RU2474917C1 - Method of separating ions of organic and bioorganic compounds in ion rotation-averaged electric field of sectioned cylindrical cell - Google Patents

Abstract

FIELD: chemistry.

SUBSTANCE: basis of the invention is separation of ions from external source according to the mass-to-charge ratio (m/z) in a mass spectrometer with an orbital ion trap without a centre electrode, where a longitudinal ion rotation-averaged electric field which is close to harmonic is formed by applying suitable potential across sections of the surface of the trap. Rotation of the analysed ions is excited when these ions are fed into the cell along its axis by a rotating electric field. Loss of ions on the wall of the cell is prevented by applying an effective potential which arises when ions pass near the surface with alternative potential on its sections. Detection of induced signals from oscillating ions is carried out when the rotating field is turned off, and processing of said signals for relatively wide ranges of m/z is carried out based on fast Fourier transformation. For the selected intervals of m/z, the registered data are approximated with exponentially damped sinusoidal curves whose frequencies characterise m/z of ions, and indices of the damped exponential characterise the average frequency of collision of the analysed ions with atoms and molecules of residual gases. Axial input of ions into the cell enables their accumulation, dissociation, energy analysis and separation according to mass, charge and mobility.

EFFECT: efficiency accumulation, separation and detection of ions.

12 cl, 5 dwg

Description

FIELD OF THE INVENTION

The invention relates to mass spectrometry. In particular, we describe below a method for separating ions introduced from an external source into a cylindrical cell (hereinafter referred to as a dynamic trap), where ions rotating around the trap axis make free, on average, close to harmonic vibrations along this axis. Registration of the induced signal from these ions and its frequency analysis, for example, based on the fast Fourier transform, gives the mass spectrum in the same way as in known mass spectrometers with an orbital ion trap or ion-cyclotron resonance devices with Fourier transform (ICR PF). The main differences from the “classical” orbital trap are the absence of a central electrode of complex shape, the sectioning of the surface of the external electrode in this case, a cylindrical shape, and the axial rather than tangential introduction of the analyzed ions into the trap. Axial input and excitation of the resonant rotation of ions with selected m / z values allows organizing the accumulation of desired ions directly inside the trap in high resolution mode, isolation and dissociation of selected ions, or additional separation of these selected ions by their energy, mobility and charge. The proposed approaches and methods are useful for qualitative and / or quantitative chemical and biological analysis.

State of the art

In recent years, the greatest successes in the separation of ions by m / z and the accuracy of determination of this basic mass spectrometric characteristic of ions have been achieved in ion-cyclotron resonance mass spectrometry with Fourier transform [1, 2]. The measuring cell of the ICR PF mass spectrometer is the Penning ion trap. In such a trap, ions are held by a strong magnetic field in the radial direction, and in the direction along the magnetic field they are held by an electric field. For a given induction of the magnetic field in which the ICR measuring cell is located, which basically sets the cyclotron rotation period of this ion, to increase the resolution and accuracy of mass measurement, it is necessary to increase the detection time of the induced current. For this, it is necessary that the ion cloud of each type of tones, performing cyclotron motion, move synchronously for as long as possible.

The main factors limiting the time during which the detected current does not decay is the pressure of the residual gases and the anharmonicity of the electric field in the cell. The collision of an ion with an atom or molecule of residual gases leads to its death or to a change in the phase and radius of rotation, differences in the distribution of electric potentials along and across the magnetic field from parabolic ones lead to the erosion of rotating ion packets. Due to recent advances in vacuum technology, the value of the first factor is decreasing, and the quality of the electric field is becoming the main source of influence on the resolution of ICR mass spectrometers. The ideal field of the ICR cell is a hyperbolic field (its equipotential surfaces are hyperbolic surfaces). A fairly large number of works are known where they are trying to create such a field with varying degrees of approximation.

Certain successes have been achieved in increasing the analytical capabilities of ICR mass spectrometers along this path, but they are still quite far from the limiting values determined by the pressure of the residual gases [3, 4].

In [5, 6], another idea was proposed and implemented to solve this problem, which consists in the fact that in order to achieve the maximum possible qualities of the recorded data, it is not necessary to strive to create an ideal hyperbolic field. Since ions make a fast cyclotron rotation, and the rest of the movements are much slower, these slow movements can be considered as occurring in an electric field averaged over cyclotron trajectories. This approach is similar to the well-known method of quasi-stationary concentrations of N. N. Semenov when considering the kinetics of reactions involving highly active particles. If the averaged field has parabolic potential distributions along the magnetic field and along the radius in the orthogonal plane, then the necessary properties of the separation of ion movements along the axes, the isochronism of their averaged trajectories will be preserved, and the recorded signal will have the necessary expected properties.

, разрезая цилиндрическую часть на сегменты и подбирая форму этих сегментов. Тогда граничные условия для потенциала

имеют вид:In the case of an ICR measuring cell, an averaged hyperbolic potential can be created

by cutting the cylindrical part into segments and choosing the shape of these segments. Then the boundary conditions for the potential

have the form:

,

,

where R is the radius of the cell, and A and B are arbitrary coefficients. This condition is satisfied if the lateral surface of the measuring cell is cut, as shown in Fig. 1. Locking voltage is applied to the thinner, concave electrodes, while the wider, convex electrodes are grounded. The sections are parabolic in shape and are defined by the equation:

where α is the angular coordinate of the cut point, N is the number of electrodes of each type, α ₀ is the coefficient that determines the distance between the cuts in the middle of the cell, and a is the half length of the cell.

In order to obtain an averaged hyperbolic field in a measuring cell of finite length, it is necessary that the boundary conditions at the end electrodes also correspond to the hyperbolic field, i.e. these electrodes would be in the form of a two-sheeted hyperboloid of revolution, and the blocking voltage mentioned above was applied to them. In the practical construction [5, 6], the hyperboloids were replaced by contacting spherical surfaces with a radius of 148.7 mm with a deviation from the hyperboloids not exceeding 0.02 mm with a cell diameter of 56 mm and a length of 150 mm. The number of sections on the cylindrical surface of the cell (N) was 8. The value α ₀ = π / 8-π / 60 was chosen so that the minimum width of the concave electrode was approximately 2 mm.

For experiments to achieve maximum resolution on a single isotope, reserpine (molecular weight 609 Da) was used in [5, 6], and bovine serum albumin, BSA (bovine albumin) was used for experiments on resolving isotopic multiplets of highly charged proteins molecular weight of about 67 kDa). When measuring the spectrum of one isotope of reserpine, the signal length reached 3 minutes. This gave a resolution of 24,000,000 on a magnet with 7 Tesla induction. A resolution of about 1200000 was obtained for the 49-charged albumin ion. Similar results are record-breaking for a magnetic field with 7 Tesla induction.

The main disadvantage of ICR mass spectrometers is their high cost, relatively large dimensions and difficulties in maintenance. All these disadvantages are associated with the presence of a superconducting magnet and the maintenance of its functioning, including a cooling system with liquid helium, a multi-tambour pumping system, etc. An alternative to ICR instruments are mass spectrometers with an orbital ion trap. For the first time, a device of this type was described by Alexander Makarov [11, 33]. A popular review of this technique is given in [12], as well as ICR mass spectrometry.

