WO2009110026A1 - Method for mass spectrometry and mass spectroscope - Google Patents

Abstract

On a revolving orbit (6) formed of a multiple revolving ion optics system (3), a detector (4) is arranged which is capable of detecting the passing-through ion in a nondestructive manner. The detector (4) acquires a signal for observing a period from launch of ion from an ion source (1) until elapse of a given signal observation time, and a signal processor section (7) calculates multiple products for auto-correlation function of such a signal to determine a periodic spectrum extracting cycles peculiar to each ion. Since the mass and revolving cycle of the ion correspond to each other, a mass spectrum is obtained from a periodic spectrum. A peak width of a cycle which is determined by the auto-correlation function is independent of the signal observation time, and thus separability of a cycle peak can be assured without increasing the signal observation time. Further, calculation of products of the auto-correlation function enables two peaks on the signal which are separated only at the time when revolving is increased in the number of times to be separated on the periodic spectrum as well, thereby improving the mass resolution.

Description

Mass spectrometry method and mass spectrometer

The present invention relates to a mass spectrometric method and a mass spectroscope, and more particularly to a mass spectrometric method and a mass spectroscope using an ion optical system that separates ions according to mass by flying ions along a circular orbit.

In general, time-of-flight mass spectrometers measure the flight time required for ions to fly a certain distance based on the principle that ions accelerated with a constant energy have a flight speed according to the mass. The mass of the ion is calculated. Therefore, it is effective to increase the flight distance in order to improve the mass resolution. Using this, a multi-round time-of-flight mass spectrometry that achieves high mass resolution by extending the flight distance of ions by increasing the number of rounds using a multi-round ion optical system that makes multiple rounds of ions along a closed orbit Devices have been developed (see Patent Documents 1, 2, and 3, Non-Patent Document 1, etc.). For the same purpose, a multi-reflection time-of-flight mass spectrometer using a multi-reflection ion optical system has been developed that extends the flight distance by reflecting ions several times with a reflection electric field.

Although the two types of time-of-flight mass spectrometers have different ion optical system configurations, the principle of increasing the mass resolution by repeatedly flying the same trajectory is the same. In the present specification, description will be made by taking a multi-turn type as an example, but all the descriptions can be applied to a multi-reflection type by simply replacing the term “round” with “reflection”. Therefore, hereinafter, “multiple circulation” used in this specification includes “multiple reflection”.

As described above, although the multi-round time-of-flight mass spectrometer can achieve a high mass resolution, there is a drawback derived from the fact that the ion flight path is a closed orbit. The drawback is that as the ions circulate, ions that fly faster due to the smaller mass overtake slower ions on a closed orbit due to the larger mass. When the time-of-flight spectrum is measured in a state where the overtaking of ions occurs in this way, peaks having different laps, that is, having different flight distances are mixed on the spectrum. In that case, since the mass of the ion and the flight distance cannot be uniquely determined, the mass spectrum cannot be obtained directly from the time-of-flight spectrum.

Because of these drawbacks, conventionally, a multi-turn time-of-flight mass spectrometer is generally used to observe only a narrow mass range in which no overtaking occurs among ions generated by an ion source. . One technique for solving this problem is proposed in Patent Document 4. In this proposed mass spectrometry, a plurality of different time-of-flight spectra are obtained by performing a plurality of measurements while changing the time from the incidence to the exit of ions to the orbit. These time-of-flight spectra include peaks for ions of the same mass that have different lap times. Therefore, by calculating the multiple correlation function of the plurality of time-of-flight spectra, the time-of-flight spectrum of a single revolution is reconstructed and converted into a mass spectrum.

However, in the above method, if the number of time-of-flight spectra used for the multiple correlation function calculation is small, there is a possibility that a false peak that does not exist originally is artificially generated. Therefore, in order to ensure sufficient accuracy, it is necessary to use a large number of time-of-flight spectra for the multiple correlation function calculation, and it is therefore inevitable that the measurement takes a considerable amount of time. In addition, it is necessary to prepare a large amount of samples in order to perform many measurements. For this reason, this method is theoretically possible, but it is not very practical.

The above-described problem related to the overtaking of ions is an unavoidable problem in terms of a configuration in which ions that have been circulated multiple times are discharged from the circular orbit and detected by a detector such as a microchannel plate (MCP). Therefore, as a different approach to solving this problem, a technique for obtaining a mass spectrum by combining a multi-circular ion optical system and an ion non-destructive detector and Fourier-transforming the signal obtained by the detector is disclosed in Patent Literature. 5 is proposed.

