Gridless ion mirrors with smooth fields

Abstract

An ion mirror 41 constructed of thin electrodes that are interconnected by resistive dividers 45 with potentials U1-U5 applied to knot electrodes to form segments 41-43 of linear potential distribution between the “knot” electrodes, yet without separating those field regions by meshes. Weak and controlled penetration of electric fields provide for a fine control over the field non linearity and over the equipotential line curvature, thus allowing to reach unprecedented level of ion optical quality: more than twice larger energy acceptance compared to thick electrode mirrors, up to sixth order time per energy focusing, ion spatial focusing and wide spatial acceptance. Novel mirrors can be formed very slim to arrange them into stacks for ion transverse displacement between ion reflections or for multiplexed mirror stacks. Printed circuit boards (PCB) are best suited for making novel ion mirrors, while novel ion mirrors are designed to suit PCB requirements.

Claims

1. An ion mirror for reflecting ions along an axis (X) comprising: a first axial segment (E2), within which the turning points of the ions are located in use, and a second axial segment (E3), wherein the first and second axial segments are adjacent each other in a direction along said axis (X); wherein at least the first axial segment comprises a plurality of electrodes that are spaced apart from each other along said axis (X), wherein the electrodes in at least the first axial segment have substantially the same lengths along said axis and adjacent pairs of these electrodes are spaced apart by substantially the same spacing such that these electrodes are arranged so as to have a pitch P along said axis; wherein said plurality of electrodes define windows arranged in a plane (Y-Z plane) orthogonal to said axis (X) through which the ions travel in use, wherein the windows have a minimum dimension H in said plane (Y-Z plane); wherein P≤H/5; and wherein the mirror has voltage supplies and is configured to apply electric potentials to the electrodes of the first axial segment for generating a first linear electric field of a first strength E2 within the first axial segment, wherein 4.3U.sub.0/D<E2<5U.sub.0/D, where U.sub.0 is equal to a mean energy K.sub.0 of an ion to be reflected in the mirror divided by the charge q of that ion, and D is the distance from the mean ion turning point to a first order energy focusing time focal point of the mirror.

2. The ion mirror of claim 1, comprising voltage supplies for applying different voltages to different electrodes of the ion mirror for generating electric fields for performing said reflecting of the ions; wherein at least the first axial segment is defined between inter-segment electrodes that are spaced apart along said axis, each of said inter-segment electrodes being an electrode to which one of said voltage supplies is connected to, wherein said plurality of electrodes in the first axial segment are arranged between the inter-segment electrodes, and are electrically connected thereto and interconnected with each other by electronic circuitry such that when the voltage supplies apply voltages to the inter-segment electrodes, this causes the plurality of electrodes to be maintained at different potentials so as to generate said electric fields.

3. The ion mirror of claim 2, wherein the plurality of electrodes in the first axial segment are interconnected to each other by a chain of resistors; wherein the chain of resistors is configured to form a substantially linear potential gradient at and along the plurality of electrodes within the segment.

4. The ion mirror of claim 2, wherein the mirror is configured such that the distance (X3) along said axis from the mean ion turning point in the first axial segment to the inter-segment electrode nearer to the mirror entrance/exit is ≤2H; ≤1.5 H; ≤IH; ≤0.5 H; in the range 0.2H≤X3≤1.7H; or in the range 0.IH≤X3≤IH.

5. The ion mirror of claim 4, comprising voltage supplies and configured to apply electric potentials to the electrodes of the first axial segment for generating a first linear electric field of a first strength E2 within the first axial segment, and to apply electric potentials to electrodes of the second axial segment for generating a second linear electric field of a second strength E3 within the second axial segment; wherein the ratio of field strengths E3/E2 is related to the distance X3 by the relationship E3/E2=A*[0.75+0.05*exp((4X3/H)−1)], where 0.5≤A≤2.

6. The mirror of claim 5, wherein the ratio E3/E2 is one of the group: (i) 0.8≤E3/E2≤2 at 0.2≤X3/H≤I; (ii) 1.5≤E3/E2≤10 at I≤X3/H≤1.5; and (iii) E3/E2≥10 at 1.5≤X3/H≤2.

7. The ion mirror of claim 2, comprising a third axial segment arranged further from an entrance end of the ion mirror than the first axial segment; and comprising voltage supplies configured to apply electric potentials to electrodes of the third axial segment for generating a third linear electric field of a third strength E1 within the third axial segment; wherein E1<E2; and wherein the mirror is configured such that the distance (X2) along said axis from the mean ion turning point within the first axial segment to the inter-segment electrode further from the mirror entrance is 0.2:S X2/H:S 1.

8. The ion mirror of claim 1, comprising voltage supplies configured to apply electric potentials to electrodes of the second axial segment for generating a second linear electric field (E3) of a second strength within the second axial segment; wherein the electrodes are configured such that the second linear electric field (E3) penetrates into the first axial segment so that the axial electric field in an axial portion of the first axial segment is non-linear where the turning points of the ions are located.

9. The ion mirror of claim 8, wherein an axial electric field strength Eo at a mean ion turning point within the first axial segment is related to the strength of the first linear electric field E2 by a relationship from the group comprising: (i) 0.01:S (Eo−E2)/E2:S 0.1; and (ii) 0.015≤(Eo−E2)/E2≤0.03.

10. The ion mirror of claim 8, wherein the electrodes are configured such that the second linear electric field (E3) penetrates into the first axial segment so that the equipotential field lines in the first axial segment are curved where the turning points of the ions are located; and/or wherein the different field strengths in said first and second axial segments produce curved equipotential field lines in a transition region between the first and second axial segments.

11. The ion mirror of claim 1, comprising a third axial segment (E1) adjacent to the first axial segment (E2) in a direction along said axis (X); wherein the third axial segments comprises a plurality of electrodes that are spaced apart from each other along said axis (X).

12. The ion mirror of claim 11, comprising voltage supplies and configured to apply electric potentials to the electrodes of the third axial segment for generating a third linear electric field (E1) of a third strength within the third axial segment; wherein the electrodes are configured such that the third linear electric field (E1) penetrates into the first axial segment so that the axial electric field in an axial portion of the first axial segment is non-linear where the turning points of the ions are located.

13. The ion mirror of claim 1, wherein the length of the first axial segment along said axis is ≤5H; ≤4H; ≤3H; or ≤2H.

14. The ion mirror of claim 1, comprising voltage supplies and configured to apply electric potentials to the electrodes of the first axial segment for generating a first linear electric field (E2) of a first strength within the first axial segment, and to apply electric potentials to electrodes of the second axial segment for generating a second linear electric field (E3) of a second, different strength within the second axial segment; so as to form a non-uniform axial electric field at the boundary between the first and second axial segments.

15. The ion mirror of claim 1, wherein at least some of the electrodes of the ion mirror are conductive strips of a printed circuit board (PCB).

16. A mass spectrometer comprising: at least one ion mirror as claimed in claim 1; an ion source for providing ions into the ion mirror; and an ion detector.

17. A method of mass spectrometry comprising: providing an ion mirror or spectrometer as claimed in claim 1; supplying ions into said ion mirror; reflecting ions at ion turning points within said first axial segment (E2); and detecting the ions.

