WO2012049731A1

WO2012049731A1 - Method for producing compound flavoring

Info

Publication number: WO2012049731A1
Application number: PCT/JP2010/067875
Authority: WO
Inventors: 美紀若松; 雄一古殿; 幸夫曽根; 裕美寒河江
Original assignee: 日本たばこ産業株式会社
Priority date: 2010-10-12
Filing date: 2010-10-12
Publication date: 2012-04-19
Also published as: JPWO2012049731A1; JP5586702B2

Abstract

Provided is a method for producing a compound flavoring comprising: analyzing and identifying aroma compounds of an odorous substance; sorting a plurality of aroma compounds in order of descending odor intensity out of the identified aroma compounds; categorizing the sorted aroma compounds into a plurality of odor quality groups based on the odor quality; selecting one aroma compound from each odor quality group; and blending the selected aroma compounds.

Description

Method for producing blended fragrance

The present invention relates to a method for producing a blended fragrance.

In recent years, there has been a demand for a blended fragrance (reconstruction) that highly reproduces the aroma of odorous substances such as natural products and natural fragrances in products in various fields. The odors of odorous substances such as natural products and natural fragrances are usually composed of a large number of fragrance components having very different odor qualities. In the production of a blended fragrance by reconstruction, it is necessary to screen among these fragrance components for fragrance components that are particularly important for expressing the fragrance characteristics of odor substances.

As such a screening method, the gas chromatographic olfactometry method (GC-O method) is used to identify the aroma component from the odorous substance by sniffing the smell of the component separated by GC with the human nose. Subsequently, a technique for screening important aroma components using an Aroma Extract Extraction Analysis (AEDA) method, a Charm Analysis Analysis, or the like is known. For example, in the AEDA method, the concentration is sequentially diluted for each identified aroma component, and smelling with GC-O is repeated for each diluted sample. The maximum dilution factor at which the odor of each component can be felt is defined as Flavor ダイ Dilution Factor (FD factor). The larger the FD factor, the more the odor of the original odor substance. Identify as a component with a high contribution of. The AEDA method is described in Patent Document 1 and Non-Patent Document 1, for example.

In the production of a blended fragrance by reconstruction, a component having a high contribution to the aroma of the original odor substance as a whole is selected according to the number of components that can be mixed in the manufacturing procedure, for example, the top 30 components.

It has been confirmed in, for example, Patent Document 1 and Non-Patent Document 2 that the blended fragrance produced by mixing the fragrance components selected by the above method shows a certain degree of similarity to the odor of the original odor substance. Has been. However, the blended fragrances produced by these conventional methods are still less similar to the odor of the entire original odor substance.

JP 2004-325116 A

As a result of studying the above prior art, the present inventors have selected the aroma component based on the element of the detection limit of the odor of each component, that is, the intensity of the odor, in selecting the important aroma component. Therefore, it has been found that among the selected important aroma components, there is a problem that the components contributing to the same odor quality with respect to the blended fragrance can be selected redundantly. Furthermore, there is a problem that a component whose odor intensity is not so strong as compared with other components is eliminated in the selection step even if it has a completely different odor quality from the other components.

Therefore, an object of the present invention is to provide the manufacture of a blended fragrance that can highly reproduce the odor of the original odorous substance even in the number of components that are restricted in the manufacturing procedure.

In selecting important fragrance components, the present inventors selected fragrance components having strong odor intensity from the analyzed and identified fragrance components, and then the fragrance components were classified into a plurality of groups based on the odor quality ( Odor quality group), and by selecting one fragrance component from each odor quality group, the diversity of the odor quality of the original odor substance is ensured even in the limited number of fragrance components, and consequently the original odor It has been found that it is possible to produce a blended fragrance that highly reproduces the odor of a substance. The present invention is based on such knowledge.

Therefore, according to the present invention, the first step of analyzing / identifying the fragrance component of the odor substance, and the second step of selecting a plurality of fragrance components from the identified fragrance component in descending order of the odor intensity. A step, a third step of classifying the fragrance component selected in the second step into a plurality of odor groups based on the odor quality, and a step of selecting one fragrance component from each of the odor groups. There is provided a method for producing a blended fragrance comprising a fourth step and a fifth step of blending the fragrance component selected in the fourth step.

According to the present invention, compared to the conventional fragrance prepared by selecting and mixing the important fragrance components based only on the odor intensity of the conventional constituent components, the mixed fragrance having higher similarity to the odor of the original odor substance. Can be manufactured.

