Multiplexing method for separators

Abstract

The present disclosure provides a method comprising providing a sample to be analysed, separating successive populations of ions from said sample in a separator, wherein said populations of ions are introduced into said separator at regular intervals, and the intervals are timed such that at least some ions in a subsequent population of ions overlap ions in a preceding population of ions, varying one or more parameters of said separator such that different populations of ions experience different separation conditions, detecting ions from said populations of ions and obtaining a convolved data set, and de¬ convolving said convolved data set using the known variance of the parameters and outputting data corresponding to the successive populations of ions.

Claims

1. A method comprising: providing a sample to be analysed; separating successive populations of ions from said sample in a separator, wherein said populations of ions are introduced into said separator at intervals, and the intervals are timed such that ions in a subsequent population of ions overlap ions in a preceding population of ions, within the separator; varying one or more parameters of said separator such that different populations, of said successive populations of ions, experience different separation conditions; detecting ions from said populations of ions and obtaining a convolved data set; and de-convolving said convolved data set using the known variance of the parameters and outputting data corresponding to the successive populations of ions.

2. A method as claimed in claim 1, wherein the populations of ions are introduced into the separator at a first frequency or in a first pattern, and the populations of ions exit the separator at a second frequency or in a second pattern, wherein the second frequency or second pattern is different to the first frequency or first pattern due to the different separation conditions experienced by each population of ions as they travel through said separator.

3. A method as claimed in claim 1, wherein said one or more parameters substantially affect a transit time of ions through said separator.

4. A method as claimed in claim 1, wherein each successive population of ions contains at least some of the same analyte compounds, and the data corresponding to the successive populations of ions comprises the mass and/or mobility peaks of said analyte compounds.

5. A method as claimed in claim 4, wherein said step of de-convolving said convolved data set comprises using a forward modelling method.

6. A method as claimed in claim 1, further comprising determining the ion mobility and/or collision cross section of ions from said data corresponding to the successive populations of ions.

7. A method as claimed in claim 1, wherein said separator is an ion mobility separator.

8. A method as claimed in claim 7, wherein said one or more parameters comprises one or more of driving force, driving DC voltage, buffer gas velocity, buffer gas composition and buffer gas pressure.

9. A method as claimed in claim 7, wherein said ion mobility spectrometer is a travelling wave ion mobility spectrometer comprising a plurality of electrodes, in which one or more transient DC voltages or potentials are applied to at least some of said electrodes in order to urge ions in a first direction through said ion mobility spectrometer to create a DC travelling wave, and said one or more parameters comprises an amplitude and/or velocity of said DC travelling wave.

10. A method as claimed in claim 7, further comprising mass analysing said ions prior to said step of detecting said populations of ions, wherein said convolved data set comprises ion mobility data nested with mass spectral data.

11. A method as claimed in claim 10, wherein said one or more parameters comprises an applied DC field strength, and said data corresponding to the successive populations of ions comprises a drift time measurement for one or more analyte compounds taken at different average field strengths.

12. A method as claimed in claim 9, further comprising determining a value of collision cross section for each of said analyte compounds using a plot of said drift time measurements against the reciprocal of the average field strength.

13. A method as claimed in claim 1, wherein said separator is configured to separate ions according to their mass to charge ratio.

14. A method as claimed in claim 13, wherein said separator is an orthogonal time of flight mass analyser, and said one or more parameters comprises a voltage associated with said orthogonal time of flight mass analyser that substantially affects the time of flight of ions in said orthogonal time of flight mass analyser.

15. An apparatus for separating and analysing ions, the apparatus comprising an ion separator, a detector and a control system, wherein the control system is arranged and adapted to: separate successive populations of ions from a sample, and introduce said populations of ions into a separator at intervals, wherein the intervals are timed such that ions in a subsequent population of ions overlap ions in a preceding population of ions, within the separator; vary one or more parameters of said separator such that different populations, of said successive populations of ions, experience different separation conditions; detect ions from said populations of ions and obtain a convolved data set; and de-convolve said convolved data set using the known variance of the parameters and outputting data corresponding to the successive populations of ions.

