Charge detection mass spectrometry

Abstract

Disclosed herein are various methods and apparatus for performing charge detection mass spectrometry (CDMS). In particular, techniques are disclosed for monitoring a detector signal from a CDMS device to determine how many ions are present in the ion trap (10) of the CDMS device. For example, if no ions are present the measurement can then be terminated early. Similarly, if more than one ion is present, the measurement can be terminated early, or ions can be removed from the trap (10) until only a single ion remains. Techniques are also provided for increasing the probability of there being a single ion in the trap (10). A technique for attenuating an ion beam is also provided.

Claims

1. A method of charge detection mass spectrometry comprising: monitoring a detector signal from a charge detector of a charge detection mass spectrometry device during a first ion trapping event within an ion trap of the charge detection mass spectrometry device to determine how many ions are present within the ion trap during the first ion trapping event; wherein when it is determined that no ions are present within the ion trap during the first ion trapping event, the method comprises terminating the first ion trapping event early and initiating a second ion trapping event; whereas when it is determined that more than one ion is present within the ion trap during the first ion trapping event, the method comprises either: (i) terminating the first ion trapping event early and initiating a second ion trapping event; or (ii) ejecting or otherwise removing one or more of the ions from the ion trap.

2. The method of claim 1, comprising ejecting or otherwise removing all of the ions from the ion trap and initiating a second ion trapping event.

3. The method of claim 1, comprising ejecting or otherwise removing one or more of the ions from the ion trap so that only a single ion remains within the ion trap.

4. The method of claim 1, where the number of ions present within the ion trap of the charge detection mass spectrometry device is determined based on the number of masses recorded in a spectrum by the charge detection mass spectrometry device and/or based on the total charge detected by the charge detection mass spectrometry device.

5. The method of claim 1, wherein the geometry of the ion trap is configured such that ion trajectories become unstable when more than one ion is present resulting in the ejection of all but one ion.

6. The method of claim 1 wherein a substantially quadratic potential is applied to the or each ion trap such that ions undergo substantially harmonic motion within the ion trap.

7. The method of claim 1, wherein the signals from the charge detection mass spectrometry device are processed using forward fitting and/or Bayesian signal processing techniques.

8. A charge detection mass spectrometry device comprising: an ion trap for holding one or more ions to be analysed; one or more charge detector(s) within the ion trap for determining a charge for the one or more ions to be analysed; and control circuitry for monitoring a detector signal from the charge detector(s) during a first ion trapping event to determine how many ions are present within the ion trap during the first ion trapping event; wherein when it is determined that no ions are present within the ion trap during the first ion trapping event, the control circuitry is configured to terminate the first ion trapping event early and initiate a second ion trapping event; whereas when it is determined that more than one ion is present within the ion trap during the first ion trapping event, the control circuitry is configured to either: (i) terminate the first ion trapping event early and initiate a second ion trapping event; or (ii) eject or otherwise remove one or more of the ions from the ion trap.

Description

BRIEF DESCRIPTION OF THE DRAWINGS

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

(2) FIG. 1 shows schematically a single charge detection mass spectrometry (CDMS) device that may be used in embodiments;

(3) FIG. 2 illustrates how the detector signal may vary when more than one ion is present within an ion trap of a CDMS device like that shown in FIG. 1;

(4) FIG. 3 shows how the rate with which good transients are obtained varies as a function of the time after which an unwanted transient can be terminated;

(5) FIGS. 4A and 4B illustrate how an ion beam may be attenuated;

(6) FIG. 5 shows schematically an ion beam attenuation device that may be used in embodiments;

(7) FIG. 6 shows the use of an ion optical device for selectively or sequentially passing respective ions to a plurality of CDMS devices;

(8) FIG. 7 shows an apparatus comprising a plurality of CDMS devices arranged within an ion guide;

(9) FIG. 8 shows an example of a CDMS device having multiple charge detectors within a single ion trap; and

(10) FIGS. 9A, 9B, 9C and 10 illustrate the operation of a SpiroTOF device that may be used according to embodiments as an ion trap for a CDMS device.

