JP2012047453A - Mass spectrometry method - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a mass spectrometry method which enables excellent identification of components from measurement data obtained from soft ionization mass analyzer, without combining analytical tools.SOLUTION: The mass spectrometry method comprises: a step A of extracting mass spectral data in which two-dimensional mass spectral data is extracted from three-dimensional measurement data obtained from the soft ionization mass analyzer; a step B of processing mass spectral data in which the extracted mass spectral data is processed into data under the same condition with database; a step C of determining matrix in which one or more components of the matrix and mixture ratio of the components are determined from the processed mass spectral data using multivariate analysis; a step D of synthesizing standard mass spectral data in which the standard mass spectral data is synthesized from the determined components of the matrix and mixture ratio and the database; and a step E of removing noise in which the standard mass spectral data is subtracted from the processed mass spectral data.

Description

  The present invention relates to a mass spectrometry method, and more particularly, to a mass spectrometry method for analyzing mass data by a soft ionization method by data analysis without using a separation means such as a chromatograph.

When using a mass spectrometer, it is necessary to ionize the measurement object.
As an ionization method of a gas chromatograph-mass spectrometer, there is an electron impact ionization method (EI method) as a commonly used ionization method. Since this EI method uses high energy, a large number of fragments are generated. Therefore, it is difficult to analyze and evaluate a mass spectrum. Therefore, it is necessary to use in combination with a separation means such as a chromatograph.
On the other hand, in mass spectrometers equipped with soft ionization methods such as ion attachment ionization method, Penning ionization method, and photoionization method, since fragment ions can be suppressed, it is possible to omit chromatographic separation means, and thus analysis time There are merits such as shortening the measurement time, avoiding denaturation and loss of the measurement object in the separating means, and enabling real-time monitoring of the change of the measurement object over time.

However, in the soft ionization mass spectrometer, there are some negative aspects caused by not using a separation means such as a gas chromatograph.
For one thing, since no separation means such as a gas chromatograph is used, the constituent components of the measurement object are complex, and in many cases, the signals of the congested component and the signal of the target component overlap, making good identification difficult. There is a case.
In particular, the measurement object has a great influence when detecting a trace component contained in a base component (hereinafter referred to as a matrix) in a polymer material such as a resin or an adhesive, oil, or a biological sample. In the case of organic chemical substances, the overlap caused by isotope signals may cause serious interference with component identification. In order to avoid such a problem, it is necessary to greatly increase the mass resolution or improve the identification accuracy by using MS / MS. However, there arises a problem that the apparatus becomes large and expensive.
Another problem is that a fine signal change hidden behind a large signal change is overlooked because no separation means such as a gas chromatograph is used. When using a gas chromatograph, each trace component is clearly separated and has a sharp enriched band, so the peak of each component in the total ion chromatogram is sharp, making it easy to check each component. It is. However, in a soft ionization mass spectrometer that does not use a gas chromatograph, the main components can be monitored on the total ion chromatogram, but it is difficult to determine the signal changes of trace components without checking individual mass chromatograms. is there.

JP 9-318599 A JP 2006-10323 A JP 2007-205745 A Reissue WO2007 / 102201

  The devices and methods related to mass spectrometry disclosed in Patent Documents 1 to 4 disclose ideas related to data processing, but all are based on a mass spectrometer equipped with a separation means such as a gas chromatograph. is there.

  Therefore, the present invention does not combine a separation means such as a chromatograph, thereby shortening the analysis means, avoiding denaturation and loss of the measured component in the separation means, and enabling real-time monitoring of changes in the measured object over time. And providing a mass spectrometric method that can remove noise from congested components, etc., even from measurement data obtained with a soft ionization mass spectrometer, and perform component identification to trace components. Let it be an issue.

The mass spectrometric method of the present invention that solves the above problems is based on three-dimensional measurement data consisting of time, mass-to-charge ratio (m / z) and signal intensity obtained by a soft ionization mass spectrometer. A mass spectrum data extraction step for extracting two-dimensional mass spectrum data composed of the ratio (m / z) and the signal intensity;
A mass spectrum data processing step of processing the mass spectrum data extracted in the mass spectrum data extraction step into the same data conditions as a database created in advance;
One or a plurality of matrices included in the processed mass spectrum data (Su) by collating the processed mass spectrum data (Su) obtained in the mass spectrum data processing step with the database using multivariate analysis means And a matrix specifying step for specifying the components of the components and the mixing ratio of the components;
A standard mass spectrum that is synthesized from the mass spectrum data of the database as standard mass spectrum data (Sr) from the matrix components and mixing ratios specified in the matrix specifying step, as mass spectrum data corresponding to the components and mixing ratios. Data (Sr) synthesis step;
A noise removing step of subtracting the standard mass spectrum data (Sr) from the processed mass spectrum data (Su) to remove duplication of signal values due to matrix components and matrix-derived components from the processed mass spectrum data (Su);
It has the 1st characteristic to have.