16. A method as claimed in claim 1, wherein different ion populations that overlap in the separator are subjected to said different separation conditions.

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 an arrangement of a mass spectrometer;

(3) FIG. 2A shows a mass-mobility spectrum based on a model for a fixed DC field ion mobility separator, and FIG. 2B shows a smaller portion of FIG. 2A;

(4) FIG. 3A shows a mass-mobility spectrum based on a model for a varying DC field ion mobility separator in accordance with the present disclosure, and FIG. 3B shows a smaller portion of FIG. 3A; and

(5) FIG. 3C shows a smaller portion of FIG. 3A but with the ion species summed to represent the expected profiles of recorded data.

DETAILED DESCRIPTION

(6) There is disclosed a method comprising providing a sample to be analysed and separating successive populations of ions from the sample in a separator. The populations of ions are introduced into the separator at regular intervals, and the intervals are timed such that at least some ions in a subsequent population of ions overlap ions in a preceding population of ions. The method includes varying one or more parameters of the separator such that different populations of ions experience different separation conditions, for example such that each successive population of ions experiences different separation conditions. The method further includes detecting ions from the populations of ions and obtaining a convolved data set. The method further comprises de-convolving the convolved data set using the known variance of the parameters and outputting data corresponding to the successive populations of ions.

(7) The populations of ions may be generated from the same sample, and may have substantially the same composition. Each population of ions may comprise one or more of the same analytes, optionally in substantially the same concentration. The data corresponding to the successive populations of ions may comprise data for each analyte.

(8) As discussed above and herein, and generally, successive populations (or packets) of ions are pulsed at regular intervals into a separation device and the conditions of separation are varied (e.g., continuously or in steps) as the successive populations of ions are introduced. Thus, different populations of ions may experience different average separation conditions. If the average separation conditions (and, hence, elution time) for ions of different properties in the different ion populations is known or otherwise determined (e.g., by pre-calibration) the complex multiplexed data produced (e.g., the convolved data set) may be de-convolved to give a simplified data set representing the elution characteristics of various species present. Any suitable method of deconvolution may be used, including those described below.

(9) As packets of ions are introduced continuously at regular intervals a larger number of ion packets may be introduced into the separator per unit time, which can maximise the duty cycle or overall space charge capacity compared to the prior art.

(10) In the case of multiplexed ion mobility separation, coupling the ion mobility separator with a high resolution time of flight mass analyser can be used to produce a nested IMS-MS data set, which may simplify the de-convolution process.

(11) FIG. 1 shows a block diagram of a mass spectrometer incorporating an ion mobility separation device.

(12) Ions may be produced in an ion source (1) and may be trapped within ion accumulation region (2). Ion populations that have accumulated in the ion accumulation region (2) may be periodically released into an ion mobility separation region (3), where they may separate according to their ion mobility. The populations of ions may be introduced into the separator at regular intervals, and the intervals may be timed such that at least some ions in a subsequent population of ions overlap ions in a preceding population of ions. In other words, the populations of ions may intermix at the detector, or otherwise overlap, such that the detected signal comprises a convolved data set.

(13) The separated ions may then pass through one or more optional components (4) (e.g., a quadrupole mass filter and/or collision or reaction cell) before passing into an optional mass analyser (5), which may be an orthogonal acceleration time of flight mass analyser. The ions can then be detected and analysed.

(14) The output signal response at the detector of the mass analyser (5) was modeled for a situation in which the ion mobility separation region (3) comprised a linear field ion mobility separation device with a drift length of 1 m, and at a pressure of 3 torr of Nitrogen. In the model, ion packets are released from the ion accumulation region (2) every 2 ms from 0 ms to 60 ms.