18. An ion mirror for reflecting ions along an axis (X) comprising: a first axial segment, within which the turning points of the ions are located in use, and a second axial segment, wherein the first and second axial segments are adjacent each other in a direction along said axis (X); and voltage supplies configured to apply electric potentials to electrodes of the first axial segment for generating a first linear electric field of a first strength within the first axial segment, and to apply electric potentials to electrodes of the second axial segment for generating a second linear electric field of a second strength within the second axial segment; wherein the voltage supplies and electrodes are configured such that the second linear electric field penetrates into the first axial segment so that the axial electric field in an axial portion of the first axial segment is non-linear where the turning points of the ions are located, and such that an axial electric field strength Eo at a mean ion turning point within the first axial segment is related to the strength E2 of the first linear electric field by the relationship 0.01≤(Eo−E2)/E2≤0.1.

19. An ion mirror for reflecting ions along an axis (X) comprising: a first axial segment (E2), within which the turning points of the ions are located in use, and a second axial segment (E3), wherein the first and second axial segments are adjacent each other in a direction along said axis (X); wherein at least the first axial segment comprises a plurality of electrodes that are spaced apart from each other along said axis (X), wherein the electrodes in at least the first axial segment have substantially the same lengths along said axis and adjacent pairs of these electrodes are spaced apart by substantially the same spacing such that these electrodes are arranged so as to have a pitch P along said axis; wherein said plurality of electrodes define windows arranged in a plane (Y-Z plane) orthogonal to said axis (X) through which the ions travel in use, wherein the windows have a minimum dimension H in said plane (Y-Z plane); and wherein P≤H/5; the ion mirror further comprising: voltage supplies for applying different voltages to different electrodes of the ion mirror for generating electric fields for performing said reflecting of the ions; wherein at least the first axial segment is defined between inter-segment electrodes that are spaced apart along said axis, each of said inter-segment electrodes being an electrode to which one of said voltage supplies is connected to, wherein said plurality of electrodes in the first axial segment are arranged between the inter-segment electrodes, and are electrically connected thereto and interconnected with each other by electronic circuitry such that when the voltage supplies apply voltages to the inter-segment electrodes, this causes the plurality of electrodes to be maintained at different potentials so as to generate said electric fields; and a third axial segment arranged further from an entrance end of the ion mirror than the first axial segment; and comprising voltage supplies configured to apply electric potentials to the electrodes of the first axial segment for generating a first linear electric field of a first strength E2 within the first axial segment, and to apply electric potentials to electrodes of the third axial segment for generating a third linear electric field of a third strength E1 within the third axial segment; wherein E1<E2; and wherein the mirror is configured such that the distance (X2) along said axis from the mean ion turning point within the first axial segment to the inter-segment electrode further from the mirror entrance is 0.2≤X2/H≤1.

20. An ion mirror for reflecting ions along an axis (X) comprising: a first axial segment (E2), within which the turning points of the ions are located in use, and a second axial segment (E3), wherein the first and second axial segments are adjacent each other in a direction along said axis (X); wherein at least the first axial segment comprises a plurality of electrodes that are spaced apart from each other along said axis (X), wherein the electrodes in at least the first axial segment have substantially the same lengths along said axis and adjacent pairs of these electrodes are spaced apart by substantially the same spacing such that these electrodes are arranged so as to have a pitch P along said axis; wherein said plurality of electrodes define windows arranged in a plane (Y-Z plane) orthogonal to said axis (X) through which the ions travel in use, wherein the windows have a minimum dimension H in said plane (Y-Z plane); wherein P≤H/5; and wherein the mirror is configured such that the distance (X3) along said axis from the mean ion turning point in the first axial segment to the inter-segment electrode nearer to the mirror entrance/exit is ≤2H; ≤1.5 H; ≤1H; ≤0.5 H; in the range 0.2H≤X3≤1.7H; or in the range 0.1H≤X3≤1H; and the ion mirror further comprising: voltage supplies for applying different voltages to different electrodes of the ion mirror for generating electric fields for performing said reflecting of the ions; wherein at least the first axial segment is defined between inter-segment electrodes that are spaced apart along said axis, each of said inter-segment electrodes being an electrode to which one of said voltage supplies is connected to, wherein said plurality of electrodes in the first axial segment are arranged between the inter-segment electrodes, and are electrically connected thereto and interconnected with each other by electronic circuitry such that when the voltage supplies apply voltages to the inter-segment electrodes, this causes the plurality of electrodes to be maintained at different potentials so as to generate said electric fields; and voltage supplies configured to apply electric potentials to the electrodes of the first axial segment for generating a first linear electric field of a first strength E2 within the first axial segment, and to apply electric potentials to electrodes of the second axial segment for generating a second linear electric field of a second strength E3 within the second axial segment; wherein the ratio of field strengths E3/E2 is related to the distance X3 by the relationship E3/E2=A*[0.75+0.05*exp((4X3/H)−1)], where 0.5≤A≤2.

Description

BRIEF DESCRIPTION OF THE DRAWINGS

(1) Various embodiments will now be described, by way of example only, and with reference to the accompanying drawings in which:

(2) FIG. 1 shows a prior art grid-covered ion mirror of SU198034 for singly reflecting time-of-flight (TOF) mass analyzer;

(3) FIG. 2 shows a prior art gridless ion mirror of GB2403063 for multi-reflecting TOF (MRTOF) mass analyzer;

(4) FIG. 3 shows a prior art gridless ion mirror of U.S. Pat. No. 6,384,410 for a singly reflecting TOF and shows ion optical properties of the ion mirror;

(5) FIG. 4 illustrates the method and the design of improved ion mirrors of embodiments of the present invention, based on merging of open and gridless segments with linear potential distribution for providing a slight and controlled non-linearity and equipotential curvature in the ion reflecting region by mutual field penetration between the segments;

(6) FIG. 5 presents axial and on-wall potential distributions for two ion mirrors, where the novel ion mirrors composed of uniform field segments is compared to conventional gridless ion mirror composed of thick electrodes;

(7) FIG. 6 compares axial distributions of field strength and higher field derivatives between mirrors of FIG. 5 to demonstrate smoother fields and smaller field variations in novel ion mirrors;

(8) FIG. 7A compares time per energy curves between novel ion mirrors, composed of uniform field segments, and conventional gridless mirrors, composed of thick electrodes and shows substantial improvement of energy acceptance in novel ion mirror;

(9) FIG. 7B plots energy acceptance of novel ion mirror as a function of normalized field strength at ion mean turning point and illustrates the need for accurate choice of field parameters for reaching high energy acceptances;

(10) FIG. 8 shows on-wall potential distributions for three novel ion mirrors, different by their lens part;

(11) FIG. 9A annotates physical parameters and field parameters for jet wider (relative to FIG. 8) variety of optimized novel ion mirrors, different by lens part; it also presents the range of “sweet spot” parameters for those optimized novel gridless ion mirrors;

(12) FIG. 9B for the same set of simulated ion mirrors as in FIG. 9A, shows the optimal range of electric field non-linearity at ion turning point and presents the link between strength and depth of penetrating fields;

(13) FIG. 10 compares energy acceptance for multiple novel mirrors and best examples of thick-electrode ion mirrors of prior art; Energy acceptance is notably higher for novel ion mirrors, in both cases—with and without isochronicity correction by non-zero lower-order time per energy aberrations;

(14) FIG. 11 shows potential distribution for one particular embodiment of a novel ion mirror composed of two field segments and presents the time per energy curve, demonstrating a compromised energy acceptance relative to above presented novel ion mirrors composed of three field segments;

(15) FIG. 12 shows that the number of power supplies can be reduced by using a precise auxiliary resistor, while the resistor precision shall be in the order of 0.1% to sustain improved energy acceptance of novel ion mirrors;

(16) FIG. 13 illustrates the generic method of forming segmented fields in novel ion mirrors by using thin electrodes, interconnected by a resistive chain, and applying potentials to “knot” electrodes separating field segments;

(17) FIG. 14 shows embodiments of novel ion mirrors, constructed of thin electrodes, and presents methods for sustaining alignment and parallelism of those thin electrodes;

(18) FIG. 15 shows embodiments of novel ion mirrors, constructed of printed circuit boards, and illustrates methods of generating antistatic features on isolating substrates; and

(19) FIG. 16 shows an embodiment of the present invention with two opposed side stacks of thin gridless ion mirrors for bypassing an ion source by long ion packets.