The combination of a mass spectrometer with an orbital ion trap with an external ion storage device, such as a linear ion trap, provides multi-level fragmentation (MS ⁿ ) to identify the analyte structure and allows conjugation with continuous ion sources, such as atmospheric pressure chemical ionization sources, sources with electrospray. The analytical capabilities (accuracy of determining the mass and resolution) of mass spectrometers of this type, combined with ease of use and small requirements for the occupied area, provide the breadth of their application, starting from the routine identification of compounds and sequencing of biopolymers, ending with the determination of trace amounts of components in complex mixtures, be it in proteomics, in drug metabolism, doping control, determination of contaminants in food and forage [13-16].

A mass spectrometer orbital trap consists of a spindle-like central electrode (16), Fig. 2, elongated along the axis, with a high voltage applied, and when ions are launched into the trap, this voltage increases further to capture these ions. The central electrode is surrounded by an external barrel-shaped grounded electrode (17). The external electrode is divided into two halves (19) and (20) in order to realize the registration of induced signals from ions. The electrodes are made in such a way that the potential distribution is quadratic-logarithmic with very high accuracy (the surfaces of these electrodes are equipotentials for potential (2)):

where k and R _m are the parameters determined by the applied voltage to the central electrode and the dimensions of the orbital trap.

When ions (18) begin their motion with a suitable energy and radius of rotation, stable trajectories are formed that combine rotation around the central electrode with oscillations along the axis and have the shape of a complex spiral. It is important to note that the axial movement of ions is completely independent of rotational motion. In an electric field (2), along the axis, the ions perform harmonic oscillations, the frequency and phase of which are constant over time. In this case, the oscillation frequency in this case depends only on the ratio of mass to ion charge (m / z):

where e is the elementary charge.

In the ICR mass spectrometer, the cyclotron frequency of rotation of ions in the first approximation is inversely proportional to the ratio of the mass of the ion to its charge, therefore, the orbital trap, usually inferior to the ICR in resolution at relatively small m / z, can have an advantage in the analysis of large biomolecules. In contrast to ICR, the orbital trap does not have the possibility of internal isolation of ions and their collision-induced dissociation and dissociation during the capture of slow electrons by multiply charged ions.

Typically, the resolving power of an orbital ion trap for an ion mass of 400 Da for a recording time of about 1 s does not exceed 100,000. For this, the ion oscillation frequency should be about 100 kHz. With a swing of about 5 cm, the average ion velocity along the axis of the orbital trap will be approximately 10 ⁶ cm / sec. At such speeds, the average number of collisions per unit time of ions with atoms or molecules of residual gases will be proportional to the speed of ions. Ions with a mass of 400 Da at this speed will have an energy of less than 250 eV. If the potential of the central electrode is about 5 kV, then the energy of ions rotating in the middle of the gap between the central and external electrode will be about 2.5 keV. Thus, the total speed of ions is approximately three times greater than their speed along the axis of the trap. This means that, at the same frequency of the recorded signal at the same residual gas pressure, the signal attenuation at ideal fields for ICR will be about three times slower than for an orbital trap, since the cyclotron rotation speed of the ions is about an order of magnitude greater than their average velocity along the magnetic fields. Thus, the reckoning for the absence of a magnetic field in the case of an orbital ion trap is a significant decrease in the achievable maximum frequency resolution for relatively small ions. The mass resolution due to the root dependence of the resonance frequency on the mass of the ion for an orbital trap is half that of the frequency resolution, while for the ICR, the values of both resolutions coincide. As already noted, for large-mass ions, an orbital ion trap can have a higher resolution than ICR mass spectrometers.

Another advantage of the orbital ion trap is the significantly smaller effect of the ion space charge on the quality of the data obtained. This is due to the fact that the rotational motion of ions in such a trap, in contrast to ICR, is not isochronous. Ions are rapidly mixed along circular trajectories and occupy a much larger volume than in the ICR cell. The relative proximity of the electrodes in the case of an orbital trap plays its role - ion image charges compensate to a large extent for the interactions of these ions, and such a phenomenon as coalescence of ion packets close to m / z is much less pronounced in this case. The phase correction of the recorded signal, which leads to approximately a doubling of the obtained resolution of the recorded spectra, is more easily achieved in the case of an orbital ion trap, because The initial phase of ion oscillations along the axis in this trap is much more defined and practically independent of the m / z ion than when the ion rotations in the ICR spectrometer are excited by a sufficiently long wave packet.

One of the important prerequisites for the present invention is the creation by us of a technique for resonant excitation of rotation of selected ions around the axis of a radio frequency quadrupole and the fragmentation of these ions due to collisions with buffer gas molecules [9, 20, 21]. This technique was new, previously not proposed by anyone. In contrast to the present invention, the rotation of the ions for their collision-induced dissociation is excited in this case during their movement along the quadrupole without preliminary isolation. This limits the possibilities for kinetic measurements and provides a limited ability to detun from signals of interfering ions. In addition, this method of resonant rotation imposes very severe restrictions on the quality of manufacturing a quadrupole. Small deviations in the diameter of the rods or in the distances between them leads to significant losses in the resolving power of the method, which turned out to be no more than 100 during actual measurements in our case.

время между столкновениями τ, если можно не принимать во внимание скорость движения самого иона, около 0,01 сек. В работе [9] получена формула для оценки ширины на полувысоте для зависимости стационарного радиуса вращения от угловой частоты вращающего поля ω_rot:In the proposed embodiment, during the accumulation of ions rotate in a relatively narrow zone (close to or slightly more than 1 cm), while they perform quasi-chaotic oscillations along this zone with an average transit time of this zone comparable to the rotation period. Thus, the field inhomogeneities are significantly averaged, and their influence on the width of the resonance curves is weakened. In this case, the resolution of the resonant excitation for given ions and a given buffer gas will be mainly determined by the density of this gas in the region of rotation. At a residual pressure of 10 ^-6 Torr for nitrogen (M = 28 Da) at room temperature with a collision cross section with an ion of about 500

the time between collisions τ, if one can not take into account the speed of the ion itself, is about 0.01 sec. In [9], a formula was obtained for estimating the width at half maximum for the dependence of the stationary radius of rotation on the angular frequency of the rotating field ω _rot :

Δω _rot = 1 / τ _V ,

,from here

,

where τ _V is the characteristic relaxation time of the velocity of ions with mass m:

(для принятых условий). Ожидаемая разрешающая способность по частоте стационарного вращения ионов (с частотой около 30 кГц≈188,4 кРадиан) на полувысоте пиков для органических ионов с массой m≈1000 Да (с относительно малой энергией при такой частоте вращения, недостаточной для диссоциации иона), должна быть около 70000. При меньших давлениях остаточных газов может быть обратно пропорциональна большая разрешающая способность, однако время установления стационарных радиусов вращения, пропорциональное τ_V, может стать неприемлемо большим.

(for accepted conditions). The expected resolution with respect to the frequency of stationary rotation of ions (with a frequency of about 30 kHz ≈188.4 kRadian) at half maximum peaks for organic ions with a mass of m≈1000 Da (with a relatively low energy at such a rotational speed insufficient for ion dissociation) should be about 70,000. At lower pressures of the residual gases, a greater resolution can be inversely proportional, however, the time to establish stationary radii of rotation proportional to τ _V can become unacceptably large.