(15) Three ions species were present in each modeled ion packet, including:

(16) Mass 350, z.sub.1, ccs=164.7 Å2, K=0.0355 m.sup.2/V/s,

(17) Mass 700, z.sub.2, ccs=231.5 Å2, K=0.0495 m.sup.2/V/s, and

(18) Mass 1050, z.sub.3, ccs=276.8 Å2, K=0.0617 m.sup.2/V/s, where “z” denotes the charge on the ion, “ccs” denotes the collision cross section, and “K” denotes the ion mobility.

(19) In the model, peak profiles are represented by a Gaussian shape with a standard deviation proportional to the expected diffusion for each ion at the pressure modeled. For the above species the z.sub.2 ion is 2 ms slower than the z.sub.3 and the z.sub.1 ion is 6 ms slower than the z.sub.3. Over 1 m for a 2 kV field, drift times were 8.1, 10.1 and 14.1 ms for z.sub.3, z.sub.2 and z.sub.1 respectively.

(20) The three species described above have identical m/z values and so cannot be distinguished by mass spectrometry. It should be recognized that this modeled situation results in a ‘worst case’ situation for a multiplexing technique of the type described.

(21) FIGS. 2A-2B show a reconstructed mass-mobility spectrum or elution profile, of m/z 350, with a fixed linear DC field. FIG. 2A shows the entire mass-mobility spectrum for the sequence of ion introductions described.

(22) As used herein, the reconstructed mass-mobility spectrum corresponds to the signal recorded at the detector within a narrow region of the mass spectrum corresponding into the ion of interest (in this case around a mass to charge ratio of 350).

(23) For a nested ion mobility-time of flight data set, this data may be generated by summing the intensities within the narrow region of the time of flight mass spectrum, and plotting the summed intensities as a function of elution time. This reduces the three dimensional data (including drift time, mass to charge ratio, and intensity) to a two dimensional spectrum comprising just drift time and intensity.

(24) The three species are shown as separate peaks for clarity. In reality, as the m/z values are identical, the output of the reconstructed mass-mobility spectrum would comprise the sum of the signal intensities for each species.

(25) Under these conditions the three species co-elute for the majority of the ion mobility separation time. This may be seen as a worst case where de-convolution would be impossible or very inaccurate.

(26) FIG. 2B shows a small portion of the elution profile illustrating the overlap of these three species. Under these conditions it is very difficult or impossible to de-convolve these three species to calculate the drift time and/or collision cross section for each of the species.

(27) FIGS. 3A-3C shows a reconstructed mass-mobility spectrum, of m/z 350 in the same model as described above and in relation to FIGS. 2A-2B, but a varying DC field is provided in place of a fixed DC field. In this model, the linear DC field ramps down (in 21 steps) from 2 kV/m to 1.5 kV/m over 60 ms.

(28) FIG. 3A shows the entire elution profile for the sequence of ion population introductions into an ion mobility separator (e.g., the ion mobility separation region (3) of FIG. 1).

(29) FIG. 3B shows a small portion of the elution profile illustrating the change in the relative elution times of the three species as the acquisition sequence proceeds. In FIGS. 3A and 3B, the three species are shown as separate peaks for clarity. In reality, the profile of recorded data will summed, and FIG. 3C shows portion of the elution profile with the signal from the three species summed to more accurately represent such an expected profile of the recorded data.

(30) In FIGS. 3A-3C the average field experienced by each population of ions introduced into the device may be different and, in contrast to the conditions of FIG. 2, the extent of the overlap between the signals from the three species may progressively change as the experiment proceeds.

(31) Based on the known parameters of the separator, and in accordance with the disclosure, the resulting complex data set or convolved data set may de-convolved to determine the ion mobility, and hence collision cross section of each component.