DETAILED DESCRIPTION

(20) Prior Art Ion Mirrors: Referring to FIG. 1, prior art grid-covered ion mirror 10 of SU198034 comprises: two mirror segments 11 and 12 (also referred as stages), formed by equal size ring electrodes; an upper cap electrode 11C; “knot” electrodes 13 with fine meshes to separate regions 11 and 12 of different uniform fields E1 and E2; power supplies 15—U1, U2 and U.sub.D, connected to electrodes 13 and 11C; and a resistive chain 14 for linear potential distribution in electrode segments 11 and 12.

(21) Mirror 10 forms uniform electric fields E1 and E2 in the core volume of segments 11 and 12 without distorting field-free (E=0) conditions in the drift space D. Plot 16 shows potential distributions: 18—at electrodes and 19—at the mirror axis. Small steps of voltage between individual electrodes appear well smoothed at sufficient distance from electrodes, usually considered equal to a spatial period of the electrode structure. To provide for second order time per energy focusing, there exists an optimal ratio of field strength E1 and E2, which depends on the segments length. In case of ultimately short stage 12, U2 is ⅔ of the ion mean specific energy per charge. As known in TOF MS field, with elongation of the stage 12, the ratio of the field strengths E2/E1 varies form E2/E1>>1 to about E2/E1=1, while reducing the ion on mesh scattering at the price of gradual reduction of the energy acceptance. The grid-covered mirror 10 has an exceptional spatial acceptance, i.e. may operate with very wide ion packets. However, if used for multi-reflecting TOF, ion passages through mesh cause devastating ion losses.

(22) Referring to FIG. 2, prior art gridless (grid free) ion mirror 20 of GB2403063 is designed for multi-reflecting TOF (MRTOF) MS. Mirror 20 comprises: a set of thick rectangular frame electrodes 23 and 23L with the window height H (in the Y-direction, corresponding to narrower dimension of electrode window) being comparable to electrodes thicknesses from L1 to L.sub.D; and a set of power supplies 25, connected to individual electrodes, denoted as U1 to U4 and U.sub.D, where U.sub.D also defines the potential of the drift space D. Plot 26 shows potential distributions: 28—near electrodes and 29—at the mirror axis. In spite of large electrode thicknesses, comparable to H, ion optical optimization of electric fields allows reaching high order isochronicity—up to fifth-order energy isochronicity (compare to second order in mirror 10) and full third-order isochronicity, including spatial, angular, and energy, both—pure and mixed term aberrations, as described in WO2013063587 and WO2014142897. Those enhanced gridless ion mirrors provide for excellent isochronicity at reasonable spatial, angular and energy acceptances simultaneously with spatial and angular ion focusing. High-order isochronicity has been obtained with one key feature of gridless ion mirrors—an attractive (accelerating) ion lens 23L is arranged by setting U4 at more attractive potential compared to the drift potential U.sub.D. Disadvantages of ion mirrors 20 are: high making cost; tight requirements on electrode straightness; wide fringing fields; and moderate energy acceptance.

(23) Referring to FIG. 3, prior art gridless (i.e. grid free) ion mirror 30 of U.S. Pat. No. 6,384,410 is a copy of the gridded mirror 10 of FIG. 1 with one difference—removing grids. Mirror 30 comprises: two mirror segments 31 and 32 (also referred as stages), formed by thin ring electrodes and an upper cap electrode 31C; boundary “knot” electrodes 33 (not having meshes!), separating regions of voltage gradient on electrodes between segments 31, 32, and field-free drift stage D; a resistive chain 34 for creating linear potential distribution at electrodes of segments 31 and 32; and power supplies 35—U1, U2 and U.sub.D, connected to the electrodes 33 and 31C. The mirror forms uniform electric fields E1 and E2 in the inner volume of segments 11 and 12, however, distinctly from FIG. 1, also having transition fields T1 and T2 at segment boundaries. Plot 36 shows potential distributions: 38—at electrodes and 39—at the mirror axis. Small steps of voltages between individual electrodes appear well smoothed at the mirror axis. U.S. Pat. No. 6,384,410 proposes optimal ratio E2/E1≅2, and a highly uniform field at ion turning point, placed deep inside the segment 31.

(24) U.S. Pat. No. 6,384,410 provides a numerical example for dimensions and voltages. We analyzed the ion optical properties of the exemplary mirror 37, as illustrated by shape of electrodes and equi-potential lines. The mirror provides for second-order time per energy focusing and allows 7% energy acceptance at 1E-5 level of time isochronicity. The design compensates for spatial focusing/defocusing of transition fields T1 and T2 (as stated in U.S. Pat. No. 6,384,410), thus, returning a non-diverging ion beam, which may be expressed as Y|Y=1. However, it does not focus initially diverging ion packets and generates a substantial second order time per space aberration T|YY, limiting packets width under 5 mm and angular divergence under 5 mrad for dT/T≤1E-5. Both shortages—absence of angular focusing and very small spatial acceptance do compromise use of mirror 30 for multi-reflecting TOF and E-traps.

(25) The reflecting field E1 in the segment 31 may be highly uniform at the ion turning region, far-spaced from the “knot” electrode 33 by distance X.sub.T, which is specifically stressed in U.S. Pat. No. 6,384,410. Simulations of the numerical example 37 have confirmed that the field E1 penetrates at 1E-6 level only at the ion turning point X.sub.T=2.5D, where D is the electrode window diameter D=25 mm. The uniform field in the vicinity of the ion turning point strongly compromises the energy acceptance of ion mirrors. Besides, by nature of electric fields, highly uniform reflecting fields have no curvature of reflecting equipotential lines, thus, not providing for any means to improve the spatial isochronicity. As known in the field of ion optics, lenses always produce positive T|YY aberrations. Mirror 30 forms lens with T1 and T2 fields but has no means for compensating their time per space aberrations. If adding spatial focusing features to mirror 31 (say by making entrance lens T2 stronger), those time aberrations would increase further. Thus, the ion mirror 30 has low ion optical quality, not suitable for multi-reflecting TOF mass spectrometers and electrostatic traps.

(26) Embodiments of the present invention improve the ion optical quality, the design and manufacturing technology of gridless ion mirrors, e.g. for MRTOF and E-Traps.

(27) Principles of novel ion mirrors: Improved ion mirrors for Multi-reflecting TOF (MRTOF) and E-traps mass spectrometers according to embodiments of the invention shall be free of grids, shall provide spatial ion focusing, and shall be highly isochronous at wide energy and spatial acceptances.

(28) Here we state that the ideal reflecting field near the ion turning point should have an optimal non-linearity of the field profile E(x) and a curvature of equi-potential lines, caused by the E(x) non-linearity to provide for two features of high quality ion mirrors: (A) compensation or minimizing of high-order time per energy aberrations; and (B) compensation of time per spatial spread aberrations. The weakly inhomogeneous field strength distribution in the area of the ion turning point leads to much better independence of the flight time with respect to energy, than both purely homogeneous and highly inhomogeneous fields of gridless ion mirrors.

