Methodology and Characterization of Isolation and Preconcentration

Subscriber access provided by University of Newcastle, Australia

Article

Methodology and characterization of isolation and preconcentration in a gas-filled digital linear ion guide. Zachary Philip Gotlib, Gregory Forrest Brabeck, and Peter T. A. Reilly Anal. Chem., Just Accepted Manuscript • DOI: 10.1021/acs.analchem.7b00356 • Publication Date (Web): 10 Mar 2017 Downloaded from http://pubs.acs.org on March 16, 2017

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

Analytical Chemistry is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 21

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Analytical Chemistry

Title:

Methodology and characterization of isolation and preconcentration in a gas-filled digital linear ion guide.

Authors:

Zachary P. Gotlib, Gregory F. Brabeck and Peter T. A. Reilly*

*Corresponding author Department of Chemistry, Fulmer 104B Washington State University Pullman, WA 99164-4630 [email protected] 509-335-0042 Abstract: Digital operation of linear ion guides allows them to operate as traps and mass filters by modulating the duty cycles of the two driving waveforms. A gas-filled (5mTorr) digitally-driven quadrupole ion guide was used to demonstrate ion isolation and preconcentration. These abilities allow ion trapping mass spectrometers to be filled to capacity with only ions in the range of interest at essentially any value of m/z. Due to the unique performance characteristics of digitally operated quadrupoles, isolation with purely duty cycle enhanced waveforms was developed with three increasingly sophisticated isolation methods. First, the guide was used as a gas-filled transmission mass filter using the waveform duty cycle to generate a narrow mass window. The second method used broadband trapping to collect ions and translationally cool along the

1 ACS Paragon Plus Environment


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 2 of 21

transmission axis before shifting the duty cycle to filter the trapped ions. A subsequent duty cycle change axially ejected the filtered population for measurement. The third method improved resolution by shifting the operating frequency during isolation. The resolving power was optimized with the shift frequency to yield a device limited resolving power of 400 (m/∆m). It is the temporal control of the duration of the isolation process that sets digital waveform based isolation apart from the current technology and that minimizes ion loss even when the mass is very large. Preconcentration by repeated trapping and isolation of an individual charge state was also demonstrated to saturate the ion guide with that charge state. These digital isolation and preconcentration technique will permit the same isolation resolution (m/∆m) at any value of mass or m/z without significant ion loss as long as the secular frequencies do not significantly overlap while in the trapping mode. It is therefore ideal for the isolation and preconcentration of single charge states of large proteins and complexes.

Introduction: Because ion traps operate in mass-to-charge ratio (m/z) space, it is natural to suggest that ion isolation depends on the m/z and not the ion mass. Consequently, one would project essentially the same isolation efficiency for the +12 charge state of cyctochrome c (m/z 1033) as singly charged bradykinin (m/z 1061). On the contrary, if one isolated singly charged bradykinin in a 4 mass unit wide window, the ion loss would be minimal whereas if one performed the same 4 mass unit isolation with the +12 charge state of cytochrome c the ion loss would 2 ACS Paragon Plus Environment

Page 3 of 21

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


be substantial and the percent ion loss at the same value of m/z gets worse with increasing charge state. In other words, if one attempted to isolate the 200+ charge state of myosin (MW = 205 kDa) at m/z 1026 with a 4 mass unit window, the ion loss would be 100 %. This is true for any of the methods of ion isolation that are currently available in commercial instruments. Most current commercially available ion isolation techniques are based on sinusoidal waveform technology and are generally performed with an ion trap or a mass filter. For example, in an ion trap, isolation is usually performed by resonance excitation or more rarely by the addition of a DC potential to eject the unwanted ions using the stability boundary. In both cases, the stable ions in the vicinity of the ions being ejected experience a severe reduction in the trapping well depth and the amplitude of their secular motion increases substantially. In mass filters, the narrowing of the mass window also yields a significant reduction in well depth and the range of secular motion also increases.1 These reductions in well depth strongly affect the probability that the stable ions remain trapped. Both narrowing of the mass window and the application of auxiliary waveforms are designed to limit the range of secular motions that can remain in the trap or filter. The secular frequency of ions is defined by the mass-to-charge ratio; however, the bandwidth of secular frequencies (i.e., the range of secular frequencies accessible by the ions of a particular value of m/z) is defined by a number of variables that include well depth, molecular mass or collision cross section, ion density and buffer gas pressure. Increasing the molecular mass, ion density (space charge) or gas pressure broadens the secular frequency distribution 3 ACS Paragon Plus Environment


