JP3931124B2 - Volatile compound identification device and method for identifying the compound - Google Patents

Description

図１３によれば、サンプル濃度に対して応答パターンは変化している。濃度レベルを揃えると応答パターンは変化しないという予測に反するものであった。これは従来型の濃縮管の場合には脱着時定数が大きいためと推量された。これは時定数が大きく、センサに到達するガス濃度の波形が、低濃度と高濃度で異なるためと推量された。
【００１６】
上記した従来型濃縮管で脱着時定数が大きくなる要因としては、チュープ形状濃縮管では吸着剤が円柱状に配置されているため匂いの脱着に時間的な幅ができること、内径が小さいためにキャリアガスの流量に制限があること、濃縮管の熱容量が大きく加熱と冷却に時間がかかること、濃縮管とセンサセルとの間での移動遅れによる影響があるものと推量した。
【００１７】
本発明者らは、鋭意検討に努めた結果従来型濃縮管の有する問題点を解決するものとして、下記するようなセンサセル直結型濃縮素子を提案するに至った。
【００１８】
図２は本発明の一例を示す濃縮素子３５の説明図を、図３は本発明のセンサセル直結型濃縮素子の一例を示す概略断面図である。漏斗３１は匂い成分等を含むガスの濃縮素子３５への供給口である。漏斗３１の上部には、網３２、通常金属製の網、好ましくはステンレススチール製の網を少なくとも中央部分に適宜の大きさで設置する。該網の上部に漏斗の拡大部よりも大きい面積の板状材３４を配置し、該板状材の中央部に貫通口を設け、該貫通口に吸着剤３３を充填し、該吸着剤の上部に網３２’を配置する。貫通口に両面を網で挟置された吸着剤を設置してもよい。吸着剤としてはTenax-TA（Alltech社製）のような疎水性吸着剤が好ましく使用される。
【００１９】
図３の場合は、濃縮素子３５を中央部の底面に配置するように設置した断面四角形状の底のない容器形状のセンサセル３６が設置されている。センサセル３６の上部には、漏斗３１の入り口３８から供給された匂い成分を含むガスが、濃縮素子３５を経て排出される開放部３９が設けられている。センサセル３６の側壁には水晶振動子ガスセンサ３７が設置されている。
【００２０】
本発明のセンサセル直結型濃縮素子を使用すると、従来提案されてきた濃縮管により生ずる問題点をいずれも解決できることが明らかとなった。以下本発明により奏される作用効果を例を挙げて説明する。
【００２１】
図４は、従来の濃縮管と本発明のセンサセル直結型濃縮素子を使用した場合の最大センサ応答値からベースラインまでの応答回復波形を比較したグラフである。吸着剤としては疎水性吸着剤であるTenax-TA（粒径２０／３０mesh）を、感応膜にはInositolを使用し、匂い成分を含むガスの流速を２００ml/min（濃縮管）、１００ml/min（センサセル直結型）とし、匂い成分として２−ヘキサノンを使用した。２−ヘキサノンに対する最大センサ応答値を１として規格化したときの各々の回復波形を示した。センサセル直結型の回復速度は濃縮管の場合に比して速く、匂いの脱着時定数は小さくなったことがわかる。
【００２２】
図５は、センサセル直結型濃縮素子を使用した場合の応答パターンの匂い濃度依存性を示したグラフである。試料としてブチルアセテート、アミルアセテート及びヘキシルアセテートを用い、吸着剤の濃縮時間を３〜１１秒まで２秒毎に５段階に変え、水晶振動子（以下ＱＣＭともいう）として２８MHzの金電極、感応膜としてInositol, Versamid 900（信和化工社製）、ガス流量２００ml/minで測定した結果である。横軸にInositolの応答ピーク値、縦軸にVersamid900の応答ピーク値を取って示した。濃度の異なる複数の測定点が原点を通る一直線上に並ぶ場合には、各センサの応答比は変わらず、応答パターンは変化しない。図中○印で囲んだ測定点での応答パターンはほぼ等しく、この測定点を利用しての匂い識別は不可能であった。なお、以下に説明するいずれの実験においても、吸着剤としてはTenax-TAを使用した。
【００２３】