In addition to the first feature, the mass spectrometric method of the present invention further includes, in the noise removal step, residual mass data obtained by subtracting the standard mass spectrum data (Sr) from the processed mass spectrum data (Su), and further the standard mass spectrum data. A second feature is that by dividing by (Sr), a minute change buried in the matrix is highlighted as residual change data.
In addition to the first or second feature, the mass spectrometric method of the present invention includes time, mass-to-charge ratio (m / z), and signal intensity in the soft ionization mass spectrometer in the mass spectrum data extraction step. For the three-dimensional measurement data, obtain an average value or total value of signal intensities for each mass-to-charge ratio (m / z) in a certain time zone, and calculate the mass-to-charge ratio (m / z) and signal intensity. The third feature is to obtain the following two-dimensional mass spectrum data.
In addition to the third feature, the mass spectrometric method of the present invention adopts a certain time zone to obtain the signal intensity at each sampling time for each mass-to-charge ratio (m / z) of three-dimensional measurement data. Performing principal component analysis as a variable, based on the obtained two-dimensional data consisting of loading data of the first to n-th principal components and time, and one or more inflection points extracted from the two-dimensional data Thus, the fourth feature is to adopt a fixed time zone.
In addition to the fourth feature, the mass spectrometry method of the present invention displays the inflection point of the two-dimensional data obtained in the mass spectrum data extraction step on the total ion chromatogram obtained from the three-dimensional measurement data. This is the fifth feature.
In addition to any one of the first to fifth features, the mass spectrometry method of the present invention performs at least one of normalization and scaling processing of the extracted mass spectrum data in the mass spectrum data processing step. Thus, the sixth feature is that the data is processed into mass spectrum data having the same data conditions as that of the database created in advance.
Further, in addition to any one of the first to sixth features, the mass spectrometric method of the present invention uses at least a modified regression analysis means as a multivariate analysis means in the matrix specifying step, and uses processed mass spectrum data (Su). ) And the mass spectrum data in the database, the seventh feature is that the components of the matrix and the mixing ratio thereof are specified.
Moreover, in addition to any one of the first to seventh features, the mass spectrometry method of the present invention includes, in the standard mass spectrum data (Sr) synthesis step, each of one to a plurality of matrix components specified in the matrix specifying step. The mass spectrum data is adopted from the database, and the standard mass spectrum data (Sr) is synthesized by combining each mass spectrum data from the adopted database corresponding to the mixing ratio of the one or more matrix components. Is the eighth feature.

According to the mass spectrometry method of the first aspect, two-dimensional mass spectrum data is obtained from the three-dimensional measurement data obtained by the soft ionization mass spectrometer through the mass spectrum data extraction step. Then, through the mass spectrum data processing step, the mass spectrum data is processed mass spectrum data (Su) under the same data conditions as the database created in advance. In the matrix specifying step, the processed mass spectrum data (Su) is analyzed by the multivariate analysis means and collated with the database, so that one or more matrix components included in the processed mass spectrum data (Su) and its components A mixing ratio is specified. In the standard mass spectrum data (Sr) synthesis step, standard mass spectrum data (Sr) corresponding to the specified matrix component of the object to be measured and its mixing ratio is synthesized from the mass spectrum data in the database. In the noise removal step, the standard mass spectrum data (Sr) is subtracted from the processed mass spectrum data (Su) of the object to be measured, so that the overlap of the matrix component and the matrix-derived component is removed from the processed mass spectrum data (Su). It is burned. Thereby, the signal intensity of components other than the matrix included in the processed mass spectrum data (Su) becomes clearer. Therefore, components other than the matrix can be identified more accurately and reliably.
According to the mass spectrometric method of claim 1, the identification and mixing of the matrix components of the object to be measured can be performed from the measurement data obtained by the soft ionization mass spectrometer without the need to combine separation means such as a gas chromatograph and a liquid chromatograph. The noise due to the matrix component with respect to the ratio and further the trace component can be removed, and the trace component can be identified effectively. Therefore, problems such as increase in analysis time due to the provision of separation means such as gas chromatograph and liquid chromatograph, denaturation and loss of the measurement object in the separation means, and real time monitoring of changes in the measurement object over time are not enough. The mass analysis of the measurement object can be performed satisfactorily while reducing.

