Technical Field of the Disclosure

The disclosure concerns electrostatic ion traps, arrays of electrostatic ion traps and methods of analyzing an ion beam.

Cross-Reference to Related Applications

This application claims priority from U.S. Patent Application 17/823,618, filed 31 August 2022, which is incorporated herein by reference.

Background to the Disclosure

High-resolution accurate-mass (HR/AM) analyzers utilizing electrostatic fields are well known. These include: Multi-reflection Time-of-Flight (mrTOF) analyzers with destructive ion detection (for instance, using secondary electron multipliers) including those described in WO2013110587, WO2019202338, WO2017087470 and references therein; orbital trapping mass analyzers, including those described in US5886346, US7767960, US7985950; and electrostatic ions traps with closed and open trajectories, including those described in US5880466, US9728384, US10453668 and D.Z. Keifer, et al., Analyst 142, 1654 (2017).

Orbital trapping mass analyzers and some other electrostatic ions traps use image current for non-destructive detection of ions. With sufficiently low capacitance of detection electrodes and transistors of a preamplifier, it is possible to detect individual ions over prolonged duration of detection. This is practiced in, for example, charge detection mass spectrometry (CDMS), as described in US11232941 , US11227759, US2022068624. Recently, detection of a single elementary charge was demonstrated by A.R. Todd et al J. Am. Soc. Mass Spectrom. 2020, 31 , 146-154.

Existing HR/AM analyzers are limited in their productivity by a few tens to few hundred spectra per second. For efficiency reasons, as many species as possible are thus crowded in a single MS or MS/MS spectrum, up to the limit of space charge. Space charge effects may ultimately limit the dynamic range and depth of analysis.

As a result, such analyzers can identify no more than 10 to 100 of the most abundant species per second. Although this level has drastically increased over the last several years, due to broad adoption of fast liquid chromatography and data-independent acquisition (DIA), new approaches are needed for the next jump of productivity, especially in proteomics.

Recent breakthroughs in the informatics foundations of DIA, including adoption of machine learning, have dramatically improved the reliability of deconvolution of mixed spectra into individual fragment spectra. This approach may work reliably for the most intense components of mixed spectra, but cannot be applied to species at the lower end of the dynamic range, as they are represented just by a few ions. This eventually limits the depth of analysis, especially in proteomics, where dynamic range of concentrations could span over 10 to 12 orders of magnitude. The problem is even more acute in single-cell proteomics, where the total number of peptides available for analysis is limited to a few billions per cell, with only a few hundred million ions actually entering a mass spectrometer.

Different techniques for enrichment have been tried either on the sample preparation side (Seer, fractionation) or ion sorting side (for instance, as shown in US9812310, US10199208). These typically work by creating a number of fractions that afterwards need to be analyzed either individually or in pools. Performance appears to be improved by these approaches, but not equally well for all species, thus leaving a number of low-abundance species unrepresented or even lost in the process. Meanwhile, individual analysis of fractions in the same mass analyzer reduces throughput proportionally, as a mass analyzer is essentially a single-channel device.

Arrays of mass analyzers have been considered to address this throughput problem. Arrays of electrostatic ion traps have been studied extensively, including for example, US5206506, US7718959, US2003089846, US6762406, W02006049623, US9779930, W02021061650 and US7985950 (in the form of orbital trapping mass analyzers). Arrays of some other analyzers have also been considered, for instance US7985950, US10049867, US10593533, US2008067349. All of these transport ions across three dimensions to make the overall multi-analyzer mass spectrometer space efficient. Nonetheless, high-resolution, accurate-mass capabilities have proved difficult in such arrays because of inherent limitations on electrode accuracy.

Improvements in electrostatic ion traps and arrays of mass analyzers are desirable to address these challenges.

Summary of the Disclosure

Against this background, there is provided an electrostatic ion trap according to claim 1 , an array of electrostatic ion traps in line with claim 26 and a method of analyzing an ion beam comprising a mixture of different analyte ions as defined by claim 32. Preferred and/or optional features are disclosed in the dependent claims.

A new type of microscale electrostatic ion trap (pEST) is proposed, with a typical length of the trap being no more than 10 mm and optionally no more than 5 mm, 2 mm, 1 mm or 0.5 mm. Additionally or alternatively, the capacitance of at least one of the electrodes (in particular, a detection electrode) to ground is no more than 1 pF. The pEST advantageously has planar electrodes distributed along a longitudinal axis (z-dimension). Each electrode extends perpendicular to the longitudinal axis (in a width or x-dimension). One set of electrodes distributed along the longitudinal axis is spaced from a second electrode set (also parallel to the longitudinal axis) to define a trapping region between the electrode sets (and defining a height or y-dimension). Electrostatic potentials are applied to at least some of the electrodes for confinement of ions received in the trapping region. For improved field, the two electrode sets may mirror one another in their configuration. Implementation of the trap electrodes on a sub-millimeter scale, preferably with nanometer tolerance, allows high resolution. The pEST may have controller, to configure its operation or a group of pESTs may share a controller. A controller may comprise a processor and computer program configured to operate on the processor.

Due to its small size and use of electrostatic potentials for ion trapping, the pEST permits confinement of a small numbers of ions, generally no more than 100, 50, 30, 20, 10, 5 or even a single charged particle in a small space. Nevertheless, high-resolution accurate-mass analysis is possible. New types of ion analysis therefore become possible. A measurement time of no more than 20 ms and/or an acceleration voltage of no more than 200V and/or a gas pressure within the electrostatic ion trap of no more than 10 ^-7 mbar may be achieved. Moreover, the pEST can be manufactured efficiently and cost-effectively, for example using lithographic techniques. The pEST could be formed as part of an integrated circuit. Modern micro- and nano-lithographic technologies may allow nanometer tolerances to be achieved on planar wafers.

One or more of the electrodes (typically towards the center of the trap along the longitudinal axis) is used for ion detection, by an induced image current. Reliable image current detection of single elementary charges is enabled by the microscale design of the trap. As noted above, the capacitance of one, some or each detection electrode to ground is kept low, typically no more (or less) than 1 pF, 500 fF, 100 fF, 50 fF, 10f F, 5fF or 1 fF. Such a low capacitance may permit single-charge detection. A transistor (for instance a FET or JFET) may be connected to the detection electrode or electrodes, which may also be formed lithographically.

Some of the electrodes (typically the outer electrodes) may be used for reflection of the ions, by application of a suitable potential. Other electrodes, generally between the reflecting electrodes and the detection electrodes may receive a suitable potential for accelerating the ions. By reducing the potentials on the reflecting electrodes, ions may be permitted to enter the trap. The potentials on the reflecting electrodes may then be raised to confine the ions to the trapping region. A gap between adjacent electrodes in the longitudinal (or z) dimension is generally no more than 100pm, preferably no more than 50pm and typically much smaller, for example no more (or less) than 20pm, 10pm or 5pm. A spacing (for instance, free space) between electrodes on different planes (in the height or y-dimension) is generally no more or less than 100pm and typically no more or less than 80pm or 70pm.

Each set of electrodes may be formed on a respective planar substrate (for instance, a wafer). To form the pEST, the two substrates may be positioned, such that the electrodes oppose one another. A spacer (or multiple spacers) may separate the sets of electrodes (or substrates or wafers). The spacer or spacers may include conductive spacers, for electrical coupling of electrodes. In some embodiments, the arrangement of each set of electrodes is substantially symmetrical between opposite sides of a center of the electrostatic trap along the longitudinal axis (z=0 line). Additionally or alternatively, one or both edges of at least some of the electrodes in the longitudinal axis has an arc shape. For example, some of the electrodes may have a curved shape, an arc shape, a circular shape or an elliptical shape. The curved shape permits improved containment of ions when using planar electrodes.

A wavy, modulated or multiply-curved shape of boundary between adjacent electrodes may be further advantageous. Since the electrostatic potential in the trap is a convolution (or blurring) of the potential on the boundary, such shapes of electrode interface may provide a greater degree of control over the electrostatic trapping field. In line with this principle, strip electrodes may be used. For example, a duty-cycle of the strip electrodes could be varied along the ion trap longitudinal axis. Such strip electrodes, particularly when held at different electrical potentials, may define one or more lenses. Using such modulated (wavy or multiply-curved) boundaries in this way, especially with strip electrodes, may reduce the number of distinct non-zero potentials required for application to individual electrodes in order to generate a trapping or confinement field. For example, two distinct non-zero potentials can readily be used and by further optimization, this may be reduced to single, non-zero electrostatic potential.

Fragmentation of ions within the pEST is possible. For example, this may be achieved by emission of a pulse from a UV or IR laser at the trapping region. The laser may emit the pulse in a direction orthogonal to longitudinal axis (that is, along the width dimension of the trap). A single laser pulse timed to match a trajectory of a target ion or multiple, unsynchronized laser pulses may be used. Light-based fragmentation techniques are known in the art. Their application to a pEST allows matching emittance of the laser to requirements of fragmentation.

The pEST may therefore be operated to confine one or more precursor ions in the trapping region (which may first be analyzed and/or selected), fragment the one or more precursor ions and then detect the fragment ions. MS/MS operation is thus possible both in data-dependent (DDA) and data-independent (DIA) acquisition modes. These steps may be repeated to provide MS ⁿ operation. The use of non-destructive image current detection permits multiple-stage analysis of a single individual ion. Data-dependent decision ion fragmentation is also possible, by controlling subsequent fragmentation based on previous detection.