図６は、センサセル直結型濃縮素子を用いた場合の一定応答レベル測定を行った結果を示すグラフである。水晶振動子として２８MHz金電極、サンプルガス流量２００ml/minで測定した。プロピオン酸をODOで４段階（原液、１：１、１：３、１：７）に希釈したものを匂いサンプルとし、吸着剤の濃縮時間を調節してinositolのセンサ応答を約３００Hzに揃えたときのinositolとVersamid900の規格化応答パターンを示す。図６の横軸は原液の希釈度と濃縮時間を示す。この場合匂い濃度に対してセンサ応答パターンは一定となっている。濃縮管の場合には脱着時定数が大きく、応答レベルを一定にした場合でも応答パターンが変化し、濃度レベルと応答レベルが一致しない。則ち、一定応答レベル測定を効果的に実施するためには、脱着時定数の小さいセンサセル直結型濃縮素子を用いる必要がある。
【００２４】
図７は水に対する希釈率を３段階（１：４、１：９、１：１９）に変えた匂いサンプル（propionic acid）について、センサセル直結型濃縮素子を用いて一定応答レベル測定を行った結果を示すグラフである。測定は各希釈率に対してInositolの応答レベルが約１４０Hzとなるように濃縮時間を変化させた。横軸は希釈率と濃縮時間を示す。希釈率が異なっても応答パターンはほぼ一定であった。
【００２５】
なお、一定応答レベル測定を行うためには、ヒータ加熱後のセンサの最大応答値が設定応答レベルの許容範囲に達するまで濃縮時間を変更して行う。この場合、濃縮時間に対してセンサ応答が比例する領域で有効であるから、初期設定濃縮時間は上記領域内に入るように設定する必要がある。本実施例では、この最大繰り返し回数は５回で、サンプルとヒーター加熱時間は３０秒〜１分間低度である。
【００２６】
上記してきたように、一定応答レベル測定を有効に実施するためには、脱着時定数の小さいセンサセル直結型濃縮素子が有用であり、湿度や匂い濃度の変動があっても、有効な匂い識別を実施するためには、センサセル直結型濃縮素子と一定応答レベル測定の併用が有効である。
【００２７】
次に、一定応答レベル以外に複数の応答レベルで応答パターンを測定し、これらの測定パターンから匂いを識別する方法について説明する。なお複数の応答レベルは一定応答レベルよりも短い濃縮時間で測定される。ただし、複数の応答レベルも濃縮時間とセンサ応答が比例する領域に入るように設定する。この方法による測定の所要時間は１サンプル当り５〜１０分である。
【００２８】
図８は、下記サンプルの第１応答レベルでの応答パターンを規格化した応答パターンの散布図を示す。
【００２９】
測定サンプルとして、各２mlの下記サンプルを使用した。
【００３０】
trans-2-hexyl acetate(原液、1:3,1:7)
trans-2-hexenal(原液、1:3,1:7)
ethyl valerate (原液、1:3,1:7)
isobutyric acid(原液、1:3,1:7)
（）内の数値はODOに対する希釈率（V/V）を示す。
【００３１】

Inositolの設定第１応答レベルを１５０Hzとした。４種類のサンプルはtrans-2-hexenalとisobutyric acidで一部オーバーラップしているが、全体的には全てのサンプルが分離されている。本図によれば、匂い濃度変動がある環境下でも本発明の濃縮素子を使用したガスセンサを用いることにより、一定応答レベルで多種類の匂いを匂い濃度依存性なく識別が可能である。なお、図中匂い濃度は原液から希釈率の小さい順にプロットの大きさを大きくした。
【００３２】
図９は、図８の場合において、第１、第２、第３応答レベルの応答パターンを識別に用いた結果を示す図面である。なお、第２レベルは一定応答レベルで決定された濃縮時間の４分の３、第３レベルは２分の１の濃縮時間で測定された応答パターンである。図中濃度の大きいもののプロットを大きく表示した。
【００３３】
式（１）で規格化した第３応答レベルまでの２次元規格化応答パターンを合わせた６次元ベクトルを用いて主成分分析を行った。図８でオーバーラップの見られた２種のサンプルについてもオーバーラップはなく、４種類のサンプルの匂いが明確に識別できた。図８、図９において、プロットの大小はサンプル濃度の大小を示すが、匂い濃度変動の影響も全く認められなかった。なお、図９横軸及び縦軸の括弧内の数値は、元の情報に対する寄与率を意味する。
【００３４】
図１０は、低濃度に希釈したpropionic acidとisobutyric acidについて、第１応答レベルでの各サンプルの規格化応答パターンの散布図を示す。プロットの大小は濃度の大小を示す
測定サンプル
propionic acid（１０％、１％水溶液）とisobutyric acid（１０％、１％水溶液）各２ml、
キャリアガス流量：２００ml/min.