It is interesting to compare such an estimate of the resolution of the process of resonant excitation of ion rotation with the dependence of the resolution of the ICR mass spectra on the recording time, which at cyclotron rotation frequencies ν _c = ω _c / 2π of about 100 kHz cannot be significantly longer than the collision waiting time τ. In the absence of attenuation of the recorded signal, its Fourier transform during time T gives [27]:

.

.

If the signal decays exponentially with a characteristic time τ, then for the Fourier transform on a semi-infinite time interval, there is [27]:

.

.

For relatively large ions, the cyclotron rotation velocity will be sufficiently small, and collisions of ions with molecules or atoms of the residual gas will no longer immediately lead to the elimination of ions from the ensemble of rotating ions. Collisions will manifest themselves in a gradual dephasing of the rotation of the ion packet and a decrease in the radius of rotation. If we assume that in this case the signal decays exponentially with a characteristic velocity relaxation time τ _V , then the frequency resolution for resonant excitation is more than three times the resolution in the frequency spectrum after the Fourier transform for the same ions, for the same residual gas pressures and rotational speeds.

Since it is necessary to introduce a compact ion packet into a trap for mass analysis of ions in an orbital ion trap, it was proposed to accumulate ions in a special type of radio frequency trap and then inject them into the orbital trap with a tangentially short high-voltage pulse [34]. Axial injection of ions in our case does not require and does not involve the use of traps of this type. The accumulation and standard input of ions from a linear radio-frequency trap is quite acceptable.

In addition, the authors of the orbital ion trap proposed creating a special quadratic-logarithmic potential of the form (2) using a cylindrical trap [35]. In this alternative design, the trap has two electrode systems. One electrode system represents a central electrode in the form of a cylinder cut into transverse segments; another electrode system is an external cylindrical electrode, also cut into transverse segments. By applying the appropriate voltage to the segments, the field of the orbital ion trap is simulated. In addition, it becomes possible to additionally manipulate the ions by pulsed switching of stresses on individual segments. In our case, the central electrode is absent, and the outer cylindrical electrode is segmented not across, but along the cylindrical surface, which, as can be assumed, should provide greater accuracy in approximating the quadratic dependence of the field potential averaged over rotations along the cell axis than for transverse segmentation.

Disclosure of invention

Features of one of the possible implementations of the proposed method are:

Ions from an external source enter the cylindrical cell, which we call a dynamic trap. It differs from the ICR mass spectrometer cell described in the previous section with dynamic harmonization of the electric field, in addition to the absence of a magnetic field, by the presence of additional cuts on convex sections of a cylindrical surface. In addition, in this case, the input end electrode of the cell is assumed to be flat with a partitioned inner surface. The opposite end electrode is absent, and the cylindrical surface of the cell is extended further than the place of its supposed localization with continued sectioning of this cylindrical surface to ensure close to the parabolic distribution along the axis of the electric potential averaged over circular paths in this region. The middle ribbon part along each convex symmetric section on a cylindrical surface is used to specify one of the alternating stresses, and the remaining meniscus parts are supported on average at 0 potential. They are cut across into two halves, from which the induced signal from ions oscillating along the axis during registration is recorded. Alternating stresses, when averaged over a circular trajectory around the axis of the trap, will not change the mean field, but they will create an effective potential that repels rotating ions from the surface, similar to that described in Marshall [7] for focusing ions moving along the axis of a cylinder divided into rings when alternating stresses are applied to the cylinder rings. Thus, if the ions entering the trap are affected by a rotating electric field, then these ions, having reached a certain limiting radius, will rotate near a cylindrical surface, without risking to recombine or die in a collision with this surface.

If the potentials applied to the concave sections of the cylindrical surface and the end electrode sections are positive, then with a suitable distribution of such potentials between the end electrode sections in such a trap, a parabolic potential distribution averaged over rotations along the axis of the trap with a minimum in the middle of the trap and over the radius of the trap is created parabolic distribution of the averaged potential, pushing positive ions to the cylindrical surface. With sufficient alternating field strength near the cylindrical surface, a potential well is created for rotating positive ions in which these ions will continue to rotate during registration without exposure to a rotating field, performing isochronous vibrations along the axis of the trap, gradually disappearing due to collisions with atoms or molecules of the residual gas. A decrease in the signal due to the loss of the phase matching of oscillations, due to the imperfection of the averaged field, will also take place, but with sufficiently accurate cuts on the trap surface and acceptable stability of the power sources, this effect will be relatively small for routine accessible pressure levels of the residual gases.

If the potentials applied to the concave sections are negative, then the averaged field along the axis of the trap will inhibit the positive ions entering the trap, and if their input energy is less than the potential difference between the beginning of the trap and its middle, then these ions will stop at some point and begin the reverse movement. If in this case the averaged field over the radius of the trap is close to harmonic, then it is possible to create resonant rotating fields for ions with known m / z. Such fields will spin the corresponding ions, while the rest will remain close to the axis and after a while leave the trap, leaving it in the opposite direction through the inlet of the trap. If the voltages applied to neighboring sections of the inner surface of the input electrode have a sufficiently large contribution of alternating voltages, then resonantly unwound ions will be reflected from this surface and accumulate in the initial part of the trap. If before entering the data recording mode the entry of new ions into the trap is stopped for some time, then all ions having resonant frequencies different from the corresponding frequencies of the rotating field will leave the trap. During this time, the resonantly rotating ions will approach the cylindrical surface of the cell, and the growth of the radius of rotation will stop. To improve the quality of the field averaged over rotations, it makes sense to increase the frequency of rotation of the ions. For this, the frequencies entering the rotational field gradually increase over time while maintaining the phase of each harmonic component of the rotational field.

If the rotating field has only one resonant frequency, then ions with the selected m / z value will be isolated in this way. In this case, after the ions reach the specified maximum speed, when the rotating field can be turned off, product ions can be formed in the collision of ions with atoms or molecules of residual gases, if you wait a while, comparable to the average time for waiting for the collision. For example, for a pressure of 10 ^-9 Torr at a rotation frequency of 300 kHz and a rotation radius of 1 cm for ions with a collision cross section of 200 Å ^2, this time will be approximately 1 second. Alternatively, by briefly turning off the alternating field near the surface of the trap, rotating ions can collide with this surface, which can lead to the formation of corresponding product ions. After this, by activating the registration mode, the mass spectrum of product ions can be recorded. To do this, potentials of the opposite sign are applied to the concave sections and end electrodes (for positive ions - positive) and the alternated voltage components on the sections of the input end electrode are turned off. Such a shutdown makes sense, since the reflection of ions from the barrier of the effective potential will lead to a change in the phase of their oscillations, while the braking in the averaged quadratic potential will preserve these phases. After such a switch, the rotating ions will begin to oscillate along the axis of the trap and the process of recording the signals induced from them will begin. In order to get rid of the remaining ions in the trap after the registration is completed, the alternating field near the surface of the trap is turned off for a sufficiently long time so that all the ions in the trap can recombine.