According to the mass spectrometric method described in claim 2, in addition to the operational effect of the configuration described in claim 1, the noise removal step subtracts the standard mass spectrum data (Sr) from the processed mass spectrum data (Su). By dividing the residual data further by the standard mass spectral data (Sr), it is supposed that the minute change of the signal intensity buried in the matrix is highlighted.
Identification and identification of components other than the matrix of the object to be measured can be performed well up to trace components.
According to the mass spectrometric method according to claim 3, in addition to the operational effects of the configuration according to claim 1 or 2, time, mass-to-charge ratio (m / z), and signal intensity in the soft ionization mass spectrometer. For the three-dimensional measurement data consisting of the above, obtain the average value or the sum of the signal intensities for each mass-to-charge ratio (m / z) in a certain time zone, Since we are trying to obtain two-dimensional mass spectrum data consisting of
It is possible to obtain mass spectrum data in a certain time zone in which analysis of components can be performed more clearly with respect to all three-dimensional measurement data in all time zones obtained with a soft ionization mass spectrometer. The signal intensity of each mass-to-charge ratio (m / z) in the spectral data can also be used as an average value or total value of the signal intensities so that the signal intensity can be compared between each mass-to-charge ratio (m / z). It can be trusted.
According to the mass spectrometric method described in claim 4, in addition to the operational effects of the configuration described in claim 3, the adoption of the certain time zone is the mass-to-charge ratio (m / m) of the three-dimensional measurement data. z) Principal component analysis is performed with the signal intensity at each sampling time as a variable for each z), and the obtained two-dimensional data consisting of loading data and time of the first to n-th principal components are extracted from the two-dimensional data. Based on one or more inflection points, a certain time zone is adopted.
It becomes possible to obtain mass spectrum data in a time zone in which the characteristics of the components contained in the measurement object appear well. Therefore, a clearer inflection point can be extracted, and a clearer component analysis can be performed.
According to the mass spectrometry method of the fifth aspect, in addition to the function and effect of the configuration according to the fourth aspect, the inflection point of the two-dimensional data obtained in the mass spectrum data extraction step is calculated from the three-dimensional measurement data. Since it is supposed to be displayed on the total ion chromatogram obtained,
It is possible to inform the measurer of the tendency of the change in the mass spectrum pattern, which is not easy to confirm on the total ion chromatogram, and to call attention.

Moreover, according to the mass spectrometry method of Claim 6, in addition to the effect by the structure in any one of the said Claims 1-5, in mass spectrum data processing process, normalization of the extracted mass spectrum data is carried out. By performing at least one of the scaling processing, it is supposed to be processed into mass spectrum data of the same data conditions as the database created in advance,
The analysis by the multivariate analysis means and the accompanying database comparison and collation can be reliably and easily performed based on the data under the same conditions.
According to the mass spectrometric method of claim 7, in addition to the operational effects of the configuration according to any of claims 1 to 6, in the matrix specifying step, at least a multiple regression analysis means as a multivariate analysis means. Is used to identify the component of the matrix and its mixing ratio by comparing the processed mass spectrum data (Su) with the mass spectrum data in the database.
Through the repeated regression analysis of the processed mass spectrum data (Su) with the same conditions as the database, it is possible to specify the components and mixing ratios of one or more matrices that are actually included in the processed mass spectrum data (Su).
Moreover, according to the mass spectrometry method of Claim 8, in addition to the effect by the structure in any one of the said Claims 1-5, in a standard mass spectrum data (Sr) synthetic | combination process, it specifies at a matrix specification process. The mass spectrum data of the one or more matrix components is adopted from the database, and the mass spectrum data from the adopted database is combined in accordance with the mixing ratio of the one or more matrix components. Since we are going to synthesize mass spectral data (Sr)
The mass spectral data to be standardized that would occur when the specified matrix component (s) are mixed at the specified mixing ratio can be created from the database, so that the standard mass spectral data (Sr ) Can be used to remove the influence of the matrix included in the processed mass spectral data (Su) of the object to be measured.

It is a flowchart explaining the mass spectrometry method of this invention. It is a flowchart explaining the mass spectrum data extraction process of the mass spectrometry method of this invention. It is a flowchart explaining the auxiliary | assistant operation | work in the mass spectrum data extraction process of the mass spectrometry method of this invention. It is the figure obtained in the auxiliary | assistant operation | work of the mass spectrum data extraction process of the mass spectrometry method of this invention, (A) is the total ion which displayed the inflection point of each chromatogram of the 1st-6th main component in (B). A chromatogram, (B), is a diagram showing a chromatogram displaying two-dimensional data composed of loading data and times of the first to sixth principal components obtained by principal component analysis. In Example 1 of this invention, it is a figure which shows the process mass spectrum data (Su) of the to-be-measured object obtained through the mass spectrum data extraction process and the mass spectrum data processing process, (A) represents signal intensity on a normal scale. (B) shows the signal intensity on an enlarged scale. In Example 1 of this invention, it is a figure which shows the analysis result about the component and mixing ratio of the matrix of the to-be-measured object obtained by the double regression analysis in a matrix specific process. The actual mixing ratio is also displayed as a theoretical value. In Example 1 of this invention, it is a figure which shows the standard mass spectrum data (Su) of the to-be-measured object obtained through the standard mass spectrum data synthetic | combination process. In Example 1 of this invention, it is a figure which shows the residual data which subtracted standard mass spectrum data (Sr) from the process mass spectrum data (Su) in a noise removal process. In Example 1 of the present invention, residual data obtained by subtracting the standard mass spectral data (Sr) from the processed mass spectral data (Su) in the noise removal step and further dividing by the standard mass spectral data (Sr) It is a figure which shows data. In Example 2 of this invention, it is a figure which shows the process mass spectrum data (Su) of the to-be-measured object obtained through the mass spectrum data extraction process and the mass spectrum data processing process, (A) represents signal intensity on a normal scale. (B) shows the signal intensity on an enlarged scale. In Example 2 of this invention, it is a figure which shows the analysis result about the component and mixing ratio of the matrix of the to-be-measured object obtained by the double regression analysis in a matrix specific process. The actual mixing ratio is also displayed as a theoretical value. In Example 2 of this invention, it is a figure which shows the standard mass spectrum data (Su) of the to-be-measured object obtained through the standard mass spectrum data synthetic | combination process. In Example 2 of this invention, it is a figure which shows the residual data which subtracted standard mass spectrum data (Sr) from the process mass spectrum data (Su) in a noise removal process. In Example 2 of the present invention, residual data obtained by subtracting the standard mass spectral data (Sr) from the processed mass spectral data (Su) in the noise removal step is further divided by the standard mass spectral data (Sr). It is a figure which shows data.