加熱時間と電流値：１．０秒−６．０A
設定応答レベル：１５０±１５Hz
初期濃縮時間：５秒
サンプルの種類によりisobutyric acidの一点を除きプロットは大きく分離していた。しかし、２つのカルボン酸の応答パターンは類似しており、精度よく匂いを識別するためには、第１応答レベルのみでは不十分であった。
【００３５】
図１１は、第１応答レベルと第３応答レベルの２次元規格化パターンを合わせた４次元ベクトルを用いて主成分分析を行った結果を示すグラフである。この結果は濃度変動がある２種の水溶液試料を明確に区別できることを示している。また複数レベルで測定することにより、濃度の異なる類似した匂いの分離が可能となることを示している。
【００３６】
【発明の効果】
本発明のセンサセル直結型濃縮素子によれば、湿度の影響を除去でき、脱着時定数が小さく、装置全体を小型化できる等の優れた利点を有する揮発性化合物の識別装置及び識別方法が提供される。
【００３７】
本発明の上記濃縮素子を使用した一定応答レベルあるいは該レベル以外に複数の応答レベルを使用した測定法を採用することにより、湿度変動の影響、匂い濃度変動の影響を排除した揮発性化合物の識別方法が提供される。
【図面の簡単な説明】
【図１】 本発明の匂い識別システムの概略説明図である。
【図２】 本発明の濃縮素子の説明図である。
【図３】 本発明のセンサセル直結型濃縮素子の説明図である。
【図４】 応答回復波形を示すグラフである。
【図５】 応答パターンの匂い濃度依存性を示したグラフである。
【図６】 一定応答レベル測定を行った結果を示したグラフである。
【図７】 湿度を付加させた時の一定応答レベル測定結果を示したグラフである。
【図８】 第１応答レベルでの応答パターンの散布図である。
【図９】 第１から第３応答レベルを用いた応答パターンの散布図である。
【図１０】 水溶液試料に対する第１応答レベルの応答パターンの散布図である。
【図１１】 水溶液試料に対する第１及び第３応答レベルの応答パターンの散布図である。
【図１２】 直接加熱型濃縮管の説明図である。
【図１３】 直接加熱型濃縮管を使用した場合の一定応答レベル測定結果を示したグラフである。
【符号の説明】
１：バイアル瓶
８：濃縮素子
９：センサセル
２２：電源
２３：濃縮管
２４：吸着剤
３３：吸着剤
３５：濃縮素子
３６：センサセル
３７：水晶振動子ガスセンサ[0001]
BACKGROUND OF THE INVENTION
The present invention relates to a volatile compound identification device and identification method using a novel concentration element. More specifically, the present invention relates to a volatile compound identification device and identification method that eliminates the influence of volatile compound concentration fluctuations or interfering substances.
[0002]
[Prior art]
In recent years, the importance of volatile substances such as odor information is increasing in the food / cosmetics industry and environmental measurement. Traditionally, odor quality control and inspection have relied mainly on sensory testing by experts. However, there is a problem in reproducibility in the sensory test. In living organisms, it is said that odors are identified by pattern recognition of response patterns from many olfactory cells with different characteristics. Therefore, an odor discriminating apparatus using a quartz vibrator gas sensor array simulating such a living body olfactory mechanism has been proposed (Japanese Patent Laid-Open No. 1-244335). The proposed scent identification device, when combining multiple types of receptors with different sensitivities to substances exhibiting a predetermined scent, the detection information obtained from this difference in sensitivity It is a device proposed by paying attention to the fact that it varies from body to body and shows a certain pattern as a whole.