The frequency spectrum of the recorded data in a wide frequency range is obtained using the fast Fourier transform. For sufficiently narrow ranges of ion oscillation frequencies, either obtained under isolation of the corresponding ion populations before recording, or by digitally filtering broadband data, an exponentially decaying sinusoidal contribution to such narrowband data can be searched. In this case, along with m / z ions, the characteristic decay times of the corresponding signals can be estimated, which for ideal fields will be determined by the average expected number of collisions of the corresponding ion with atoms or molecules of the residual gas per unit time during the registration process.

Additional possibilities for ion separation in a dynamic trap arise if, in the mode of ion arrival, the braking electric field is reduced so that part of the ions passes through the trap and is registered at its exit, for example, using a secondary electron multiplier. The registration of the ion current with a change in the height of the potential barrier in the middle of the dynamic trap allows, after numerical differentiation, to obtain the energy spectrum of the incoming ions. To increase the stability of numerical differentiation, one can use the approximation of the initial data by smoothing splines [17] or the quasispline approximation [18, 19].

The frequency and amplitude of the rotational voltage can also influence the flow of ions passing through a dynamic trap. In this case, the pressure of the residual gases in the trap is better to have increased, otherwise the radius of rotation of the ions will depend only on the discrepancy between their m / z and the resonant value for the frequency of the rotational voltage, the amplitude of this voltage and the time it is exposed to this ion. At a noticeable frequency of collisions of ions with atoms or molecules of a gas, the radius of rotation of the ion at a given amplitude of the rotational voltage at the resonant frequency of rotation will also depend on the mobility of the ions. Two other factors that can affect the passage of ions through the trap are the amplitudes of the alternating stresses on the corresponding sections of the cylindrical surface or on the input electrode of the trap. As they decrease, ions suitable for surfaces can die and not contribute to the recorded ion current. Thus, it is possible to record the ion current values in the form of a four-dimensional or even five-dimensional array depending on the four or five parameters listed above: the delay potential, frequency, amplitude of the rotating field, and the amplitude of one or two alternating voltages. Analysis of such data can be organized in the same way as described in our patent application of the Russian Federation [31] and as briefly repeated in the next section.

Brief Description of the Illustrations

For a more complete understanding of the present invention, the following description is related to the corresponding illustrations, in which:

Fig. 1. The described scheme of a dynamically harmonized cell of an ICR mass spectrometer.

Fig. 2. Schematic representation of an orbital ion trap.

Fig. 3. The proposed dynamic trap scheme.

Fig. 4. Illustration of the capture of ions in a dynamic trap.

Fig. 5. The circuit for supplying voltage to the dynamic trap sections.

All these illustrations are explanatory in nature and do not impose any restrictions on the possible implementation of the invention.

The implementation of the invention

A new approach for transporting ions from the high-pressure region at the exit of the mobility cell of the ion mobility spectrometer to the vacuum part of the mass spectrometer through the formation of a supersonic gas flow is described in our US patent No. 7,482,582 of January 27, 2009 [22]. It was further developed to provide additional analytical capabilities due to the resonant excitation of ion rotation around a supersonic flow in a radio-frequency quadrupole at the input of a time-of-flight mass spectrometer with orthogonal ion input (ortho-VPMS) in our next US patent No. 7.547.878 of June 16, 2009 [23 ]. The specific development of these approaches to ensure effective quantitative determination of the presence of impurities in gas mixtures and structural analysis of multiply charged ions of biomolecules without preliminary separation of ions by mobility is described in our patent of the Russian Federation No. 2402099 dated 10.20.2010 [24] and in our patent of the Russian Federation [25] and patent application of the Russian Federation No. 20111119310 dated 05.16.2011 [26]. An effective method for separating and introducing ions from an external ion source into a supersonic gas stream is described in our patent application of the Russian Federation No. 2011123281 dated 06/09/2011 [31]. The implementation of all these inventions usually involved the use at the final stage of ortho-VPMS as a mass analyzer. However, this analyzer may well be a device based on an orbital ion trap, and it seems that the variant of the orbital ion trap or dynamic trap that is proposed in the present invention can be particularly effective in this role. Having the expected resolution that is several times higher than that which is routinely achieved on the ortho-VPMS classical design, based on the one described in our long-standing invention [28], the proposed dynamic ion trap will be able to realize functions that are not available, such as this option ortho-VPMS, as well as the currently widely used orbital ion trap.

As already indicated, the starting point of the present invention is a dynamically harmonized cell of the ion-cyclotron resonance mass spectrometer, which is schematically shown in Fig. 1 with the sectional shape described by formula (1) for N = 4 (for simplicity of image). A 0-volt voltage is applied to the convex sections (11), and to the concave (12) in the recording mode, the blocking voltage is positive for positive ions (15) entering the cell in the previous cycle. The same voltage is applied to the spherical end electrodes (13) and (14). These voltages create, as the corresponding calculations show and confirm the corresponding experimental data (for a cell variant with 8 sections of each type) [5, 6], an electric field inside the cell, providing close-harmonic vibrations of rotating ions along the cell axis. In this case, an electric field in a plane orthogonal to the cell axis with a quadratic potential distribution along the radius pushes the rotating ions to the inner cylindrical surface, which is prevented by the Lorentz force of the magnetic field. In the absence of a magnetic field, these ions would have to die on this surface. A possible way to prevent such a death is to create an electric field near the surface that is sufficiently strong and sharply changes for ions rotating near the surface. A situation will arise similar to that described in [7], when for ions moving in a pipe drawn from rings, focusing was observed on the pipe axis when constant stresses of different signs were applied to neighboring rings.

If the convex sections of the cylindrical surface shown in Fig. 1, to which the 0 potentials were applied, cut additionally, as shown in Fig. 3, and apply equal in magnitude but opposite in sign voltage to parts (31) and (32) , then the potentials averaged over circular trajectories on the surface do not change. For this, it is necessary, of course, that the sizes of parts (31) and (32) coincide. At the same time, for rotating ions (in the coordinate system associated with the ion) an alternating field will be created near the surface. As is known [8], such a field creates an effective potential that pushes ions out of the region of a strong field. For each harmonic component of such a field with frequency ω, the value of this effective potential in the plane orthogonal to the cell axis (in an approximation that preserves the terms of the second order of smallness) will be:

,

,

- средние квадраты напряженности компонент электрических полей с частотой ω. Основная частота такого поля для секционирования, показанного на рис.2, будет равна удвоенной частоте вращения ионов. При экспериментальной реализации имеет, по-видимому, смысл удвоить число секций каждого типа, как это сделано в работах [5, 6]. Это улучшит качество усреднения полей и в то же время повысит точность приближения, которое приводит к формализму эффективного потенциала.where m is the mass of the ion, q is its charge,

- the average squares of the intensity of the components of the electric fields with a frequency ω. The fundamental frequency of such a sectioning field, shown in Fig. 2, will be equal to twice the ion rotation frequency. In the experimental implementation, it seems to make sense to double the number of sections of each type, as was done in [5, 6]. This will improve the quality of field averaging and at the same time increase the accuracy of the approximation, which leads to the formalism of the effective potential.

At the edges of each convex section in Fig. 3, parts with normally 0 potential are left, as previously this entire section. These parts are cut into two halves (23) and (21), which are respectively electrically connected for different sections, as shown in Fig. 5. The potential difference between these halves, arising from induced charges oscillating along the axis of the ion cell, is fed through an appropriate amplifier to the ADC of the recording system (50), Fig. 5.