Embodiments of the present invention will be described below with reference to the drawings.
Referring to the flow diagram of FIG. 1, the mass spectrometry method of the present invention includes a mass spectrum data extraction step A, a mass spectrum data processing step B, a matrix identification step C, and a standard mass spectrum data (Sr) synthesis step D. And a noise removal step E.

<Mass spectrum data extraction process>
With reference to the flow diagram of FIG. 2, in the mass spectrum data extraction step A, a measurement target is obtained from three-dimensional measurement data including time, mass-to-charge ratio (m / z) and signal intensity obtained by a soft ionization mass spectrometer. Extract mass spectrum data of the object.
The soft ionization mass spectrometer includes ion attachment ionization method, Penning ionization method, photoionization method, field ionization method, field desorption ionization method, electrospray ionization method, nanoelectrospray ionization method, probe electrospray ionization method, matrix A mass spectrometer equipped with means of a so-called soft ionization method capable of suppressing the generation of fragment ions of an object to be measured, such as a assisted laser desorption ionization method, and detecting it as a molecular ion.
The measurement data of the object to be measured obtained by the soft ionization mass spectrometer is three-dimensional data. A fixed time zone indicating a characteristic mass spectrum pattern is adopted from the measurement data (step A1).
An average value of signal intensity (average signal intensity) is calculated for each mass-to-charge ratio (m / z) in the adopted fixed time zone (step A2).
Then, two-dimensional mass spectrum data including each mass to charge ratio (m / z) and average signal intensity is obtained (step A3).

Referring to the flow diagram of FIG. 3, as ancillary work of the mass spectrum data extraction step, in order to designate a certain time zone in the step A1, each mass-to-charge ratio (m / For each z), the loading data of the first to n-th principal components is obtained by performing principal component analysis with the signal intensity at each sampling time as a variable (step A4).
Using the two-dimensional data composed of the loading data of the first to n-th principal components obtained in step A4 and time, the inflection points of the respective two-dimensional data are extracted (step A5).
The certain time period includes two-dimensional mass spectrum data composed of the obtained loading data of the first to n-th principal components and time, and one or more variables extracted from the two-dimensional mass spectrum data. Based on the inflection points, a fixed time zone in which it can be determined that the characteristics of the object to be measured often appear is selected from the total measurement time and adopted.
More specifically, as the measurement data, mass data of the generated gas component from the sample to be measured depending on the environmental state (for example, temperature, moisture content, gas type, gas flow rate, etc.) in each time zone is recorded. For this reason, individual inflection points or average mass spectrum data between the inflection points are caused by the environmental conditions in each time zone and the sample to be measured. However, the vaporization component of the medium volatile organic compound in the time zone B, the thermal decomposition component of the non-volatile or nonvolatile organic compound in the time zone C, the residual component in the analyzer in the time zone D, etc. It is a direct reflection of the state in the time zone. For this reason, it is easy and accurate to adopt, for example, the time zones B and C as the time zones that directly represent the properties of the sample to be measured from the characteristic average mass spectrum data in each time zone. Is possible.

The position of the inflection point extracted from the two-dimensional mass spectrum data is indicated by a vertical line as the peak position on each chromatogram, as shown in FIG.
And as said ancillary work, as shown to FIG. 4 (A), the obtained inflection point is displayed on a total ion chromatogram (step A6).
The total ion chromatogram consists of a time axis and a total signal intensity axis, and can be easily shown in the figure by calculating the total signal intensity for each sample time from the measurement data (three-dimensional data) obtained by the mass spectrometer. Can do.
Then, as shown in FIG. 4 (A), by displaying the inflection point on the total ion chromatogram (indicated by a triangle), the measurement can be performed within the entire large change indicated by the total ion chromatogram. The specific points that the person should be aware of can be informed.