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 4 of 21

while increasing the well depth yields a decrease.2,3 Consequently, the efficiency of isolation decreases significantly with increasing molecular mass because the ions have more access to unstable motion within the device during the isolation process. During resolved ion isolation with current technology, the ion excitation processes compete with the relaxation processes because the isolation process has to be applied for a relatively long period of time. In other words, the isolation cannot be rapidly switched on for a short duration and then just as rapidly switched off. It is not agile. Digital waveform technology changes that paradigm. Digital waveforms can be instantaneously switched from a broadband trapping waveform to an isolation waveform for as little as one RF period and then switched back. It allows the user to take advantage of the differing rates of excitation on either side of the stability boundary. It means isolation processes can be optimized to reduce loss of the ions of interest during isolation. It is this ability to precisely control the duty cycle, frequency and the duration of the applied waveforms that make the digital waveform isolation process so effective in external traps. In this work we demonstrate digital waveform base ion isolation in a 5 mTorr home built ion guide. The reader is taken through the process of selecting the waveform duty cycle and frequency for precise ion isolation and then shown that the process can be repeated to saturate the linear ion guide with ions of a selected charge state. It is our contention that this process can be used to quickly populate high resolution mass analyzers with ten million charges in only one ion 4 ACS Paragon Plus Environment

Page 5 of 21

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


charge state and facilitate the analysis of very large proteins. A comparison to the current technology will be presented. Methods: All spectra were taken with a custom-built digitally-driven quadrupole ion guide. The instrument is schematically depicted in Figure 1. It consists of a 100 µm diameter flow limiting orifice and a plenum chamber with a cone to help aerodynamically focus the ions passing through to exit

Figure 1: Instrument schematic of the digitally-driven gas-filled linear ion guide

into the quadrupole ion guide for analysis. Ions exiting the guide were detected with a 5 kV conversion dynode and a Channeltron electron multiplier for mass spectral measurements. The ion guide consisted of four 2.54 cm diameter, 15.24 cm long rods with a 1.11 cm radius. The insulators were stock polymer spacers. The rods and spacers were not precision ground and the rods were also not shimmed to adjust the spacing. Slanted wire electrodes were inserted along the asymptotes of the ion guide to create a field along the quadrupole axis.4 The field created by applying a negative potential to the wire electrodes pulled the ions toward the exit even when the 5 mTorr buffer gas (air) has completely damped out the expansion-induced forward motion.5



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 6 of 21

The sample used in all experiments was cytochrome C (Sigma Aldrich, St. Louis, MO). A 5.0 µM solution of cytochrome C in 50% acetonitrile, 50% water and 1% acetic acid (Fisher Chemical, Fair Lawn, NJ) was electrosprayed through a 100 µm ID fused silica capillary (SGE Inc., San Diego, CA). The voltage drop from the capillary to the inlet was held constant at +2800V throughout all experiments. The electrosprayed solution was pneumatically driven with 6 psi filtered air to ensure a constant flow rate. The high voltage waveforms used to drive the guide were produced by first creating low voltage waveforms and then amplifying them with high voltage pulsers.6 The low voltage waveforms were produced with an externally clocked field programmable gate array (FPGA, National Instruments, PXIe-7961R with adapter module NI 6581) operated with an in house written LabVIEW program. The clock was provided by a 40 MHz function generator with 0.355 ߤHz frequency resolution (National Instruments, PXI-5406).7 The low voltage waveform was used to gate two high voltage pulsers (Directed Energy Inc., Model PVX 4150) to create the high voltage waveform applied to the rods. The voltages applied to both rod sets were defined by two high voltage DC power supplies (Matsusada, Model AU-1R300) connected to the pulsers. The electrode voltages were a constant 250 V0-P throughout this work. There was no DC potential applied between the rod sets. Slanted wires inserted between the electrodes were held at a constant potential of approximately -250 V. The exit end cap was held at a ground potential when the guide was operated as a trap; however, a -25 V potential was applied when it was operated as a mass filter. When not operated as 6 ACS Paragon Plus Environment