To ensure the harmony of the averaged field inside the cell, you need to set the appropriate boundary conditions on the end electrodes of the cell. In Fig. 1, these electrodes had a spherical shape closest to the rotation hyperboloids, which are equipotential surfaces of an ideal hyperbolic field:

where Φ ₀ is the field potential for z = r = 0, and γ is its increment along the cell axis for z = ± 1. The same potential was applied to the end electrodes as to the concave sections (12). In our case, as will be seen from what follows, for the input electrode, a flat shape with an opening for introducing ions is preferable (33). To connect with the registration system for ions passing through the trap (34), it makes sense to leave the output end of the dynamic trap open. The possibility of recording such ions may provide additional opportunities for their separation. To create the necessary averaged boundary conditions, the input electrode (28) must be partitioned. To ensure proximity to the field averaged over rotations (4) at the right end of the region of ion oscillations, the cylindrical surface of the dynamic trap is extended by a certain length compared to the cell in Fig. 1 and is also partitioned. In order to ensure a quadratic dependence of the potential averaged over concentric circles (36) on their radius r, the boundaries of curved sections on the input electrode (28) in this case, for the four petals of the application of two potentials that determine the averaged boundary conditions, will be determined by the equations:

where, as in equation (1), α is the angular coordinate of the cut point, R is the electrode radius (28).

To ensure the averaged boundary conditions for the field (4), the concave sections (30), (37) of the cylindrical surface and the input electrode (28) are supplied with the voltage of the edge of the input electrode and the cylindrical surface:

and the convex sections (38) of the input electrode (28) should be supplied with the expected voltage of the center of the input electrode in the absence of the inlet (35):

U ₀ = γ2a ² + Φ ₀ .

The values of the parameters of the potential γ and Φ _{0 are} determined from equation (6) and from the expression for the average potential on the middle circular line passing through transverse sections (36) on a cylindrical surface, bearing in mind the formula for section sections (1), for the variant with N = 4 shown in Fig. 2:

.

.

From where, using (6):

;

.

;

.

From where the voltage value U ₀ is determined through U _R :

The opposite end of the dynamic trap is assumed to be completely open. To maintain an acceptable distribution of the fields inside the trap, its cylindrical surface (36) is extended beyond the boundary of the end of the cuts, similar for sections (11) shown in Fig. 1, and these sections are extended (25). To preserve the quadratic dependence of the potential averaged over annular paths on a cylindrical surface (36), the potentials of sections (26) are equal to twice the potential value on sections (30) - 2U _R.

When ions are introduced into the cell in order to stop the ions inside the cell, the potential value U _R must be negative. If the potential of sections (26) is maintained equal to 2U _R , and the potentials of sections of the input electrode are equal to U _R and U ₀ in accordance with formula (7), then the distribution of potential averaged over rotations along the axis of the dynamic trap will be close to parabolic (48). To create an effective potential that prevents the death of ions on the inner surface of the input electrode (28), radial cuts (29) are made that divide all the curved sections of the electrode (28) in half. In addition to the potentials U _R and U _0, alternating potentials of a sufficiently large absolute value are applied to each such half. For this effective potential to arise, similarly to our work [9], the rotational motion of incoming ions (33) is excited by applying phase-shifted π / 2 sinusoidal voltages to some sections, for example, (31) and (32) and to two of the same ones on the opposite side of the cylindrical surface. In this case, the distribution of the averaged potential energy of ions rotating at a certain average distance (50) from the axis of the dynamic trap (45), Fig. 4, will look approximately as shown (46).

To spin all the ions (33), the frequency of the rotational voltage can be quite high, going far beyond the resonant frequencies of the incoming ions (33). For a good averaging of electric fields, it should be significantly higher than the frequency of radial vibrations, which will then be used in the registration of ions. After the desired accumulation time of all ions (33), the rotating field is turned off. A positive voltage U _R is applied to the concave sections of the cell surfaces, and a voltage U ₀ (7) is applied to the convex sections of the input electrode. The shape of the potential distribution along the axis of the dynamic trap (45) becomes approximately the same as (47) in Fig. 4. Thus, the registration mode of the induced signals on the parts (21) and (23) of the convex sections of the cylindrical surface is turned on. Rotating ions (45) will perform quasi-harmonic oscillations with frequencies (3), according to which, after the Fourier transform of the induced signal, after the appropriate calibration, m / z of these ions are determined.

Since the averaged distribution of the potential over the radius for axial distributions (48) and (46) is parabolic with a minimum on the axis of the trap (45), the resonant spin-up of selected ions (42) by a rotating field is possible. These ions, having acquired a sufficient spin radius during potentials of reciprocating motion in potentials (48) - (46), when the surface of the input electrode (28) is reached, they will be reflected from it by the effective potential of the alternating voltages of sections (37) and (38). The effective potential created by the alternate potentials of sections (31) and (32) will prevent approaching the cylindrical surface of the trap (41). Thus, these ions (42) can accumulate over time. Ions (44), not subject to resonance spin and possessing energy insufficient to overcome the potential barrier (48), will return back through the inlet of the electrode (28). Ions (43) with energy exceeding the barrier (48) will pass through the trap (41) and can be detected if, for example, a secondary electron multiplier or an additional mass analyzer is installed at the output of the trap (41). Such registration will also be useful for monitoring the accumulation of resonantly spun ions (42), because upon accumulation of a sufficiently large number of them, they, under the action of a space charge, will begin to exit the trap (41) in the forward direction.

At a maximum rotation radius of 0.8 cm at a rotation frequency of 10 kHz, the ion velocity is about 500 m / s. This is less than the rms velocity of the nitrogen molecules at room temperature. This means that the stationary internal temperature of ions due to collisions with buffer gas molecules will be about 200 ° C [10]. Such a temperature for most organic and bioorganic ions will not lead to their effective decay within a reasonable time of measurements. Thus, rotational speeds of about 10 kHz and less are acceptable for the accumulation of selected ions over virtually any time. In this case, a sufficiently high selectivity of accumulation can be achieved only for ions whose kinetic energy is slightly less than that which is sufficient to overcome the potential barrier (48). Such ions, approaching the apex (48) or (46), slow down their movement, and they will have enough time to spin the resonant rotating field to a radius exceeding the radius of the inlet of the end electrode. With a relatively short accumulation time, when collisions of ions with atoms or molecules can be neglected, the rotation frequencies can be greater, and in this case, the stringency of the requirements for the input energy of the accumulated ions decreases. On the other hand, if with an increase in the accumulation time there will be a noticeable number of ion collisions leading to the formation of product ions, when isolating ions with a single value m / z, this can be used as a source of information on the structure of the ion. The product ions themselves, having m / z that does not correspond to the frequency of the rotating field, will gradually reduce the speed and radius of their rotation. In this case, the likelihood of subsequent dissociation processes of these product ions decreases, but over time, the possibility arises of the loss of these product ions due to their reverse exit through the electrode inlet (28).