In this method, a detector capable of detecting passing ions electromagnetically and non-destructively (maintaining ions as they are) on a periodic orbit by a multi-circular ion optical system is installed, and a signal for a predetermined period of time is installed. To get. If the initial energy of ions is constant regardless of the mass, the frequency at which the ions circulate depends only on the mass. Therefore, the detection signal includes a plurality of periodic signals having a frequency depending on the mass of ions. This signal can be converted into a frequency spectrum by Fourier transform, and the frequency spectrum can be easily converted into a mass spectrum from the correspondence between the frequency and the mass. Thereby, the target mass spectrum can be obtained by subjecting the detection signal acquired by only one measurement to Fourier transform.

This method can be said to be an excellent mass spectrometry method for avoiding the problems of the multi-round time-of-flight mass spectrometer. Hereinafter, in this specification, this method will be referred to as “Fourier Transform Multiple Loop Mass Spectrometry (FT-MT / MS)”.

In addition to FT-MT / MS, mass spectrometry (FT / MS) using Fourier transform is conventionally known. A typical example is Fourier transform ion cyclotron resonance mass spectrometry (FT-ICR / MS) in which mass analysis is performed using the frequency of ion cyclotron motion in a uniform magnetic field (see Non-Patent Document 2). In recent years, an apparatus for reciprocating ions in an electrostatic field of a specific shape and performing mass spectrometry using the frequency has been developed.

As described above, FT-MT / MS is an excellent method for avoiding the problems of the multi-round time-of-flight mass spectrometer, but has one problem compared to other FT / MS. It is a problem related to the signal observation time required to obtain a predetermined mass resolution. The problem will be described below.

Generally, in FT / MS, the mass resolution obtained depends on the frequency resolution of the apparatus. As will be described below, the signal observation time required for obtaining a predetermined frequency resolution depends on the frequency of ion motion included in the signal.

The mass resolution R (= m / Δm) in FT / MS is proportional to the frequency resolution f / Δf of the apparatus and can be expressed by the following equation (1).
R = m / Δm = α (f / Δf) (1)
Here, f is the frequency of motion for ions of mass m, and Δf is the frequency peak width. Α is a constant depending on the device, and is 1 for FT-ICR / MS and 0.5 for FT-MT / MS.

Here, it can be confirmed by simple calculation that the frequency peak width Δf obtained by Fourier transform is in the relationship of the signal observation time S and the following equation (2).
Δf = 1 / S (2)
That is, the frequency peak width Δf is inversely proportional to the signal observation time S. This is a fundamental and mathematical relational expression that does not depend on the characteristics of the apparatus. From the equations (1) and (2), the relationship between the signal observation time S necessary for obtaining a constant mass resolution R and the ion motion frequency f is given by the following equation (3).
S = R / (αf) (3)
From this, it can be seen that the signal observation time S necessary for obtaining a constant mass resolution R is inversely proportional to the frequency f of the ion motion.

Here, the frequency of each motion in FT-MT / MS and FT-ICR / MS will be compared. First, for FT-MT / MS, if the flight distance per circular orbit by a multi-circular ion optical system is L [m] and the acceleration voltage of ions is V [kV], the circulation for a monovalent ion of mass m The frequency f _MT is given by the following equation (4).
f _MT = (1 / L) √ (2 eV / m) (4)
Here, e is an elementary charge. As a common-sense value for the apparatus conditions, if the flight distance L of one lap is 1 [m] and the acceleration voltage V is 10 [kV], the equation (4) becomes the following equation (5).
f _MT = 1.3882 × 10 ⁶ / √m [Hz] (5)

Next, for FT-ICR / MS, if the magnetic flux density of the uniform magnetic field is B [T], the frequency of the cyclotron motion is given by equation (6).
f _ICR = eB / (2πmL) (6)
At a magnetic flux density B = 10 [T] which is a typical value in a general apparatus, the following equation (7) is obtained.
f _ICR = 1.5356 × 10 ⁸ / m [Hz] (7)
FIG. 5 shows the relationship between the mass m and the motion frequency f for each of FT-MT / MS and FT-ICR / MS, which is obtained by the equations (5) and (7).