(32) The step of de-convolving the convolved data set may comprise one or more known de-convolution techniques, including methods based on forward fitting of model data such as non-negative least squares, maximum likelihood (least squared), maximum entropy, Bayesian (probabilistic) methods and filter diagonalisation. Such methods may be known in the art as “forward modeling methods”.

(33) In various embodiments, the multiplexed ion mobility data may be coupled with time of flight mass spectrometry to produce an ion mobility-mass spectrometry (“IMS-MS”) data set. The mass to charge ratio resolution of time of flight mass spectrometry can allow simplification of the ion mobility spectrum for each mass to charge ratio range. Each mass to charge ratio region of the IMS-MS data set may be de-convolved separately, or the entire data set (e.g., a 3D data set comprising mass to charge ratio, drift time and intensity) may be de-convolved as a multi-dimensional array. The known correlation between mass to charge ratio, charge state and ion mobility may be used to assist or guide the de-convolution algorithm, which can speed up the processing time.

(34) The step of de-convolving the convolved data set may comprise using a forward modeling method, such as an iterative forward modeling algorithm. Such an algorithm may comprise: (i) modifying the amplitude and/or frequency of at least some of the model signals, (ii) superimposing the modified model signals, (iii) comparing the resulting composite signal to the signal output from the detector, and (iv) calculating a goodness of fit between the composite signal and the ion signal output from the detector; wherein steps (i)-(iv) are repeatedly performed in an iterative manner until a termination criterion is satisfied, or until the goodness of fit between the composite signal and the ion signal output from the detector is within said predetermined probability or tolerance. The termination criterion may be maximum likelihood, maximum entropy, or maximum a posteriori (MAP).

(35) The iterative process may be a Markov Chain Monte Carlo method or a nested sampling method, producing samples from a probability distribution where each sample represents a possible reconstruction of the data.

(36) The deconvolution technique may comprise a least squares or non-negative least squares algorithm, or a filter diagonalisaltion method.

(37) In forward modeling methods, it is desired to determine a set of model signals, corresponding to particular ion mobility values, that when superimposed match the experimentally observed signals generated by the ion mobility separator. The method iterates different combinations of modeled signals having differing ion mobilities and amplitudes until the best match for the experimentally obtained signal is determined. The model signals making up the best match are then used to determine the ion mobilities and intensities of the ions.

(38) Such methods have previously been too computationally intensive to be of practical application, however, advances in computational electronics and methods have made these techniques more practical.

(39) In addition, the coupling of the multiplexed IMS device with mass spectrometry facilitates application of these forward fitting techniques with practical timescales. For example, a correlation between mass to charge ratios of the ions and the drift times may be known or determined and used to simplify the modeling process. In particular, the mass to charge ratios of the ions may be determined and the correlation may then be used to determine the drift times of the ions that would be expected under the conditions of the multiplexed separation. The forward modeling need then only model signals having characteristics that correspond to the expected drift times of the ions. Other model signals need not be considered as they would correspond to drift times of ions that are not present. This process significantly simplifies the modeling. The mass to charge ratios may be determined by mass analysing the ions downstream of the IMS device and/or by providing a mass filter upstream of the IMS device that mass selectively transmits only certain ranges of mass to charge ratios.

(40) Forward fitting of model data may be applied to each narrow mass to charge ratio region in which far fewer species exist and therefore the signal is greatly simplified resulting in more precise results in far shorter timescales. The model data may be obtained from calibration standards or sufficiently pure species within the analyte.

(41) In an ion mobility separator, the ion mobility peak widths will typically be slightly wider for ions experiencing lower linear DC driving potential. This deterministic difference or progression of difference in ion mobility peak line width and/or maximum intensity may also be used to assist de-convolution and can be incorporated into the model data.

(42) The effect of continuously varying a DC field in an ion mobility separator as successive ion populations are introduced may be considered analytically, and an example is set out below.

(43) The velocity with respect to time of an ion of mobility K in a changing electric field E(t) is given by:
u(t)=E(t).Math.K  (1)

(44) Considering a linear field change the field is given by:
E(t)=E.sub.0+bt  (2)
where E.sub.0 is the initial field.