DETAILED DESCRIPTION

(11) Various embodiments are directed towards methods of charge detection mass spectrometry (CDMS). It will be understood that CDMS generally involves a simultaneous measurement of both the mass-to-charge ratio (m/z) and the charge (z) of an ion. In this way, the mass (m) of the ion can then be determined (indirectly). The charge of an ion may typically be measured directly using a charge detection electrode. For example, when an ion is caused to pass through (or by) a charge detection electrode, the ion will induce a charge on the charge detection electrode which can then be detected, for example, by suitable detection (signal processing) circuitry connected to the charge detection electrode. The mass-to-charge ratio of the ion can generally be determined in various suitable ways. For example, the mass-to-charge ratio may be determined from the time-of-flight of the ion within the CDMS device or the ion velocity (so long as the energy per charge is known). Thus, various examples of CDMS experiments are known and it will be appreciated the embodiments described herein may generally applied to any suitable CDMS experiment, as desired.

(12) However, typically, the mass-to-charge ratio may be determined from the frequency of oscillation of the ion, for example, within a trapping field. Thus, the CDMS device may generally comprise an ion trap within which ions to be analysed are contained. Ions are thus analysed in discrete ‘ion trapping events’. Thus, in each ion trapping event, the ion trap is opened to allow ions to enter the ion trap for analysis. At the end of an ion trapping event those ions may then be ejected and a new ion trapping event initiated.

(13) For example, in some CDMS experiments such as that described in Keifer et al. “Charge Detection Mass Spectrometry with Almost Perfect Charge Accuracy”, Anal. Chem. 2015, 87, 10330-10337 (DOI: 10.1021/acs.analchem.5b02324), single ions are analysed in an ion trap for periods of up to about three seconds. In the CDMS experiment described by Keifer et al. ions are caused to pass repeatedly through a metal cylinder at the centre of the ion trap which is connected to an amplifier and digitiser. When ions are at the centre of the cylinder, the magnitude of the charge induced on the cylinder is equal to the charge on the ion.

(14) FIG. 1 shows schematically a single CDMS device according to an embodiment. As shown in FIG. 1, the device comprises an electrostatic ion trap in the form of a cone trap 10 formed by a pair of spaced-apart conical electrodes 10A, 10B to which suitable electric fields can be applied in order to confine ions within the cone trap 10. A charge detector 12 is provided within the cone trap 10 comprising a metal cylinder that acts as a charge detecting electrode. The movement of one or more ion(s) through the electrodes of the charge detector 12 generates a signal indicative of the charge of the ion(s). Ions can thus be injected into the cone trap 10, and confined thereby (an ion trapping event), and caused to move between the electrodes of the charge detector 12 in order to perform a CDMS measurement. Once the CDMS measurement has been performed, any ions currently within the cone trap 10 can be ejected and a new ion trapping event initiated (by injecting a new set of ions).

(15) However, other arrangements would of course be possible. Thus, whilst FIG. 1 shows a cone trap 10, it will be appreciated that any other suitable ion trap may be used. Similarly, any suitable arrangement of charge detecting electrode(s) may be used in combination with such ion traps.

(16) In a well-calibrated system, the amplitude of the recorded signal can therefore be used to measure the charge on the ion. However, because the signal to noise ratio is low, many ion passes may typically be required to make an accurate charge measurement. Current state of the art instruments are capable of producing better than unit-charge resolution, for example, so that the charge on almost all of the trapped ions can be determined exactly. The frequency of oscillation of the ion in the trap is related to its mass to charge ratio. Although the signal is typically significantly non-sinusoidal, a Fourier transform of the recorded transient allows a measurement of the mass-to-charge ratio (albeit at low resolution). Taken together, the measurements of the mass-to-charge ratio and charge allow the mass of the ion to be determined.