In another aspect, an array of microscale electrostatic ion traps (pESTs) may be provided, each ion trap having a longitudinal length of no more than 10 mm and/or at least one electrode (preferably, a detection electrode) with a capacitance to ground of no more than 1 pF. Each pEST can therefore analyze a small number of ions, but with a large array, parallel analysis on a significant scale is possible. Overall throughput and sensitivity can thereby be improved. Optionally, the capacitance of one, some or each detection electrode to ground is kept low, typically no more (or less) than 1 pF, 500 fF, 100 fF, 50 fF, 10f F, 5fF or 1 fF. The array typically has an inlet for receiving an ion beam and a portion of the ion beam may be trapped in each of the pESTs. Each pEST of the array typically has a design according to the general details discussed above. Such an array may be especially useful for peptide analysis, as will be discussed below. The same calibration mixture may be used for calibration of multiple pESTs of the array.

In particular, the array may have a geometry based on one or more parallel planes, each plane being defined by two opposing substrates (wafers), having electrodes formed thereupon to oppose one another and thereby define one or more pEST in the same plane. In other words, traps may be formed using parallel wafers, stacked in multiple levels to achieve massively parallel operation. This may be advantageous, because a single laser pulse may be able to fragment ions in multiple trapping regions. Additionally or alternatively, three or more substrates may define multiple distinct planes, each with one or more pEST. Where fragmentation is caused by a UV or IR laser, a splitter arrangement may be used for spatially dividing the laser pulse output to the different planes. A lens array (particularly using miniature lenses, optionally with anti-reflective coating) may then focus each part of the pulse into a respective collimated beam.

In operation, ions from a single ion beam may be distributed to multiple pESTs of the array, which may then process the received ions in parallel. Statistical and/or machine-learning methods may be applied to the detection outputs from the pESTs. This may assist in identification of a composition of the single ion beam. Optionally, a chromogenic tag or a tandem mass tag are used to improve analysis.

A further aspect may be considered in relation to analyzing an ion beam comprising a mixture of different analyte ions. Ions from the ion beam are directed to multiple electrostatic ion traps of an array of electrostatic ion traps. Each trap has a longitudinal length of no more than 10 mm (and preferably smaller, as discussed above) and/or at least one electrode (preferably, a detection electrode) with a capacitance to ground of no more than 1 pF (and preferably smaller, as discussed above). Mass and charge data in respect of the ions can then be obtained (for example by analysis of the ions and/or derivatives of the ions) from the multiple traps, preferably by at least some simultaneous analysis. The mass and charge data can be combined to identify components of the mixture. This process may be performed by computer control and can be implemented in a computer program, for execution by a processor configured to control such an array of electrostatic ion traps or a mass spectrometer comprising such an array.

Advantageously, the ions (and/or their derivatives) can be fragmented in the traps. The precursor ions and/or fragment ions can be analyzed, advantageously to obtain mass and charge data from multiple traps and multiple degrees of fragmentation. This MS/MS operation can be extended to MS ⁿ operation by repeated fragmentation, as discussed above. Other features and/or aspects discussed herein may also be implemented within this approach. According to all aspects of the disclosure, higher productivity, sensitivity and dynamic range are possible. Combinations of aspects and/or features from within aspects are possible. Further benefits may be attained by combining aspects of the disclosure with ion sorting devices (for example, ion mobility analysis followed by an ion trap array).

Brief Description of the Drawings

The disclosure may be put into practice in a number of ways, and preferred embodiments will now be described by way of example only and with reference to the accompanying drawings, in which:

Figure 1 A schematically shows a basic design for a pEST according to an embodiment;

Figure 1 B shows an example distribution of potentials along the z dimension for the pEST of Figure 1A, for injection of ions;

Figure 2A shows a schematic diagram of a second embodiment of pEST in the plane of one set of electrodes;

Figure 2B shows the embodiment of Figure 2A in a perpendicular plane;

Figure 3A plots an example signal induced on the detection electrodes of the pEST of Figure 2 without noise;

Figure 3B depicts a simulated frequency spectrum from a single positive charge oscillating within the pEST of Figure 2 with thermal noise;

Figure 4 illustrates an example wafer having electrodes formed thereupon at different levels of magnification, showing an edge roughness of significantly less than 10 nm;

Figure 5 schematically depicts a pEST in three consecutive steps of operation;

Figure 6A shows a stack of wafers comprising a plurality of pESTs;

Figure 6B depicts a magnified portion of Figure 6A to demonstrate ion fragmentation by irradiation according to an implementation;

Figure 7 shows a schematic layout of a wafer and electrical connections for a single row of traps within an array;

Figure 8 illustrates a schematic diagram of an experimental apparatus used for testing and using a pEST array;

Figure 9 schematically depicts two planar electrostatic trap electrode designs, with (a) showing a side view applicable to both designs, (b) showing a top view of a first design with straight (or singly curved) electrode interfaces and (c) showing a top view of a second design with wavy electrode interfaces;

Figure 10 plots the electrostatic potential along the boundary created by a sinusoidally varying electrode interface in an example, for four different distances from the boundary;

Figure 11 plots variations of the electrostatic potential created by the sinusoidal electrode interface shown in Figure 10 for increasing distance from the electrode; Figure 12 shows different curved electrode interfaces, specifically: (a) a sinusoidal variation of inter-electrode boundary; (b) a square-wave interface with curvature of boundary midline; and (c) an elongated strip-electrode design with modulated duty-cycle along the longitudinal axis of the trap;

Figure 13 plots electrostatic field induced by a straight and sinusoidally varying electrodes interfaces along the boundary for four different interface lengths;

Figure 14A plots out-of-plane electric field along the midline of a planar electrostatic trap for an axially misaligned trap with sinusoidally varying electrode interfaces, for four different interface lengths;

Figure 14B plots out-of-plane electric field along the midline of a planar electrostatic trap for an axially misaligned trap with square-wave varying electrode interfaces, for four different interface lengths;

Figure 15 plots a boundary potential due to a square-wave type electrode interface;

Figure 16 plots electrostatic potential arising from a strip electrode design for four different displacements from the electrode boundary;

Figure 17A plots electrostatic potential of strip electrode design at (a) boundary and (b) midline for an example with linearly increasing duty-cycle;

Figure 17B plots electrostatic potential of strip electrode design at (a) boundary and (b) midline for an example with linearly increasing duty-cycle, where a circular arc is engineered into the induced midplane potential by adjusting displacing the duty-cycle variation of each strip;

Figure 18 plots electrostatic potential along the midline of the trap due to the strip electrode design of Figure 12(c);

Figure 19 depicts, for an electrostatic trap design based on electrodes held at just two distinct potentials: (a) boundary potentials of the electrodes with positive voltage electrodes, grounded electrodes and negative voltage electrodes; (b) a potential induced in the midplane of the trap, (c) a potential along the midline of the trap;

Figure 20 schematically shows an approach for aligning two substrates of a pEST based on a ball positioned in grooves; and

Figure 21 illustrates a schematic structure for an alternative approach for alignment, in which the surface metallization is deposited on a dielectric intermediate layer.

Detailed Description of Preferred Embodiments

The disclosure provides a new type of electrostatic ion trap, having a longitudinal length of no more than (and preferably less than) 10mm. This is termed a micro-scale electrostatic trap (pEST) and in some implementations, the longitudinal length may even be no more than (or less than) 5mm, 2mm, or 1 mm. Such an electrostatic ion trap can be formed by a set of planar electrodes, preferably two sets of opposing planar electrodes (parallel to each other). Such electrodes can reliably be formed with nanometer tolerance, for example using lithographic techniques. A pEST may thus be adapted for confinement and/or analysis of small numbers of ions, for instance less than 10 or even a single ion. Advantageous designs of pESTs will be discussed further herein.

The use of electrostatic potentials for trapping and/or analysis allows high-resolution accurate mass analysis. Non-destructive and reliable detection is possible by image current measurement, even allowing detection of single elementary charges. The latter is facilitated by the small size of the pEST. Indeed, an image current signal V induced by a charge q is proportional to q/C, where C is the capacitance between detection electrodes (which in turn, is broadly proportional to the size of the detection electrodes). For detection electrodes smaller than 1 mm, this results in image current signal V in the pV range (that is, less than 1 mV and typically significantly less than 1 mV, but generally higher than 10OnV). Keeping the capacitance of the detection electrodes to ground less than 1 pF and typically significantly smaller (for example, 500 fF, 100 fF, 50 fF, 10fF, 5fF or 1 fF) is advantageous. Meanwhile, the noise of modern transistors used to amplify the signal from those detection electrodes could be made well below 10 nV/^Hz. Also, the capacitance of those transistors could be also made below (or at least in the order of) the detection electrode capacitance C.

The inventors have realized that with such a low noise level, it is possible to achieve acquisitions within a time duration of 10ms to 100ms with the accuracy of charge determination better than one elementary charge (e=1 .602*1 O' ¹⁹ Coulomb). This allows not only determination of mass-to-charge ratio of ions, but also their charge and hence their mass directly, without the use of isotopic distribution, for example. This in turn opens opportunities for high-throughput analysis of individual ions.

An array of pESTs offers the opportunity to analyze large numbers of ions at the same time, without the limitation of a single channel and without a large volume for the instrument. Particularly beneficial designs of pEST permit space efficient arrays and may also allow parallel fragmentation of ions, as will be described below. MS/MS and MS ⁿ analysis is also possible thereby.

The disclosure thereby essentially overcomes the issues due to the single-channel nature of HR/AM mass spectrometers. Operating large numbers (for instance, thousands) of pESTs in parallel, each with a throughput of 10 to 100 analyses per second permits new modes of operation. For example, multiple fractions may be pre-separated by such ion storage arrays. Higher specificity of analysis is also possible.

This approach is distinct from known micro-scale RF traps. These include US8362421 , US8835839, Tolmachev et.al., Anal. Chem. 86 18 9162-9168 (2014), WO2017062102, US9984861 , US8975578, US8841611 . Although the dimensions of the ion trap may be similar to those described herein (and some of the manufacturing techniques may overlap), the structure of the traps are different and different voltages are applied.