For sufficient accuracy of averaging the electric field over ion rotations, the rotation frequency should be significantly greater than the frequency of axial ion vibrations. To do this, the frequency of the rotating field (and if there are several, then the frequencies) should increase for some time until the desired value is reached. In this case, two conditions must be met. Firstly, the input ion flow must be suspended, and secondly, to exclude significant losses of accumulated ions, there should be no significant discontinuities in the phase of the changing rotating field. If you wish, after the ions are unwound to the required rotation speed, you can wait some time for the formation of product ions in this case with increased energy of collisions with an atom or molecule of residual gases. It is also possible for a short time to turn off the alternating stresses in sections (32) and (30) for the collision of ions with the surface, as a result of which corresponding product ions can also be formed.

After that, by switching on the locking voltage U _R on the sections (30) of the cylindrical surface (41) and the corresponding stresses on the sections (26) and on the electrode (28), a potential distribution of the form (47) is realized along the trap (41) and oscillations of rotating ions begin (45) ), the induced signals of which from sections (23) and (21) are recorded.

с фазами φ, величины которых показаны в левой части рисунка:Figure 5 shows the scan sections of the cylindrical surface (51) of the trap (41) for the number of sections of each type N = 8. Inside the reamer (51), the solid lines show the electrical connections between the sections. Thin lines with dots inside the sections show the connection of the corresponding voltages to the sections. The dashed lines denote lines for measuring induced signals to the corresponding sections using a recording system (50). These sections are connected to ground through suitable resistances (53). To the left of the sweep (51) through the capacitors (52) is shown the connection of the voltages creating a rotational field. This is one or the sum of sinusoidal voltages with a certain frequency or with the corresponding number of frequencies ω _k with amplitudes

with phases φ, the values of which are shown on the left side of the figure:

.

.

Alternating voltages U _A or - U _A, respectively, are applied to the same sections together with rotational voltages through resistances (54) or through - (55). A voltage U _R is supplied to the concave sections through a resistance (56), and a double voltage 2U _{R is} applied to the triangular sections on the right side of the scan (51) through a resistance (57). The resistances and capacitors shown in this figure are selected so as to provide a sufficiently low level of noise contributions and distortions to the applied voltage and the recorded data.

The frequency spectrum of the recorded data in a wide frequency range is obtained by applying the fast Fourier transform, as is usually done in ICR mass spectrometry or an orbital trap. Data for relatively narrow frequency ranges of ion oscillations can be obtained by isolating the corresponding ion populations before recording. Alternatively, this data can be obtained, for example, by the inverse Fourier transform, cut to the selected interval of the frequency spectrum of the broadband data. After that, the search for exponentially attenuated sinusoidal contributions to such narrow-band data and their decomposition into the corresponding components can be performed. This possibility becomes real, since the main factor in the attenuation of the recorded signal during ion oscillations in the averaged over rotations of an almost parabolic electric field will be collisions with atoms and molecules of residual gases. Moreover, for relatively high frequencies of rotations and oscillations and sufficiently heavy gases, almost every collision will lead to dissociation of the ion and, consequently, to its loss. For example, for a frequency of 200 kHz and a circumference of rotation of 5 cm, the average speed will be 10 ⁶ cm / s, which is about 8 times higher than the average thermal speed of helium atoms at room temperature. This corresponds to a collision energy of about 2.4 eV. Even such energy may already be enough to dissociate a peptide bond in a biomolecule. For a nitrogen molecule, this energy will already be about 17 eV. It is likely that in such a collision almost any ion will be dissociated.

With the exponential attenuation of the sinusoidal contributions to the source data, which can be exhaustively described by a weighted sum of such contributions, along with m / z ions, the characteristic attenuation times of the corresponding signals can be estimated, which for ideal fields will be determined by the average expected number of collisions of the corresponding ion with atoms or molecules residual gas per unit time during the registration process. For data of this kind:

,

,

where n is the number of sinusoidal contributions, ω _i are the frequencies, a _i , b _i are the values that determine the amplitudes and phases of the contributions, k _i are the decay rates of the contributions, an accurate linear forecast of the subsequent value from the previous ones takes place [29, 30]:

,

,

where Δt is the step between the points in the forecast array, a multiple of the time interval of the data measurement, A _l are the forecast coefficients independent of the time t of the forecast point. To find the values of these coefficients, one can require a minimum of the total forecast error for a certain set of N forecast points (N> 2n), for example, for successive measurement points with an interval δt:

Having written the necessary conditions for the extremum, we obtain a system of linear algebraic equations, solving which, we can find the desired values of A _i .

.

.

To determine the optimal number of forecast coefficients 2n, one can calculate the forecast error, for example, for a point that did not participate in the determination of the coefficients A _i :

,

,

for a successively increasing number of points 2n in the forecast array (2, 4, 6, ...) and stop when the minimum error δ _{0 is} reached.

After finding the forecast coefficients, by analogy with the theory of linear differential equations with constant coefficients, a characteristic equation is compiled that is equivalent to performing this forecast for the exponential function e ^λt :

. Таким образом, после нахождения корней уравнения (8) величины w_l будут определять частоты синусоидальных вкладов, а κ_l - скорости затухания этих вкладов. Чтобы найти амплитуды этих вкладов (a_l, b_l), нужно потребовать минимума суммарной квадратичной ошибки аппроксимации исходных данных во всех точках регистрации и решить соответствующую систему линейных алгебраических уравнений, получающуюся приравниванием 0 частных производных этой ошибки по всем искомым коэффициентам:which is an algebraic equation of degree 2n with respect to the variable e ^-λΔt . For real quantities A _{i, the} roots of this equation will form complex conjugate pairs:

. Thus, after finding the roots of equation (8), the quantities w _l will determine the frequencies of the sinusoidal contributions, and κ _l will determine the decay rates of these contributions. To find the amplitudes of these contributions (a _l , b _l ), one needs to require a minimum of the total quadratic error of the approximation of the initial data at all registration points and solve the corresponding system of linear algebraic equations obtained by equating 0 partial derivatives of this error with all the sought coefficients:

where N is the number of all registration points, and the time of the starting point is taken equal to 0.

Additional opportunities for ion separation in a dynamic trap arise if, in the mode of ion arrival, the braking electric field is reduced so that some of the ions pass through the trap and register at its exit, for example, using a secondary electron multiplier or a subsequent mass analyzer. The registration of the ion current with a change in the height of the potential barrier in the middle of the dynamic trap allows, after numerical differentiation, to obtain the energy spectrum of the incoming ions. To increase the stability of numerical differentiation, one can use the approximation of the initial data by smoothing splines [17] or the quasispline approximation [18, 19].

The flux of ions passing through a dynamic trap can also be affected by the frequency and amplitude of the rotational voltage precisely because of the increase in the divergence of the local electric potentials with the ion radius of the ion from the trajectory averaged over the trajectory in a given section of the trap. In this case, the pressure of the residual gases in the trap is better to have increased, otherwise the radius of rotation of the ions will depend only on the discrepancy between their m / z and the resonant value for the frequency of the rotational voltage, the amplitude of this voltage and the time it is exposed to this ion. At a noticeable frequency of collisions of ions with atoms or molecules of a gas, the radius of rotation of the ion at a given amplitude of the rotational voltage at the resonant frequency of rotation will also depend on the mobility of the ions. Two other factors that can affect the passage of ions through the trap are the amplitude of the alternating voltage across the corresponding sections of the cylindrical surface of the trap and / or at the input electrode. With a decrease in each of these voltages, ions approaching the surface can die and not contribute to the recorded ion current. The rate of their death will be determined not only by the average distance of the maximum approximation of the ions to the surface, but also by the diffusion spreading of the ion cloud, which depends on the effective temperature of the ions and their charge. As shown in [31], this can lead to a significant increase in the separation of ions according to their mobility and will allow separation of ions with the same m / z and mobilities, but with different charges, which can be especially important in the analysis of large multiply charged ions of biomolecules.