From FIG. 5, it can be seen that although depending on the mass, the frequency of ion motion in FT-MT / MS is smaller by one to two digits than that in FT-ICR / MS. This leads to the following important conclusions: Assuming that the signal observation times necessary for obtaining a constant mass resolution R in each of FT-MT / MS and FT-ICR / MS are S _MT and S _ICR , (3), (5) and (7 From equation (8), equation (8) is obtained.
S _MT / S _ICR = 2.2 × 10 ² / √m (8)
In other words, depending on the observed mass, in order to obtain the same mass resolution with FT-MT / MS as with FT-ICR / MS, the signal observation time is one to two digits longer than with FT-ICR / MS. It is understood that is necessary.

As an example, assuming that an ion having a mass of 1000 is observed at a mass resolution of 100,000, a signal observation time of 6.5 seconds is required for FT-ICR / MS, whereas 46 for FT-MT / MS. A signal observation time as long as 2 seconds is required. Thus, FT-MT / MS requires a signal observation time that is one to two digits longer than other FT / MS (specifically, FT-ICR / MS) to achieve the same mass resolution. Therefore, the measurement throughput is significantly reduced. This is one problem in FT-MT / MS.

One of the causes of the above problems lies in the characteristics of signal analysis by Fourier transform, in which the width of the frequency peak by Fourier transform depends on the signal observation time. In FT-MT / MS, when the signal observation time is extended, that is, when the number of laps is increased, ion packets (a set of ions taking into account the spread in the time direction) of two types of ions close to each other in mass are separated. However, due to the characteristics of Fourier transform, even if the ion packets are separated, the frequency corresponding to the two ion packets cannot be separated on the frequency spectrum unless sufficient signal observation time is taken. That is, although the signal observation time is extended to improve the spatial separation, the improvement in the separation does not necessarily lead to the improvement in mass resolution.

JP 11-1335060 A Japanese Patent Application Laid-Open No. 11-135031 JP 11-195398 A JP 2005-79049 A JP-A-2005-79037 M. Toyoda and three others, “Multi-turn time-of-flight mass spectrometers with electrostatic sectors”, Journal of・ Mass Spectrometry (J. Mass Spectrom.), 38, pp. 1125-1142, 2003 Marshall (AGMarshall) and two others, “Fourier Transform Ion Cyclotron Resonance Mass Spectrometry: Fourier Transform Ion Cyclotron Resonance Mass Spectrometry: A Mass Spectrom. Rev.), 17, pp.1-35, 1998

The present invention has been made to solve the above-described problems. In a mass spectrometer that combines a multi-circular ion optical system and an ion non-destructive detector, the problems of conventional signal analysis by Fourier transform are solved. The main purpose is to overcome and achieve high resolution and high throughput (or high speed).

The mass spectrometric method according to the present invention, which has been made to solve the above-mentioned problems, spatially separates ions derived from a sample according to the mass by repeatedly flying ions derived from a sample along a circular orbital orbit. A mass spectrometric method using a mass spectroscope comprising a multi-circular ion optical system and a detection means for non-destructively detecting ions flying on the circular orbit or reciprocating orbit,
By calculating an auto-correlation function (ACF: Auto-Correlation Function) for the observation signal obtained within a predetermined time by the detection means, the periodicity of the observation signal is extracted to obtain a periodic spectrum, and the mass is calculated from the periodic spectrum. The spectrum or the mass of each ion is calculated.

A mass spectrometer according to the present invention made to solve the above problems is an apparatus for carrying out the mass spectrometry method according to the present invention,
a) a multi-circular ion optical system that spatially separates ions according to mass by repeatedly flying ions derived from a sample along a circular or reciprocating orbit;
b) non-destructive detection means for detecting ions flying on the orbit or reciprocating orbit,
c) an arithmetic processing means for extracting a periodicity of the observed signal and calculating a periodic spectrum by calculating an autocorrelation function for the observed signal obtained within a predetermined signal observation time by the detecting means;
d) conversion processing means for calculating the mass spectrum or the mass of each ion from the periodic spectrum;
It is characterized by having.

The signal observed by the detection means from when the ions are emitted from the ion source arranged outside the multi-circular ion optical system until the predetermined signal observation time elapses indicates information on the ions passing therethrough. Includes all. That is, every time an ion having a certain mass makes a round orbit (or travels one way along the reciprocating orbit), a peak for that ion should appear, and this peak has a period corresponding to the mass. When a plurality of types of ions having different masses are mixed, the peaks having the periodicity as described above are mixed or sometimes overlapped. In the present invention, signal analysis is performed on such an observation signal using an autocorrelation function instead of Fourier transform. Intuitively, the autocorrelation function is a function that indicates how much an observed signal at a certain time is correlated with a signal after a certain period of time has elapsed. Therefore, by finding a signal having a high autocorrelation, it is possible to extract the periodicity of the peak according to the mass of the various ions described above.