Referring first to Figure 1 A, there is schematically shown a basic design for a pEST according to an embodiment. The pEST 10 comprises a first set of electrodes 20 and a second set of electrodes 30. Both the first set of electrodes 20 and the second set of electrodes 30 are planar. For example, the first set of electrodes 20 may comprise a substrate 22 (for instance, a wafer) and electrodes 25 formed on the substrate 22 (for instance, using lithographic techniques). Similarly, the second set of electrodes 30 may comprise a substrate 32 (which may also be a wafer) and electrodes 35 formed on the substrate 32. The electrodes 25 and electrodes 35 are elongated in a plane perpendicular to that of the drawing (in other words, extending in a width dimension). The height and sometimes the width are typically smaller (and preferably significantly smaller) than the length. The use of two opposing sets of electrodes D may provide high quality of electric field.

A longitudinal axis 40 of the pEST 10 is also shown (which may be considered to define a ‘z’ dimension). The electrodes 25 and electrodes 35 can therefore be considered as distributed along the longitudinal axis 40, with their width dimension (along a ‘x'-axis) thus being perpendicular to the longitudinal axis 40 (and not entirely in the plane of the drawing). As noted above, a maximum length of the pEST 10 along the longitudinal axis 40 is no more than 10 mm, preferably no more than 2 mm and more preferably no more than 1 mm. The gap between the two sets of electrodes 20, 30 defines a height dimension (along a ‘y’-axis), which may be set by the use of precision spacers (not shown).

The electrodes 25 and electrodes 35 each have respective functions, with opposing electrodes of the first set of electrodes 20 and the second set of electrodes 30 having the same function. The two central electrodes D are used for ion detection. At least some of the electrodes are configured to receive an electrostatic potential for confinement of received ions. The outermost electrodes R are used for reflection of ions (ion mirrors) and a suitable DC potential is applied to these. The electrodes between the detection electrodes D and the reflection electrodes R are acceleration electrodes A, to which a suitable electrostatic potential is applied for accelerating the ions and for their spatial focusing. An example ion trajectory 45 is shown, with the ions thereby being confined within a trapping region of the pEST 10, formed in the gap between the first set of electrodes 20 and the second set of electrodes 30.

Referring next to Figure 1 B, there is shown an example distribution of potentials along the z dimension for the pEST of Figure 1 A, for injection of ions. In this mode of operation, a voltage on the entrance of the trap (applied to the reflection electrodes R) is reduced to let in a short packet of ions of interest (which optionally might be mass pre-selected). The voltage applied to the reflecting electrodes R is then elevated to a reflecting level before ions return back to the entrance, to provide a trapping mode. In Figure 1 A, injection is shown as done from the top along the z-axis. After the voltage applied to the reflection electrodes R is increased (such that the trap is closed) and the ion trajectory is stabilized, detection takes place using the detection electrodes D. An on-chip differential amplifier 50 receives the image current from the detection electrodes D. The resulting signal is routed towards multiplexing and signal processing electronics, some of which may be off-chip. For example, part of the signal processing could be also implemented on the same circuitry, with remaining part (or parts) of the signal processing carried out on an application-specific integrated circuit (ASIC) or field-programmable gate array (FPGA).

Detection is thus carried out non-destructively based on image current and therefore multiple stages of mass analysis become possible, as will be discussed further below. The small size of the trap (with a typical gap between electrodes being less than 50 to 100 pm) facilitates a significantly lower capacitance for the detection electrodes than in existing electrostatic traps. This enables single-charge sensitivity and also allows simultaneous measurement of both charge and mass-to-charge ratio of individual ions. In this respect the electrostatic trap may be considered similar to, for example, a Cone-trap as described in D.Z. Keifer, E.E. Pierson, M.F. Jarrold, Analyst 142, 1654 (2017), but features much higher sensitivity and greater independence of the oscillation period on the initial parameters of incident ions.

Using a trap scaled to sub-millimeter dimensions (as shown in Figure 2) yields oscillation frequencies approximately in the range of 1 to 10 MHz. This enables resolving powers in the range of 30,000 to 100,000 to be achieved with less than 20 ms measurement times for peptide ion acceleration of 100 to 200 V. For this, a mean free path of greater than 100 m can be achieved with vacuum of less than 10 ^-7 mbar (10 ^-5 Pa). It may be understood that the frequency spread and hence accuracy of mass determination of the ions of interest is related to the tolerance of the design (flatness, parallelism, line accuracy, etc.). The exact relationship may depend on the beam and optics.

In a general sense of the disclosure and according to one aspect, there may be considered an electrostatic ion trap, comprising: a first set of planar electrodes distributed along a longitudinal axis of the ion trap; and a second set of planar electrodes distributed along the longitudinal axis of the ion trap, each of the electrodes of the second set arranged to be spaced apart from and oppose a corresponding electrode of the first set. A length of the first and second sets of planar electrodes along the longitudinal axis is advantageously no more than 10 mm (and optionally no more than 5 mm, 2 mm, 1 mm or 0.5 mm). Additionally or alternatively, a capacitance of at least one of the electrodes (preferably, one, some or each detection electrode) to ground is no more than 1 pF. Beneficially, at least some of the electrodes of the first and second sets are configured to receive an electrostatic potential for confinement of ions received in the space between the first and second sets of planar electrodes. The electrostatic ion trap may be configured to receive no more than 100 (optionally 50, 40, 30, 20, 10 or 5) ions injected into the electrostatic ion trap and preferably, the electrostatic ion trap is configured to receive a single ion injected into the electrostatic ion trap. A controller and/or processor may be provided to control operation of the electrostatic ion trap and/or to receive one or more outputs from the electrostatic ion trap. A computer program comprising instructions to implement a method of operation as herein disclosed may be configured to operate on such a processor.

A corresponding method of manufacturing and/or operating such an electrostatic ion trap may also be considered (for example, including a method of analyzing an ion beam), having steps of forming and/or providing and/or using the features of this device. For instance, the first and second sets of electrodes may be manufactured by lithographic techniques.

Preferably at least one of the electrodes (and more preferably some of the electrodes) is a detection electrode, configured to detect an image current of confined ions. By use of such a small electrostatic trap size (as discussed above), detection of image current is possible with resolution better than 1 elementary charge. The detection electrode or electrodes may be located toward the center of the ion trap along the longitudinal axis. The capacitance of one, some or each detection electrode to ground is advantageously kept low, typically no more than 1 pF, 500 fF, 100 fF, 50 fF, 10fF, 5fF or 1 fF. This may be implemented and/or advantageous even without a longitudinal length of no more than 10 mm, although the parameters may be closely connected.

In an embodiment, some of the electrodes are reflecting electrodes, configured to receive a reflecting potential. The reflecting electrodes are beneficially located at the ends of the ion trap along the longitudinal axis. In embodiments, the potentials on the reflecting electrodes are configured selectively to be at: a lower level, to allow ions to enter the ion trap; and a higher level, to confine the ions to the ion trap.

Optionally, some of the electrodes are accelerating electrodes, configured to receive an accelerating potential. The accelerating electrodes are preferably located between the reflecting electrodes and the detecting electrodes along the longitudinal axis.

Advantageously, the first set of electrodes are formed on a first planar substrate. In embodiments, the second set of electrodes are formed on a second planar substrate opposing the first planar substrate. The first and second substrates may form part of an integrated circuit. A gap between adjacent electrodes is preferably no more than 500pm, 200pm or 100pm (and more preferably no more than 50pm). One or more spacers may be provided between the first and second sets of electrodes. This may define a height of the device.

In embodiments, the arrangement of the first set of electrodes is substantially symmetrical between opposite sides of a center of the electrostatic trap along the longitudinal axis. Additionally or alternatively, the arrangement of the second set of electrodes is substantially symmetrical between opposite sides of a center of the electrostatic trap along the longitudinal axis.

The electrostatic ion trap may be configured such that one or more of: a measurement time is no more than 20 ms; an acceleration voltage is no more than 200V; and a gas pressure within the electrostatic ion trap is no more than 10 ^-7 mbar. Details of specific embodiments will now be discussed. Further description according to this general sense and other general senses of the disclosure will be provided below.

In practice, more complex electrode configurations than those shown in Figure 1 A may be employed to provide more optimized pEST properties with reduced spatial and time-of-flight aberrations. In this respect, reference is next made to Figure 2A, in which there is shown a schematic diagram of a second embodiment of pEST in the plane of one set of electrodes (using the dimensions discussed with reference to Figure 1 A, xz plane, y=0). Reference is also made to Figure 2B, showing the embodiment of Figure 2A in a perpendicular plane (yz plane, x=0). Both Figures 2A and 2B are annotated with dimensions, which are all in mm, and further annotated with voltages (voltages and dimensions are rounded to the nearest digit).

The design shown in Figures 2A and 2B differs from that shown in Figure 1 A in a number of respects. First, a larger number of electrodes are provided, including a central electrode Eo between the central detection electrodes (D1 , D2). Second, the shape of the electrodes differs. The electrodes on one side of the z=0 line have a generally rectangular shape (in the xz plane), whereas the electrodes on the other side of the z=0 line have shapes that are defined by arcs. The central electrode Eo has an edge on one side of the z=0 line that is defined by a straight line and an edge on the other side of the z=0 line defined by an arc. This central electrode E _o acts as a curved lens, with some of the other curved electrodes having a mirror function. Nonetheless (and as can be seen from Figure 2B), the arrangement of the electrodes on both sides of the z=0 line is otherwise symmetric. As is shown in Figure 2B, the gaps between all electrodes are 0.005 mm.