(45) By substitution:
u(t)=(E.sub.0+bt).Math.K  (3)

(46) By integration, the position of an ion within the separator at time t is given by:

(47) $\begin{array}{cc}x\left(t\right)=K\left(E_{0}t+\frac{{\mathrm{bt}}^2}{2}+C\right)& \left(4\right)\end{array}$
as the position of the ions within the ion mobility separator x=0 at time t=0 the integration constant C=0.

(48) The time taken for ions to exit the device of length L is t.sub.L, and:
x(t.sub.L)=L  (5)
as well as:

(49) $\begin{array}{cc}x\left(t_L\right)=K\left(E_0{t}_L+\frac{{\mathrm{bt}}_L^2}{2}\right)& \left(6\right)\end{array}$

(50) Therefore:

(51) $\begin{array}{cc}L=K\left(E_0{t}_L+\frac{{\mathrm{bt}}_L^2}{2}\right)& \left(7\right)\end{array}$

(52) Equation (7) may be rewritten as:

(53) $\begin{array}{cc}0={\mathrm{KE}}_0{t}_L+K\frac{{\mathrm{bt}}_L^2}{2}-L& \left(8\right)\end{array}$

(54) The solution to this quadratic equation is:

(55) $\begin{array}{cc}{t}_L=\frac{E_0}{b}\left[-1\pm \sqrt{1+\frac{2.Math.L.Math.b}{E_0^2.Math.K}}\right]& \left(9\right)\end{array}$

(56) The solution for which the driving field does not reverse direction is:

(57) $\begin{array}{cc}{t}_L=\frac{E_0}{b}\left[-1+\sqrt{1+\frac{2.Math.L.Math.b}{E_0^2.Math.K}}\right]& \left(10\right)\end{array}$

(58) Considering n multiple injections of ions with an interval between injections of Δt the field experienced by the nth ion will be:
E(t)=E.sub.0+n.Math.b.Math.Δt+b.Math.t  (11)

(59) Defining the initial field for each of the n ion packets as:
E.sub.0(n)=E.sub.0+n.Math.b.Math.Δt  (12)

(60) The time from injection to elution for each of the n injections is given by:

(61) $\begin{array}{cc}{t}_n=\frac{E_0\left(n\right)}{b}\left[-1+\sqrt{1+\frac{2.Math.L.Math.b}{{E_0\left(n\right)}^2.Math.K}}\right]& \left(13\right)\end{array}$

(62) However, the measured arrival times (At) are recorded relative to the injection of the first packet of ions in the sequence. Therefore the observed drift times for the n injections of ions is given by:
At.sub.n=t.sub.n+n.Math.Δt  (14)
and:

(63) $\begin{array}{cc}{\mathrm{At}}_n=\frac{E_0\left(n\right)}{b}\left[-1+\sqrt{1+\frac{2.Math.L.Math.b}{{E_0\left(n\right)}^2.Math.K}}\right]+n.Math.\Delta t& \left(15\right)\end{array}$

(64) Re-Arranging for K yields:

(65) $\begin{array}{cc}K=\frac{2b.Math.L}{{\left(E_0\left(n\right)+b\left({\mathrm{At}}_n-n.Math.\Delta t\right)\right)}^2-{E_0\left(n\right)}^2}& \left(16\right)\end{array}$

(66) In combination with a down-stream analyser there is often a mass to charge ratio dependent offset, D(m/z), in the measured ion mobility arrival times due to the time of flight of ions from the exit of the ion mobility separator to the detector.

(67) This may be included in the expression above to give.