(17) It will be appreciated that this approach may be particularly useful for producing mass spectra of high molecular weight species (such as in the range of mega Dalton and above) as traditional (for example) electrospray mass spectra can be hard to interpret in this regime as different charge states are often poorly resolved from each other. However, CDMS techniques can be relatively slow. For instance, thousands of ion trapping events may typically be required to build up a useful mass spectrum. Methods of shortening the time required to produce a spectrum are therefore of particular interest.

(18) Various examples of the present disclosure will now be described.

(19) Single Ion Selection

(20) In some embodiments, it may be desired to select a single ion (N=1) for analysis for efficient operation of the CDMS device. According to the techniques described in Kiefer et al., the mean of the ion arrival Poisson distribution is set to one ion (in a fill period of ˜0.5 ms). However this means that in a majority of cases (˜63%) the fill will result either in no ions (N=0) or more than one ion (N>1). When N=0, the (long) acquisition time (up to ˜three seconds) is wasted. Furthermore, when more than one (N>1) ion is held in the ion trap, the signal may be badly contaminated due to space charge effects.

(21) Thus, in embodiments, the detector signal may be monitored in real time, and if after a period of time (for example, 10 or 50 or 100 ms) signal processing suggests N=0 or N>1, the current acquisition may be terminated early and a new fill event started, resulting in increased throughput. For instance, the acquisition may be terminated by applying suitable electric fields to (rapidly) remove all of the ions from the CDMS device. For example, by removing the trapping fields and/or applying one or more ejection fields the ions can then be “ejected” (or otherwise removed) from the trap and lost to the system or to collisions with the electrodes.

(22) Alternatively, in other embodiments, when it is determined that N>1, it may be possible to excite ions in the trap to eject N−1 ions (such that these ions are then lost, as above), leaving only a single ion for analysis. This may be done deterministically or further monitoring may be performed to check that only one ion remains. It will be appreciated that ejecting ions from the trap may be advantageous compared to starting a new fill event since in that case the success rate may be close to 100% (whereas a new fill would generally succeed in only 37% of cases—that is there is a ˜63% chance that the new fill will result either in no ions or more than one ions).

(23) Similarly, in this way, if an ion is lost during a trapping period (so that N=0), for example, due to scattering with the residual gas, or an unstable trajectory, the acquisition may be terminated early allowing a new fill event.

(24) Thus, by contrast to more conventional approaches where a fixed ion trapping period is used for CDMS measurement (even if there are no ions being measured, or wherein multiple ions are present compromising the signal), in embodiments, an ion trapping event can be terminated early if the signal processing suggests N=0 or N>1. Alternatively, if the signal processing suggests N>1, the operation of the CDMS device can be adjusted until N=1. Thus, the CDMS device can be dynamically controlled based on a determination of how many ions are present in the device.

(25) The detector signal may be monitored using any suitable techniques. For instance, in some embodiments, real time signal processing may consist of a series of overlapping apodised fast Fourier transforms. Estimation of the number of ions present in the trap may, for example, be based on the number of masses present in the spectrum above a noise threshold, or the total charge detected, or a combination of these.

(26) Embodiments are also contemplated for tuning the ion arrival rate to maximise the probability of N=1. For instance, in some examples, one or more dynamic range enhancement (DRE) lenses may be used to control the flux of the ion beam in real time over a wide dynamic range. For example, a configuration involving multiple DRE lenses separated by gas filled cells at collision cell pressure for beam remerging may assist with control of the flux of the ion beam in real time over a wide dynamic range to help maximise the probability of N=1 ions arriving at the CDMS device.

(27) In some embodiments, instead of exciting ions from the ion trap when it is determined that more than one ion is present, the ion trap itself may be designed such that the ion trajectories become unstable when more than one ion is present, resulting in ejection of all but one ion. In other words, the ion trap may be designed as a so-called “leaky” single ion trap. For instance, this may be achieved using an appropriately designed geometry and/or by applying one or more appropriate electric fields to the ion trap. In embodiments, the ion trap(s) may be of the type described in U.S. Pat. No. 8,835,836 (MICROMASS) wherein once the charge capacity of the ion trap has been reached the force on the ions due to coulombic repulsion is such that excess ions will leak or otherwise emerge from the trap.