The edge of the central electrode Eo defined by an arc and the adjacent edge of the detection electrode D1 are both defined by concentric circles, centered at z=-1 . The radius of the edge of the detection electrode D1 is indicated as Ri. The edges of all of the other electrodes on that side of the z=0 line (except for the far edge of the end electrode) are defined by concentric circles, centered at z=+5.3. The radius of the edge of the sixth electrode nearest the z=0 line is also indicated as Re. The central electrode or curved lens E _o and the curvature of the electrodes in the z<0 part of the design (left side) keeps the ions along the z-axis. Otherwise, the ions may drift in the x-direction (along the width of the trap). In addition, the curvature of the central electrode or curved lens E _o may compensate for TOF aberrations from the curvature of the electrodes on the left. This curvature may allow containment of ions in the X-direction, preferably well within the x-dimension of detection electrodes D1 and D2. Instead of curvature, additional electrodes above and below of Fig. 2A could be also used (not shown), providing such containment by fringing fields.

The DC voltages applied to the electrodes are also shown. These voltages are applied symmetrically to electrodes on either side of the z=0 line and also symmetrically to opposing electrodes of the two sets, with the exception of the potential Uo applied to the central electrode E _o that has an opposing offset (labelled ±A) applied to the opposing electrodes for tuning.

Symmetry of the power supplying electrodes with respect to each trap is desirable (these are not shown, but typically are adjacent the far edges of the shown electrodes in the x-dimension and extending in the z-dimension), at least where the trap width in the x-dimension (width) is not more than 8 times h (where h is the height of the trap in the y-dimension, in this example 60 pm). Otherwise, an uncompensated ion drift could appear in this direction, which could distort the time- of-flight. This may be particularly relevant in a variant implementation, in which only electrodes without curvature are used (for example, as on the right side of Fig. 2A, where z>0 and mirroring this on the left side). In this case, fringing fields may be used to limit ion motion along the x- dimension.

The configuration shown in Figures 2A and 2B thus shows a more advanced planar pEST geometry that constraints ions in all directions, at the same time providing resolving power of up to 100,000 (or mass accuracy below 10 ppm) for ions with energy spread +/-1% within an incoming beam that occupies 40% of the gap between electrodes. This geometry may be seen to combine planar electrostatic mirrors with curved mirrors and a curved cylindrical lens in the middle.

In operation, ions oscillate between the electrodes receiving the Ue potential. The electrode receiving the U2 potential is the accelerating electrode. The electrodes receiving the potentials U3, U4 and U ₅ are finely tuned to reduce or minimize higher-order aberrations both for spatial and time-of-flight focusing. This may allow mass resolution that is one or more orders of magnitude higher than possible with the design of Figure 1 . In addition, demanding ion focusing properties, such as weak dependence on ion energy, angular spread and/or beam size are possible.

Referring now to Figure 3A, there is plotted an example signal induced on the detection electrodes of the pEST of Figure 2 without noise. Reference is also made to Figure 3B, in which there is depicted a simulated frequency spectrum from a single positive charge oscillating within the pEST of Figure 2 with thermal noise. From these plots, the strongly anharmonic nature of oscillations can be seen, resulting in first and third harmonics of the signal having similar amplitudes. Consequently, every ion could be detected at one or both of these frequencies, beneficially with signal-to-noise ratio high enough to determine its charge from either harmonic. Such a signal could be obtained with low-capacitance JFET transistors wire-bonded on the wafer (substrate), or with transistors manufactured in a single CMOS process with the wafer. With top and bottom wafer separated by precision spacers, the first stage of amplification could take place on one or both wafers. Discrepancy between separations provided by spacers for the same trap is preferably less than 5%, 2%, 1% and most preferably at or no more than 0.5% or 0.2% of the ion beam size (for instance, largest cross-section dimension, which in this example is about 40% of the gap between wafers, h). If the beam size is unknown, this range may correspond to roughly at least 0.1% and no more than 2% of the gap between wafers (h). For trap sizes considered in this disclosure, this may correspond to a parallelism of 0.1 pm to 1 pm.

In one advantageous implementation, opposing detector plates can be connected by conductive precision spacers. These may be structured by MEMS processes, for example in proximity to the sensitive areas. In this case, one readout circuit may be used for a single detector cell. Where non-conductive spacers are used, two separate read out circuits may be required for the upper and lower substrate of a single detector cell. The signals can be added electrically after amplification in this case.

High resolution and mass accuracy can be provided by high-order time-of-flight focusing within the electrostatic trap, combined with a nanometer-scale accuracy of manufacturing on a planar wafer. Referring to Figure 4, there is illustrated an example wafer having electrodes formed thereupon at different levels of magnification, showing an edge roughness of significantly less than 10 nm. This may be achieved more easily for very thin metallization, for instance of around 75 nm as shown in Figure 4. A high fidelity of edge definition by lithographic techniques may determine the ion-optical quality of the pEST, the ultimate mass accuracy and mass resolution. Generally, a ratio of the gap size between wafers to an accuracy of edges of gaps between electrodes (specifically between reflecting electrode R and accelerating electrode A and/or between accelerating electrode A and detection electrode D) should be at least 1000, preferably at least 2000, 5000 or 7000 and most preferably at least or greater than 10000, 20000 or 50000. It is also desirable that both wafers are well aligned, particularly within at most 2% of the ion beam size and preferably within at most 0.5% of the ion beam diameter (which in the example of the disclosure is about 40% of the gap between wafers, h). For trap sizes considered in this disclosure, this corresponds to an alignment accuracy of 0.1 pm to 1 pm.

Mass accuracy in the range of several parts per million may allow narrowing down the list of candidate compounds (for instance, peptides in bottom-up proteomics) from existing databases.

Returning to the general sense of the disclosure discussed above, further optional features may be considered. For example, one or both edges of at least some of the electrodes (on a side of the electrode extending perpendicular to the longitudinal axis, that is in the width dimension) may have an arc shape (and optionally, both edges may have arc shapes, for instance defined by the same circle, or the edges may have arc shapes defined by different circles). Additionally or alternatively, at least some of the electrodes may have a curved shape, an arc shape (for example, this may be applied to reflecting electrodes) or a circular or elliptical shape (for example, applied to detecting electrodes, which may be based on two circular arcs). This may apply to electrodes only on one side of a center of the electrostatic trap along the longitudinal axis (z=0 line). A central electrode may be provided between the central detection electrodes. The central electrode may be configured to act as an ion lens. Preferably, the central electrode has at least one curved edge (in the width dimension).

In embodiments, a transistor (for instance, a FET or JFET) is connected to the detection electrode (or electrodes). Advantageously, the transistor may be formed by lithographic techniques and/or form part of the integrated circuit. Preferably, the transistor is wire bonded onto or integrated with the substrate, for instance by a CMOS manufacturing process. A capacitance of the transistor is preferably comparable (for example, within one order of magnitude) of a capacitance between detection electrodes and/or the capacitance between a detection electrode and ground.

Where one or more spacers are provided between the sets of electrodes, at least one of the one or more spacers may be conductive, for coupling of at least one electrode on the first set of electrodes with at least one electrode on the second set of electrodes (for example, detection electrodes of each set). A parallelism of 0.1 pm to 1 pm for the spacers is preferred.

In some implementations, ions may be selected by application of a pulsing voltage to one of the reflecting electrodes, thereby causing at least some ions (for instance, unwanted ions) to leave the trap.

Beneficially, a ratio of the gap size between planar substrates to an accuracy of edges of a gap between adjacent electrodes of the first set and/or the second set is at least 1000, preferably at least 2000, 5000 or 7000 and most preferably at least or greater than 10000, 20000 or 50000. Additionally or alternatively, alignment of the planar substrates may be within an accuracy of 0.1 pm to 1 pm.

Further specific details will again be discussed, before returning to the general sense of the disclosure.

Fragmentation of ions in a pEST is possible by irradiation techniques, which may be adapted to the planar geometry of the pEST. Referring next to Figure 5, there is schematically depicted a pEST in three consecutive steps of operation: a) precursor analysis; b) selection and fragmentation; and c) fragment analysis. This represents a simple form of multi-stage analysis. In step a), the ion trajectory 45 is a normal confinement pattern, as described above with reference to Figure 1 A. This allows analysis of precursor ions. In step b), the ion trajectory 46 is affected by irradiation by a focused Ultraviolet (UV) or Infrared (IR) laser pulse until the ion fragments form (as will be discussed below). The resulting fragments resume a confinement trajectory 48 in the pEST and are re-analyzed in step c). Repetition of steps b) and c) enables MS ^r operation.

UV photodissociation is frequently used in conventional mass spectrometry (see for example WO-2007/109672). Advantageously, it could be used to provide additional structural information about ions within pESTs. Ions of interest could optionally be selected in step b) by pulsing down a voltage applied to an upper reflecting electrode of the pEST to allow all unwanted ions to leave the trap (shown by the pulse voltage added to Us in Figure 2A). When an ion of interest reaches a turning point 47, it is subjected to a focused pulse of UV light (for example, from a 213 nm Nd:YAG or a 193 nm ArF laser). A timing of the UV pulse could be matched to the trajectory of the target ion for better selectivity. Alternatively, the timing could be unsynchronized, but multiple pulses could then be used to increase the irradiation probability of the ion of interest (preferably to 100%). A UV laser pulse of a few nanoseconds in duration can readily produce fragmentation of most biologically relevant organic molecules (for instance, see J.S. Brodbelt, et al., Chem. Rev. 120, 3328 (2020)). IR can be used instead of UV. As noted above, the process of selection and fragmentation could be repeated several times (MS ⁿ approach), for example depending on probability of formation and capture of charged fragments. Fragmentation in the turning point may offer the advantage of minimizing initial energy of fragment ions, therefore reducing energy spread and improving analysis accuracy.