<Mass spectrum data processing process>
In the mass spectrum data processing step B, the two-dimensional mass spectrum data extracted in the above-described mass spectrum data extraction step A is processed under the same data conditions as a database created in advance.
The processing of the mass spectrum data performs at least one of data normalization and scaling processing. This matches the data format of the database.
The database obtains mass spectrum data for a plurality of components expected as matrix components in advance and uses them as a database. For this database, at least one of data normalization (normalization) and scaling processing is applied in order to give universality. For this reason, the mass spectrum data extracted in the mass spectrum data extraction step is also subjected to at least one of normalization and scaling processing to obtain processed mass spectrum data (Su) that matches the data format of the database. In the normalization, the maximum value is 1 and the minimum value is 0. In the scaling process, any one or more of auto scaling, distributed scaling, range scaling, and the like are used.
Of course, the purpose of this process is to make the data conditions of the database and the mass spectrum data extracted in the mass spectrum data extraction process the same, so the content of the specific data processing is not necessarily the standardization and scaling described above. It is not limited to processing.
The database may be created by a person who uses the method of the present invention, or may be used by adopting an existing database.

<Matrix identification process>
In the matrix specifying step C, the processed mass spectrum data (Su) is collated with the database using the multivariate analysis means, thereby specifying one or more matrix components included in the processed mass spectrum data (Su). To do.
As a multivariate analysis means, at least a double regression analysis means is used. By performing the double regression analysis using the two-dimensional processed mass spectrum data (Su), the matrix component of the object to be measured can be collated with the database and specified. Further, even when two or more types of matrix components are mixed, it is possible to specify the two or more types of matrix components including their mixing ratio.
By using the multiple regression analysis means, for example, by analyzing the correlation between all combination patterns of a plurality (for example, nine types) of matrix components registered in the database and the processed mass spectrum data (Su), A coefficient related to the abundance ratio of a plurality of matrix components registered in the database, a multiple correlation coefficient indicating the analysis accuracy, a multiple determination coefficient, and others are obtained. Here, both the processed mass spectrum data (Su) and the database data to be compared are normalized so that the signal intensity term is excluded. An apparent mixing ratio of the matrix components can be obtained.
When there is only one matrix component, the principal component analysis or cluster analysis is performed as a multivariate analysis means, so that the matrix component can be visually matched and judged to match or not match the components registered in the database. Thus, the matrix can be specified. On the other hand, when there are two or more kinds of matrix components, the matrix components cannot be sufficiently identified by the principal component analysis or cluster analysis.
In addition, when identifying the components and mixing ratios of a matrix using double regression analysis, the analysis accuracy of double regression analysis is affected by the object to be measured and measurement conditions, making it difficult to identify components with a concentration below a certain level. There is a case. Therefore, when specifying the matrix components and the mixing ratio by the multiple regression analysis, the matrix components analyzed as a mixing ratio of less than a certain concentration, for example, a concentration of less than several percent, such as a concentration of less than 5%, are used as matrix components. May not be adopted. A component that is not employed as a matrix component is not used as a matrix component in a standard mass spectral data (Sr) synthesis process described below.

<Standard Mass Spectral Data (Sr) Synthesis Process>
In the standard mass spectrum data (Sr) synthesis step D, the mass spectrum data corresponding to the component and the mixing ratio is set as the standard mass spectrum data (Sr) from the matrix components and the mixing ratio specified in the matrix specifying step. Synthesize from mass spectrum data in database.
Standard mass spectrum data (Sr) corresponding to one or more matrices specified in the matrix specifying step can be adopted from a database.
On the other hand, for example, combining two types of matrix data corresponding to the mixing ratio cannot be simply performed because the signal output in each matrix component is not uniform. For this reason, the mass spectrum data of the two types of matrixes that are mixed at 50% and 50% in advance (more specifically, the two matrixes obtained by the multiple regression analysis of these mass spectrum data) Signal intensity ratio) must be registered in the database. And it is necessary to synthesize data using the intensity ratio of the signal intensities of both matrices in the registered mixed data.
Therefore, the database includes a plurality of matrix components (for example, five types a, b, b, d, and e) that are predicted in advance, and combinations of all two (ab, ac, ad, ae, bc, bd, and be , Cd, ce, de), mass spectrum data of 50% and 50% mixed are registered as a database.
Of course, after the matrix components are determined in the matrix specifying step C, mass spectrum data of 50% and 50% mixture for all the two combinations is collected for each of the plurality of determined matrix components. It is also possible to obtain an intensity ratio of signal intensity at The reason for mixing 50% and 50% is to obtain the intensity ratio of the mass spectrum between the two components when the same amount is added. Of course, in order to obtain the intensity ratio between the two components, it is not always necessary to use a mixture of 50% and 50%, and the signal intensity ratio can also be obtained using another mixture ratio.
As a database, it is sufficient to know the intensity ratio of signal intensities for all two combinations of matrices. This is because even if there are three or more types of matrices specified in the matrix specifying step C, the signal intensity ratio between the plurality of types of matrices can be clarified.