(28) Ion Trap—Space Charge Effects

(29) FIG. 2 shows a series spectra obtained by simulating the motion and detection of two identical ions with energies of 100 eV in a cone trap configured for CDMS after 0.05s, 0.08s, 0.2s and 1s respectively. The transients were sampled at a rate of 1.25 MHz. Spectra were obtained from the raw transients using a Fast Fourier Transform (FFT). The ions have mass of 100 kDa and a charge of 100 so that their mass to charge ratio is 1000 Th.

(30) In particular, FIG. 2 compares the ideal data that would be obtained if the ions did not interact with each other with the data obtained when realistic space charge effects are taken into account. The ideal data is essentially the same as would be obtained for a single ion, and shows a steady increase in resolution as the time is increased, as expected, with the peak centered on the correct mass to charge ratio. On the other hand, where the two ions are able to interact, it can be seen that even after 0.05s there is already a deviation from the correct mass to charge ratio, and by 0.08s the signal has split into two distinct peaks. By 0.2s these two peaks have collapsed and by the end of the transient at 1s, the data are completely compromised.

(31) By providing and analysing these data while the transient is still in progress, then by 0.08s or even earlier it is possible to determine whether more than one ion is present in the trap. This determination could be made using statistical or Bayesian model comparison (comparing the probability that one peak is present with the probability for two peaks or more than two peaks) or hypothesis testing or by simply counting peaks in a smoothed version of the spectrum, or by measuring the full width of the spectrum at a fraction of the maximum intensity compared with the expected width for a single peak, or by a wide variety of other possible methods. In this case, since the full transient length is 1s, terminating trapping after 0.2s (allowing 120 ms for data processing) saves 0.8s of wasted acquisition time.

(32) FIG. 2 thus shows that it is possible to identify very quickly when the ion trap contains more than ion, to allow the transient to be terminated early, or for the ion trap to be controlled to eject one or more ion(s). Clearly, it can also be identified very quickly when no signal is present, in which case the transient may also be terminated early.

(33) More generally, if the full transient time is T.sub.L and a transient is ended after time T.sub.S if it contains no ions or more than one ion then the rate with which good transients are obtained is:

(34) $R_{\mathrm{good}}=\frac{\lambda }{\left(T_L-T_S\right)\lambda +e^{\lambda }{T}_S}$
where λ is the average number of ions that enter the trap during a trap filling period. R.sub.good is maximised when Δ=1 regardless of the values of T.sub.L and T.sub.S so that the intensity of the ion beam supplying the trap should be optimised to obtain this rate as nearly as possible. For Δ=1,

(35) $R_{\mathrm{good}}=\frac{1}{T_L+\left(e-1\right)T_S}$

(36) FIG. 3 shows how R.sub.good changes for a fixed value of T.sub.L=1 and T.sub.S is varied. For T.sub.S=0.2, good, single ion transients are obtained with a rate R.sub.good=0.74 which is more than double the rate obtained when bad transients cannot be terminated early (i.e. T.sub.S=T.sub.L=1).

(37) High Dynamic Range Ion Beam Attenuation

(38) As mentioned above, embodiments are contemplated for controlling the flux of the ion beam in real time over a wide dynamic range to help maximise the probability of N=1 ions arriving at the CDMS device. However, it will be appreciated that there are many scenarios in which it is desirable to reduce the intensity of an ion beam in a controlled, quantitative, unbiased manner. That is, the degree of attenuation should not depend on m/z, ion mobility, propensity to fragment or charge reduce or any other ion characteristic within a relevant range for each property.