[0003]
However, in general, the concentration of odors contained in the atmosphere is low, and it is usual that interfering gases such as water vapor are mixed. In this case, fluctuations in temperature, odor concentration, humidity, and the like become disturbances and affect the identification accuracy.
[0004]
In order to exclude the influence of humidity, a method has been proposed in which a scent component and humidity are separated using a hydrophobic concentrating tube to separate a lean scent component (WAGroves et al. Anal. Chem. Acta. 371). , (1998) 131). The concentration tube can concentrate a dilute odor component.
[0005]
The odor molecules adsorbed on the adsorbent can be desorbed by applying a heat pulse, and if an appropriate desorption temperature is selected, the odor discrimination ability can be improved by applying a heat pulse at a different temperature (T. Nakamoto et al. al. Sensors & Actuators B, 71 (2000) 155-160). However, the conventional concentration tube has a large desorption time constant, and the dependency of odor concentration variation cannot be eliminated.
[0006]
[Problems to be solved by the invention]
An object of the present invention is to provide an identification device and an identification method capable of effectively identifying an odor component even when the odor concentration varies.
[0007]
It is another object of the present invention to provide an identification device and an identification method that can effectively identify an odor component even in a high-humidity environment or when there is a variation in humidity and the odor concentration varies.
[0008]
It is another object of the present invention to provide an odor discriminating apparatus and a discriminating method that are resistant to fluctuations in odor concentration even in a high humidity environment by increasing the speed of the adsorption / desorption cycle and making the sensor response level constant.
[0009]
[Means for Solving the Problems]
In the present invention, a volatile compound is adsorbed on a plurality of quartz vibrator gas sensors whose electrode surfaces are coated with a volatile compound sensitive film, and a change in the resonance frequency of each quartz vibrator due to the adsorption is recognized by pattern recognition. In an apparatus for identifying a sex compound,
The apparatus includes a concentration element of a volatile compound and a plurality of quartz vibrator gas sensors, and the concentration element includes a net and an adsorbent packed in a flat plate sandwiched between the nets, and the adsorbent A heating means is provided on one side or both sides of the mesh, and the quartz resonator gas sensor is provided at a short distance from the concentrating element so as to be in direct contact with the volatile compound desorbed from the concentrating element. A volatile compound identification device is provided. In the present invention, the direct contact means that the concentrating element and the crystal resonator gas sensor are integrated, and the volatile compound desorbed from the concentrating element is not brought into contact with the crystal resonator gas sensor by a pipe or the like. Means you can.
[0010]
Furthermore, the present invention allows a volatile compound to be adsorbed to a plurality of quartz resonator gas sensors whose electrode surfaces are coated with a sensitive film of a volatile compound, and changes in the resonance frequency of each quartz resonator due to the adsorption are pattern-recognized. In a method for identifying volatile compounds,
The method includes an adsorbent packed in a flat plate sandwiched between nets and provided with a heating element on one or both sides of the adsorbent, provided at a distance close to the quartz crystal gas sensor. It is characterized in that the concentration time to the adsorbent or the dilution rate of the carrier gas is made variable, and the response pattern obtained with the response of the crystal resonator gas sensor as a constant level is recognized by pattern recognition to identify volatile compounds. It provides a method for identifying and / or identifying volatile compounds.
[0011]
The present invention further includes adsorbing a volatile compound to a plurality of quartz resonator gas sensors whose electrode surfaces are coated with a sensitive film of a volatile compound, and measuring a change in the resonance frequency of each quartz resonator due to the adsorption. In a method for identifying a sex compound,
The method includes an adsorbent packed in a flat plate sandwiched between metal meshes and provided with a heating means on one or both sides of the adsorbent provided at a distance close to the quartz resonator gas sensor. Using the element, the concentration time to the adsorbent or the dilution rate of the carrier gas is variable, and the response pattern of the volatile compound is measured with the response of the crystal resonator gas sensor as a constant level. The present invention provides a method for identifying a volatile compound, characterized in that a response pattern is measured at a response level, and the response pattern is recognized to identify a volatile compound.