FIG. 1 shows a visual analog scale. FIG. 2 is a diagram illustrating a similarity profile of one fragrance component with respect to a reference fragrance component. FIG. 3 is a cluster diagram by performing multivariate analysis on the result of the similarity profile.

The method for producing a blended fragrance of the present invention belongs to a so-called reconstruction technique that includes reproducing the odor of the original odorous substance by blending.

In the present invention, first, the aroma component of the odorous substance is analyzed and identified (first step). Examples of odorous substances include plant raw materials such as tobacco, fruits, vegetables, cereals, teas, herbs, and processed products thereof, natural flavors, coffee, alcoholic beverages, and other favorite beverages. . Among these, preferred are tea leaves such as tobacco, green tea and black tea, and beverages.

In the first step, there is no particular limitation on the method for analyzing and identifying the aroma component of the odor substance, but the GC-O method can be used.

In the GC-O method, a sample (odorous substance) is injected into a separation column of a gas chromatograph together with a carrier gas (for example, helium gas) as a mobile phase. In the separation column, the fragrance component is separated based on the difference in the distribution coefficient of each fragrance component between the stationary phase packed in the separation column and the mobile phase. At the outlet of the separation column, each separated fragrance component is sent to the mass analyzer (MS) on the one hand and to the scent device on the other side, and the qualitative and quantitative analysis of the fragrance component is performed by the MS, and the scent device The smell of each fragrance component is detected by the human sense of smell. In this way, the aroma component of the odor substance can be identified.

Next, the odor intensity of the fragrance component identified in the first step is measured, and the fragrance component is selected in order of increasing odor intensity (second step).

In the second step, the method for measuring the odor intensity is not particularly limited, but the AEDA method can be used.

In the AEDA method, each of the identified aroma components is sequentially diluted, for example, 2 ² times, 2 ³ times, 2 ⁿ times, or 4 ² , 4 ^3, 4 ⁿ times, The GC-O method is performed on each diluted sample. The maximum dilution factor at which the odor of the fragrance component is felt is recorded as the FD factor of the fragrance component. Since an odor component having a large FD factor can detect odor even when diluted thinly, the contribution of the original odor substance to the fragrance is large.

In this way, the odor intensity is measured, and the aroma components are selected in order of increasing odor intensity.

Next, the aroma components selected in the second step are classified into a plurality of odor groups (odor groups) based on the odor quality (third step).

In this third step, there is no particular limitation on the method of classifying the odor component into a plurality of odor groups in the odor quality, but for example, the documents RH 文献 WRIGHT, KM MICHELS, Ann NY Acad Sci, 116, The similarity profile described in -551 (Jul 30, 1964) can be used.

More specifically, from the fragrance components selected in the second step, the odor quality can be easily stored by the panelists, and the odor quality can be clearly distinguished from each other. Select some (for example, 10 components). Next, the similarity between the fragrance component (including the selected reference component) selected in the second step and the reference component is determined by a panel using a visual analog scale (VAS) (see FIG. 1). evaluate. This VAS has dissimilar (handled as 0 on numeric data) and similar (handled as 100 on numeric data) as descriptors at both ends.

Next, a similarity profile is created by using the similarity between all components as the value obtained by arithmetically averaging the similarity values obtained by all panelists. An example of a similarity profile for 10 reference components (reference components 1 to 10) for one aroma component is shown in FIG. A cluster diagram is obtained by multivariate analysis of the obtained similarity profile result. Then, in this cluster, for example, the 10 reference components selected first are separated into positions (dotted line a in FIG. 3) where the number of groups is the smallest among the different clusters. Thus, the aroma components selected in the second step are classified into a plurality of odorous groups (10 odorous groups A to J in FIG. 3). The number of odor groups is not particularly limited, but when it is less than 10, the degree of similarity to the natural aroma is low, while when it is more than 30, the productivity decreases, so it is preferably 10 to 30.

Thereafter, one fragrance component is arbitrarily selected from each odor group (fourth step).

In the fourth step, the selection of fragrance components from each odor group can be performed arbitrarily, but preferably the odor quality distance is determined for all pairs of fragrance components in the same odor group and the total odor is selected. By selecting the component with the smallest mass distance, a blended fragrance having a higher degree of similarity to the original odor substance can be produced. More specifically, the odor quality distance is determined using the VAS method for all pair combinations of fragrance components within the same odor quality group. Then, among the components in the same odor group, the component having the smallest sum of odor distances (in other words, when each component in the same odor group is placed in space, it is compared with all other components. The component that is close to the target, that is, the component closest to the center of the space is selected.