(68) 0$\begin{array}{cc}K=\frac{2b.Math.L}{(E_0\left(n\right)+{b\left({\mathrm{At}}_n-n.Math.\Delta t-D\left(\mathrm{mz}\right)\right)}^2-{E_0\left(n\right)}^2}& \left(17\right)\end{array}$

(69) Operating an ion mobility separation device in this manner may be analogous to the ‘direct’ method that collision cross section values can be calculated, that is using a linear field ion mobility separation device. In the direct method a series of separate ion mobility separations are performed for the same analyte, at different field strengths and a plot of drift time against the reciprocal of the linear field strength can be used to determine the time offset D(mz). Through knowing the buffer gas pressure, temperature, drift length and field strength, mobility and hence collision cross section may be calculated using the known Mason-Schamp equation:

(70) $\begin{array}{cc}K=\frac{3}{16}\frac{q}{N}\sqrt{\frac{2\pi }{\mu \mathrm{kT}}}\frac{1}{\Omega }& \left(18\right)\end{array}$
where q is the charge on the ion, N is the drift gas number density, μ is the reduced mass, T is the gas temperature and k is Boltzmann's constant.

(71) If the same analyte is present in each of the populations of ions introduced into the separator in the manner described in this disclosure, a series of drift time measurements for that analyte can be produced at different average field strengths, since each population of ions will experience a different average field strength.

(72) This information may be used along with the known parameters of the ion mobility separator to directly calculate collision cross section for that particular analyte. The average field strength for each of the n injections Eav.sub.n is given by:

(73) $\begin{array}{cc}{\mathrm{Eav}}_n=\frac{b*t_n}{2}+E_0\left(n\right)& \left(19\right)\end{array}$

(74) Therefore:

(75) $\begin{array}{cc}\left({\mathrm{At}}_n-n\Delta t\right)=\frac{L}{{\mathrm{Eav}}_{n}K}& \left(20\right)\end{array}$

(76) As indicated in equation (20), the recorded arrival time adjusted for the introduction delay (At.sub.n−nΔt) may be plotted against the reciprocal of the calculated average field strength (calculated from the arrival time for each packet of ions recorded at the detector in sequence), multiplied by the length of the device (L). From this plot, the time offset D(mz) may be determined from the intercept of the straight line produced, and the mobility from the gradient of the line. Hence, the collision cross section of a particular species may be determined directly from a single series of multiplexed injections of the same analyte.

(77) The disclosed technology may therefore provide a rapid method of producing data for calculation of collision cross section for a particular analyte or multiple analytes. In combination with time of flight mass spectrometry, sufficient mass resolution may be provided to allow reconstructed mass-mobility chromatograms of individual species to be generated from simple mixtures. The disclosed method can therefore provide data for ‘direct’ measurement of collision cross section in chromatography time scales.

(78) In various embodiments, any parameter affecting ion separation may be altered instead of, or in addition to the driving force in an ion mobility separator. For example, where there is a counter flow of buffer gas one or more of its velocity, pressure and composition may be altered during the separation, for example between introduction of each population of ions into the separator.

(79) The separator may be a travelling wave ion mobility separator, for example comprising a plurality of electrodes, wherein one or more transient DC voltages or potentials are applied to at least some of the electrodes in order to urge ions in a first direction through the ion mobility spectrometer (i.e., to create a DC travelling wave). A control system may be arranged and adapted to apply the one or more transient DC voltages or potentials to the plurality of electrodes so that said successive populations of ions are translated along the ion mobility separator with a given velocity.

(80) The travelling wave parameters (velocity, amplitude) may be altered to change the elution times of different ion populations. Pre-calibration of mobility and drift time for ions of known mass to charge ratio and ion mobility may be performed under each condition for single pulses at the appropriate point in the ion mobility separation cycle and this information may be used to produce model data to use in the subsequent de-convolution or decoding.

(81) It is recognised that introducing a series of populations of ions into a travelling wave ion mobility separation device while the travelling wave is on may, in some cases, cause losses in transmission of ions. This is because some ions entering the travelling wave device may experience the leading edge of a transient DC pulse as they enter the separation region. This leading edge may either prevent ions from entering the separation region or cause ions to be accelerated out of the entrance region in a reverse direction.