Again considering the general sense of the disclosure, further advantageous features may be considered. For example, the electrostatic ion trap may further comprise a UV or IR laser, configured to emit a pulse at a trapping region of the electrostatic ion trap, to fragment ions confined in the trapping region thereby. Beneficially, the laser may be configured to emit the pulse in a direction orthogonal to longitudinal axis of the electrostatic ion trap. In some embodiments, the laser is configured to emit a single laser pulse with timing matched to a trajectory of a target ion. Alternatively, the laser may be configured to emit multiple laser pulses unsynchronized to a trajectory of a target ion.

The laser may be controlled for fragmentation of ions in the trapping region. Then the controller or processor may receive a signal from the detection electrodes for detection of ions fragmented in the trapping region. Preferably, detection and/or selection of ions is additionally performed before fragmentation.

Another embodiment of fragmentation could include the same UV laser used to illuminate the surface of electrodes 22, to emit electrons via photoeffect. This may allow other methods of fragmentation of general electron dissociation (ExD) type, for example electron-capture dissociation (ECD), electron-induced dissociation (EID), etc. A weak magnetic field from external permanent magnets could be added to increase residence time of electrons in the trap and hence efficiency of fragmentation.

In preferred implementations, the electrostatic ion trap may be controlled to effect repetition of fragmentation and detection. MS ⁿ operation may be provided in the electrostatic ion trap thereby. Optionally, fragmentation may be controlled in response to information determined from a previous detection. Data-dependent decision ion fragmentation may be effected in this way.

Additional details of specific embodiments will again be discussed, before further discussion of the general sense of the disclosure. The design of pEST at a small scale and particularly on a substrate, for instance a wafer, allows placement of a respective set of electrodes for each of a plurality of traps on a single wafer. Using two such wafers, a two-dimensional array of traps may be created. Using three or more wafers stack in multiple levels may thereby provide multiple such two-dimensional arrays and permit a three-dimensional array of pESTs. In this way, a large number of pESTs may be provided within a small volume, thereby making massively parallel operation possible.

Referring next to Figure 6A, there is shown a stack of wafers comprising a plurality of pESTs. Each of the wafers may have electrodes formed on both top and bottom sides, along the lines discussed above. The stack 80 includes a first wafer 81 and an adjacent second wafer 82, separated by a height of 50 pm.

Reference is now made to Figure 6B, depicting a magnified portion of Figure 6A to demonstrate ion fragmentation by irradiation according to an implementation. Multiple ion confinement areas 100 are illustrated, each of which is defined by a respective pEST comprising individual electrodes formed on the wafers. Ions enter each of the pESTs along axes 110 from the edge of wafers (which are also longitudinal axes of each pEST). Thus, N pESTs may be provided in a single row. A laser pulse 120 is directed along the plane of the ion confinement in a direction perpendicular to the axes 110 of ion entry. In this way, a single laser pulse 120 can be used to fragment ions in multiple pESTs simultaneously.

For an array of N traps in a row with M rows above each other, a single laser pulse could be split in M sub-rays, each of them focused by an array of miniature lenses with anti-reflective coating into collimated beam. With N=10 to 20, traps span over 5 to 10 mm and the diameter of the beam could be kept within the required 20 to 30 urn along all this length and without touching electrodes. Angular divergence should be within around ±1 mrad to achieve this. With a fragmentation threshold for peptides of 1 mJ/mm ² , less than 1 pj of pulse energy per row may be required, although with tight collimation.

Reference is next made to Figure 7, in which there is shown a schematic layout of a wafer for a single row of traps within an array. Eight traps are shown, comprising: a first trap 131 ; a second trap 132; a third trap 133; a fourth trap 134; a fifth trap 135; a sixth trap 136; a seventh trap 137; and an eighth trap 138. Each of these traps is spaced apart from each other, preferably with at least (or greater than) 6h distance between their longitudinal axes, most preferably at least (or greater than) 10h distance, where h represents the spacing between wafers (in other words, the height of the trap). As fringing fields exponentially reduce when moving from the trap longitudinal axis, anything at least about 3h away from the ions on the z-axis is unlikely to affect the field. For these reasons, for example transistors for image current detection, wires or similar can be so placed.

Also shown in Figure 7 are wires for connection to the traps. In particular, DC voltage wires 141 , 142, 143, 144, 145 and 146 provide DC voltages to and between the traps (smaller, similar connection wires are provided between adjacent traps, but these may not clearly be visible from the drawing). Pulsed signal wires 151 , 152, 153, 154, 155 and 156, provide pulsed signals to the traps (similar wires to provide pulsed signals to other traps are present but not highlighted). Output signal lines 160 (connected to the drain and source of the detection JFETs) are also indicated.

Following standard mass spectrometry notation (for instance, see P. Roepstorff, J.

Fohlman, Biomed. Mass Spectrom. 11 (11 ), 601 (1984)), the majority of UVPD fragments for tryptic peptides are a- and y- ions, followed by x- and z- species. Low energy charged fragments formed during this pulse are typically accelerated back to the pEST from the turning points, while uncharged fragments and unselected precursors are lost from analysis. Following capture of charged fragments in the pEST and their accurate mass measurement, vital structural information (sequence tags) about the ion of interest could be deduced to pinpoint the correct compound (for example, peptide) against the candidates from a library that may be short-listed after filtering on precursor mass.

For MS/MS and MS ⁿ approaches using pESTs, modification of the existing tools used in conventional mass spectrometry may be desirable. First, increased ambiguity may be caused by the presence of only one isotope in the spectrum, resulting in an unknown isotope abundance (for instance, with ¹³ C or ¹⁵ N). With multiple traps and higher throughput, the identification process may greatly benefit from applying statistical and/or machine learning methods. The use of chromogenic tags appended to peptides can also be considered to increase the identification rate. TMT and TMTPro reporter ions could be used for analyzing pooled samples.

The output of each pEST in an array may be individually digitized and processed. This permits mass/intensity pairs to be extracted. Intensities can be converted into an integer number of charges (charge state) of the ion. This allows determination of the mass from m/z measurement with low ppm accuracy. Data from MS/MS and/or MS ⁿ can be linked to the single stage (MS) data of the precursor. This data may then be used for a search (for example in a database).

It is likely that only one isotopologue will be detected per trap. Moreover, the average number of fragments will typically be lower than in conventional (macro-scale) mass spectrometry. Hence, the search space for any output data will generally be wider than in conventional proteomic analysis. In contrast, the ability to link fragments directly to precursors (since the precursor will be a single ion) may reduce the search space compared with macro-scale proteomic analysis. Data from other traps could be also taken into account, for example using machine learning to identify precursors better.

The pEST may even be useful without fragmentation. For example, this may be useful if a particular sample can have different charge states for the same mass-to-charge ratio, m/z, for instance in the case of analysis of intact proteins and top-down proteomics. In a general sense of the disclosure (which may be combined with, or a part of the general senses discussed above), there may be considered an array of electrostatic ion traps, each ion trap having a longitudinal length of no more than 10 mm and/or with a at least one of electrode (preferably, one, some or all detection electrodes) having a low capacitance to ground, generally no more than 1 pF (and optionally lower, as identified above). The longitudinal length may optionally be no more than 5 mm, 2 mm, 1 mm or 0.5 mm. Advantageously, the array comprises an inlet for receiving an ion beam, configured such that a portion of the ion beam can be trapped in each of the ion traps. A controller and/or processor may be provided to control operation of the array and/or to receive one or more outputs from the array. Also, a corresponding method of manufacturing and/or operating such an array of electrostatic ion trap may also be considered, having steps of forming and/or providing and/or using the features of this device.

In an aspect, there may be considered a method of analyzing an ion beam comprising a mixture of different analyte ions. The method comprises: directing ions from the ion beam to multiple electrostatic ion traps of an array of electrostatic ion traps (for example, the array as discussed above and elsewhere herein, optionally comprising any features of the electrostatic ion traps herein described), each electrostatic ion trap having a longitudinal length of no more than 10 mm; obtaining signals indicative of mass and charge data in respect of the ions from said multiple electrostatic ion traps; and combining the mass and charge data obtained from the signals for identification of components of the mixture. The ion beam may be generated from a mixture of analytes.

The signals are advantageously obtained by analyzing the ions and/or derivatives of the ions. For example, the ions or their derivatives in each trap may be considered precursor ions. Then, the precursor ions may be fragmented in the multiple traps to obtain fragment ions. The obtaining signals may include obtaining signals by analyzing the precursor ions in the multiple traps and/or obtaining signals by analyzing the fragment ions in the multiple traps. Then, the combining the mass and charge data may combine mass and charge data in respect of the precursor ions and/or mass and charge data in respect of the fragment ions. Beneficially, the steps of fragmenting, obtaining signals and combining the mass and charge data are repeated in sequence (to effect MS ⁿ operation).

Obtaining signals may comprise simultaneously analyzing ions in each of the multiple traps. This may include simultaneous analysis of one set of ions in the multiple traps and/or simultaneous analysis of fragment ions in the multiple traps. Simultaneous fragmentation of ions in the multiple traps is also possible.

In line with this method aspect (and any other method aspect provided herein), a computer program may be provided, comprising instructions that, when executed by a processor, cause the processor to perform the corresponding method. For instance, the processor may be configured to control an array of electrostatic ion traps (or a mass spectrometer comprising such an array). A (non-transitory) computer readable medium comprising or storing thereon such a computer program may also be provided.