Now, for example, the case where the matrix components specified by the matrix specifying process C are olive oil and soybean oil and the apparent mixing ratio is specified as 8 to 2 is taken as an example.
The mass spectrum data (signal intensity data) of olive oil shown in the database is L, the mass spectrum data (signal intensity data) of soybean oil shown in the database is M, and the ratio of the signal intensity of olive oil and soybean oil shown in the database is 2. When it is 1: 1, the synthetic mass spectrum data (S) using olive oil and soybean oil as a matrix is synthesized as follows.
S = L × (8/10) ÷ (2/3) + M × (2/10) ÷ (1/3)
Where S: synthetic mass spectral data
L: Mass spectral data (database) of olive oil
M: Mass spectral data of soybean oil (database)
In order to make the synthetic mass spectrum data obtained in this way the same as the data condition of the mass spectrum in the database, for example, standardization is performed by setting the maximum value to 1 and the minimum value to 0, thereby standard mass spectrum data (Sr) It becomes.

One method of using the standard mass spectral data (Sr) is to convert each signal of the processed mass spectral data (Su) of the measured object into signals derived from the matrix and signals other than the matrix based on the standard mass spectral data (Sr). It can be used to classify. This makes it possible to avoid erroneously determining a matrix-derived signal as a signal that should be the original target.
In addition, when using a mass spectrometer capable of measuring accurate mass such as TOFMS, if it can be confirmed by using standard mass spectrum data (Sr) that the processed mass spectrum data (Su) is not affected by matrix components, Therefore, it can be judged that each component has been sufficiently separated only by mass separation using a mass spectrometer, so that it is as accurate as evaluating the accurate mass of components separated using a separation means such as a gas chromatograph. It becomes possible to evaluate the mass of each component.

<Noise removal process>
In the noise removal process E, the standard mass spectrum data (Sr) is subtracted from the processed mass spectrum data (Su) using the standard mass spectrum data (Sr) synthesized in the standard mass spectrum data (Sr) synthesis process D. By doing so, signal value duplication due to matrix components and matrix-derived components is removed from the processed mass spectrum data (Su).
The standard mass spectrum data (Sr) is mass spectrum data including only the specified matrix component. On the other hand, the processed mass spectrum data (Su) is mass spectrum data including the specified matrix component and other components.
Therefore, by subtracting the standard mass spectrum data (Sr) from the processed mass spectrum data (Su), it is possible to remove the matrix-derived signal that has had a great influence on the mass spectrum data.
This makes it possible to more clearly analyze the signals of components other than the specified matrix component, and to specify the component and further specify the content.

In the noise removal step E, residual change data obtained by subtracting the standard mass spectral data (Sr) from the processed mass spectral data (Su) and further dividing the residual data (Su-Sr) by the standard mass spectral data (Sr). (Su-Sr) / (Sr).
In the residual data (Su-Sr) obtained by subtracting the standard mass spectral data (Sr) from the processed mass spectral data (Su), in addition to signals of components other than matrix components, noise in measurement data obtained by a mass spectrometer is included. It is mixed.
Therefore, as means for further reducing the influence of noise in the residual data (Su-Sr), in the present invention, residual change data (Su-Sr) is further divided by standard mass spectral data (Sr). Su-Sr) / (Sr) is obtained.
By obtaining the residual change data (Su-Sr) / (Sr), it becomes possible to highlight the change in signal intensity due to the components other than the matrix component, while reducing the influence of noise.

  An unknown vegetable oil was analyzed as a measurement object X using an ion attachment ionization mass spectrometer. Actually, a vegetable oil having a mixing ratio of 99.5% olive oil and 0.5% coconut oil was used as an object to be measured X for analysis.

  As database, mass spectrum data of 9 kinds of vegetable oils such as olive oil, sesame oil, sweet almond oil, soybean oil, grape seed oil, rapeseed oil, argan oil, diacylglycerol oil, coconut oil (maximum value is 1 and minimum value is 0) Normalized) and signal intensity ratios for all two combinations of the nine vegetable oils were constructed.