Page 7 of 21

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


a mass filter, the entrance end cap electrode was toggled between 0 V and 18 V to gate ions into the quadrupole after the sampling period ended so that all ions in the guide were translationally cool before digital filtering was applied. The detector was a shielded electron multiplier (DeTech, Model 2300) placed approximately 2 cm from the exit endcap. Results and Discussion: Initially the quadrupole was set up for operation as a mass filter with a constant 60.5/0/39.5 waveform to define the mass window. The stability diagram for this waveform is shown in Figure 2 (a). No axial trapping or ejection was used when the guide was operated as a mass filter. The mass was scanned by stepping through a frequency list that yielded constant m/z increments. The duration of ion transmission or ion trapping at each mass step was always the same for all mass isolation methods discussed. This approach

Figure 2: (a) the duty cycle base stability diagram that was used to create a mass filter. (b) The spectrum generated using the duty cycle based isolation procedure in a gas-filled (5 mTorr) quadrupole. Ions in the green region are stable

on both axes, blue are stable along x only and yellow are stable along y only.

is directly comparable to ramping the RF voltage of a traditional mass filter operated near the apex of the stability diagram. If one looks closely at the stability diagram in Figure 2 (a), it can be seen that the mass window is very large as defined by the vertical width of the green 7 ACS Paragon Plus Environment


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 8 of 21

stability region at any fixed frequency. This width is about 100 units in the vicinity of 180 kHz. This selection was made for two reasons. First, linear quadrupoles are not generally used for ion selection in the presence of a buffer gas because ion transmission is usually minimal when the mass window is narrowed.1 As evidenced by the signal-to-noise ratio observed in Figure 2 (b), the mass window had to be this wide in order to achieve appreciable transmission. Increasing the difference between t1 and t3 to narrow the mass window yielded no recognizable transmission. The poor transmission through such a large mass window illustrates the need to cool the radial motion of the ions before the isolation process begins. Narrowing the mass window with the duty cycle greatly reduces the trapping well depth.1 When entering the quadrupole, the ions generally have significant components of velocity along the radial axes that can overcome the shallow radial trapping well even without the presence of a buffer gas. With the presence of a buffer gas, the probability of escaping the radial trapping well during injection is exacerbated. Even though the ion throughput is poor, the baseline width of the peaks correlate well with the mass widths defined by the stability diagram in Figure 2 (a). Moreover, the obtained resolving power is on a par with that obtained by Fenn et al.8 for ESI of intact proteins analyzed with a mass filter in vacuum. Analysis of the resolving power and sensitivity of Fenn’s mass spectra8 shows that the resolving power dropped significantly with increasing molecular mass and illustrates that a mass filter does not provide a good method of ion preselection for large intact proteins.


Page 9 of 21

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


The above result revealed the necessity of trapping and collisionally cooling the ions before isolation is performed. Fortunately, the duty cycle can also be manipulated to turn an ordinary transmission ion guide into a linear ion trap without adding a DC potential to the entrance and exit plates of the quadrupole.9 The advantage of digitally driven traps and guides is the ability to switch the waveform duty cycle and/or frequency on the microsecond timescale. This ability allows the ions to be axially collected in the quadrupole under broadband trapping conditions that maximize ion capture efficiency. After the ions settle to the central axis of the quadrupole, the duty cycle can be instantaneously switched to narrow the range of radial stability and destabilize the ions beyond the newly created stable mass window. This process has been dubbed “Trap and Isolate” or TI by the members of our laboratory.