[0012]
DETAILED DESCRIPTION OF THE INVENTION
FIG. 1 is a schematic explanatory diagram of an odor identification system used in the odor identification apparatus or identification method of the present invention. A plurality of vials 1 containing the liquid sample 2 are arranged in a thermostatic chamber 3. The flow rate of the dry air 4 that is the carrier gas is controlled by the mass flow controller 5. A predetermined flow rate of air is supplied into the vial 1 through the electromagnetic valve 6 to extrude the headspace gas, and the gas reaches the concentrating element 8 through the electromagnetic valve 6 ′. The concentrating element is filled with a hydrophobic adsorbent, where odor components and water vapor are separated. Desorption of odor molecules adsorbed on the adsorbent is performed by applying a heat pulse. The desorbed odor component reaches the sensor cell 9, and the oscillation frequency of the sensor gas crystal sensor changes. This frequency change is measured by a reciprocal frequency counter 10 and loaded into the computer 12. When one sample measurement is completed, the XYZ stage 7 moves to the next vial and performs the next measurement operation. If these operations are controlled by a computer, a large number of samples can be automatically measured.
[0013]
FIG. 12 is a drawing showing the structure of a direct heating type concentrating tube that has been proposed as a conventional concentrating tube. A gas containing an odor component is supplied from a pipe 21, preferably a pipe 21 made of Teflon tube (Teflon: registered trademark), to the concentrating tube 23 through a connector 25. The supplied gas is adsorbed by the hydrophobic adsorbent 24 filled in the concentrating tube 23, and concentration of odor molecules and removal of moisture and interfering gases are performed. The concentrating tube is directly heated by applying a current pulse from the power source 22 to the wall surface of the concentrating tube. However, when such a concentration tube is used, there is a problem that the concentration waveform of the odor that reaches the crystal resonator gas sensor differs depending on the sample concentration, and the concentration level and the response level do not match. However, even if a direct heating type concentrating tube is used, if the concentration level is changed by changing the concentration time, the odor concentration supplied to the quartz resonator gas sensor becomes equal, so the response pattern is expected to remain unchanged. It was done.
[0014]
Figure 13 shows odor samples prepared by diluting 2-hexanone with odorless diluent medium-chain fatty acid triglyceride (hereinafter referred to as ODO) in three stages, ie, undiluted solution, 1: 3, 1: 7. The normalized pattern of the adjusted sensor peak response is shown. This normalization pattern shows the response pattern of the sensitive membrane Ucon75HB-9000 (hereinafter referred to as Ucon90000, manufactured by Shinwa Chemical Industries) and Inositol when the sensor peak response of phosphatidyl inositol (hereinafter referred to as Inositol) is aligned at about 80 Hz. Normalization was performed according to the following formula.
[0015]
[Expression 1]

According to FIG. 13, the response pattern changes with respect to the sample concentration. This is contrary to the prediction that the response pattern does not change when the concentration levels are made uniform. This is presumed to be due to the large desorption time constant in the case of the conventional concentrating tube. This is presumed to be because the time constant is large and the waveform of the gas concentration reaching the sensor differs between low and high concentrations.
[0016]
The factors that increase the desorption time constant in the above-mentioned conventional concentrating tube are as follows. In the tube-shaped concentrating tube, the adsorbent is arranged in a columnar shape, so that the odor can be desorbed in time, and the inner diameter is small. It was inferred that there was a restriction on the gas flow rate, the heat capacity of the concentrating tube was large, and it took time to heat and cool, and that there was an influence due to the movement delay between the concentrating tube and the sensor cell.
[0017]
As a result of diligent investigations, the present inventors have proposed a sensor cell direct-coupled concentration element as described below as a solution to the problems of the conventional concentration tube.
[0018]
FIG. 2 is an explanatory diagram of the concentration element 35 showing an example of the present invention, and FIG. 3 is a schematic cross-sectional view showing an example of the sensor cell direct-coupled concentration element of the present invention. The funnel 31 is a supply port for the gas concentrating element 35 containing odor components and the like. On the upper part of the funnel 31, a mesh 32, usually a metal mesh, preferably a stainless steel mesh, is installed at an appropriate size at least in the central portion. A plate-like material 34 having an area larger than the enlarged portion of the funnel is disposed at the top of the net, a through-hole is provided in the center of the plate-like material, and the adsorbent 33 is filled in the through-hole. A net 32 'is placed on the top. You may install the adsorption agent by which both surfaces were pinched | interposed into the through-hole with the net | network. As the adsorbent, a hydrophobic adsorbent such as Tenax-TA (manufactured by Alltech) is preferably used.