Finally, the aroma component selected in the fourth step is blended to produce a desired blended fragrance (fifth step).

In the fifth step, the blending ratio of the fragrance component selected in the fourth step can be arbitrarily set, but so that all odor qualities are perceived equally in the manufactured blended fragrance, It is preferable to mix each aromatic component with equal strength.

As described above, it is possible to produce a blended fragrance that emits an odor very similar to the odor of the original odor substance.

Hereinafter, the present invention will be described with reference to examples, but the present invention is not limited to these examples.

Example 1
<Aroma component analysis and selection of aroma components>
Eight cabin mild cigarettes with the filter removed were burned using an automatic smoking device. The crude tar adhering to the Cambridge filter was extracted with 100 mL of diethyl ether, and then concentrated to 2 mL at room temperature and normal pressure using a rapid solvent extraction device to prepare a sample.

The above samples were subjected to qualitative and quantitative analysis (odor sniffing (ODP) GC / MS analysis) by GC-O under the following conditions. Further, under the same conditions, the sample was sequentially diluted by 4 times by the AEDA method, and the FD factor of each aroma component was measured. The top 100 components (component 1 to component 100) having a large FD factor were selected as important aroma components (Table 1 below).

[GC-O analysis conditions]
Equipment: Agilent 6890GC / 5973MSD
Column: J & W DB-WAX (0.25 μm (film thickness) × 25 mm (inner diameter) × 60 m (length)
ODP: MS split ratio: 1: 1
Inlet mode (CIS4): Solvent Vent
Injection volume: 1 μL
CIS temperature: -100 ° C (0.1 min) → temperature rise (12 ° C / sec) → 300 ° C (3 min)
Mobile phase: Helium gas Column temperature: 50 ° C. (0 min) → Temperature rise (5 ° C./min)→250° C. (20 min)
ODP transfer line temperature: 280 ° C
MS transfer line temperature: 250 ° C
MS source temperature: 230 ° C
MS quadrupole temperature: 150 ° C.

<Classification into odorous groups>
Ten components were selected from the 100 components shown in Table 1 as reference components. The reference components 1 to 10 were selected by an experimenter on the basis that the odor quality can be easily memorized and the odor qualities can be clearly distinguished from each other. Subsequently, the similarity between 100 components shown in Table 1 (including reference components 1 to 10) and each of the reference components was measured by the VAS method by sensory evaluation by 20 panelists. The VAS used has dissimilar (handled as 0 on numeric data) and similar (handled as 100 on numeric data) as descriptors at both ends (see FIG. 1). A similarity profile was created with the arithmetic mean of the similarities obtained by 20 panelists as the similarity between each component. This similarity profile is shown in FIG. The cluster diagram shown in FIG. 3 was obtained by multivariate analysis of the similarity profile results. In this cluster, the clusters were cut at positions (dotted line a in FIG. 3) where the number of groups was the smallest among the reference components 1 to 10 that were set first in different clusters. As a result, it was classified into 10 odor groups A to J shown in FIG. 3 (see also Table 1). Groups A to J were of Onion, Sulfur, Gassy, Animal, Roast, Sour, Caramel, Floral, Fruity and Medicinal odors, respectively.

<Selection of aromatic components from odor group and preparation of fragrance>
By arbitrarily selecting one component from each of the above ten odor groups A to J (indicated by a circle in Table 1), and blending at a concentration at which the odor intensity of each component is equal, A blended fragrance having components as shown in 1 was obtained.

Example 2
Of the 100 components obtained in Example 2, based on intensity information as usual, the top 10 components of the FD factor are selected (indicated by a circle in Table 1), and the odor intensity of each component is equal. A blended fragrance having components as shown in Table 1 was obtained by blending at a concentration of

The 20 panelists were investigated which of the blended fragrance obtained in Example 1 and the blended fragrance obtained in Example 2 felt closer to the smell of tobacco when smoking. As a result, 17 panelists evaluated that the blended fragrance obtained in Example 1 felt closer to the smell of tobacco when smoking. This indicates that there is a significant difference at a risk rate of 1%. In the method of the present invention, by selecting important fragrance components according to the quality of odor, it is possible to produce a blended fragrance having a higher similarity with the original odor of cigarettes in the same number of fragrance components than in the conventional method. all right.