It is possible to record the values of the ion current, for example, in the form of a four-dimensional array depending on the four parameters listed above: the delay potential, frequency, amplitude of the rotating field and the amplitude of the alternating voltage. It is possible, of course, how to reduce the dimension of the data array by eliminating the variation of any of the listed voltages, and increase this dimension to five dimensions by introducing a step-by-step change in both alternating stresses for sections of the cylindrical surface and the input electrode. It is only necessary to bear in mind that the introduction of an additional dimension, increasing the separation ability of the method, leads to a multiple increase in the measurement time. Analysis of such data can be organized in the same way as described in our patent application of the Russian Federation [31]. The local maxima of the moduli of the gradients of change in the registered total ion currents are found over the entire array of these measurements. The relative values of the components of this gradient at the maximum point of its modulus are accepted as corresponding to a certain type of ion. The position of the maximum modulus or square of the gradient and the relative values of the gradient components in this position characterize the corresponding ions and are used to identify them. To increase the stability of the results of calculating the gradients of the distribution of total ion currents, the construction of smoothing splines [17] or the quasispline approximation of these distributions over all coordinates of the mentioned distributions can be used [18, 19].

, максимально подавляющий в среднем все векторы градиентов

в этом P - мерном массиве с номерами k₁, k₂…k_P при заданном подавлении вектора градиента целевого соединения J_µ, полученного при калибровочном измерении задержки ионов этого соединения в точке максимума квадрата градиента величины пропускаемого ионного тока. Т.е. решается задача на условный экстремум - найти такие F_µ, которые дают минимально возможное значение дляIn this case, a specific version of the selective digital filtering method can be used to effectively search for target ions [32]. After registering the described multidimensional data array and calculating the gradient array with the components with respect to the coordinates numbered by the index μ from 1 to P, including using the spline or quasispline approximation, a linear digital filter F _µ is constructed with a unit sum of squared filter coefficients

maximally suppressing on average all gradient vectors

in this P - dimensional array with numbers k ₁ , k ₂ ... k _P for a given suppression of the gradient vector of the target compound J _µ obtained by the calibration measurement of the ion delay of this compound at the maximum square of the gradient of the transmitted ion current. Those. the conditional extremum problem is solved - to find those F _µ that give the minimum possible value for

where F _trans is the transparency of the filter, showing the degree of attenuation of the resulting signal for the target compound compared to the sum of the intensities of all components of its gradient. This value is key to quantifying the relative share of the determined component in the analyzed mixture.

The formulated problem can be solved by minimizing the corresponding Lagrange function:

The values of the parameters λ and η can then be found using expressions for the corresponding conditions (9). After equating the 0 first derivatives with respect to the values of F _{µ of} function (10), we obtain the following system of linear algebraic equations:

,

,

F_eff=1/λ, θ=-η/λ. Величина F_eff может быть названа эффективностью фильтра, так как она связана со способностью фильтра подавить компоненты

(для F_eff=0 величины F_µ=θJ_µ не зависят от градиентов

). Величина θ может быть найдена из условия

. Величина F_eff также могла бы быть найдена при заданной величине F_trans из последнего условия (9). Однако, проще использовать фильтры с различными значениями F_eff и выбрать тот, который дает приемлемый результат, и использовать полученное из последнего равенства (9) значение F_trans для количественной оценки относительной доли искомого вещества в анализируемой смеси.Where

F _eff = 1 / λ, θ = -η / λ. The value of F _eff can be called the filter efficiency, since it is related to the filter’s ability to suppress the components

(for F _eff = 0, the values of F _µ = θJ _µ are independent of the gradients

) The value of θ can be found from the condition

. The value of F _eff could also be found for a given value of F _trans from the last condition (9). However, it is easier to use filters with different values of F _eff and choose one that gives an acceptable result, and use the F _trans value obtained from the last equality (9) to quantify the relative fraction of the desired substance in the analyzed mixture.

As an optimal filter, one can take the one that provides the maximum signal-to-noise ratio for the maximum value of the square of the gradient after filtering, taking the noise level as the average value of the squares of the gradient after filtering in the P-dimensional array while moving away from the localization of the specified maximum value of the gradient square, greater a given distance, or taking for such averaging the values of the squares of the gradient that are less than the mentioned maximum value by a given number of times, for example, 10. K The detection criteria for the target compound are close to the expected localization of the maximum value of the square of the gradient after filtering and the achieved signal-to-noise ratio for it. Quantitative parameters of these criteria and estimates of the relative content of the target compound in the analyzed mixture are generated based on the described analysis of calibration mixtures of known composition containing the corresponding target compounds. This concerns the average values of localizations and their dispersions for the maximum value of the square of the response of the digital filter to the data recorded and transformed as described above for calibration mixtures. The levels of the ratio of the described signal-to-noise ratio corresponding to certain concentrations of the target compounds in the calibration mixtures are also determined, and the corresponding calibration curves are constructed.

Claims (12)