As an autocorrelation function used here, an autocorrelation function C () defined by the following equation (9) with respect to an observation signal f (t) (0 ≦ t ≦ S) obtained within the signal observation time S: τ) (0 ≦ τ ≦ S) can be used. Also, since the denominator of equation (9) is not important, only the numerator of the following equation can be used.

Although the theoretical explanation will be described later in detail, the peak width appearing in the autocorrelation function is twice the peak width in the original time domain and does not depend on the signal observation time in terms of the calculation principle. On the other hand, the width of the frequency peak by the Fourier transform used in the conventional FT-MT / MS depends on the signal observation time even in terms of the calculation principle. Therefore, in the mass spectrometry method and the mass spectrometer according to the present invention, the signal observation time is extended with respect to two ion packets having close masses, and at least the ion packets are separated in space. There is a possibility that the respective masses can be calculated by separating the periods corresponding to the ion packets.

However, there are cases in which the period for two types of ions that are separated on the observation signal cannot be separated by simply performing processing using the autocorrelation function. This is a case where the masses of the two types of ions are very close and separation can be confirmed on the observation signal only after many rounds. Therefore, in order to improve the resolution of the period for ions separated on the observation signal, in the present invention, for example, the following arithmetic processing is preferably performed on the autocorrelation function of the observation signal.

　ここで、Ｗ（ｎ_T）は重み関数であり、１又はｎ_Tを変数とした所定の関数を用いることができる。[ｘ]はｘを超えない最大の整数を表す。 That is, as one of arithmetic processing methods, the autocorrelation function product F (T) may be obtained from the following equation (10) for the result obtained in the autocorrelation function.

Here, W (n _T ) is a weight function, and a predetermined function with 1 or n _T as a variable can be used. [x] represents a maximum integer not exceeding x.

This is because C (2T), C (3T),... When the autocorrelation function C (s) of the signal observed by the multi-circular ion optical system has a significant value C (T) at s = T. , C (NT) also has a significant value. If two peaks that are not separated in the vicinity of τ = T in C (τ) are separated in the vicinity of τ = NT, the periodicity of the signal calculated that the two peaks are not separated is It will be negated by the operation. In other words, periods corresponding to two very close ions that are not separated only by the calculation of the autocorrelation function are separated. In this way, the periodic resolution can be improved to the same degree as the time-of-flight resolution achieved by the multi-circular ion optical system.

If the above arithmetic processing is more generalized, when obtaining a periodic spectrum based on the autocorrelation function for the observation signal, a small function is used by using the value of the autocorrelation function in a time delay of an integral multiple period for one period. It can be said that this is an arithmetic process in which a larger weight is given to the value and the function value is referred to.

The autocorrelation function product directly represents the signal intensity with respect to the flight period of the multi-turn ion optical system, that is, the periodic spectrum. When the acceleration voltage of ions is constant, the cycle of one round depends on the mass of ions. Therefore, the periodic spectrum based on the autocorrelation function product is substantially equivalent to the mass spectrum, and the mass can be easily calculated.

According to the mass spectrometry method and the mass spectrometer according to the present invention, the signal observation time is shorter than that of the conventional FT-MT / MS by about two digits or three digits, which is sufficiently high to meet the extended flight distance. Mass resolution can be obtained. Further, multiple rounds of measurement need only be performed once, and it is not necessary to repeat the measurement for the same sample many times as in the case of MT / MS using a conventional multiple correlation function. Accordingly, the total measurement time can be shortened, and samples need only be prepared as much as necessary for one measurement.

It is a schematic block diagram of the mass spectrometer by one Example of this invention, (A) is an example using a multiple circulation ion optical system, (B) is an example using a multiple reflection ion optical system. The figure which shows the simulation result of the ion passage signal observed with a detector. The figure which shows the periodic spectrum obtained by calculating an autocorrelation function product with respect to an observation signal, and the figure which shows the periodic spectrum obtained based on original data (B). The figure which shows the calculation result of the mass identification of all the ion packets. The figure which shows the relationship between the mass and the motion frequency about each of FT-MT / MS and FT-ICR / MS. The schematic of the observation signal with respect to the single ion obtained with a multi-circular ion ion system.