[0019]
In the case of FIG. 3, a sensor cell 36 having a rectangular shape with no bottom and having a quadrangular section is installed so that the concentrating element 35 is disposed on the bottom surface of the central portion. In the upper part of the sensor cell 36, there is provided an open part 39 through which the gas containing the odor component supplied from the inlet 38 of the funnel 31 is discharged through the concentration element 35. A quartz oscillator gas sensor 37 is installed on the side wall of the sensor cell 36.
[0020]
It has been clarified that the use of the concentrating element directly connected to the sensor cell of the present invention can solve all the problems caused by the conventionally proposed concentrating tube. Hereinafter, the operation and effects achieved by the present invention will be described with reference to examples.
[0021]
FIG. 4 is a graph comparing response recovery waveforms from the maximum sensor response value to the baseline when the conventional concentrating tube and the sensor cell direct connection type concentrating element of the present invention are used. As the adsorbent, Tenax-TA (particle size 20/30 mesh), which is a hydrophobic adsorbent, is used as the sensitive membrane, and the flow rate of gas containing odor components is 200 ml / min (concentration tube), 100 ml / min. (Sensor cell direct connection type) and 2-hexanone was used as an odor component. Each recovery waveform when the maximum sensor response value with respect to 2-hexanone is normalized as 1 is shown. It can be seen that the recovery rate of the sensor cell direct connection type is faster than that of the concentration tube, and the odor desorption time constant is smaller.
[0022]
FIG. 5 is a graph showing the odor concentration dependence of the response pattern when a sensor cell direct-coupled concentration element is used. Using butyl acetate, amyl acetate, and hexyl acetate as samples, the concentration time of the adsorbent was changed from 5 to 11 every 2 seconds from 3 to 11 seconds, and a 28 MHz gold electrode and a sensitive film as a quartz oscillator (hereinafter also referred to as QCM) As a result of measurement with Inositol, Versamid 900 (manufactured by Shinwa Kako Co., Ltd.) and gas flow rate of 200 ml / min. The horizontal axis represents the response peak value of Inositol, and the vertical axis represents the response peak value of Versamid 900. When a plurality of measurement points having different densities are arranged on a straight line passing through the origin, the response ratio of each sensor does not change and the response pattern does not change. The response patterns at the measurement points surrounded by circles in the figure are almost equal, and odor discrimination using this measurement point was impossible. In all the experiments described below, Tenax-TA was used as the adsorbent.
[0023]
FIG. 6 is a graph showing the results of performing a constant response level measurement when using a sensor cell direct-coupled concentration element. The measurement was performed using a 28 MHz gold electrode as a crystal resonator and a sample gas flow rate of 200 ml / min. Propionic acid diluted with ODO in four stages (stock solution, 1: 1, 1: 3, 1: 7) was used as an odor sample, and the concentration of the adsorbent was adjusted to adjust the sensor response of inositol to about 300 Hz. The standardized response pattern of inositol and Versamid900 is shown. The horizontal axis of FIG. 6 shows the dilution and concentration time of the stock solution. In this case, the sensor response pattern is constant with respect to the odor concentration. In the case of a concentration tube, the desorption time constant is large, and even when the response level is kept constant, the response pattern changes, and the concentration level and the response level do not match. In other words, in order to effectively perform the constant response level measurement, it is necessary to use a sensor cell direct connection type concentration element having a small desorption time constant.
[0024]
FIG. 7 shows the result of measuring a constant response level for a odor sample (propionic acid) in which the dilution ratio with respect to water is changed in three stages (1: 4, 1: 9, 1:19) using a sensor cell direct-coupled concentration element. It is a graph which shows. In the measurement, the concentration time was changed so that the response level of Inositol was about 140 Hz for each dilution ratio. The horizontal axis shows the dilution rate and concentration time. The response pattern was almost constant even at different dilution rates.