(39) For example, this may be desirable to avoid unwanted problems arising from high ion flux including overfilling of traps including those used in ion mobililty experiments (resulting in uncontrolled and biased loss of ions or unwanted fragmentation), space charge effects, detector saturation (resulting in loss of quantitative accuracy, mass accuracy and artificial peaks) and charging of surfaces inside an instrument resulting in further loss of ions or distortion of the onwardly transmitted ion beam in a range of applications including but not limited to producing controlled low ion fluxes to be used in experiments involving single ions or few ions such as CDMS.

(40) When a beam has been attenuated in a quantitative and unbiased manner it is often possible to recover many of the properties of the ideal signal that would have been obtained from the original un-attenuated beam by simply rescaling or otherwise adjusting the data produced by the instrument in question (for example the intensity of a mass spectral peak produced by a mass spectrometer).

(41) The degree of attenuation can be constant for the duration of an experiment or it may vary in a predetermined way, or in response to information obtained from data that has already been acquired during the experiment (in a data dependent way).

(42) Beam attenuation can also result in loss of small signals which fall below a detection threshold following attenuation. For this reason, an instrument may alternate between two or more modes of operation utilizing different degrees of attenuation. A final combined data set may then be reconstructed from the two or more datasets by taking small signals from data that is less attenuated, and larger signals from data that is more attenuated.

(43) U.S. Pat. No. 7,683,314 (MICROMASS) discloses methods of attenuation of an ion beam which operate by alternating between a mode in which transmission is substantially 100% (for time ΔT.sub.2) and a mode in which transmission is substantially 0% (for time ΔT.sub.1). For example, this may be achieved by alternating a retarding voltage to repeatedly switch the ion beam between the two states.

(44) FIG. 4A shows the ideal beam intensity as a function of time following this attenuation step. Since the resulting beam is discontinuous, or chopped, it is possible to operate such a device upstream of an ion guide or gas collision cell in order to convert it into a substantially continuous beam that has been reduced to a fraction ΔT.sub.2/ΔT.sub.1 of its original intensity as shown in FIG. 4B.

(45) However, since it inevitably takes a finite time for the ion beam to fully respond to changes in voltage intended to switch between the on and off states, when the duration of the on state ΔT.sub.2 becomes too short, there is insufficient time to recover 100% transmission before the next voltage change and attenuation is no longer linear or quantitative. On the other hand, when the time interval ΔT.sub.1 becomes comparable with the time to pass through the downstream gas cell or ion guide, it is no longer possible to restore the beam to a substantially continuous beam.

(46) This means that there is a practical limit to the degree of quantitative attenuation that can be achieved by such a device (e.g. attenuation to 1% of the original intensity in a typical device).

(47) According to an embodiment of the present disclosure, there is provided a method of attenuation using two attenuation devices of the type described above, separated by a gas cell or ion guide designed to convert the ion beam into a substantially continuous beam.

(48) FIG. 5 shows an example of an attenuation device according to an embodiment. As shown, the device includes a first attenuation device 50 comprising a plurality of electrodes defining an electrostatic lens and a second attenuation device 52 of the same type. The first and second attenuation devices 50,52 are separated by a first ion guide or gas collision cell 54. The incoming ion beam can thus be attenuated by the first attenuation device 50 (for example according to a scheme like that shown in FIG. 4A). As the chopped ion beam passes through the first ion guide or gas collision cell 54 the interactions of the ions with the gas molecules cause the ions to spread out and the beam is converted back into a substantially continuous beam (as shown in FIG. 4B). The beam is then passed to the second attenuation device 52 where it is attenuated again before being passed through a second ion guide or gas collision cell 56.