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 ... Ion source 2 ... Ion entrance switch 3 ... Multiple round ion optical system 3 '... Multiple reflection ion optical system 4 ... Detector 5 ... Incidence track 6 ... Round track 6' ... Round track 7 ... Signal processing part

The principle of the mass identification method in the mass spectrometry method according to the present invention will be described below.

First, the signals observed when an ion nondestructive detector is used in a multi-circular ion optical system will be described. In general, a multi-circular ion optical system is designed to have a first-order or higher time convergence point with respect to the initial position of ions, the incident angle of ions, and the energy of ions. Depending on the structure of the ion optical system and the initial conditions, the time width of the ion packet after going around 100 laps, that is, the temporal spread of the same type of ions, is on the order of several tens [ns] at the time convergence point. It is. On the other hand, the period when the ions make one round is approximately on the order of several to several tens [μs] under a common-sense acceleration voltage condition. Therefore, if the ion non-destructive detector is ideal and the passage time width of the ion packet can be accurately observed, the ion passage signal observed for the circular motion of one kind of ion is: A peak having a width of about 10 to several tens [ns] is a periodic signal that appears every 10 to several tens [μs]. FIG. 6 is a schematic diagram showing the state of an observation signal for a single type of ion.

これは、半値幅がΔｔであるピークが周期Ｔでｎ周回分、観測された場合の信号のモデルである。ここで、Δｔ＜＜Ｔ、である。 As an analytical model of this signal, a peak is given by a Gaussian distribution, and an ion passage signal f (t) according to the following equation (11) is assumed.

This is a model of a signal when a peak having a half-value width of Δt is observed for n cycles in a period T. Here, Δt << T.

ここでＡは定数である。このより、次の２つの項目を確認することができる。
　（１）自己相関関数では、信号に含まれる周期Ｔについて、信号観測時間を越えないその整数倍の周期に対してピークが得られる。
　（２）自己相関関数に現れるピーク幅は、元の信号に観測されるピーク幅の２倍となる。 Next, an autocorrelation function of the ion passage signal f (t) is calculated. Of the autocorrelation functions, the denominator is a constant factor introduced by definition and is not important for the purpose here.

Here, A is a constant. From this, the following two items can be confirmed.
(1) In the autocorrelation function, a peak is obtained with respect to a period T included in the signal with respect to an integer multiple of the period not exceeding the signal observation time.
(2) The peak width appearing in the autocorrelation function is twice the peak width observed in the original signal.

The above (1) is that the autocorrelation function has the property of extracting the periodicity included in the signal, and the observed signal is correlated with the time delay of iT (1 ≦ i ≦ n−1) due to its property. It is due to that there is. On the other hand, (2) shows that the peak width of the autocorrelation function depends only on the peak width of the observation signal. This is a significant feature in contrast to the fact that the frequency peak width in Fourier transform depends on the signal observation time.

Here, it is assumed that there are two ion packets that can be separated for the first time by overlapping n turns. It is defined that the peak full width of each ion packet is ΔT, and the cycle period is T and T + ΔT / n. As described above, this is a state in which two ion packets have a peak width and a period difference that are separated by each other for the first time after n rounds. It is assumed that a signal while these ion packets are repeated 2n times or more is observed by an ion non-destructive detector, which is f (t). At this time, according to the above definition, two peaks separated by the tail are observed in the vicinity of t = nT, and the peak interval is further opened by ΔT in the vicinity of t = 2nT.

The autocorrelation function is calculated for this observation signal f (t). Then, as described above, since the peak width of the autocorrelation function is twice the peak width of the observed signal, two peaks separated by a half-value width appear near t = nT, and have a tail near t = 2nT. Two separated peaks appear at the positions of τ = 2nT and τ = 2nT + 2ΔT. However, the physical meaning of this result is only that the periodicity of 2nT and 2nT + 2ΔT exists in the observed signal f (t) with a width of 2ΔT, respectively, and exists in the original signal f (t). The periods T and T + ΔT / n can not be analyzed. The ultimate purpose is to analyze the periodicity of T and T + ΔT / n included in the observation signal f (t) with the shortest signal observation time possible. Therefore, further ingenuity is required to increase the periodic resolution.

