller 25, the quadrupole mass filter 23 repeatedly performs mass scanning in a predetermined mass range, and the ion detector 24 obtains a detection signal which is a source of a mass spectrum for each scanning.

  The detection signal obtained by the ion detector 24 is processed by the data processing unit 61 to repeatedly create a mass spectrum having the mass number on the horizontal axis and the signal intensity on the vertical axis, and without paying attention to the mass number. A total ion chromatogram (TIC) is created by plotting time on the horizontal axis and signal intensity on the vertical axis. Furthermore, a mass chromatogram is created by focusing on a certain mass number and setting the horizontal axis to time and the vertical axis to signal intensity. In order to detect the timing at which each component emerges from the column 10, it is sufficient to create a TIC, but a mass spectrum or mass chromatogram may be created if necessary. In the data processing unit 61, peaks are extracted from the created TIC and the time at which the peaks appear (peak time), that is, the timing at which each component appears is stored.

  Next, based on the peak time stored in the data processing unit 61, the central control unit 62 controls the drive motor of the heart-cut flow path 13 to generate the sample gas and the emission gas obtained by removing a predetermined component from the sample gas. It is introduced into the gas recovery unit 5. Specifically, when the sample gas is recovered in the gas recovery unit 5, the sample gas discharged from the column 10 when the analysis target gas is passed through the column 10 is all collected in the gas recovery unit 5 without heart-cutting. Introduce. When recovering the omission gas in the gas recovery unit 5, the gas to be analyzed is passed through the column 10 again, and heart-cutting is performed at a timing when a predetermined component comes out from the column 10 to remove the predetermined component. Is introduced into the MS part 2, and the other components are introduced into the gas recovery part 5. As a result, the sample gas and the emission gas are sequentially stored in the sample bag SB attached to the gas recovery unit 5.

  The peak time and the peak intensity are stored in the data processing unit 61 in association with each other, and only the components corresponding to the peaks are stored in the MS unit 2 in descending order of peak intensity, and the sample gas from which the components are removed The heart cut flow path 13 may be switched so that it is introduced into the gas recovery unit 5. Further, simply, in order of the peak time, only the component corresponding to the peak is introduced into the MS part 2, and the sample gas from which the component is removed is introduced into the gas recovery part 5, so that the heart cut flow path 13 is formed. You may make it switch.

When the collection of the omission gas excluding all the components contained in the gas to be analyzed one by one is completed, then the odor measuring unit 3 measures the odor of the sample gas and the omission gas (hereinafter referred to as measurement gas). To execute.
Specifically, under the control of the odor measurement control unit 39, the measurement gas is sequentially drawn into the sensor cell 34 from the sample bag SB of the gas recovery unit 5 by the pump 35. As a result, as described above, when the components contained in the measurement gas come into contact with the odor sensors 341a to 341j, the odor sensors 341a to 341j respectively output detection signals.

The data processing unit 61 creates an odor vector representing the detection result of each measurement gas in the 10-dimensional space based on the detection signal from the odor measurement unit 3. Then, using the odor vector of the sample gas as a reference vector, an index value indicating the degree of similarity between the odor vector of each omission gas and the reference vector is calculated, and the result is displayed on the display unit 64. At this time, the output values from the ten odor sensors 341a to 341j of each measurement gas may be displayed.
Further, the result obtained by the MS unit 2 and the result obtained by the odor measuring unit 3 may be displayed in association with each other. For example, the total ion chromatogram may be displayed in advance and the contribution of the peak component to the odor formation may be displayed in association with each peak appearing in the chromatogram.

Next, the operation of recovering the omission gas and the operation of determining the similarity between the sample gas and the omission gas will be described with reference to specific examples.
First, it is assumed that the TIC shown in FIG. 3A is obtained as a result of the sample gas containing all of the components coming out of the column 10 being introduced into the MS part 2. This TIC has four peaks from A to D, and these four peaks are A, C, D, and B when they are arranged in descending order of their intensities.

  Assuming that the omission gas removed from the sample gas in order from the component corresponding to the peak with the highest peak intensity is recovered, in the first omission gas recovery process, the heart-cut flow path is generated according to the timing at which peak A appears. The valve of No. 13 is switched, and only the component corresponding to peak A (component A) is introduced into the MS part 2, and the other component (“cut A” component) is introduced into the gas recovery part 5. At this time, the TIC shown in FIG. 3B is obtained from the result obtained by the MS unit 2 in which the component A is introduced.

In the second omission gas recovery step, the valve of the heart-cut flow path 13 is switched according to the timing at which the peak C appears, and only the component corresponding to the peak C (component C) is transferred to the MS section 2. The other components (“cut C” components) are introduced into the gas recovery section 5. At this time, the TIC shown in FIG. 3C is created from the result obtained by the MS unit 2 in which the component C is introduced.
Although not shown, the components corresponding to the peaks D and B are also similarly branched and introduced into the MS part 2 and the gas recovery part 5 to obtain the respective TICs.

When the recovery of the sample gas and the omission gas is completed, the sample gas and the omission gas are successively drawn into the sensor cell 34. As a result, the components contained in each gas come into contact with the ten odor sensors 341a to 341j, and the odor sensors respectively output detection signals. Based on these detection signals, the odor vector of each gas is converted into the data processing unit 61. Created by.
FIG. 4 shows the output values of the 10 odor sensors for the sample gas and the emission gas (“cut A” gas, “cut C ” gas). Further, FIG. 5 shows odor vectors of the sample gas and the omission gas (“cut A” gas, “cut C ” gas) obtained from these output values.