[0025]
In order to perform a constant response level measurement, the concentration time is changed until the maximum response value of the sensor after heating the heater reaches the allowable range of the set response level. In this case, since the sensor response is effective in a region where the concentration time is proportional to the concentration time, the initial concentration time needs to be set so as to fall within the region. In this example, the maximum number of repetitions is 5, and the sample and heater heating time is low for 30 seconds to 1 minute.
[0026]
As described above, a sensor cell direct-coupled concentration element with a small desorption time constant is useful for effective measurement of a constant response level, and effective odor identification is possible even when there are fluctuations in humidity and odor concentration. In order to carry out, the combined use of the sensor cell direct-coupled concentration element and the constant response level measurement is effective.
[0027]
Next, a method for measuring response patterns at a plurality of response levels in addition to a constant response level and identifying an odor from these measurement patterns will be described. The plurality of response levels are measured with a concentration time shorter than the constant response level. However, the plurality of response levels are also set so as to fall within a region where the concentration time and the sensor response are proportional. The time required for measurement by this method is 5 to 10 minutes per sample.
[0028]
FIG. 8 shows a scatter diagram of response patterns obtained by normalizing response patterns at the first response level of the following samples.
[0029]
As measurement samples, 2 ml of the following samples were used.
[0030]
trans-2-hexyl acetate (stock solution, 1: 3, 1: 7)
trans-2-hexenal (stock solution, 1: 3, 1: 7)
ethyl valerate (stock solution, 1: 3,1: 7)
isobutyric acid (stock solution, 1: 3,1: 7)
The numbers in parentheses indicate the dilution ratio (V / V) with respect to ODO.
[0031]

The first response level set for Inositol was 150 Hz. The four types of samples partially overlap with trans-2-hexenal and isobutyric acid, but all samples are separated as a whole. According to this figure, by using the gas sensor using the concentration element of the present invention even in an environment where the odor concentration varies, it is possible to distinguish many kinds of odors at a constant response level without depending on the odor concentration. In the figure, the size of the plot was increased in the order of decreasing odor concentration from the stock solution.
[0032]
FIG. 9 is a diagram illustrating a result of using the response patterns of the first, second, and third response levels for identification in the case of FIG. The second level is a response pattern measured at three-quarters of the concentration time determined at a constant response level, and the third level is a response pattern measured at one-half concentration time. In the figure, the plot of the one with a high concentration is displayed in a large size.
[0033]
Principal component analysis was performed using a 6-dimensional vector obtained by combining two-dimensional standardized response patterns up to the third response level normalized by Equation (1). There were no overlaps in the two types of samples in which overlap was seen in FIG. 8, and the odors of the four types of samples could be clearly identified. 8 and 9, the magnitude of the plot indicates the magnitude of the sample concentration, but no influence of the odor concentration fluctuation was observed. Note that the numerical values in parentheses on the horizontal axis and the vertical axis in FIG. 9 mean the contribution rate to the original information.
[0034]
FIG. 10 shows a scatter diagram of the normalized response pattern of each sample at the first response level for propionic acid and isobutyric acid diluted to low concentrations. The size of the plot is the measurement sample showing the concentration
2 ml each of propionic acid (10%, 1% aqueous solution) and isobutyric acid (10%, 1% aqueous solution)
Carrier gas flow rate: 200ml / min.
Heating time and current value: 1.0 sec-6.0A
Setting response level: 150 ± 15Hz
Initial concentration time: 5 seconds Depending on the type of sample, the plot was largely separated except for one point of isobutyric acid. However, the response patterns of the two carboxylic acids are similar, and the first response level alone is insufficient to accurately identify the odor.
[0035]
FIG. 11 is a graph showing the result of principal component analysis using a four-dimensional vector obtained by combining the two-dimensional normalized patterns of the first response level and the third response level. This result shows that two types of aqueous solution samples with concentration fluctuations can be clearly distinguished. In addition, it is shown that similar odors with different concentrations can be separated by measuring at a plurality of levels.
[0036]
【The invention's effect】
According to the sensor cell direct-coupled concentration element of the present invention, there is provided a volatile compound identification device and identification method that have excellent advantages such as the removal of the influence of humidity, a small desorption time constant, and the miniaturization of the entire device. The
[0037]
Discrimination of volatile compounds that eliminates the influence of humidity fluctuation and odor concentration fluctuation by adopting a measurement method using a constant response level using the concentration element of the present invention or a plurality of response levels in addition to this level. A method is provided.