(49) The first attenuation device 50 alternates between full transmission mode (for time periods of length ΔT.sub.A2) and low transmission mode (for time periods of length ΔT.sub.A1). The resulting beam is then preferentially converted to a substantially continuous beam by the subsequent ion guide or gas collision cell 54, with a fraction ΔT.sub.A2/ΔT.sub.A1 of its original intensity. Similarly, the second attenuation device 52 operates with high transmission and low transmission time periods ΔT.sub.B2 and ΔT.sub.B1 respectively so that the average transmission through the second device 52 is ΔT.sub.B2/ΔT.sub.B1. Preferentially, the beam may be subsequently converted to a substantially continuous beam by a second ion guide or gas collision cell 56. The overall result of the above arrangement is that the ion beam is reduced to a fraction (ΔT.sub.A2ΔT.sub.B2)/(ΔT.sub.A1 ΔT.sub.B1) of its original intensity.

(50) If each of the first and second attenuation devices 50,52 are independently capable of quantitatively reducing the ion beam to a fraction p of its original intensity, the combined device can quantitatively achieve a fraction p.sup.2 of the original intensity. For example if the maximum quantitative attenuation for an individual device is 1%, then the combined device can achieve 0.01%.

(51) Clearly the concept can be extended to include more than two devices separated by ion guides or gas collision cells designed to produce substantially continuous beams. For instance, when N devices, each individually capable of reducing the ion beam to a fraction p of its original intensity, are combined in this manner, a fraction p″ of the original beam intensity may be achieved quantitatively. This power law behaviour means that extremely high attenuation factors can be achieved quantitatively using relatively few devices. This may be required, for example, to achieve the low ion arrival rates necessary to yield a high probability of populating a trap with a single ion.

(52) In practice, it is not necessary for the attenuation devices or the associated gas cells to be arranged contiguously in an instrument. They may be separated by other devices such as reaction cells, mass filters, ion mobility devices etc. Each of these additional devices may serve several purposes or operate in several different modes, and may be configured to react, fragment or filter ions, or (possibly simultaneously) to convert a pulsed ion beam to a substantially continuous ion beam.

(53) Additionally, one or other or both of the attenuation devices may be operated continuously in full transmission mode, with attenuation only activated as required.

(54) Space Charge Tolerance of Trap

(55) In embodiments, it may be desired for the CDMS device to be able to analyse multiple ions simultaneously to increase throughput. However, as mentioned above, with conventional CDMS devices, such as that described in Kiefer et al., space charge effects may significantly affect the performance when more than one ion is present in an ion trap.

(56) Thus, in some embodiments, it is contemplated the CDMS device may comprise a plurality of ion traps. For example, the CDMS device may comprise a plurality of parallel ion traps, each having an associated one or more charge detection electrodes, arranged to receive a plurality of ions from an upstream device. In this example, multiple ions from the upstream device may be shared between the plurality of ion traps using appropriate ion optics (for example, ion lenses or beam splitting devices). Thus, the system may be arranged so that (single) ions are sequentially or selectively passed to one of a plurality of different ion traps.

(57) FIG. 6 shows an example of such an arrangement wherein two CDMS devices of the general type shown in FIG. 1 are arranged in parallel and wherein an ion optical device 60 such as an ion lens, or other beam splitting device, is provided upstream of the CDMS devices for selectively or sequentially passing ions to the respective CDMS devices. In general, any suitable ion optical device may be used for directing the ions to the respective devices. For instance, US Patent Publication No. 2004/0026614 (MICROMASS) describes various techniques for ion beam manipulation. Of course, although FIG. 6 shows only two CDMS devices, this can be extended to any number of parallel CDMS devices, as desired. Furthermore, the CDMS devices need not be physically arranged in parallel, and can be arranged in any suitable fashion. For example, the devices could be arranged substantially opposite or orthogonal to one another.

(58) As another example, the CDMS device may comprise a series of “leaky” ion traps, with each ion trap having a geometry that is configured such that trajectories become unstable when more than one ion is present. In this case, provided that the ions are suitably confined within the CDMS device, the ions will naturally distribute themselves along the series of traps as a result of space charge effects. The series of ion traps may therefore be contained within an ion guide such as a stacked ring ion guide.