Various options will now be considered according to any of the above aspects and/or implementations. For example, each electrostatic ion trap may be (and preferably is) in accordance with any of those described above. For example, the electrostatic ion trap may comprise: a first set of planar electrodes distributed along a longitudinal axis of the respective ion trap; and a second set of planar electrodes distributed along the longitudinal axis of the respective ion trap, each of the electrodes of the second set arranged to be spaced apart from and oppose a corresponding electrode of the first set. A length of the first and second sets of planar electrodes along the longitudinal axis is no more than 10 mm and/or at least one electrode (preferably, a detection electrode) may have a capacitance to ground of no more than 1 pF (and optionally lower, as identified herein). At least some of the electrodes of the first and second sets may be configured to receive an electrostatic potential for confinement of ions received in the space between the first and second sets of planar electrodes.

In preferred embodiments, the array is configured with multiple electrostatic ion traps in the same plane. Additionally or alternatively multiple electrostatic ion traps may be provided in distinct planes. Where multiple electrostatic ion traps are provided in the same plane, a single laser pulse may be used to fragment ions in multiple electrostatic ion traps (specifically, their respective trapping regions) in that plane. A single laser pulse may also be used to fragment ions confined in different planes. For example, a UV or IR laser may be provided, configured to generate a pulse. Then, a splitter arrangement may be configured to split the generated pulse spatially into parts and to direct each part along a respective, different plane. A lens array comprising a plurality of (miniature) lenses may be used. Each lens is beneficially configured to focus a respective part of the generated pulse into a respective collimated beam. Anti-reflective coating may be applied to the lenses.

Ions may be received beam at multiple electrostatic ion traps of the array from a single ion. The received ions at the multiple electrostatic ion traps may be processed in parallel (particularly, where the ion beam comprises ions of a peptide).

Detection outputs from each of the multiple electrostatic ion traps may be provided to a processor. The processor may then be configured to apply statistical and/or machine-learning methods to the detection outputs for identification of a composition of the single ion beam.

Ions may be injected to multiple electrostatic ion traps of the array from the same calibration mixture, for calibration of the multiple electrostatic ion traps.

An ion beam may be transmitted to at least one electrostatic ion trap of the array, the ion beam including at least one ion with a chromogenic tag or a tandem mass tag.

Additional details of specific embodiments will again be discussed, before further discussion of the general senses of the disclosure. Referring next to Figure 8, there is illustrated a schematic diagram of an experimental apparatus used for testing a pEST array. This comprises: an existing macroscale mass spectrometer 200 (marketed under the name of Orbitrap™ Exploris™ 480 instrument by Thermo Fisher Scientific, Inc.) as a front-end instrument; an adaptor chamber 300; and a pEST array 400. The macroscale mass spectrometer 200 comprises: a high capacity transfer tube 210; an electrodynamic ion funnel 220; an internal calibrant source 230; an active beam guide 240; a quadrupole mass filter 250; an ion gate 260; a curved ion trap (C-trap) 270; an orbital trapping mass analyzer 280; an ion routing multipole 290; and a transport multipole 295.

The adaptor chamber 300 comprises: a gate valve 310; an ion funnel 320 with a gas line 325; a Z-lens (line-of-sight blocker) 330; a X, Y deflector 340; and lenses 350.

Prior to analytical runs, each pEST may be calibrated using the same calibration mixture as the front-end instrument. In operation, the internal calibrant of the mass spectrometer from source 230 may be used to monitor the mass accuracy of each trap within the array and, if needed, employed as a lock mass.

In operation, each pEST is filled sequentially (although parallel fill is possible), for example in a raster. This may be achieved by pulsing of the reflecting electrode to receive an ion of a particular m/z ratio (as discussed above with reference to Figure 2, with the pulse added to the Ue potential). Each pEST can thus be used to detect a single charge (due to its small size).

Single ion sensitivity even at high repetition rate allows the pESTs to measure even very low abundance species within a mixture. This, in turn, may enable deeper proteome coverage or similar. On the other hand, the small size of the pEST, which makes such ultra-sensitive detection possible, may also limit the number of ions analyzed to a maximum of few tens per injection.

Alternative designs for planar electrostatic ion traps based on modulated, or wavy, interfaces between adjacent electrodes are also now considered. These are applicable to and/or can be combined with any other design or implementation disclosed herein. A wavy interface between a pair of electrodes held at distinct potentials may have the effect of smoothing the transition between the potential regions, compared with the straight electrode case, and can offer improved control over the electrostatic field confining ions. This design may also allow ion traps to be tuned for greater performance and can reduce the required number of electrodes at distinct potentials. Additionally, these designs may help to reduce the sensitivity of trap performance to the misalignment of the electrode plates. This may be of particular concern in the assembly and operation of pESTs.

Referring next to Figure 9, there are schematically depicted two planar electrostatic trap electrode designs, with (a) showing a side view applicable to both designs, (b) showing a top view of a first design with straight (or singly curved) electrode interfaces and (c) showing a top view of a second design with wavy electrode interfaces. The design in Figure 9(b) is a planar electrostatic trap design based on straight electrode interfaces, as can be seen for: a first set of electrodes 301 ; a central electrode 302; and a second set of electrodes 303. As discussed above, the electrodes are held at carefully chosen potentials to create a trapping field, while the pickup electrodes measure the image charge induced by the ion. The axis of the trap is indicated by the dashed line along the z-axis.

As can be seen from Figure 9(a), a planar electrostatic ion trap comprises two opposing plates of electrodes. As discussed above, each electrode is held at a fixed potential and the combined electrostatic field of the electrodes confines the ion. In a multi-reflecting design, ions are inserted along the axis of trap by lowering and raising a gate electrode. Ions are then reflected between a pair of electrostatic mirrors at each end of the trap, confining the ion for multiple cycles and enabling precise measurement of the oscillation frequency.

The accuracy of the mass to charge ratio prediction may be limited by the variation of the ion oscillation frequency (or equivalently time-of-flight) as a function of ion energy, insertion angle and position. To improve (or potentially) maximize the accuracy of the mass to charge ratio prediction therefore, the geometry and potential of each electrode is optimized to minimize variation of the time-of-flight due to uncontrolled experimental variables.

Conventional designs for planar electrostatic ion traps are based on sets of electrodes with uniform boundaries along the traverse axis of the trap (which is the x-axis in Figure 9). The potentials and lengths of the electrodes can then be carefully chosen to minimize the variation of the time-of-flight for trajectories that are closely aligned with the axis of trap. This optimization can be performed by taking a Taylor expansion for the time-of-flight about the principal trajectory and minimizing the Taylor series coefficients (often called aberration coefficients) with respect to ion energy, angle and position. This process may be continued to the highest order possible until the degrees of freedom offered by the electrodes are insufficient to proceed further. More electrodes can be added to obtain additional degrees of freedom in this optimization process, however, minimum electrode size restrictions and the number of voltage supplies required can quickly become an issue.

Referring to Figure 9(c), a planar electrostatic trap utilizing wavy electrode interfaces and strip electrodes is shown. As can be seen, the first set of electrodes 311 comprises a high potential electrode 311 a and a negative potential electrode 311 b with a wavy interface between. Similarly, the central electrode 312 comprises a high potential electrode 312a and a negative potential electrode 312b with a wavy interface between and the second set of electrodes 313 comprises a high potential electrode 313a and a negative potential electrode 313b with a wavy interface between. Grounded pickup electrodes 315 are also shown. This design thus utilizes electrodes held at just three distinct potentials high potential, negative potential, and grounded.

In this approach, planar electrostatic trap designs are extended by curving the interfaces between adjacent electrodes. Specifically, wavy interfaces are introduced between adjacent electrodes. In so doing, the transition between potential regions can effectively be smoothed on the boundary. This is possible since the electrostatic potential in the trap is a convolution, or blurring, of the potential on the boundary. Using electrode interfaces of arbitrary geometry may therefore provide a greater degree of control over the electrostatic trapping field.

Designs utilizing these, so called, wavy electrodes may have several advantages over designs using straight electrode geometries. Firstly, as each electrode provides greater control over the electrostatic field, fewer electrodes may be required to achieve the same level of trap performance. This may be particularly desirable in the design and construction of large arrays of pESTs, since fewer connections are required to external voltage supplies.

Planar electrostatic ion traps may also be sensitive to misalignment of the electrode plates, particularly for relative displacements along the axis of the trap. This sensitivity may be driven by the sharp changes in the electrostatic field along the axis. Consequently, when the electrode plates are not precisely aligned, an unbalanced electric field may be created, which pushes ions out of the plane of the trap (which is the xz-plane in Figure 9). Wavy electrode interfaces may smooth the sharp transitions in the electrostatic field and reduce the magnitude of the out-of-plane electric field (which is the y-axis in Figure 9), due to this misalignment. The high tolerances for optimal ion trap performance may make the construction and alignment of the electrode plates for pEST challenging. Thus, wavy electrode interfaces may be ideally suited for this application.

The electrostatic fields produced by wavy and straight electrode interfaces can be compared, as will now be discussed. It can be seen that small-scale interface features are rapidly suppressed with increasing distance from the boundary. Moreover, the gradient of the electrostatic potential within the trap may be precisely controlled by varying the electrode interface. It can also be established that wavy electrode interfaces produce smaller out-of-plane electric fields (which is the y-axis in Figure 9) when the electrode plates are misaligned.

The electrostatic field within the trap is dictated by the potential along the boundaries of the domain. If a designer were able to continuously vary the potential over this boundary, an arbitrary electrostatic field could be induced within the domain of the trap. In essence, this unrealistic level of control over the boundary potential could recreate the electrostatic fields induced by three- dimensional sets of electrodes (for example quadrupole guides). As planar electrostatic trap designs are limited to a discrete set of electrodes at fixed potentials, this level of ultimate control cannot be realized.