Here, attention is again paid to the nature of the autocorrelation function for the ion passage signal observed in the multi-circular ion optical system. In this autocorrelation function, as shown in (1) above, it is known that peaks appear not only in the cycle of ions, but also in an integer multiple cycle within a range not exceeding the signal observation time. . In other words, even if a significant value is calculated for a certain period, if a significant value is not calculated for all integer multiple periods within the range not exceeding the signal observation time, the calculated value is calculated. It can be determined that the period is simply within the width of other adjacent periods or is due to some noise. Specifically, in the above example, the state where the peaks separated at τ = 2nT and τ = 2nT + 2ΔT appear is that the observation signal does not include the periodicity of τ = 2nT + ΔT, and this is the cause. It suggests that the periodicity of τ = T + ΔT / 2n cannot exist.

Based on the above consideration, in one form of the mass spectrometry method according to the present invention, the multiple product given by the above equation (10) is calculated for the autocorrelation function. As is clear from the expression of this equation, it can be said that this is an operation for taking a geometric mean for all values of an integer multiple period that can be observed. By referring to the values over all integer multiple cycles, the reliability and accuracy of the determination are improved. Furthermore, by calculating the autocorrelation function product, it is possible to increase the influence of the determination of the autocorrelation function at the maximum number of rotations showing the highest resolution.

In the above calculation, the signal observation time required to obtain a predetermined resolution (periodic resolution) is about twice that when a multi-turn ion optical system is used as a time-of-flight mass spectrometer. This is because the peak width of the autocorrelation function is twice the peak width of the observation signal. This signal observation time is much shorter than the value required for the conventional FT / MS as described above. A general multi-turn time-of-flight mass spectrometer requires several hundred rounds in order to achieve a mass resolution of 100,000, and the flight time required is on the order of several to several tens [ms]. Therefore, also in the mass spectrometer according to the present invention, the required flight time is on the order of 10 to 100 [ms]. On the other hand, for example, in FT-ICR / MS, several [s] of signal observation time is required to achieve the same mass resolution. Therefore, it can be said that the signal observation time required by the present invention is a value that is about 1 to 2 digits smaller.

As described above, the mass spectrometry method and the mass spectrometer according to the present invention overcome the problems of the conventional FT-MT / MS with respect to the signal observation time required to achieve a predetermined mass resolution, and Compared with FT / MS, it provides an advantage in terms of signal observation time. Specifically, in the present invention, measurement with almost the same resolution is possible with a signal observation time shorter by one to two digits than other FT / MS, and the measurement throughput is greatly improved. .

The configuration of an embodiment of a mass spectrometer to which a mass spectrometry method according to the present invention is applied and the operation based on simulation calculation will be described with reference to the accompanying drawings. FIG. 1 is a schematic configuration diagram of a mass spectrometer according to the present embodiment, where (A) is an example using a multi-circular ion optical system, and (B) is an example using a multiple reflection ion optical system.

The mass spectrometer shown in FIG. 1 (A) gives an initial kinetic energy to various ions and outputs them all at once, that is, an ion source 1 that starts flight, a plurality of electrodes (not shown), and voltages to the electrodes. A multi-circular ion optical system 3 that repeatedly flies ions along the same circular orbit 6 by the action of a plurality of electric fields formed by electrodes, and an incident orbit 5 emitted from the ion source 1. The ion incident switch 2 for introducing ions traveling along the circular orbit 6 by the multi-circular ion optical system 3 and the number (quantity) of ions passing along the circular orbit 6 of the multi-circular ion optical system 3 and passing therethrough. ), And a signal processing unit 7 that receives the detection signal from the detector 4 and executes the arithmetic processing as described above.

The ion injection switch 2 is an orbital deflection element that can be driven in a pulsed manner. While the ion incident switch 2 is on, the ion trajectory is deflected so that ions are introduced from the incident trajectory 5 to the circular orbit 6, and when the switch 2 is off, it can be considered that the switch 2 is not present. The flying ions pass through the switch 2 as they are. The detector 4 can output an electrical signal corresponding to the passage amount of ions that are charged particles by using, for example, electromagnetic induction.

In the above configuration, the number of ion packets having different periods (that is, masses), the mass of each ion, and the ion intensity are generated by random numbers, and these ion packets are emitted from the ion source 1 and repeated on the orbit 6. A simulation calculation was performed on the ion passage signal obtained by the detector 4 when it was made to fly. As calculation conditions, the length Lin of the incident orbit 5 is 0.6 [m], and the length L of one turn of the circular orbit 6 is 1.0 [m]. The distribution of ion packets is an ideal Gaussian distribution, and the ion acceleration voltage is 10 [kV]. The detector 4 is ideal so that the passing signal can be accurately observed without any loss of ions. Further, the sampling rate of the ion passage signal is 1 [GHz]. The simulation result of the ion passage signal at this time is shown in FIG. The signal observation time is 0 to 100 [μs].