(29) The inventor has found that the quality of ion mirrors can be improved compared to the prior art by merging open regions of uniform fields, where mutual field penetration between segments allows the production of a monotonous and nearly uniform reflecting field at the ion turning point, with a controlled optimal non-linearity (of few percent) in order to provide for high order energy focusing and wider energy acceptance, also accompanied by providing spatial isochronicity. For yet better ion optical quality, the length of the ion reflecting segment shall be limited to allow for a sufficient field penetration from both ends, this way maximizing the energy acceptance.

(30) Referring to FIG. 4, one embodiment of an ion mirror 40 of the present invention comprises two parallel and identical rows 46 spaced by distance H. Each row 46 comprises a plurality of thin (<<H) conductive electrodes that are spaced apart along the X-axis. Schematic 40 shows an enlarged view of the portion of row 46 that is circled in schematic 40. Individual potentials U1, U2, U3 etc. are applied to different ones of the spaced apart electrodes. These electrodes are referred to herein as “knot” electrodes 44 (or inter-segment electrodes), and they define the axial boundaries of the axial segments 41, 42, 43 etc of the ion mirror. Each axial segment comprises a plurality of the electrodes arranged between the “knot” (or inter-segment) electrodes. These plurality of electrodes are interconnected with each other by resistive chains 45, and the electrodes at the axial ends are connected to the adjacent “knot” electrodes by the resistive chain. As such, when potentials U1, U2, U3 etc. are applied to the “knot” electrodes 44, this causes potentials to be applied to the plurality of electrodes therebetween. The structure thus forms a set of openly merged axial segments 41, 42, 43 etc with individual linear field strengths E1, E2 E3 etc along the electrode row 46. The electrodes may have substantially the same lengths along the X-axis and every adjacent pair of these electrodes may be spaced apart by substantially the same spacing such that these electrodes are spatially arranged at a certain pitch P along the X-axis.

(31) The axial segments described herein may be denoted by their fields Ei.

(32) The structure of openly merged segments 41,42,43 etc. forms potential distribution U(x) 47 at the mirror symmetry axis (at Y=0, i.e. away from the electrodes) with nearly uniform fields in the axially central part of individual segments, and with transition fields at segment boundaries. The potential distribution 47 is characterized by an accelerating lens around the segment E5 for spatial ion focusing in the Y-direction, so as by a reflecting field in segments E1 to E4 to provide for isochronous ion reflection in the X-direction.

(33) Alternative electrode structures may be used to generate the same structure of electrostatic field. Those structures may comprise a set of thin electrodes with rectangular or circular windows, a pair of parallel printed circuit boards (planar ceramic, epoxy or Teflon PCB, or a flexible kapton PCB, rolled into a cylinder) with conductive stripes and with high-Ohmic antistatic coating, a pair of resistive plates (or a cylinder) with conductive stripes for knot electrodes, or an insulating (planar or cylindrical) support with resistive coating, separated into segments by conductive stripes. While understanding that multiple known technologies may be used to form the desired fine electrode structure, embodiments of the invention are primarily concerned with the properties of the desired electrostatic field itself to form the optimal non linearity 48 and the optimal curvature 49 of the electrostatic field near the ion turning region.

(34) The ion mean turning point is defined by the potential U=U.sub.0=K.sub.0/q at the mirror axis, corresponding to the full stop of ions with mean kinetic energy K.sub.0 and charge q. In the embodiment 40, let us distinguish one core segment (a first axial segment 42) with the field E2, wherein ions of mean energy are turned: U2>U.sub.0>U3. An important feature of embodiments of the present invention is the controlled penetration of surrounding uniform fields E1 and E3 (from second and third axial segments 41,43) into the E2 segment (42) and particularly to the location of the ion turning point (at X=0). As we found at ion optical modeling, the ion optical quality of the ion mirrors may be improved due to the penetration of the E3 field (from the second axial segment 43) into the E2 segment (42) to the location of the ion turning point. This provides for both: (a) slight and controlled non-linearity of E(x) curve as shown in icon 48; and (b) spatial curvature of equipotential lines in the region, surrounding the ion turning point at optimal X=0, as shown in the icon 49. Both non-linearity 48 and curvature 49 are mutually related by the nature of electrostatic fields. The optimal penetration of the E3 field corresponds to approximately 1-3% of E(x) variation=(E.sub.0−E2)/E2. In other words, the penetration of fields into the E2 segment (42) to the location of the ion turning point (X=0) may cause the field at that point E.sub.0 to differ from E2 by approximately 1-3% of E2. Allowing penetration of yet another field E1 (from the third axial segment 41) into the E2 segment (42) to the location of the ion turning region allows further improvement of the ion optical quality and provides for higher flexibility of controlling the field non-linearity in the E2 segment. Accordingly, the E1 and/or E3 field may be caused to penetrate to the ion turning region.

(35) Comparing novel and prior art gridless mirrors: Referring to FIG. 5, field distributions are compared between the prior art mirror 20 of FIG. 2 and the exemplary novel ion mirror 40 of FIG. 4. Individual thick electrodes of the mirror 20 define a stepped potential (U-steps) distribution 52 on electrode walls, smoothed at the axis to the axial distribution 54 by nature of electrostatic fields. Segmented linear potential distribution 53, forming steps of the field strength E (E-steps) on electrodes of the mirror 40 according to the embodiments, provides a closer initial approximation to the axial distribution 55, thus, forming smoother axial distribution 55.

(36) The difference between axial distributions 54 and 55 is barely visible on a crude scale. However, let us highlight one difference: the axial potential distribution 55 of ion mirrors 40 according to embodiments is much more linear near the ion turning point at U/U.sub.0=1, i.e. field strength variations E/E.sub.0 at the ion turning point are much smaller and more monotonic as compared to prior art mirrors 20 constructed of thick electrodes.

(37) Referring to FIG. 6, the above described difference between the axial U(x) distributions 54 and 55 of mirrors 20 and 40 becomes more apparent when looking at higher order derivatives of the potential distribution U(x)−field strength E(X)=dU/dX, first dE/dX, second d.sup.2E/dX.sup.2 and third d.sup.3E/dX.sup.3 field derivatives, normalized to specific (to charge) mean ion energy U.sub.0 and to the distance D from the ion turning point to the time-focal point (D is shown in FIG. 4). Dashed lines correspond to prior art mirror 20 composed of thick electrodes, generating steps of wall potential, denoted as U-step in the drawing. Solid lines correspond to segmented linear potential distributions obtained in the mirror 40 according to embodiments of the invention, and denoted as E-steps in the drawing. As apparent from the graphs, stepped E of novel mirror 40 provides smaller variations of field strength E/E.sub.0 around the ion turning point at X=0, so as to achieve monotonous and much smoother distributions of higher field derivatives compared to the prior art stepped U mirror 20.

(38) Referring to FIG. 7A, plot 71 compares flight time per ion energy curves (T−T.sub.0)/T.sub.0 Vs (K−K.sub.0)/K.sub.0 for prior art mirror 20 with a stepped wall potential (Step U) and for the ion mirror 40 of embodiments with a stepped field strength (step E) on electrodes, as denoted on the legend 71. The curve 72 for stepped U corresponds to the third-order time per energy focusing with ΔK=6% energy acceptance at 1E-5 level of isochronicity. The curve 74 for stepped E corresponds to the forth-order time per energy focusing with ΔK=14% energy acceptance at 1E-5 level of isochronicity. Fine tuning of mirrors potentials allows leaving minor residual coefficients for lower order time per energy aberration (shown in the figure legend 71) in order to reach a wider energy acceptance, as shown by curves 73 for stepped U and 75 for stepped E. Then the mirror 20 with stepped U provides for ΔK=9% energy acceptance, while the mirror 40 provides for a larger ΔK=22.5% energy acceptance. Thus, the ion mirror of the embodiments with stepped field strength E provides for substantially (2.5 times) wider energy acceptance. Experts in TOF MS are aware that the energy acceptance of a TOF analyzer limits the maximal usable field strength in accelerators, in turn limiting the minimal achieved turn around time, currently being the major limit for resolution in TOF MS. Thus, wider energy acceptance nearly directly translates into resolutions per flight path in MRTOF, where novel ion mirrors are expected to gain 2.5 fold higher resolution per flight path.