Example 3
In the same manner as in Example 1, the 100 important aroma components are classified into 5 odor groups (indicated by dotted line b in FIG. 3), and one component is arbitrarily selected from each group, and the odor intensity of each component Synthetic fragrances were blended so as to have equal strength.

Example 4
In the same manner as in Example 1, the 100 important aroma components are classified into 30 odor groups (indicated by dotted line c in FIG. 3), and one component is arbitrarily selected from each group, and the odor intensity of each component Synthetic fragrances were blended so as to have equal strength.

Example 5
The 100 important fragrance components were directly used as 100 odor groups, and synthetic fragrances were prepared so that the odor strength of each component was equal.

For all the combinations of Example 1 and Examples 3 to 5, 20 panelists were investigated which felt close to the smell of tobacco when smoking. The results are shown in Table 2.

In Table 2, each symbol has the following meaning.
+: Significantly selected at 5% risk rate; ++: Significantly selected at 1% risk rate;-: Not selected significantly at 5% risk rate;-: Significantly at 1% risk rate Not selected; NS: no significant difference.

It was found that the more the number of odor groups, the closer to the smell of tobacco when smoking. On the other hand, the productivity of blended fragrances decreased as the number of odor groups increased. Table 3 shows the degree of similarity, productivity, and their overall evaluation. As can be seen from Table 3, when the number of odorous groups was in the range of 10 to 30, the degree of similarity to the odor of cigarettes at least as in Example 2 was shown, and the productivity was sufficiently secured.

In Table 3, the evaluation criteria for similarity are as follows.
A: Significantly higher similarity to Example 1; O: Similar to Example 1; ×: Significantly lower similarity to Example 1.

In Table 3, the productivity evaluation criteria are as follows.
◎: Can be easily prepared by manual operation; ○: Manual preparation is possible, but requires a certain amount of time; ×: Preparation by manual operation takes an enormous amount of time;
Furthermore, in Table 3, the comprehensive evaluation criteria are as follows.
○: Suitable as the number of odorous groups in the composition of the reproduced fragrance; ×: Not suitable as the number of odorous groups in the structure of the reproduced fragrance.

Example 6
In Example 1, after classification into odorous groups, the odorous distances for all pairs of components within the same odorous group were determined using the VAS method. And the synthetic | combination fragrance | flavor was obtained by selecting the component from which the sum total of an odor quality distance becomes the minimum among each component in the same odor quality group, and preparing these by the density | concentration which becomes equal intensity | strength.

Investigated 20 panelists as to which of Example 1 and Example 6 felt closer to the smell of tobacco when smoking. As a result, 15 panelists evaluated that the blended fragrance obtained in Example 6 felt closer to the smell of tobacco when smoking. This indicates that there is a significant difference at a risk rate of 5%. It was found that the components selected from each odorous group are closer to the smell of tobacco when smoking than selecting the components located in the center of each odorous group rather than selecting them arbitrarily.

Example 7
A cigarette was prepared by adding 10 μL of the blended fragrance obtained in Example 6 to the tobacco part of the mild seven one cigarette using a microsyringe. A smoking evaluation was performed on the panelists. As a result, 18 panelists evaluated that the cigarette added with the blended fragrance obtained in Example 6 felt that the smell of tobacco was stronger when smoking. This indicates that there is a significant difference at a risk rate of 1%. It has been found that the odor of tobacco during smoking can be enhanced by adding the blended fragrance produced by the method of the present invention.

Claims

A first step of analyzing / identifying the fragrance component of the odor substance; a second step of selecting a plurality of fragrance components in order of increasing odor intensity from the identified fragrance component; and the second step. A third step of classifying the fragrance components selected in step 4 into a plurality of odor groups based on the odor quality, a fourth step of selecting one fragrance component from each of the odor groups, and the fourth step The manufacturing method of the mixing | blending fragrance | flavor containing the 5th process of preparing the fragrance | flavor component selected at the process.
The method according to claim 1, wherein the odor group consists of 10 to 30 groups.
In the fourth step, the odor distance is determined for all pairs of aroma components in the same odor group, and the sum of the odor distances is minimized among the components in the same odor group. The method according to claim 1 or 2, wherein the components are selected.
A blended fragrance obtained by the method according to any one of claims 1 to 3.
The blended fragrance according to claim 4, which reproduces the smell of tobacco when smoking.
Cigarette to which the blended fragrance according to claim 5 is added.