(59) FIG. 7 shows an example of such an arrangement wherein two CDMS devices 72, 74 of the general type shown in FIG. 1 are formed within a single ion guide 70 with the electrodes of the ion guide thus providing the ion traps and charge detectors for the CDMS devices. For instance, suitable RF and/or DC potentials can then be applied to the electrodes of the ion guide 70 in order to (radially) confine ions within the ion guide 70 and also to define one or more axial trapping regions along the length of the ion guide with the electrodes in the centre of the trapping region(s) then providing a charge detector for performing CDMS measurements. Ions can thus be injected into the ion guide 70 and allowed to naturally distribute between the ion trapping regions defining the CDMS devices 72,74. A CDMS measurement can then be performed in each CDMS device 72, 74 in parallel before ejecting the ions from each of the ion traps (and from the ion guide 70). Although FIG. 7 shows only two CDMS devices 72,74 it will be appreciated that any number of CDMS devices may be used in such an arrangement.

(60) In these embodiments, each of the ion traps within the CDMS device may be arranged to analyse only a single ion. For example, N ion traps (wherein N>1) may be provided for analysing N ions.

(61) However, embodiments are also contemplated wherein multiple ions (N>1) are analysed within a single ion trap. For example, if it can be arranged for trajectories to diverge (fan out) outside the region of the charge detector electrode, it may be possible to increase the capacity of the ion trap beyond a single ion (whilst still providing sufficient signal quality). For example, in three dimensions, the trajectories could occupy a “dumbbell” (or rotated “H”) shape. In this case, ions would tend to be to be furthest apart when they are moving slowly, and therefore space charge effects would be reduced. Thus, in embodiments, multiple ions (N>1) may be analysed simultaneously, with the ion trajectories for the ions being arranged to diverge outside the region of the charge detector electrode.

(62) Alternatively, or additionally, the ion trap may be extended to contain more than one charge detection electrode. For example, ions may be caused to take a folded flight path like trajectory within the ion trap, for example, wherein ions are caused to repeatedly pass back and forth between two reflecting electrodes in a multi-pass operation, for example, so as to travel along a substantially zigzagged, or “W”-shaped, path. Charge detection electrodes may then be periodically placed along the folded flight path (for example, in place of the periodic focussing elements that may be found within a folded flight path instrument). Each ion may thus pass through each of the multiple charge detection electrodes (so that multiple measurements can be made for each ion, thus potentially improving the signal quality). As another example, instead of using a folded flight path type geometry, a multi-detector configuration could be wrapped round in a circle to give a cyclic CDMS device with multiple charge detection electrodes. The signal from each charge detection electrode could be analysed separately or, if more convenient, some may be electronically coupled and the combined signal deconvolved in post-processing.

(63) As yet another example, the device could be linear or circular with no orthogonal trapping and with many charge detection electrodes arranged along the flight path (for example, in a similar manner to ion velocity Fourier transform mass spectrometry techniques).

(64) For instance, FIG. 8 shows an example of a CDMS device wherein multiple independent charge detecting electrodes are provided within a single cone trap 10. Although FIG. 8 shows four charge detectors 82,84,86,88 it will be appreciated that any number of charge detectors may be used, as desired. In embodiments, this device may be used for analysing single ions (with an increased resolution). However, provided that the ion trajectories are sufficiently separated, the device of FIG. 8 can also be used to perform simultaneous measurements on a plurality of ions. As shown, the charge detectors are decoupled from each other. This allows more information to be extracted. For instance, whilst the four (in this example) signals could be analysed separately and the results combined, in embodiments, the inference of the mass to charge ratio and charge values may be carried out simultaneously using the separate, uncombined signals. Various methods for analysing the data are possible. For example, the signals may be analysed using maximum likelihood (least squares), maximum a posteriori, Markov chain Monte-Carlo methods, nested sampling, and the like. Various other arrangements would of course be possible.