(39) Referring to FIG. 7B, plot 75 presents energy acceptance ΔK/K at 1E-5 level of isochronicity as a function of normalized field strength E.sub.0D/U.sub.0 at ion mean turning point with X=0, E=E.sub.0, and U=U.sub.0, using annotations of FIG. 4. Optimum is observed within about +/−1% of E.sub.0D/U.sub.0 variation. Thus, ion mirrors 40, built of segmented fields, notably improve energy acceptance ΔK/K, however, their field structure and parameters shall be accurately set and controlled. Embodiments of the invention provide a combination of segmented fields with optimal “sweet spot” mirror parameters.

(40) Referring back to FIG. 6, let us relate the obtained improvement of energy acceptance in FIG. 7 to the field structure of FIG. 6 to offer an intuitive explanation. Steps of wall potential (U step in FIG. 6) within prior art thick electrode ion mirror 20 allow reaching the desired field properties and compensating multiple time aberrations in the close vicinity of the ion mean turning point at X=0, as witnessed by curve 72 in FIG. 7. This is achieved by the optimized and adjusted field penetration from thick electrodes, surrounding the ion turning point at X=0 and U=U.sub.0. However, by fields nature, such penetration of stepped U produce larger field variations and non monotonous higher field derivatives in a somewhat wider region around the turning point, thus, not sustaining the desired ion optical properties for wider energy spreads of ion packets, corresponding to longer spans of ion turning points. In contrast, the ion mirror 40 according to embodiments of the invention with a stepped field strength E on the wall generates an initially constant field strength E≅E.sub.0 in the wider vicinity of ion turning point X=0, while field penetration from surrounding field segments allows adding a desired and optimal degree of the field non uniformity and curvature of equi-potential lines, thus, providing for a wider spatial span of ion reflecting points, where time aberrations are compensated, this way providing for a wider energy acceptance of ion mirror.

(41) Optimizing novel mirrors: To accelerate the analysis and the optimization of ion mirrors according to embodiments of the invention, the inventor came up with an analytical expression for the axial distribution of the electric field E(x) in the planar two-dimensional gap with height H, where two segments with the field strengths E1 and E2 are openly merged at X=0:
E(x)=E1+(E2−E1)*(2/π)*arctan(exp[−π*x/H])

(42) At |X/H|>0.1 the expression may be approximated by:
E(x)=E1+(E2−E1)*(2/π)*exp(−πX/H)*[1+⅓*exp(−2πX/H)+⅕*exp(−4πX/H)]

(43) Having an analytical expression strongly accelerates ion optical simulations and optimization procedures. Now we could vary parameters—channel height H, segments lengths Li and segments field strengths Ei at the walls, while optimizing a large set of low-order and high-order time and spatial aberrations for a variety of mirror systems which differ by the entrance lenses.

(44) Optimization criteria: In optimization procedure we were setting acceptance criteria, comprising: spatial ion focusing (Y|Y=0 per one reflection); at least third-order time per energy (T|K=T|KK=T|KKK=0) focusing with low or zero higher order time per energy terms; full compensation of at least second-order time per spatial, angular and energy aberrations, including cross terms; and wider spatial and angular acceptances of model ion mirrors at about 1E-5 level of isochronicity.

(45) Variety of novel mirrors: To provide for spatial ion focusing, the mirrors according to embodiments of the invention may have an entrance lens, preferably at an attracting potential |U.sub.L|<|U.sub.D|, which can be either a single stage lens or a multi-stage lens, or an immersion lens. The entrance lens part can be formed either with stepped field segments of thick electrodes. The reflecting fields of mirrors according to embodiments of the invention were constructed with segmented fields (stepped E) and were individually optimized per specific entrance lens. Varying the lens part of the ion mirror leads to minor adjustments of the mirror reflecting part if optimizing those ion mirrors for lowest aberrations and highest energy acceptances.

(46) Referring to FIG. 8, diagram 80 presents potential distributions (U/U.sub.0) Vs X/D at electrode walls for another three variants of ion mirrors according to embodiments of the invention with stepped field (step E) reflecting parts. Plot 81 corresponds to an ion mirror with accelerating lens formed with segmented fields (stepped E), 82—with a long accelerating lens, formed with thick electrodes (stepped U), and 83—with a decelerating lens, formed with segmented fields (stepped E). Obviously, the two field segments, denoted by their fields E1 and E2 in the reflecting part of ion mirrors are quite similar for all three variants.

(47) Table of comparison of ion-optical parameters of E-Steps and T-Steps mirrors.

(48) TABLE-US-00001 U-steps, E-steps, E-steps, Mirr 20 Mirr 40 MirrB (T|KKKK)/T.sub.0 11.5 0 0 (T|KKKKK)/T.sub.0 8.7 −4.67 −7.70 (T|KKKKKK)/T.sub.0 116.7 31.2 45.9 (T|BBK)/T.sub.0 6.4 13.3 9.7 Acceptance ΔK/K 6% 16% 14% at ΔT/T.sub.0 = ±1E−5 Aberration corrected ΔK/K 9% 22% 17% ΔB, mrad, at [(T|BBK)/T0]* 14.4 5.35 6.97 B.sup.2(ΔK/K/2) = 1E−5

(49) Sweet spot: While varying the lens part of novel ion mirrors, optimizing ion mirror aberrations, and analyzing parameters of field segments, we arrived to the following conclusions and rules:

(50) 1. Qualitative rules:

(51) Ion mirrors composed of segmented fields allow reaching substantially better ion optical quality than prior art thick electrode mirrors. In particular, mirrors according to embodiments of the invention provide for about twice larger energy acceptances, which allows strong improvement of MRTOF resolution per flight path, while not compromising or moderately compromising other properties of thick electrode mirrors, such as spatial ion focusing and wide spatial acceptance at high (1E-5) isochronicity; The optimum for ion mirrors according to embodiments of the invention appears when the field E2 in the segment containing the ion turning point has a weak field non-linearity (E0−E2)/E2 in the range of a few percent, primarily produced by penetration of a stronger field E3. Then such mirrors provide for all the desired properties, listed in the section “optimization criteria” and provide for substantial improvement the of energy acceptance compared to prior art thick electrode mirrors; The optimal non linearity of the electric field near the ion turning point in E2 region is obtained by a weak penetration of electric fields from the upstream (i.e. towards the mirror entrance/exit) adjacent segment E3, where further improvements are obtained by penetration of the downstream field segment E1; While using uniform field segments around the ion turning point is important, the rest of the ion mirror may be composed of either uniform field segments or may use conventional thick electrodes. The lens part may be chosen as per particular requirements of the ion mirror, where: (a) an accelerating lens provides for highest energy acceptance, normalized to ion mean kinetic energy; (b) decelerating lenses are capable of providing the same absolute energy acceptance at the same maximal mirror voltage, but at notably higher ion kinetic energies; (c) longer, multi-stage, and immersion lenses reduce time per spatial aberrations at a cost of higher lens complexity; (d) similar lens constructed with segmented fields require lower absolute voltages and provide for smaller aberrations.
2. Sweet spot parameters for two-dimensional (2D) mirrors of planar symmetry. Exact optimal parameters may slightly vary between ion mirrors with different entrance lenses, however, all systems fall into the below described range of parameters, as illustrated in FIG. 9: The required electrode density in the ion reflecting segment E2 shall be supporting smoothness of the generated field better than 1%, which is achieved if the period between thin electrodes in the E2 segment is less than 0.2 of the window height H: P≤H/5; Optimal strength of the reflecting field E.sub.0 at the ion turning point is linked to the specific (per charge) ion mean energy U.sub.0=K.sub.0/q and to the distance from the ion turning point to the time focal point D as 4.3<E.sub.0*D/U.sub.0<5; Optimal height H of ion mirror window relates to distance D: 0.04<H/D<0.06 with best results obtained in the range: 0.045<H/D<0.055; The useful (for improved energy acceptance) non-linearity (E.sub.0−E2/E2)|X=0 of electric field E2 at the ion turning point X=0 is between 0.1% and 10%, with better results obtained in the range between 0.5% and 5% and with very best results obtained in the range from 1% to 2%. The distance X3 from the ion turning point (X=0) to the knot electrode (inter-segment electrode) U3 appears linked with the ratio of fields E3/E2 (where E3>E2) for reaching the desired field penetration and for the desired range of field non linearity in the E2 field segment, as shown in FIG. 9F; While energy acceptance already improves when using at least two field segments, shown as E2 and E3 in FIG. 8, however, adding segment E1 with E1<E2 further improves energy acceptance. Usually, optimal E2/E1 ratio varies in the range from 1.01 to 1.1 with best results obtained in the range from 1.02 to 1.05.

(52) Referring to FIG. 9, the above expressed sweet “spot rules” are illustrated by a set of diagrams 91 to 99, with annotations being presented in the scheme 90 (also matching those in FIG. 4). Simulations were made for a number of novel ion mirrors, denoted on drawing as E-steps, and composed of field segments E1, E2, E3. The lens part was varied between mirror variants, where simulated cases comprise short and long lenses, accelerating and decelerating lenses, thick electrode and segmented field lenses. Parameters of various simulated ion mirrors were normalized to the window height H, to the distance D from the ion turning point to the time focal point, and to the potential of the ion turning point U.sub.0 (assuming grounded drift region). Similar normalization have been made for a number of prior art thick electrode (U-steps) ion mirrors, referred to in the introduction.

(53) Diagram 91 shows the normalized field strength at the ion turning point E.sub.0D/U.sub.0 for novel ion mirrors (E-steps) 92, and for prior art thick electrode mirrors (U-steps) 93. Data points are aligned by the ratio X2/H, which can not be defined in thick electrode systems and is set to 0 for displaying purposes. While E.sub.0D/U.sub.0 may widely vary for thick electrode mirrors, the optimal range is narrow and well defined for novel mirrors: 4.5<E.sub.0D/U.sub.0<5, with most of points clustered around E.sub.0D/U.sub.0=4.6. The result means that all novel mirrors reproduce similar optimal field distributions in the ion reflecting part.

(54) Diagram 94 shows the normalized window height H/D for novel ion mirrors (E-steps) 95, and for prior art thick electrode mirrors (U-steps) 96. Data points are aligned by the ratio X2/H. While H/D ratio may widely vary for thick electrode mirrors, the optimal range is narrow and well defined for novel mirrors: 0.04<H/D<0.06, with most of points clustered around H/D=0.055, again meaning that novel mirrors reproduce similar optimal field distributions in the ion reflecting part.

(55) Diagram 97 plots the field non linearity (E.sub.0−E2)/E2 for novel ion mirrors at ion mean turning point (X=0), aligned with the X2/H ratio (same as in diagrams 91 and 94). The plot illustrates the central point of the invention—novel ion mirrors composed of field segments should have a non-zero optimal non-linearity at the ion turning point to provide for a notable improvement of the energy acceptance. The useful range of the reflecting field non-linearity appears 0.01<(E.sub.0−E2)/E2<0.04 for all simulated cases of novel mirrors. Comparing energy and angular acceptances of all simulated cases, best results are obtained in the range 0.015<(E.sub.0−E2)/E2<0.03.

(56) Diagrams 97 and 98 illustrate that to reach the optimal non-linearity of diagram 97, the steps in the surrounding field shall be linked to the depth of mutual field penetration. According to diagram 98, field strength of E1 segment shall be slightly smaller than E2: E1<E2; 1.02<E2/E1<1.08. E2−E1 step grows at deeper field penetration X2/H. The useful range of penetration depth X2/H is limited to 0.8.

(57) According to diagram 99, the field strength E3 should be in general larger than E2 (E3>E2), and the E3/E2 ratio is linked to the penetration depth X3/H by an empirical formula: E3/E2=[0.75+0.05*exp((4X3/H)−1)], that is E3/E2 grows with deeper X3/H penetration. The penetration depth X3/H is limited to 1.7.

(58) In some exceptional cases, where the penetration depth X3/H is small, E3 can be somewhat smaller that E2; in this case the proper sign of the field strength non-linearity at the ion turning point is provided by penetration of the field E4 from the next (4-th) segment. Thus, in the most general case the ratio of field strengths E3/E2 is E3/E2>0.8 and is linked to the X3 distance by the relation E3/E2=A*[0.75+0.05*exp((4X3/H)−1)], where 0.5<A<2 to provide for a controlled non-linearity of the axial field distribution, demonstrated to enhance the energy acceptance of the ion mirror.

(59) The above presented graphs and empirical rules tell that in all simulated cases novel ion mirrors reproduce a similar structure of ion reflecting field, characterized by a weak though controlled field non-linearity 0.01<(E.sub.0−E2)/E2<0.04 at the ion turning point X=0. This non-linearity is achieved by a field penetration from adjacent field segments with E1 and E3 fields, where steps in field strength E1/E2 and E3/E2 appear linked with the depth of field penetration X2/H and X3/H for improving the ion mirror energy acceptance.

(60) Referring to FIG. 10, energy acceptances ΔK/K.sub.0 are presented for novel ion mirrors (E-steps) and for best known prior art thick electrode mirrors (U-steps). The set of analyzed ion mirrors matches the one used in FIG. 9.

(61) Similar to FIG. 7, energy acceptances are calculated at exactly zero T|K(n) aberrations (101 and 103) and for the case of intentionally left minor residual low-order aberrations (102 and 104), maximizing energy acceptance at a given level of isochronicity, here at ΔT/T=1E-5 level. Data points are aligned with X2/H, similar to graphs of FIG. 9. One can see that the energy acceptance of novel mirrors (E-steps) is about twice higher than for prior art thick electrode systems (U-steps) in both non-compensated and compensated cases. It is also apparent that novel mirrors optimize (for higher energy acceptance ΔK/K.sub.0) at either small X2/H or small X3/H (either X2/D<0.3 or X3/D<0.3), meaning that ion mean turning point shall be close to at least one field boundary to provide for a sufficient non-linearity and curvature of reflecting field at the ion turning point (X=0).

(62) It must be understood that the range of sweet spot parameters presented in FIG. 9 may be somewhat wider if softening requirements onto the ion optical quality of novel ion mirrors. FIG. 11 and FIG. 12 present cases of compromised novel ion mirror with reduced number of power supplies. FIG. 16 presents a case of a compromised novel ion mirror with a reduced relative width H/D and with a different balance between mirror aberrations.