A well-known and interesting property of electrostatic fields (that is, harmonic functions) is that each point within the domain is an average of the behavior of the field along the boundary. In formal terms, the field at each point within the planar electrostatic trap can be expressed as a convolution over the boundary potential. This convolution process acts as a blurring operation, so that the potential in the midplane of an electrostatic trap is a blurring of the boundary potential. For example, consider a square-wave variation in in the boundary potential of a two- dimensional planar electrostatic trap with wavelength, , and duty-cycle, 0 < a < 1. The peak-to- peak variation of the n-th harmonic decays as, where n > 1 and Ay is displacement from the electrode boundary. This shows that higher spatial frequencies of the boundary potential decay exponentially for increasing displacement from the boundary. Notably, for three-wavelengths into the domain, the contribution from each spatial frequency of the boundary decays by roughly 1/1000. Referring now to Figure 10, there is plotted the electrostatic potential along the boundary created by a sinusoidally varying electrode interface in an example, for four different distances from the boundary. The electrostatic potential is created by a sinusoidally varying electrode interface with wavelength A = 12 im and transition length L _interface = 120|im for increasing distance from the boundary. The trap height is taken to be 60 \im.

Figure 10 thus illustrates this blurring effect, which shows the electrostatic potential for increasing distance from the boundary electrode. It demonstrates that sharp variations of the boundary potential are almost completely gone after a displacement on the order of a single wavelength.

Referring next to Figure 11 , there are plotted variations of the electrostatic potential created by the sinusoidal electrode interface shown in Figure 10 for increasing distance from the electrode. Figure 11 (a) plots standard deviation of the potential in the transverse x-direction along the longitudinal z-axis of the trap. The “ideal” of zero variation is also shown (thick-solid black). Figure 11 (b) plots the same data on logarithmic-linear scale. It can be seen that the standard deviation rapidly approaches the limit of numerical simulations ~10 ^-5 with increasing distance from the electrode. This standard deviation of the electrostatic field, that is, deviation from an “ideal” (constant) potential, is thus plotted as a function of distance from the electrode. By exploiting this blurring property of electrostatic fields, the limit of ultimate control can be approached over the boundary potential described above, using only a discrete set of electrodes.

Reference is now made to Figure 12, in which there are shown different curved electrode interfaces, specifically: (a) a sinusoidal variation of inter-electrode boundary; (b) a square-wave interface with curvature of boundary midline; and (c) an elongated strip-electrode design with modulated duty-cycle along the longitudinal axis of the trap. Figures 12(a) and 12(b) thus show an additional level of control over the electrostatic field offered by wavy electrode interfaces. This is advantageous in the design of electrostatic traps which exhibit minimal variation in the time-of- flight of ions with respect to energy, insertion angle and position. Referring next to Figure 13, there is plotted electrostatic field induced by a straight and sinusoidally varying electrodes interfaces along the boundary for four different (increasing) interface lengths, L _in t _er f _ace , f° ^r comparison. The trap height is taken to be 60 im. This demonstrates enhanced control, showing that the gradient of the electrostatic potential can be adjusted using the length of the electrode interface.

Reference is then made to Figure 14A, in which there is plotted out-of-plane electric field, E _y , along the midline of a planar electrostatic trap (x = y = 0) for an axially misaligned trap (Ad _axia | = 1.0 (im) with sinusoidally varying electrode interfaces, for four different (increasing) interface lengths, L _int erface- The trap height is taken to be 60 nm. Reference is also made to Figure 14B, in which there are plotted out-of-plane electric field along the midline of a planar electrostatic trap for an axially misaligned trap with square-wave varying electrode interfaces, for four different interface lengths, for comparison with Figure 14A. Thus, the same parameters are otherwise used in both Figures 14A and 14B.

Now referring to Figure 15, there is plotted a boundary potential due to a square-wave type electrode interface. A further benefit of wavy electrode interfaces is the reduction of the magnitude of the out-of-plane electric fields (which is the y-axis in Figure 9) induced by a misalignment of the electrode plates along the longitudinal direction of the trap. This is shown in Figures 14A, 14B and 15.

The concept of wavy electrode interfaces can be extended to colinear strips of electrodes held at distinct potentials. This is specifically shown in Figure 12(c), for example. By varying the duty-cycle of the strips along the traverse direction of the trap, the effect of varying between intermediate potentials along the boundary can be established.

Referring next to Figure 16, there are plotted electrostatic potential arising from a strip electrode design for four different displacements from the electrode boundary, Ay _boundary . The wavelength of the strip design is X = 15 .m and the duty-cycle is periodically varied between 0.1 < a < 0.9. The trap height is taken to be 60 im. This drawing shows the boundary potential due to a set of strip electrodes and the electrostatic potential for increasing displacement from the boundary, illustrating that the features of individual strips are exponentially suppressed for increasing displacement into the trap.

Reference is now made to Figure 17A, in which there are plotted electrostatic potential of strip electrode design at (a) boundary and (b) midline for an example with linearly increasing dutycycle. Specifically, the duty-cycle is increased from a = 0.1 -» 0.85 over 180 [im. The wavelength of the strip-design in (b) is A = 15 im. The potential is induced in the midplane of the trap. The trap height is 60 m.

Reference is further made to Figure 17B, in which there are plotted electrostatic potential of strip electrode design at (a) boundary and (b) midline for an example with linearly increasing duty- cycle, where a circular arc is engineered into the induced midplane potential by adjusting displacing the duty-cycle variation of each strip. Specifically, the duty-cycle is increased from a = 0.1 -» 0.85 over 180 [im. The wavelength of the strip-design in (b) is A = 15 im. The potential is induced in the midplane of the trap. The trap height is 60 .m.

Referring also to Figure 18, there is plotted electrostatic potential along the midline of the trap due to the strip electrode design of Figure 12(c). Figures 17A, 17B and 18 demonstrate that electrostatic lenses of arbitrary shape and potential gradient can be fabricated using this principle.

Next referring to Figure 19, there are depicted, for an electrostatic trap design based on electrodes held at just two distinct potentials: (a) boundary potentials of the electrodes with positive voltage electrodes, grounded electrodes and negative voltage electrodes; (b) a potential induced in the midplane of the trap, (c) a potential along the midline of the trap. This corresponds with the design shown in Figure 9(c).

This strip-electrode design principle may dramatically reduce the number of distinct electrode potentials required for the fabrication and operation of high-performance electrostatic ion traps. Figures 9(c) and 19 thereby outline an electrostatic trap design requiring only two nongrounded electrode potentials. This design may recreate the electrostatic field induced by the straight electrode array of Figure 9(b), which uses electrodes held at six distinct non-grounded potentials. Reducing the required number of distinct potentials may present a significant advantage in the construction of high-performance microscale electrostatic ion traps, due to strict constraints in the fabrication and operation of these devices.

The design outlined in Figures 9(c) and 19 is an example of the strip electrode technique only and should not be understood to be the only possible formulation. Moreover, the strip electrode approach can easily be extended to a design using only a single non-grounded electrode potential.

A further benefit of strip electrodes may be the enhanced control over the electrostatic potential within the trap. Consequently, they may allow for the design of electrostatic traps with smaller time-of-flight variations for ions with different energy, insertion angle and position. As an example of this design process, suppose that a particular potential is desired along the midline of the trap (Figure 9, x = 0,y = 0) denoted by v|/(z). As the electrostatic field within the trap is determined by the boundary potential, the desired potential can be expressed as a convolution, i (z) = (0,0, z) = J dx' dz' K(x - x',y - y',z - z') f(x' ,z')\ _{x=0 y=0} , where K x.y.z) is the kernel and f(x,z~) is the, as yet, undetermined boundary potential. Along the transverse -axis, each strip electrode may be determined by the duty-cycle at that longitudinal position, a(z). This may allow the boundary potential to be expressed in terms of the unknown duty cycle as where S(x; a) is a square-wave of unit period and duty-cycle a. Substituting this equation into the preceding equation, a nonlinear Fredholm integral problem of the first kind may be established for the unknown duty cycle as where the nonlinear kernel is given by,

T(z, a) = J dx' K(x', y, z) S(x'/ , a)| _y = ₀ .

Fast Fourier transform techniques may allow for rapid evaluation of the convolution forms given in these latter two expressions. This may allow a solution to be developed for the duty-cycle by discretizing and using standard regularization methods for Fredholm integral problems of the first kind. An example is the regularized non-linear least squares formulation, a = argmin _aeB \\rp(z) - T[a] (z) || ² + p|| a" (z) || ² , where the minimization is taken over a finite-dimensional basis set B and g is the regularization parameter.

Returning to the general senses of the disclosure discussed above, further details may be considered. For example, a boundary between at least some of the electrodes adjacent to one another may have a wavy, modulated or multiply-curved shape. In some embodiments, at least some of the electrodes are strip electrodes. Then, a duty-cycle of the strip electrodes may vary along the longitudinal axis of the ion trap. Optionally, the strip electrodes may be configured to define a lens. In this case, the strip electrodes may be configured to be held at different electrostatic potentials. In embodiments, the electrostatic potential received by the at least some of the electrodes of the first and second sets for confinement of ions may be a single, non-zero electrostatic potential. Alternatively, the electrostatic potential received by the at least some of the electrodes of the first and second sets for confinement of ions may be a discrete number of different, non-zero electrostatic potentials, typically 2 or 3 different, non-zero electrostatic potentials.

An alternative implementation of pEST will now be considered (which can be combined with any other implementation disclosed herein). As noted above, a pEST is advantageously fabricated using micro- lithographic and/or nano-lithographic technologies, to achieve tolerances in nanometer range on planar wafers. The detection electrodes beneficially have low capacitance (below 1pF) to allow single-charge detection. Each electrode set is formed on a planar substrate and two substrates oppose each other to form the trap or a linear arrangement of traps.