Although the characteristic signal analysis calculation as described above is performed on such an ion passage signal, it is wasteful to perform integration of the autocorrelation function over the entire time domain. Therefore, it is possible to roughly extract a period for performing a series of signal analysis operations by performing a small-scale and simple period determination on the observation signal in the multi-circular ion optical system before calculating the autocorrelation function. desirable. This point will be described in detail.

First, in the multi-circular ion optical system 3, it will be described that the mass range that can be introduced into the circular orbit 6 is limited by the size of the apparatus due to its structure. Ions ejected in a pulse form from the ion source 1 at a constant acceleration voltage are introduced into the orbit 6 through the incident trajectory 5 and the ion incident switch 2. The ions simultaneously emitted from the ion source 1 also vary spatially according to the mass before reaching the ion injection switch 2. Therefore, while the ion injection switch 2 is kept on in order to introduce ions having a low speed into the orbit 6, the ions introduced earlier at a higher speed orbit the orbit 6 and reach the ion injection switch 2 again. A situation can occur. In this case, only one of light ions previously introduced into the circular orbit 6 or heavy ions to be introduced into the circular orbit 6 later can continue to fly on the circular orbit 6. Thus, the mass range that can be introduced into the orbit 6 is limited by the length of the incident orbit 5 and the orbit 6.

Assuming that the minimum mass that can be introduced into the orbit 6 is mmin and the maximum mass is mmax, it is understood from the simple consideration that the relationship of the following equation (13) holds.
mmax / mmin = {1+ (L / Lin)} ² (13)
Assuming that the sampling of the ion passage signal in the signal processing unit 7 starts at the same time as the ion emission from the ion source 1, the time when the peak is first observed in the ion passage signal, that is, the passage of the ion packet having the minimum mass is passed. When the time is t ₁ , the passage time t _f of the maximum mass ion packet that can be introduced into the orbit 6 is given by the following equation (14).
t _f = {1+ (L / Lin)} t ₁ (14)
Therefore, it can be seen that the time range of 0 ≦ t <t _f in the observation signal in the detector 4 corresponds to a normal time-of-flight spectrum without overtaking.

With respect to an arbitrary peak observed in this time range, assuming that the time, i.e., the flight time until reaching the detector 4, is t _i , the cycle period T _i of the ion packet is from the ion source 1 to the detector 4. Can be easily obtained from the following equation (15).
T _i = (L / L ′) t _i (15)
Automatic peak detection is easy. Therefore, based on the flight time of each peak observed within the time range, all of the values that can exist as the cycle of the ion packet introduced into the orbit 6 are defined by the observed peak width. It is possible to calculate within an error range in advance, that is, before calculating an autocorrelation function or the like as described later. Specifically, it can be calculated using a time-of-flight spectrum in a time range of 0 ≦ t <t _f obtained when ions pass through the detector 4 for the first time.

For periods that are not calculated by this preliminary period determination process, even if the autocorrelation function and the autocorrelation function product are calculated later, only noise is generated. Therefore, by performing the preliminary cycle determination in this way and excluding the cycle estimated to be nonexistent from the calculation, the calculation amount can be greatly reduced and the calculation cost can be reduced. It also has a great effect in preventing the occurrence of artifacts due to computation.

FIG. 3A shows a periodic spectrum obtained by calculating the autocorrelation function product for the observation signal after performing the preliminary period determination as described above for the observation signal shown in FIG. Here, the weighting function W (n _T ) is 1. For comparison with this, FIG. 3B shows a periodic spectrum corresponding to the number and mass of ion packets generated by random numbers. In the periodic spectrum by the autocorrelation function product shown in FIG. 3A, it can be confirmed that all the generated periods are calculated without omission.

Furthermore, the mass was obtained from the period obtained from the periodic spectrum by the autocorrelation function product, and mass identification of all ion packets was attempted. FIG. 4 shows the calculation result. It can be seen that the mass identification result obtained from the autocorrelation function product agrees very well with the original data generated. Thereby, it has confirmed that high mass identification accuracy was realizable with the mass spectrometry method and mass spectrometer which concern on this invention.