(63) Two reflecting segments: Referring to FIG. 11, graph 110 shows the potential distribution U(x)/U.sub.0 Vs X/D for a simplified novel mirror; curve 111—at the electrodes, and curve 112—at the symmetry axis (Y=0). The simplified novel mirror is composed of fewer field segments to reduce the number of high voltage supplies to three, not accounting drift space supply. The reflecting part uses only two field segments E2 and E3. The non-linearity and the curvature of the E2 field at the ion mean turning point (X=0, U=U.sub.0) are formed by penetration of the E3 field only, where the distance X3 from the turning point to the field boundary is about 0.075 D and is smaller than 1.5H. Graph 113 shows time per energy plot at some residual lower-order time per energy aberrations, shown in the icon 114, optimized to expand the energy acceptance ΔK/K.sub.0 to 12% at ΔT/T.sub.0<1E-5 isochronicity. The achieved energy acceptance 12% of the mirror 111 is notably lower than ΔK/K.sub.0=21% of the mirror 40 in FIG. 4. Thus, reducing number of power supplies and leaving field penetration from one side only compromises parameters of segmented ion mirror.

(64) Referring to FIG. 12, there is shown an electrical scheme 121 for a more efficient way of reducing the number of power supplies. Accounting that the field strengths E1 and E2 are close in optimal novel mirrors (see plot 98 in FIG. 9), it is preferable omitting the U2 supply, while adjusting the E2/E1 ratio by an additional resistor 122. While using a shunt divider is an obvious step, however, it is not obvious whether reducing the number of adjustable parameters still allows mirror tuning. In practice, setting of E2/E1 ratio by the resistor 122 may be achieved within 1% routine accuracy. Plot 122 shows that inaccuracy of E2/E1 setting in the ion mirror 40 of FIG. 4 may be compensated by tuning voltages U1 and U3. Plot 123 shows that accuracy of E2/E1 setting shall be maintained with 0.1% precision in order to sustain an improved energy acceptance ΔK/K.sub.0=22% of novel ion mirrors. Thus, using shunt resistors in prior art was not supported by the knowledge of optimal mirror parameters and did not account the requirements on the divider precision.

(65) Novel ion mirror embodiments: Referring to FIG. 13, embodiment 130 presents the “generic” electrode structure and electrical scheme for energizing of novel ion mirrors of the embodiments of the present invention. Stepped fields of novel ion mirrors are generated by forming several segments of linear potential distributions E1 . . . E4 at thin (per X-direction) electrodes 131, while the segments remain open to each other, i.e. not separated by grids. Thin electrodes may be formed with sheet frames or by parallel electrode rows.

(66) Uniform fields between electrodes within each segment are supported by resistive chains 134, say, using commercially available resistors with 0.1%-1% precision and 10 ppm/C thermal coefficients. Potentials 135, denoted as U0, U1 . . . and U.sub.D are then applied to “knot” electrodes (inter-segment electrodes) 133 only. The power supply U2 may be omitted and the ratio of the field strengths E1 and E2 adjusted by additional shunt resistors Rs with at least better than 1% precision. Diagram 136 shows potential distributions: 138—at the electrodes, and 139—at the mirror axis. It is of practical importance that minor variations of individual electrode thickness or voltages are expected to be smoothed and compensated by potential tuning. To provide a reasonably uniform field at least within the E2 segment, the electrode period P in this segment shall be at least 5 times finer than the window height H: P≤H/5. Since the optimal window height H is about 1/40 to 1/50 of cap-cap distance Lcc≅2D in MRTOF, design 130 requires making physically narrow electrodes. Say, for Lcc=50 cm the above requirement converts into P<2 mm, while electrodes 131 shall be yet thinner to allow for insulating gaps. Thus, making and assembly methods shall provide for mechanical stability and straightness of electrodes 131.

(67) Thin electrodes designs: Referring to FIG. 14, novel ion mirrors 140, 143, 145 and 148 may be constructed of thin (0.5-3 mm) electrodes 131, which may be either stamped or EDM machined from a metal sheet, or made from metal coated PCB plates, or from carbon filled epoxy rods made by protrusion. Parallelism of thin electrodes is sustained by features, being particular per exemplary design.

(68) In embodiment 140, the straightness of electrodes 131 is sustained with slots in the substrate 142, where the substrate may be either plastic, ceramic, glass, Teflon, or epoxy (say, G-10) material. A pair of opposite substrates 142 may be aligned by pins or shoulder screws in thick electrodes, such as the cap 131C electrode and the thick entrance electrode 132.

(69) In embodiment 143, straightness of electrodes 131 is sustained by precise insulating spacers 144 at electrodes clamping with screws (e.g. made of plastic threaded rods or metal screws with PTFE sleeve). Spacers 144 may be either ring spacers or insulating sheets, both made of either plastic, PTFE, PCB, or ceramic. Electrode side shift is controlled by assembly with technological jigs and electrode displacement is prevented by tight clamping. Note that the design 143 is least preferred for accumulating inaccuracies in stack assembly and for being susceptible to electrode bend if spacers' surfaces are not highly parallel.

(70) In embodiment 145, straightness of electrodes 131 is ensured by: (a) making initially flat electrodes (e.g., EDM made or stamped and then improved with thermal relief in stack); (b) aligning electrodes 131 with a side technological fixtures (not shown jig); and then (c) fixing electrodes 131 to the substrate 147 with connecting features 146. Preferred substrate 146 is PCB with metal coated vias. Other insulating substrates are usable, including plastic, ceramic, PTFE, glass and quartz. Preferred methods of attachment are epoxy gluing or soldering. When soldering, the preferred material for electrodes 131 is nickel 400 material, so as nickel or silver coated stainless steel. When gluing, the preferred electrode material is stainless steel. Electrodes 131 are preferably EDM machined or stamped with multiple connecting pins. Alternatively, electrodes 131 may be attached by brazing or spot welding to metal coated vias or pins in ceramic PCB. Yet alternatively, electrodes may be attached by rivets or connected by side clamps to plastic or PCB substrates.

(71) In embodiment 148, electrodes 131 are made of carbon filled epoxy protrusion, optionally coated by metal for reducing chips and dust. The material provides an exceptional initial straightness, not achievable with metal rods. Electrodes 131 are aligned by technological jigs on each support plate 147 (PCB, plastic, ceramic, PTFE, or glass) for gluing or soldering via standoffs 146. Epoxy based PCB (like FR-4) are preferred for matching and low thermal coefficients TCE=4-5 ppm/C.

(72) In case of using PCB supports 147, dividing chains may employ surface mount (SMD) resistors or a resistive strip generated with resistive inks, in particular developed for ceramic substrates.

(73) PCB designs: Referring to FIG. 15, another and more preferred family of ion mirror embodiments comprises an open box 150 (2D view 151), composed of printed circuit boards (PCB) 152, exampled with PCB variants 152-A to D. Optionally, the box is enclosed with side PCB boards 152s. PCB technology provides standard methods of making thin conductive stripes 154 (down to 0.1 mm thick) with high precision and parallelism, specified better than 0.1 mm. Conductive stripes may be curved as shown in PCB embodiment 152-D. PCB substrate 153 may be made of epoxy resin (FR-4), of ceramic, quartz, glass, PTFE or of kapton (useful for cylindrical mirror symmetry).