The measurement object X was applied to an ion attachment ionization mass spectrometer, and three-dimensional measurement data including a mass-to-charge ratio (m / z), time, and signal intensity was obtained.
In the mass spectrum data extraction step A, the measurement data was used to perform principal component analysis with the signal intensity at each mass to charge ratio (m / z) as a variable at each sampling time. Then, referring to the inflection points of the two-dimensional chromatogram of the loading data and time of the first to n-th principal components obtained by the principal component analysis, the measurement time showing the characteristic mass spectrum pattern of the measurement object X A time zone of 0 minutes to 8 minutes and 30 seconds was selected and adopted as a certain time zone, and average mass spectrum data during this measurement time of 0 minutes to 8 minutes and 30 seconds was extracted.
In the mass spectrum data processing step B, the extracted mass spectrum data was normalized with the maximum value being 1 and the minimum value being 0, and processed mass spectrum data (Su) having the same conditions as the database conditions.
Processed mass spectrum data (Su) of the measurement object X is shown in FIGS. The horizontal axis is m / z, and the vertical axis is the normalized signal intensity. (B) is an enlargement of the normalized signal intensity scale of (A). The data shown in FIG. 5 (A) generally shows an olive oil signal and no coconut oil signal. In the enlarged data shown in FIG. (B), as shown by the arrows, a slight signal of coconut oil appears, but in reality, it is buried in another signal.

In the matrix specifying step C, the processed mass spectrum data (Su) was subjected to a double regression analysis and collated with a database to obtain a matrix component and a mixing ratio of the measurement object X. The analysis result by the double regression analysis is shown in FIG.
In the analysis result of FIG. 6, the matrix of the measurement object X is 100% olive oil. Actually, 0.5% coconut oil was mixed as described above, but since the concentration was as low as less than a few percent, it was not adopted as a matrix component within the error range.

In the standard mass spectrum data synthesis step D, based on the analysis result by the above-described regression analysis, that is, the analysis result that the matrix of the component X to be measured is 100% olive oil, the mass spectrum data of olive oil is 100% from the database, the other matrix The standard mass spectral data (Sr) was synthesized with the component as 0% (in fact, it was not necessary to synthesize).
FIG. 7 shows standard mass spectrum data of the object X to be measured. This data is nothing but the mass spectrum data of olive oil in the database.

In the noise removal step E, the standard mass spectrum data (Sr) is subtracted from the processed mass spectrum data (Su) of the object X to obtain residual data [(Su) − (Sr)].
FIG. 8 shows residual data [(Su) − (Sr)]. According to FIG. 8, in addition to olive oil indicated by white arrows, a signal of coconut oil indicated by black arrows is also visually recognized. That is, it can be visually recognized that coconut oil is contained in the measurement object X.
Further, in the noise removal step E, the residual data is divided by the standard mass spectrum data (Sr) to obtain residual change data [(Su) − (Sr)] / (Sr).
FIG. 9 shows residual change data [(Su) − (Sr)] / (Sr) of the measurement object X. This residual change data can weaken signals due to matrix components and noise, and can make signal changes due to components other than matrix components stand out. That is, in FIG. 9, the signal by coconut oil shown with the black arrow clearly appears.

  An unknown vegetable oil was analyzed as an object to be measured Y using an ion attachment ionization mass spectrometer. Actually, vegetable oil having a mixing ratio of 20% olive oil and 80% soybean oil was used as the measurement object Y.

  As in the case of Example 1, the database includes mass spectral data of nine kinds of vegetable oils such as olive oil, sesame oil, sweet almond oil, soybean oil, grape seed oil, rapeseed oil, argan oil, diacylglycerol oil, and coconut oil (maximum A standardized value with a value of 1 and a minimum value of 0) and signal intensity ratios for all two combinations of the nine vegetable oils were used.

As in the case of Example 1, the object to be measured Y was subjected to an ion attachment ionization mass spectrometer, and three-dimensional data including a mass-to-charge ratio (m / z), time, and signal intensity was obtained. Then, two-dimensional mass spectrum data was extracted through the mass spectrum data extraction step A, and processed mass spectrum data (Su) was obtained through the mass spectrum data processing step B.
FIG. 10 shows processed mass spectrum data (Su) of the object to be measured Y. The horizontal axis is m / z, and the vertical axis is the normalized signal intensity. The data shown in FIG. 10 generally shows the signals for olive oil and soybean oil. Whether or not other components are mixed is not certain from this processed mass spectrum data.

Further, in the matrix specifying step C, the processed mass spectrum data (Su) of the object to be measured Y is subjected to repeated regression analysis and collated with a database to obtain a matrix component and a mixing ratio of the object to be measured Y. The analysis result by the double regression analysis is shown in FIG.
In the analysis result of FIG. 11, the matrix of the measurement object Y is 19.4% for olive oil and 80.6% for soybean oil. Actually, 20% olive oil and 80% soybean oil are mixed as described above.

In the standard mass spectrum data synthesis step D, based on the analysis result by the above-described regression analysis, that is, the analysis result that the matrix of the component Y to be measured is 19.4% olive oil and 80.6% soybean oil, Standard mass spectral data (Sr) containing 19.4% olive oil and 80.6% soybean oil was synthesized from the signal intensity ratio between the matrices (signal intensity ratio between olive oil and soybean oil).
FIG. 12 shows standard mass spectrum data (Sr) of the measurement object Y.