[Brief description of the drawings]
FIG. 1 is a schematic explanatory diagram of an odor identification system of the present invention.
FIG. 2 is an explanatory diagram of a concentration element of the present invention.
FIG. 3 is an explanatory diagram of a sensor cell direct connection type concentration element of the present invention.
FIG. 4 is a graph showing a response recovery waveform.
FIG. 5 is a graph showing the odor concentration dependency of a response pattern.
FIG. 6 is a graph showing the results of performing a constant response level measurement.
FIG. 7 is a graph showing measurement results of a constant response level when humidity is added.
FIG. 8 is a scatter diagram of response patterns at a first response level.
FIG. 9 is a scatter diagram of response patterns using first to third response levels.
FIG. 10 is a scatter diagram of a response pattern of a first response level with respect to an aqueous solution sample.
FIG. 11 is a scatter diagram of response patterns of first and third response levels for an aqueous solution sample.
FIG. 12 is an explanatory diagram of a direct heating type concentrating tube.
FIG. 13 is a graph showing measurement results of a constant response level when a directly heated concentrating tube is used.
[Explanation of symbols]
1: vial 8: concentrating element 9: sensor cell 22: power supply 23: concentrating tube 24: adsorbent 33: adsorbent 35: concentrating element 36: sensor cell 37: crystal oscillator gas sensor

Claims (8)

Volatile compounds are adsorbed to a plurality of quartz resonator gas sensors whose electrode surfaces are coated with a volatile compound sensitive film, and changes in the resonance frequency of each quartz resonator due to the adsorption are recognized as patterns to identify volatile compounds. In the device to
The apparatus includes a concentration element of a volatile compound and a plurality of crystal resonator gas sensors, and the concentration element is a planar metal net and an adsorbent packed in a flat plate sandwiched between the nets. The adsorbent is provided with heating means on one or both sides of the adsorbent so that the quartz crystal gas sensor can be in direct contact with the volatile compound desorbed from the concentration element. An apparatus for identifying a volatile compound, comprising: The volatile compound identification device according to claim 1, wherein the quartz resonator sensor is disposed on a side surface of an open cylindrical or polygonal container having a flat concentrating element on the bottom surface. Volatile compounds are adsorbed to a plurality of quartz resonator gas sensors whose electrode surfaces are coated with a volatile compound sensitive film, and changes in the resonance frequency of each quartz resonator due to the adsorption are recognized as patterns to identify volatile compounds. In the way to
The method includes an adsorbent packed in a flat plate sandwiched between metal meshes and provided with a heating means on one or both sides of the adsorbent provided at a distance close to the quartz resonator gas sensor. Using the element, the concentration time to the adsorbent or the dilution rate with the carrier gas is variable, and the response pattern obtained with the response of the crystal resonator gas sensor as a constant level is recognized to identify volatile compounds. A method for identifying a characteristic volatile compound. The method for identifying a volatile compound according to claim 3, wherein the adsorbent is hydrophobic. Identification of volatile compounds by adsorbing volatile compounds to a plurality of quartz resonator gas sensors whose electrode surfaces are coated with a sensitive film of volatile compounds and recognizing changes in the resonance frequency of each quartz resonator due to the adsorption. In the way to
The method includes an adsorbent packed in a flat plate sandwiched between metal meshes and provided with a heating means on one or both sides of the adsorbent provided at a distance close to the quartz resonator gas sensor. The element is used to measure the response pattern of the volatile compound with the concentration rate to the adsorbent or the dilution rate being variable depending on the carrier gas, the response of the quartz crystal gas sensor is set to a constant level, and then different from the constant response level. A method for identifying a volatile compound, comprising: measuring a response pattern at a response level, and recognizing the response pattern to identify a volatile compound. 6. The method for identifying a volatile compound according to claim 5, wherein the adsorbent is hydrophobic. 6. The method for identifying a volatile compound according to claim 5, wherein the plurality of response levels are response levels measured at a concentration time shorter than a constant response level. 6. The method for identifying a volatile compound according to claim 5, wherein the plurality of response levels are response levels measured at a dilution ratio larger than a constant response level.

Images