(82) In a non-multiplexing mode, this problem may typically be overcome by interrupting the travelling wave while ions are injected into the separation region. The travelling wave amplitude is typically set to zero for a time between 100 μs and 500 μs. During this time ions can enter far enough into the device that they only experience a forward driving force. In addition, sufficient DC potential gradient may be applied between the upstream accumulation region and the ion mobility separation region to overcome any adverse effects from the transient DC voltage. The amplitude of the travelling wave may also be ramped from a low to a higher value during the experiment such that the amplitude is at the lowest value at the point ions are injected into the separator.

(83) In multiplexed operation it is possible, although undesirable to interrupt the travelling wave or drop the wave amplitude at the point each ion packet is introduced. This may be undesirable as it can be more difficult to optimise the separation conditions of ions if the driving force is reduced discontinuously at such regular intervals.

(84) Referring back to FIG. 1, the separator may be a travelling wave ion mobility separator, and a DC potential gradient may be applied when introducing a population of ions in order to urge that population of ions from the accumulation region (2) to the ion mobility separation region (3), and this DC potential gradient may be increased. However, in some cases this can lead to unwanted rises in the energy of the ions, which can lead to fragmentation.

(85) An optimal method may be to arrange for the amplitude of the travelling wave in a first, upstream portion of the ion mobility separation region (3) (e.g., an entrance region) to be relatively low, and then increase in amplitude along some proportion, for example 5%, 10%, 20%, 30%, 50%, 75% or 100% of the length of the device.

(86) The amplitude of the travelling wave in the first, upstream portion (e.g., an entrance region) of the ion mobility separation region (3) may remain invariant such that every ion population experiences the same, relatively low amplitude of the DC travelling wave as they enter the separator. Once ions are driven away from or leave the first, upstream portion (e.g., an entrance region) the travelling wave amplitude may be restored or increased. The amplitude and/or velocity of the DC travelling wave and/or the overall driving force may be varied between successive populations of ions as described, but not in the first, upstream portion.

(87) It should be noted that the methods disclosed herein may be used regardless of the multiplexing scheme utilised. For example, encoding the delay between introductions of ion packets without varying the travelling wave separation parameters in a manner described in the prior art in relation to separators using DC fields.

(88) In various embodiments, the parameters may be ramped or changed continuously, or in a series of steps.

(89) In the example provided in respect of FIGS. 3A-3C ion populations may be pulsed into the separator (e.g., 30 pulses at 2 ms apart). At the end of the sequence of pulses time may be allocated for the lowest mobility ion of the final pulse to elute. This can reduce the overall duty cycle of the experiment as during most of this time no ions are accumulated in the upstream trap to avoid space charge saturation.

(90) In various embodiments the ion introduction sequence may run continuously and the DC field or other parameter may be ramped back to its original value after a predetermined number (e.g, greater than 10, 20 or 30) of populations of ions have been introduced into the ion mobility separation region (3), while the data corresponding to the successive populations of ions may comprise multiple two dimensional IMS-MS spectrums (each corresponding to the predetermined number of populations of ions), and/or ion mobility data may continuously be acquired in a single large data set. In these embodiments no inter-scan delay or minimal inter-scan delay may be required maximizing duty cycle.

(91) In various embodiments described above, a monotonically changing DC field is described, however other non-monotonic changes to the ion mobility separation parameters may be envisaged. As used herein, a monotonically changing DC field may correspond to a field that is always either increasing or decreasing during the analysis and does not reverse direction during the experiment. For example, the field may be decreased for part of the sequence, maintained at a constant value for part of the sequence and increased for part of the sequence. The increases and/or decreases in the applied field may be non-linear.

(92) Generally, as long as the relationship between the average field experienced by each population of ions and their mobility is known (or otherwise generated) the data may be de-convolved into a single spectrum (e.g., ion mobility spectrum) as described above.