(74) Preferably, PCB plates 152 and side PCB plates 152s are attached to thick supports 132 with aligning pins or shoulder screws, though thick plates may be replaced by metal coated PCB 159 for better thermal match and lower weight. In this case, the overall assembly 150 is fixed by technological jigs and soldered or glued. Preferably, stiffness of boards 152 is improved with PCB ribs 158. Preferably, SMD resistors 134 are soldered on outer PCB surfaces, where connection of conductive stripes 154 to power supplies 135 and to dividing resistors 134 may be arranged either with vias 156, or with edge conductive strips, or with rivet holes, or with side clamps. SMD resistors may be replaced by a distributed resistor, formed by a paste with resistance in MOhm/square range, with the resistive paste being applied between and on top of electrodes 154. Then the dividing chain may be placed on inner box surface without making vias 156. PCB 152 may further comprise conductive lines to connecting pads for convenient connection to vacuum feedthroughs, or may have an intermediate multi-pin connector for connecting assembly 150 by a ribbon cable. PCB 152 may further comprise mounting and aligning features for assembling the overall MRTOF analyzer.

(75) Antistatic PCB features: It is advantageous to provide antistatic properties to the inner PCB surfaces (in box 150) that may be exposed to stray ions. On one hand, it is desired that the antistatic features shall not distort the accuracy of the resistive dividers 134, at least at 1% precision, meaning that the resistance between strips may be above 100 MOhm, which corresponds to approximately 10 GOgm/square minimal surface resistance, accounting about 100:1 length to width ratio of insulating strips. On the other hand, ions scattered from nA beams may produce up to 10 fA/mm2 currents onto the insulating support. To maintain potential distortions well under 0.1V, the antistatic surface resistance may be under 10 TOhm/square. Thus, antistatic coatings do not have to be precise and uniform but could be maintained in a wide range from 1E+10 to 1E+13 Ohm/square. This is 10-100 fold lower relative to standard resistance of FR-4 PCB boards, specified at 1E+14 to 1E+15 Ohm/square.

(76) One solution is to use ceramics substrates having lower own resistance, such as Zr02, Si3N4, BN, AlN, Mullite, Frialite and Sialon. However, ceramics are less attractive as they are higher cost and have a fragile overall construction. More favorable solutions are shown in FIG. 15. They are based on deposition of an antistatic layer or using a finer electrode structure.

(77) Again referring to FIG. 15, PCB embodiment 152-A employs a structure of fine (0.1 mm wide) intermediate conductive strips 157 between relatively thicker conductive strips 155. Optionally, the potential drop between fine strips may be distributed by a resistive coating 155. Making local coating for a crude potential distribution is less challenging than coating the entire PCB. Besides, numerical estimates show that in the case of using fine strips 155, the self conductance of PCB in 1E+14 Ohm/square range may be sufficient even without using the resistor layer 155. Experimental tests shall be made to confirm that the PCB conductivity is reproducibly sufficient from batch to batch.

(78) PCB embodiment 152-B shows an example of antistatic coating 155 deposited on top of PCB 153 conductive stripes 154. The coating may be then made after PCB manufacturing. Antistatic coating 152 may be formed by exposing epoxy or ceramic PCB to glow discharge with deposition of copper, aluminium, tin, lead, zirconium, or titanium. Alternatively antistatic coating may be produced by depositing conductive particles (say carbon powder) with thin polymer coating. Embodiment 126 shows example of resistive layer (similar to one used in electron tubes and scopes) under conductive stripes 121, which may be preferred for better adhesion on ceramic, quartz and glass substrates.

(79) PCB embodiment 152-C presents a reversed case, where the antistatic coating 155 is deposited on top of PCB 153 before depositing conductive stripes.

(80) Solving antistatic PCB properties opens an opportunity of using economy PCB for making ion mirrors. PCB technology provides an advantage of forming thin and sufficiently parallel electrodes, so as provides a convenient method of making fine resistive dividers by using economy and compact SMD resistors. PCB technology is a perfect match for novel ion mirrors. We can state that novel ion mirrors are designed for PCB technology and PCB technology is the best way of making novel ion mirrors composed of field segments.

(81) Mirror stack: Referring to FIG. 16, a stack 160 of slim PCB mirrors (like 150 in FIG. 15) is proposed for constructing a multi-reflecting TOF with an orthogonal accelerator (OA) at a very large duty cycle. Embodiment 160 comprises: an ion source S, here shown with a gas filled RF ion guide followed by set of lenses; an elongated orthogonal accelerator 161 with the OA storage region having ion confining means 162; a trans-axial lens 163 at the OA exit; two stacks of slim PCB mirrors 166; a detector 167; and optional two pair of deflection plates 165. Exemplary ion confining means 162 are described in the co-pending application GB 1712618.6 and may include various electrostatic or RF ion guides, such as periodic lens, quadrupolar electrostatic guide, alternated quadrupolar electrostatic ion guide.

(82) In operation, ions from the ion source S are ejected into the OA 161 and travel along the confining means 162 at a moderate energy, say, 20 to 50 eV. Periodically, pulses are applied to (not shown) Push and Pull electrodes of the OA 161, optionally accompanied by switching voltage on the confining means 162. Long ion packets (50-150 mm long) 164 are extracted from the OA, spatially focused by a trans-axial (TA) lens 163 in the Z-direction and enter a field-free space between the ion mirrors 166 at a moderate inclination angle, expected in the order of 3 to 5 degrees. Two stacks of slim PCB ion mirrors 166 are arranged for opposed ion reflections. The opposed stacks are half-period shifted in the Y-direction. Ion packets 168 get side displaced in the Y-direction at every ion mirror reflection, while being spatially focused in the Z-direction by one of the following actions: (i) either by the action of TA-lens 164 alone; (ii) or being assisted by spatial focusing of PCB mirrors with curved strips as in embodiment 152-D of FIG. 15; or (iii) by a combined action of spatially focusing TA-lens and isochronicity compensating field bows arranged within at least one PCB ion mirror. Electrostatic wedge field of PCB mirror may be used for compensating possible mirror misalignments within the XZ plane, in other words, compensating components minor rotation around the Y axis.

(83) As a result, the long ion packet 168 does not interfere with the OA after the first ion mirror reflection, even though the ion drift displacement AZ per mirror reflection is much shorter compared to the Z-length of the ion packet 168. Ion packets are spatially focused in the Z-direction (by a TA lens, optionally assisted by curved fields in PCB mirror) at prolonged flight path, corresponding to several ion mirror reflections to focus (in the Z-direction) ion packets when they hit the ion detector 167. Thus, the novel embodiment achieves multi-reflecting TOF separation of long ion packets at fully static operation of MRTOF. Absence of deflecting pulses preserves the full mass range of mass analysis.

(84) The embodiment 160 also illustrates that the ion injection from wider (in Y-direction) OA and into slim ion mirrors 166 may be assisted by using two pair of deflection plates 165 for side ion deflection in the Y-direction at a relatively small angle and moderate time-of-flight aberrations associated with the Y-steering. Large duty cycles of OA in the order of 20-30% are expected at static ion beam operation, and the duty cycle may be further improved to nearly unity if accumulating ions in the RF ion guide and synchronizing pulsed ion ejection with OA 161 pulses.

(85) The stack 166 of slim (in Y-direction) and low cost PCB based TOF and MRTOF analyzers allows various known multiplexing solutions, such as: E-trap with enhanced dynamic range, as described in WO2011086430; using multiple ions sources, or increasing pulsing rate of single ion source, and using multiple channels for MS2 analysis in MS-MS tandems as described in WO2017091501 and WO2017042665.

(86) Although the present invention has been describing with reference to preferred embodiments, it will be apparent to those skilled in the art that various modifications in form and detail may be made without departing from the scope of the present invention as set forth in the accompanying claims.