The data processing unit 61 obtains an angle formed by the odor vector of the sample gas and the odor vector of the omission gas as an index showing the similarity between the sample gas and the omission gas, and displays the angle together with the odor vector on the display unit 64. . For example, in the example shown in FIG. 5, the angle formed by the sample gas and the omission gas (“cut A” gas) is θA, and the angle formed by the sample gas and the omission gas (“cut C ” gas) is θC, (θA < θC). From this result, the user can determine that the component C has a greater contribution to the odor formation of the gas to be analyzed than the component A.

In addition, here, the detection results of the odor sensors 341a to 341j of the sample gas and the omission gas are represented as odor vectors, and the angles formed by these are used as the index values indicating the similarity, but the sample gas is represented as the odor vector, For the omission gas, the output values of the 10 odor sensors may be represented as one measurement point having coordinate components. Then, the distance between the measurement point and the odor vector of the sample may be used as an index value indicating the degree of similarity.
Also, the omission gas except the component corresponding to the peak is collected in the order of the peak intensity, but the omission gas excluding the component corresponding to the peak may be collected in the order of the peak appearance. good.

  Furthermore, in the above-mentioned embodiment, the one in which only one component is removed from the sample gas is used as the omission gas, but the one in which a plurality of components are removed may be used as the omission gas. For example, if it is found that the odor of the sample gas and the omission gas have a high degree of similarity and a component that is not contained in the omission gas does not contribute to the formation of the odor of the gas to be analyzed, that component and another component ( Omission gas may be obtained by removing both components (whether or not it contributes to odor formation). In this case, if there are a plurality of components that are known not to contribute to the formation of odors, those excluding the plurality of components (that is, three or more components) may be used as the omission gas.

Further, the heart-cut flow path 13 can be configured to include a pressure sensor that measures the pressure at the outlet of the column 10, a valve, and a flow controller (APC) that switches the valve based on the measurement value of the pressure sensor. When a certain component contained in the gas to be analyzed comes out of the column 10, the outlet pressure of the column 10 rises. Therefore, in this configuration, based on the measured value of the pressure sensor exceeding the preset threshold value, the valve is switched to heart-cut a certain component in the MS unit 2 and the remaining component in the gas recovery unit 5. Introduce. Which of the components contained in the gas to be analyzed is heart-cut (that is, at which timing the valve is switched) can be determined in the same manner as in the above embodiment.
Furthermore, the heart cut 13 can be configured to include a flow rate sensor that measures the flow rate at the outlet of the column 10, a valve, and a flow controller (AFC) that switches the valve based on the measurement value of the flow rate sensor. . In this configuration, the valve is switched based on the outlet flow rate of the column 10 exceeding the preset threshold value.

  Further, although the mass spectrometric means (MS unit 2) is adopted as the detection means in the above-mentioned embodiment, the present invention is not limited to this, and a hydrogen flame ionization detector may be used as the detection means. Furthermore, although the sample separated into components in the GC section 1 was introduced in parallel to the MS section 2 and the odor measuring section 3, the odor evaluator directly sniffed the odor to input information such as intensity. You may provide a smelling part. Also, in other respects, it is clear that appropriate changes and modifications can be made within the scope of the present invention.

1 ... Gas chromatograph section 10 ... Column 13 ... Switching valve (heart-cut flow path)
14 ... GC control part 2 ... Mass spectrometric part 3 ... Odor measuring part 31 ... Suction port 32 ... Diluting part 33 ... Concentrating part 34 ... Sensor cells 341a-341j ... Odor sensor 35 ... Pump 36 ... A / D conversion part 37 ... Signal processing Part 373 ... Storage part 39 ... Odor measurement control part 4 ... Interface part 5 ... Gas recovery part 51 ... Mounting port 52 ... Outlet port 53 ... First flow path switching part 54 ... Auto sampler 55 ... Second flow path switching part 56 Control unit 6 Personal computer 61 Data processing unit 62 Central control unit 63 Input unit 64 Display unit

Claims (2)

A gas chromatograph having a separation column that separates a plurality of components contained in an analysis target gas having an odor in a time direction,
Detection means for detecting the timing at which each of the plurality of components emerges from the separation column;
Based on the detection result by the detection means, by passing the analysis target gas through the separation column a plurality of times, a sample gas that is a gas containing all the components that come out of the separation column, and a predetermined amount from the sample gas. Recovery means for recovering each of the omission gas excluding the components,
An odor measuring means including m (m is an integer of 2 or more) odor sensors having different response characteristics for measuring the odor of gas,
The measurement result of the sample gas and the measurement result of the omission gas by the odor measuring means are represented as one odor vector in an m-dimensional space formed by the detection outputs of the m odor sensors, respectively. An arithmetic processing means for calculating an index value representing the similarity between the sample gas and the omission gas, based on the positional relationship between the odor vector of the odor vector and the odor vector of the omission gas,
The detection means is a mass spectrometer,
The recovery means comprises a plurality of sample bags, and a recovery flow path switching means for switching the recovery flow path so as to recover the sample gas and the omission gas into different sample bags,
The omission gas from said sample gas, Ri gas der the component is removed emerging from the separation column to the time peak appears in the mass chromatogram obtained by the mass spectrometer,
A component that comes out of the separation column at the time when the peak of the mass chromatogram obtained by the mass spectrometer appears and an omission gas from which the component has been removed from the sample gas are branched into two different flow paths. Characterized by comprising a branching means,
Odor evaluation device. The recovery means further comprises an introduction flow path switching means for switching an introduction flow path for introducing the sample gas and the omission gas collected in the sample bag into the odor measuring means, Item 1. The odor evaluation device according to item 1 .

Images