In the noise removal step E, the standard mass spectrum data (Sr) is subtracted from the processed mass spectrum data (Su) of the workpiece Y to obtain residual data [(Su) − (Sr)].
FIG. 13 shows residual data [(Su) − (Sr)] of the object to be measured Y. According to FIG. 13, in the residual data, it is possible to determine that signals other than the signals due to olive oil and soybean oil are small and other components are not included. But not enough to make a clear decision.
Therefore, in the noise removal step E, the residual data is divided by the standard mass spectrum data (Sr) to obtain residual change data [(Su) − (Sr)] / (Sr) of the measurement object Y. .
FIG. 14 shows residual change data [(Su) − (Sr)] / (Sr) of the measurement object Y. This residual change data can weaken signals due to matrix components and noise, and can make signal changes due to components other than matrix components stand out. That is, in FIG. 14, since the residual change data is small and uniform in each m / z ratio, it can be determined that no other component is mixed.

  The mass spectrometric method according to the present invention enables the analysis time to be shortened, the denaturation and loss of the component to be measured in the separating means, and the real-time monitoring of the change in the measured object can be achieved without combining the separating means such as chromatograph. While maintaining the merits, etc., it is possible to remove noise from congested components, etc., even from measurement data obtained with a soft ionization mass spectrometer, and to perform component identification well up to trace components There is great industrial applicability in the analytical field.

Claims (8)

From the three-dimensional measurement data consisting of time, mass-to-charge ratio (m / z) and signal intensity obtained with a soft ionization mass spectrometer, the mass-to-charge ratio (m / z) and signal intensity for the object to be measured A mass spectrum data extraction step for extracting dimensional mass spectrum data;
A mass spectrum data processing step of processing the mass spectrum data extracted in the mass spectrum data extraction step into the same data conditions as a database created in advance;
One or a plurality of matrices included in the processed mass spectrum data (Su) by collating the processed mass spectrum data (Su) obtained in the mass spectrum data processing step with the database using multivariate analysis means And a matrix specifying step for specifying the components of the components and the mixing ratio of the components;
A standard mass spectrum that is synthesized from the mass spectrum data of the database as standard mass spectrum data (Sr) from the matrix components and mixing ratios specified in the matrix specifying step, as mass spectrum data corresponding to the components and mixing ratios. Data (Sr) synthesis step;
A noise removing step of subtracting the standard mass spectrum data (Sr) from the processed mass spectrum data (Su) to remove duplication of signal values due to matrix components and matrix-derived components from the processed mass spectrum data (Su);
A mass spectrometric method characterized by comprising:   In the noise removal process, the residual data obtained by subtracting the standard mass spectral data (Sr) from the processed mass spectral data (Su) is further divided by the standard mass spectral data (Sr), so that a minute change buried in the matrix is obtained. The mass spectrometric method according to claim 1, wherein: is made to stand out as residual change data.   In the mass spectrum data extraction step, each mass-to-charge ratio (m / z) in a certain time zone is obtained from three-dimensional measurement data consisting of time, mass-to-charge ratio (m / z) and signal intensity in a soft ionization mass spectrometer. 3. The two-dimensional mass spectrum data comprising a mass-to-charge ratio (m / z) and signal intensity is obtained by obtaining an average value or a sum value of signal intensity for each z). Mass spectrometry method.   The adoption of a fixed time zone is the principal component analysis in which the signal intensity at each sampling time is a variable for each mass-to-charge ratio (m / z) of the three-dimensional measurement data. The fixed time zone is adopted based on two-dimensional data composed of component loading data and time and one or more inflection points extracted from the two-dimensional data. The mass spectrometry method described.   5. The mass spectrometry method according to claim 4, wherein the inflection point of the two-dimensional data obtained in the mass spectrum data extraction step is displayed on a total ion chromatogram obtained from the three-dimensional measurement data.   In the mass spectrum data processing step, the extracted mass spectrum data is processed into mass spectrum data having the same data conditions as a database created in advance by performing at least one of normalization and scaling processing. Item 6. The mass spectrometry method according to any one of Items 1 to 5.   In the matrix identification process, the matrix components and their mixing ratios are identified by collating the processed mass spectrum data (Su) with the mass spectrum data in the database using at least the modified regression analysis means as the multivariate analysis means. The mass spectrometry method according to claim 1, wherein the mass spectrometry method is performed.   In the standard mass spectrum data (Sr) synthesis process, each mass spectrum data of one or more matrix components specified in the matrix specifying process is adopted from the database, and each mass spectrum data from the adopted database is used as the 1 The mass spectrometry method according to any one of claims 1 to 7, wherein the standard mass spectrum data (Sr) is synthesized by combining corresponding to mixing ratios of a plurality of matrix components.

Images