(93) The methods disclosed herein may be applied to any ion separator, and are not limited to ion mobility separators. For example, a mass to charge ratio separator may benefit from the broadest aspects of the present disclosure. An example is an orthogonal time of flight mass to charge ratio (“TOF”) separator, which may comprise a pusher electrode arranged and adapted to pulse successive populations of ions into the TOF separator, and/or a reflectron arranged and adapted to reflect ions that are pulsed into the TOF separator.

(94) In this disclosure, the interval between successive populations of ions being introduced into the separator is substantially constant, and the parameter to be varied may comprise a pusher electrode voltage and/or reflectron voltage, and/or other voltage associated with the TOF separator. The parameter may be a parameter of the TOF separator that substantially affects the time of flight of ions in said orthogonal time of flight mass analyser. Similar methods of de-convolution as disclosed above may be used.

(95) In embodiments where the separator is a mass to charge ratio separator, the apparatus and/or control system thereof may be configured such that the mass to charge ratio resolution of the separator does not substantially or appreciably change over the range of the parameter that is varied. Pre-calibration of the time of flight of ions in the populations of ions, e.g., using pure standards may be used to construct a model for de-convolution and to maintain the mass measurement accuracy of the de-convolved data.

(96) As a further example, the methods disclosed herein may be used in capillary electrophoresis chromatography. Capillary Electrophoresis may be described as liquid phase ion mobility separation, so the separator as described generally herein may be a capillary tube filled with an electrolyte solution. The parameter to be varied may be the magnitude of the electric field applied along the length of the capillary. The field may urge analyte ions in solution from one end of the capillary to the other with a velocity which is dependent of their electrophoretic mobility. Analyte solution may be introduced electro-kinetically at regular intervals as the separation proceeds, which may produce a multiplexed, convolved data set analogous to that described above in respect of gas phase ion mobility. The multiplexed, convolved data set may be de-convolved as described above.

(97) The methods disclosed herein may be applied to gated Fourier transform ion mobility spectrometry methods (as described above). Instead of the gate frequency being varied, variation of the applied DC field or gas flow may result in the same mobility dependent modulation. Fourier transform or other de-convolution technique may be used to de-convolve (i.e., reconstruct) the mobility spectrum from the frequency of intensity modulation recorded with respect to changing field or gas flow applied.

(98) The method is not limited to a separator having an upstream accumulation region, and may be used with gated drift tubes and/or non-RF-confined drift tubes without pre-accumulation of ions.

(99) The multiplexing techniques described herein may be combined with feedback control of the ion populations accumulated for each impulse of ions into the separator, for example to control space charge effects. Rescaling of intensity may be applied before or after de-convolution based on the known fill time for each ion population released.

(100) The disclosed techniques may be combined with a pseudo random packet timing introduction system of the prior art. For example, in the method described in Anal. Chem. 2007, 79, 2451-2462 (entitled “Multiplexed Ion Mobility Spectrometry-Orthogonal Time-of-Flight Mass Spectrometry”), ions are released into an ion mobility separation region in a pseudo-random sequence, such that the time between release pulses varies. Conventionally, the DC field applied to the separation device is invariant. This method may be modified in accordance with an embodiment of the present disclosure such that, in addition to ions being introduced in a pseudo-random sequence, the magnitude of the DC field applied to the separator is also varied during the separation period. This will produce a complex multiplexed (and convolved) data set, the form of which would depend on both the sequence of ion packet introduction, but also the change in field during the ions progress along the separator device. This system may be analytically modeled to produce model data, or the model data may be experimentally measured using standard compounds of known ion mobility, such that a deconvolution technique may be applied, as described above. This embodiment may be seen as a variant on other embodiments on this disclosure, in that the populations of ions are not introduced into the separator at regular intervals, but in a pseudo-random sequence.

(101) 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.