(65) Improved Trajectories for Higher Resolution or Faster Operation

(66) The Applicants have further recognised that the use of an approximately quadratic potential within the ion trap may result in improved energy tolerance of the device, for example, in that ions of the same mass-to-charge ratio but differing energy will produce signals having a more similar (or substantially the same) shape. More harmonic (sinusoidal) signals may give rise to cleaner spectra (with reduced harmonics). Thus, in embodiments, a substantially quadratic potential is used to confine the ions within the ion trap so that the ions undergo substantially harmonic motion within the ion trap (and through the charge detector electrode(s)). In this case the charge detector electrode may be located at the centre of the substantially quadratic potential. However, other arrangements would of course be possible.

(67) Various existing geometries having suitably substantially quadratic potentials could be utilised. For example, it is contemplated that an Orbitrap type device or a SpiroTOF device (for example, as described in U.S. Pat. No. 9,721,779 (MICROMASS) or US Patent Application Publication No. 2017/0032951 (MICROMASS)) may be used. Devices with a central electrode (particularly the Orbitrap) have a relatively high space charge tolerance.

(68) FIGS. 9A, 9B, 9C and 10 illustrate the operation of a SpiroTOF device that may be used according to embodiments as an ion trap for a CDMS device. As shown in FIG. 9A, ions are injected into an annular region defined between an inner cylinder 100 and an outer cylinder 102, each comprising an axial arrangement of electrodes. The ion beam may be expanded along the axis of the device during the injections (for example as described in U.S. Pat. No. 9,245,728 (MICROMASS)). The potentials that are applied between the inner and outer cylinders are selected to allow the ions to form stable circular orbits 104 within an entrance region of the device, as shown in FIG. 9B. Once the ions have been injected into a stable circular orbit, the ions can then be initially accelerated along the axis of the device, as shown in FIG. 9C.

(69) A substantially quadratic axial potential can then be set up along the device to cause the ions to begin to oscillate axially with substantially simple harmonic motion, as shown in FIG. 10. The conditions may be chosen so that the orbits remain circular (as shown in FIG. 10), or the ions may be allowed to oscillate radially (by imparting some radial excitation during the initial acceleration). A charge detector 1100 may then be positioned within the device, for example in the center thereof, so that the ions repeatedly pass close to the detector electrodes to generate a signal. The charge detector 1100 may comprise one or more of the segments chosen from the existing electrodes used to fix the substantially quadro-logarithmic potential in the device, or they may be additional electrodes with geometries and voltages designed to minimise perturbations to that potential.

(70) This arrangement has the advantage that, even for a small number of ions, the average initial separation between the ions can be increased by beam expansion during the initial injection, reducing space charge effects. Furthermore, the inner electrodes 100 help to shield the ions from each other. Additionally, when ions of the same mass to charge ratio are moving slowly (at the extremes of their axial motion), and are therefore most susceptible to space charge effects, their average separation is largest owing to beam expansion.

(71) However, other arrangements would of course be possible. For instance, an Orbitrap-type geometry using a substantially quadro-logarithmic potential may also provide similar advantages. This may also be the case, for instance, for Cassinian orbits such as those described in U.S. Pat. No. 8,735,812 (BRUKER DALTONIK GMBH), depending on the trajectory chosen.

(72) Signal Processing

(73) The use of Fourier Transform processing on anharmonic signals is well known to produce artefact “harmonics”. However, in embodiments, forward fitting/Bayesian signal processing using model peak shape, or shapes, may be used. This may significantly reduce the intensity of harmonics and improve signal-to-noise in the inferred spectrum. Thus, this may in turn provide a higher mass resolution in a fixed time (or similarly the same resolution to be achieved in a shorter time). For instance, the Applicants have recognised similar techniques such as those described in US Patent Application Publication No. 2016/0282305 (MICROMASS) for processing ion mobility data may also advantageously be used for processing the CDMS signals obtained according to various embodiments described herein. For example, by using similar such techniques, it may be possible in embodiments to extract a charge value from the fitted amplitude. Especially if space charge limitations are reduced, such signal processing approaches may thus be capable of extracting high quality spectra from trapping events including more than one ion.

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