As noted previously, the structures on each side are separated (in a z-direction) by a spacer, which may typically have a 60pm dimension. The structures are aligned in planar direction (perpendicular to the direction of separation, that is x-direction and y-direction) with a precision better than 0.5pm and a parallelism on the system level that is better than 0.5pm. Achieving this accuracy is challenging. High accuracy positioning tools can be used, but these tools typically require special expertise and the time-consuming alignment process is done at chiplevel. The approach described below makes use of a self-alignment method using high precision silicon anisotropic etching in combination with a dielectric substrate.

Silicon can be etched with high precision using Potassium Hydroxide (KOH) or Tetramethylammonium Hydroxide (TMAH) solutions. These anisotropic processes use the monocrystalline property of silicon. The (111) plain in this monocrystal is a natural etch stop in that process. In standard silicon (100) wafers the (111) plain is oriented under 54.735° from the surface. When a rectangular shape is unmasked and etched by the such solutions, an inverse pyramid or V-groove is created. The angles of these V-grooves have atomic precision. The position of the groove however is limited by the lithography steps. The basics of these etching processes are described in “KOH anisotropic silicon etching for MEMS accelerometer fabrication”, Petteri Kilpinen, Doctoral thesis Department of Materials Science and Engineering, Aalto University (2012).

This inverse pyramid appears to be a perfect pitch for a precision ball. Such balls are available in many materials such as Sapphire, Aluminum Oxide, or Stainless Steel at diameters of greater than 0.2mm and variation from the perfect shape of less than 0.08pm and lot diameter variations of less than 0.13pm (see “Prazisionskugeln - mit Maximalnote”, Datasheet, Saphirwerk).

Referring next to Figure 20, there is schematically shown an approach for aligning two substrates of a pEST based on a ball positioned in grooves (particularly, V-grooves). A first groove or pitch 401 is provided in a first substrate and an opposing second groove or pitch 402 is provided in a second substrate. A precision ball 403 positioned between the first pitch 401 and the second pitch 402 forms a precise alignment system. When force is applied, orthogonal to the substrate plane both wafers are precisely aligned to each other in all dimensions. This approach is similar to that disclosed in US-5,639,387, but which is used for alignment in optical applications.

Various techniques are available to fix the substrates permanently in this position. Suitable bonding techniques may include, for instance, gluing, soldering or thermal compression bonding. The accuracy is limited by the accuracy of the balls, the etch depth, and the lithography. The gap g between the substrates can be calculated by the following equation. where g represents the gap distance between the substrates, d represents the depth of the groove, r represents the radius of the ball (not shown in Figure 20) and a represents the orientation angle of the (111 ) plane.

With suitable balls, for example Sapphire balls having a radius of 150pm +/-0.125pm and precise groove etching of, for example 229.8pm +/- 0.5pm depth, a gap precision of 60pm +/- 1 .5pm is achievable. When V-grooves are used, the etching rate is reduced by a factor of 400 when the (111) planes fall together. The precision of the depth is mainly defined by the lithography and not by the etching rate in this case. Parallelism is expected to be even better as the etch depth and lithography at chip-level is much better than at wafer-level.

The lateral precision may depend on the precision of the lithography. This may allow a lateral precision of better than 0.5pm using a stepper.

In this approach, the alignment is done at chip-level, but it is also possible to perform the alignment at wafer-level. A minimum of three balls per chip may be used to define the lateral position and to form a stable gap when alignment is done at chip-level. At waver-level, either three balls per chip or a minimum of three balls per wafer can be used.

These fixtures may be applied after alignment to fix the alignment position of wafers. These connections can also be used to form electrical connections between the wafers.

Variations to this general alignment approach are also possible. Anisotropic dry etching techniques can be used to form precise pitches. Fibers can be used instead of balls with similar results. Any object having a spacing dimension of precision to match that of the groove may be suitable. Silicon is preferred for the reasons given above, but alternative may be considered, using appropriate substrate materials for which an etch resulting in grooves of suitable precision is possible.

Although this approach is based on the use of silicon as a substrate, this may cause other issues. A silicon substrate may lead to high parasitic capacitances of the detection pads, which may make single ion detection difficult. That said, other materials may not allow for anisotropic etching techniques. With this in mind, another method may be to deposit on silicon, the surface metallization on a dielectric intermediate layer, for example SiO2 or SiN4, with a thickness of up to several pm. In other words, a dielectric layer interposes between a silicon substrate and a surface metallization. When the silicon is etched, the dielectric layer may remain in place. This may reduce parasitic capacitances, whilst allowing alignment by silicon etching, as described above. Thus, this approach may be combined with the alignment technique based on silicon etching.

Referring now to Figure 21 , there is illustrated a schematic structure for an alternative approach for alignment, in which the surface metallization is deposited on a dielectric intermediate layer. This effectively shows a cross-section in the xy-plane, showing a silicon substrate 410, from which a portion 415 has been removed (underneath critical structures) from the backside using anisotropic etching. The metal features are aligned 440 from the top silicon substrate to the bottom silicon. The substrate is etched to oxide layer 430 to meet the low capacitance desideratum of the detection areas. Electrical contacts 420 are provided for connectivity purposes.

This approach allows precise alignment together with the desired electrical properties based on a combination of techniques to form an improved electrostatic trap with single ion sensitivity. Although embodiments according to the disclosure have been described with reference to particular types of devices and applications (particularly mass spectrometry) and the embodiments have particular advantages in such case, as discussed herein, approaches according to the disclosure may be applied to other types of device and/or application. In particular, the devices according to the disclosure may be used for other applications. The specific structure, arrangement and operational details (for example, parameters) of the pEST and/or array of pESTs, whilst potentially advantageous (especially in view of known configurations and capabilities), may be varied significantly to arrive at modes of operation with similar or identical performance. The scale of the pEST could also vary widely, for example from nanometer to millimeter range. Certain features may be omitted or substituted, for example as indicated herein. Each feature disclosed in this specification, unless stated otherwise, may be replaced by alternative features serving the same, equivalent or similar purpose. Thus, unless stated otherwise, each feature disclosed is one example only of a generic series of equivalent or similar features.

In this detailed description of the various embodiments, for the purposes of explanation, numerous specific details are set forth to provide a thorough understanding of the embodiments disclosed. One skilled in the art will appreciate, however, that these various embodiments may be practiced with or without these specific details. Furthermore, one skilled in the art can readily appreciate that the specific sequences in which methods are presented and performed are illustrative and it is contemplated that the sequences can be varied and still remain within the scope of the various embodiments disclosed herein.

Although Figure 1 A shows a basic design for a pEST, some modifications are possible. For example, it may not be necessary to have two detection electrodes on each substrate. In one alternative approach, detection electrodes may be provided on only one substrate. The detected signal may be halved as a result of this, but the capacitance is also lowered (which may affect noise pickup), so signal-to-noise may only be reduced by 2. As a further alternative, a single detection electrode could be used. This is less preferred, since two electrodes are better for differential detection and common-mode rejection.

The number and positioning of the accelerating electrodes and/or reflecting electrodes may be varied. It is expected that at least one accelerating electrodes and at least one reflecting electrode is used on each side (in the z-dimension) of each set of electrodes (that is, four accelerating electrodes and four reflecting electrodes). It may be possible to use fewer accelerating electrodes than this in some configurations. With a different design of pEST, it may also be possible to reduce the number of reflecting electrodes.

A wide variety of manufacturing technology can be used, including: glass on silicon, cryogenic silicon, wafer polishing, accurate alignment techniques and similar. Variation of the pEST shown with reference to Figure 2 may also be considered. For example, the number of electrodes and/or the potentials applied may be changed significantly. Curvature of the electrodes in the design of Figure 2 is only on the electrodes in the z<0 part (left side), but in some implementations, the other electrodes (on the z>0 or right side) could be curved. The number of possible variations on this basic design is essentially infinite.

As used herein, including in the claims, unless the context indicates otherwise, singular forms of the terms herein are to be construed as including the plural form and vice versa. For instance, unless the context indicates otherwise, a singular reference herein including in the claims, such as "a" or "an" (such as an ion multipole device) means "one or more" (for instance, one or more ion multipole device). Throughout the description and claims of this disclosure, the words "comprise", "including", "having" and "contain" and variations of the words, for example "comprising" and "comprises" or similar, mean "including but not limited to", and are not intended to (and do not) exclude other components. Also, the use of “or” is inclusive, such that the phrase “A or B” is true when “A” is true, “B is true”, or both “A” and “B” are true.

The use of any and all examples, or exemplary language ("for instance", "such as", "for example" and like language) provided herein, is intended merely to better illustrate the disclosure and does not indicate a limitation on the scope of the disclosure unless otherwise claimed. No language in the specification should be construed as indicating any non-claimed element as essential to the practice of the disclosure.

The terms “first” and “second" may be reversed without changing the scope of the disclosure. That is, an element termed a “first” element may instead be termed a “second” element and an element termed a “second” element may instead be considered a “first” element.

Any steps described in this specification may be performed in any order or simultaneously unless stated or the context requires otherwise. Moreover, where a step is described as being performed after a step, this does not preclude intervening steps being performed.

It is also to be understood that, for any given component or embodiment described herein, any of the possible candidates or alternatives listed for that component may generally be used individually or in combination with one another, unless implicitly or explicitly understood or stated otherwise. It will be understood that any list of such candidates or alternatives is merely illustrative, not limiting, unless implicitly or explicitly understood or stated otherwise.

All literature and similar materials cited in this disclosure, including but not limited to patents, patent applications, articles, books, treaties and internet web pages are expressly incorporated by reference in their entirety for any purpose. Unless otherwise described, all technical and scientific terms used herein have a meaning as is commonly understood by one of ordinary skill in the art to which the various embodiments described herein belongs.