As shown in FIG. 1B, the same applies when a multiple reflection ion optical system 3 ′ is used instead of the multiple circulation ion optical system 3. In this case, the ions pass through the detector 4 on the forward path and the return path of the reciprocating orbit 6 '. Therefore, the length L of one turn of the orbit 6 corresponds to the reciprocal distance from passing through the detector 4 to passing through the detector 4 in the opposite direction again, as shown in FIG. The length of one round trip of the inner round trip track 6 ′ is 2L.

As can be seen from a comparison between FIGS. 3A and 3B, the peak positions, that is, the periods are in good agreement, but the relative relationship of the intensity of each peak changes. Since the peak intensity does not affect the qualitative property at all, there is no problem when performing qualitative analysis by mass identification. On the other hand, when quantitative accuracy is required, it is necessary to improve the reproducibility of the peak intensity. For this purpose, the following equation can be used as a weighting function when calculating the autocorrelation function product.
W (n _T ) = (1 / n _T !) ^{1 / nT} (16)
This can be regarded as normalizing the influence of factors appearing when calculating the autocorrelation function.

It should be noted that any of the above-described embodiments is merely an example of the present invention, and it is obvious that changes, corrections, and additions may be made as appropriate within the scope of the present invention.

Specifically, in the above description, the resolution of the periodic spectrum has been improved by calculating the multiple product of the autocorrelation function, but in order to achieve the same effect, another method such as harmonic averaging should be used. You can also.

Claims (8)

By repeatedly flying samples-derived ions along a circular orbit orbit, a multi-circular ion optical system that spatially separates the ions according to the mass, and ions that fly on the orbit or the orbit A mass spectrometric method using a mass spectroscope equipped with non-destructive detecting means,
By calculating the autocorrelation function for the observation signal obtained within a predetermined time by the detection means, the periodicity of the observation signal is extracted to obtain the periodic spectrum, and the mass spectrum or the mass of each ion is calculated from the periodic spectrum. A mass spectrometric method characterized by: 2. The mass spectrometric method according to claim 1, wherein an autocorrelation function C (τ) (0 ≦ τ ≦) with respect to an observation signal f (t) (where 0 ≦ t ≦ S) obtained within the signal observation time S. A mass spectrometric method characterized by calculating only the molecule of the following formula or the following formula as S):

3. The mass spectrometric method according to claim 2, wherein when a periodic spectrum is obtained based on an autocorrelation function for an observation signal, a value of an autocorrelation function at a time delay of an integral multiple period is used for one period. A mass spectrometric method characterized in that a calculation is performed by giving a large weight to a small function value and referring to the function value. 4. The mass spectrometric apparatus according to claim 3, wherein a periodic spectrum is obtained by calculating an autocorrelation function product F (T) according to the following equation for the value of the autocorrelation function.

Here, W (n _T ) is a weight function, and a predetermined function with 1 or n _T as a variable is used. [x] represents a maximum integer not exceeding x. a) a multi-circular ion optical system that spatially separates ions according to mass by repeatedly flying ions derived from a sample along a circular or reciprocating orbit;
b) non-destructive detection means for detecting ions flying on the orbit or reciprocating orbit,
c) an arithmetic processing means for extracting a periodicity of the observed signal and calculating a periodic spectrum by calculating an autocorrelation function for the observed signal obtained within a predetermined signal observation time by the detecting means;
d) conversion processing means for converting the periodic spectrum into a mass spectrum;
A mass spectrometer comprising: 6. The mass spectrometer according to claim 5, wherein the arithmetic processing unit performs an autocorrelation function C (τ on the observation signal f (t) (where 0 ≦ t ≦ S) obtained within the signal observation time S. ) (0 ≦ τ ≦ S), and only the molecules of the following formula are calculated:

7. The mass spectrometer according to claim 6, wherein the arithmetic processing unit uses a value of an autocorrelation function in a time delay of an integral multiple period for one certain period, and gives a large weight to a small function value. And a periodic spectrum is obtained by executing a calculation referring to the function value. 8. The mass spectrometer according to claim 7, wherein the arithmetic processing unit obtains a periodic spectrum by calculating an autocorrelation function product F (T) according to the following equation for the value of the autocorrelation function. Mass spectrometer

Here, W (n _T ) is a weight function, and a predetermined function with 1 or n _T as a variable is used. [x] represents a maximum integer not exceeding x.

Images