1. The method of separation of ions of organic and bioorganic compounds from an external source in a mass analyzer containing an orbital ion trap, characterized in that said trap (hereinafter referred to as a dynamic trap) does not have a central electrode and includes a hollow circular cylinder with a partitioned inner surface at least one end electrode with a spherical surface convex inside the trap, or with a flat sectioned surface; the type, arrangement of the sections and the distribution of electric potentials over the sections of the aforementioned cylindrical and flat surfaces, as well as the potential and radius of the aforementioned spherical surface during registration of induced signals from ions oscillating along the axis of the trap, on some sections of the aforementioned cylindrical surface are such as to ensure an almost parabolic distribution of potential electric field along the axis of the trap with a minimum in the middle of the trap when averaging along circular paths around the axis of the trap, that causes quasi-harmonic oscillations along the said axis of rapidly packets of ions; constant or variable potentials with relatively large fluctuations in values are applied to some sections of the mentioned trap surfaces, which ensure the emergence of a sufficient repulsive effective potential that prevents rotating ions from approaching and dying on the said trap surfaces. 2. The method according to claim 1, characterized in that the input of the analyzed ions inside the dynamic trap is made along its axis through a circular hole in the center of one of the mentioned end electrodes; at the time of introducing the mentioned ions by selecting potentials on the sections of the inner surface of the dynamic trap and the end electrode or sections of the end electrode, the distribution of the electric field potential along the axis of the dynamic trap is organized, which ensures that the movement of a certain fraction of the mentioned ions entering the dynamic trap in the initial motion of the mentioned ions ( input) half of the dynamic trap; the remaining ions passing through the dynamic trap are detected, for example, by a secondary electron multiplier or a subsequent mass analyzer; with a controlled change in the maximum potential level on the axis of the dynamic trap (delaying potential), the function of energy analysis of incoming ions is realized; by numerically differentiating the recorded distribution of the ion current by the values of the retarding potential, we obtain the energy spectrum of ions passing through a dynamic trap; To increase the stability of the calculation results, the construction of smoothing splines or a quasispline approximation of this ion current distribution can be used. 3. The method according to claim 2, characterized in that said selection of potentials at the time of introduction of said ions creates a nearly parabolic radial distribution of electric potentials averaged over circular trajectories around the axis of the dynamic trap with a minimum on this axis to enable resonance spin of ions. 4. The method according to claim 2, characterized in that for some time from the start of said input of the said ions, by applying the corresponding phase-shifted alternating voltages to some sections of the inner surface of the dynamic trap, a rotating field is created in the dynamic trap, which displays the said ions in stable orbits of rotation near the inner surface of the dynamic trap; the registration process of induced signals from moving ions on some sections of the inner surface of the dynamic trap is turned on after turning off the potentials that create the rotating field, and setting constant (holding) potentials on some sections of the inner surface of the dynamic trap and on the mentioned end electrodes providing the mentioned quasi-harmonic vibrations along the axis of the dynamic traps of rapidly rotating ion packets. 5. The method according to claim 4, characterized in that the inner surface of the said end electrode, through which the analyzed ions are introduced, is sectioned flat, to the sections of which are applied, along with constant voltages, creating, together with the potentials of the sections of the said cylindrical surface, a parabolic distribution along the radius of the dynamic trap averaged over circular trajectories around the axis of the trap of the electric potentials of claim 3, alternating stresses creating repulsive effective potential for ions rotating along the surface; Before the registration process of claim 4, together with the inclusion of the said holding voltages of the electrode sections, said alternating voltages on the inner surface of the said end electrode can be turned off. 6. The method according to claim 5, characterized in that the said rotating field is created by adding harmonic rotating fields with frequencies close to resonant for the selected ions; such a choice of rotational field frequencies ensures the accumulation of selected ions in a dynamic trap, ions that differ in mass-to-charge ratios (m / z) from the selected ions receive a relatively small spin radius and exit back from the dynamic trap through an outlet in the end electrode; after such accumulation for a specified time, the input ion flow into the dynamic trap is interrupted by setting the corresponding potentials on the ion path into the dynamic trap; to improve the averaging of the fields over the ion rotations, the frequencies of the rotating field gradually increase while maintaining "continuity" changes in this field until the frequencies are several times higher than the expected frequencies of the ion vibrations along the axis of the dynamic trap, after which the recording mode of step 4 is turned on; when setting a single frequency of the rotating field close to the resonance frequency for the selected ion, thus, the isolation of ions of the corresponding m / z value in a dynamic trap is carried out, the isolation of the selected ions is controlled by recording data according to claim 4. 7. The method according to claim 6, characterized in that when the isolation of the selected ions with a single value of m / z is achieved and the frequency of the rotational voltage is established sufficient to produce product ions in a single collision of the ion with an atom or residual gas molecule, the rotational field is turned off, and with the delay necessary for the required number of ion-neutral collisions, the ion registration mode of claim 4 is turned on, and the mass spectrum of the ion products of collision-induced dissociation is recorded. 8. The method according to claim 6, characterized in that when the isolation of the selected ions with a single value of m / z is achieved for a short time substantially shorter than the ion rotation period, alternating stresses on sections of the inner surface of the trap are turned off, creating a repulsive effective potential for rotating ions; with the subsequent disappearance of the mentioned effective potential, these ions will have a predominantly single oblique collision with the surface, as a result of which with some probability product ions will be formed; after restoration of the mentioned effective potential with the minimum possible delay, the ion registration mode of claim 4 is turned on and the mass spectrum of the ion-products of dissociation induced by a collision with the surface is recorded. 9. The method according to claim 6, characterized in that when the ions are unwound with a single-frequency rotating field, the total currents of ions emerging from the dynamic trap are recorded, for example, using a secondary electron multiplier of rapidly rotating ion packets conjugated with a dynamic trap, or ion currents with m / z corresponding to the frequency of the aforementioned rotational field, using the subsequent mass analyzer in the form of a three-, four- or five-dimensional array with step-by-step selective scanning of the frequency, amplitude of the rotational voltage, magnitude of yn янут янут янут янут янут янут янут ом The curved delayed potential in the middle part of the dynamic trap and the amplitudes of alternating stresses create effective potentials that repel rotating ions from the trap local maxima of the moduli of the gradients of variation of the aforementioned ion currents are found over the entire array of these measurements; the relative values of the components of the mentioned gradient at the maximum point of its modulus are taken as corresponding to a certain type of ion; the position of the mentioned maximum and the relative values of the components of the mentioned gradients in this position characterize the corresponding ions and are used to identify them; To increase the stability of the results of calculations of the gradients of the distribution of ion currents, the construction of smoothing splines or quasispline approximation of these distributions over all coordinates of the mentioned distributions can be used. 10. The method according to claim 9, characterized in that to search for the target compound after registering the aforementioned array of ion currents and calculating the gradient vectors for this array, including when using the aforementioned spline or quasispline approximation, a linear digital filter with a unit sum of squares is constructed filter coefficients, which suppresses on average all the gradient vectors in this array of ion currents at a given suppression of the gradient vector for the ion current of the target compound at the maximum square of this gradient the degree of the last suppression is determined by obtaining the maximum signal-to-noise ratio for the maximum value of the square of the gradient after filtering, taking as the noise level the average value of the squares of the gradients after filtering in the said array, not exceeding a fraction (for example, 10%) of the mentioned maximum square of the gradient; the criteria for the detection of the target compound are close to the expected localization of the aforementioned maximum value of the square of the gradient or its local maximum, because the global maximum of said gradient square may correspond to some close isomer of the target compound, and the achieved level of said signal to noise ratio; quantitative parameters of these criteria and estimates of the relative content of the target compound in the analyzed mixture are generated on the basis of the described analysis of mixtures of known composition containing the target compounds and their isomers. 11. The method according to claim 4, characterized in that after the registration of the induced signals from the ions is completed to obtain an ion oscillation frequency spectrum in a relatively wide frequency range, a fast Fourier transform is performed and the frequency scale is converted to the m / z ion scale using the standard procedure, adopted in spectrometers with an orbital ion trap; for selected sufficiently narrow m / z intervals or corresponding frequencies, the recorded data is converted to suppress frequencies beyond the selected frequency range, for example, by performing a fast inverse Fourier transform of the data after the direct Fourier transform with 0 equal to their values outside the selected frequency range ; decomposition of the data thus transformed into exponentially damped sinusoidal contributions is performed; the found frequencies of the aforementioned sinusoidal contributions and the characteristic times of their attenuation are converted by known methods into m / z ions and their cross sections for collisions with atoms and molecules of the residual gas, if the pressure and composition of the residual gases are known. 12. The method according to claim 11, characterized in that for the decomposition of said transformed data into said exponentially decaying sinusoidal contributions, the order and coefficients of the optimal linear prediction are found, which predict, with minimal and stable error, the last in the series of equally spaced values of the aforementioned converted data from the previous values; a characteristic equation is compiled with the found forecast coefficients and its complex roots are found; through the real and imaginary parts of these roots, the frequencies of the mentioned sinusoidal contributions are calculated and the characteristic times of the exponential decay of these contributions are found; the amplitudes of these contributions are found from the best in the rms approximation of the aforementioned converted data.

Images