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 10 of 21

The process is illustrated by the stability diagrams in Figure 3 (a) and (b). Figure 3 (a) reveals the m/z, F stability diagram for a 45/10/45 waveform. Because t1 and t3 are equal, there is no high mass cutoff (the green stable region is only bounded on the low mass/low frequency side). For 10 % of the waveform cycle, both rod sets are in the low potential state. Because the rods were operated between ± 250 V and t2 was set to 10%, a time averaged axial well depth of -25 V relative to the endcaps was created to axially trap the incoming ions with high efficiency in the 5 mTorr buffer gas. Increasing the duration of t2 stretches the q, a stability diagram along q (compare Figure S2 (a) and (b) for an example). In the m/z, F stability diagram, it has the effect of slightly shifting the low mass cut off to lower m/z and frequency values.10 The ions were collected for 200 ms

Figure 3: (a) Stability diagrams used for broadband trapping and cooling (b) the stability diagram used for mass isolation in a broad window. It illustrates the narrowing of the mass window part of the isolation procedure. (c) The mass spectrum that illustrates the efficacy of the TI isolation procedure.

and allowed to cool for 500 µs after the ion


Page 11 of 21

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


collection was halted by application of an 18V potential to the entrance end cap electrode of the quadrupole. The isolation waveform (60.5/0.5/39) was then applied for 200 RF periods (approximately 1 ms) to eject the ions outside of the boundary. The stability diagram for this waveform is shown in Figure 3 (b). The difference between t1 and t3 is slightly greater than in Figure 3 (a) and so the mass window is narrower at approximately 66 units across the spectral range. After the isolation waveform was applied, the waveform was switched back to 45/10/45 to let the ions cool for another 500 µs before the isolation waveform was applied a second time. The waveform was then switched to 7.5/85/7.5 for prompt axial ejection of the ions with an averaged potential of 212 V (250 x 0.85 = 212 V) out of the quadrupole and into the awaiting detector. The mass spectrum from this isolation method shown in Figure 3 (c) was also created by stepping through the same list of frequencies that provides consistent mass steps in Figure 2 (b). The baseline width of the charge state peaks is consistent with the width of the waveform generated stability window. It should be noted that even though the stability window is narrower, the signal-to-noise ratio is significantly greater when the ions are trapped and cooled before the isolation process is applied. Our current low voltage waveform generator limits the ability of the duty cycle alone to set the width of the stable window. Duty cycle resolution in the current system is frequency dependent.7 In the range of multiply charged cytochrome C, the duty cycle resolution is on the order of 0.1 %. Even so, duty cycle control improves as the waveform period increases. Shifting the duty cycle 11 ACS Paragon Plus Environment


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 12 of 21

alone will not produce a very good mass filter because the duty cycle resolution at this time is regrettably insufficient.11 In contrast, frequency is effectively a continuous variable that may be specified with microhertz precision even at high frequencies. Fortunately, the frequency and duty cycle can be rapidly jumped to provide the needed mass resolution. Applying the duty cycle defined mass window at two separate frequencies sequentially results in a net mass window that is defined by the overlap of the two. This procedure is illustrated in Figure 4 (a). After the 200 ms collection followed by the 500 µs cooling step, the 60.5/0.5/39 isolation waveform at frequency F1 is applied for 200 RF periods. The mass window created at F1 is projected onto the m/z axis as shown by the blue lines. The ions are then cooled for another 500 µs

Figure 4: (a) Stability diagram illustrating the Jump procedure used to precisely narrow the range of masses trapped. (b) Mass spectra of multiply charged cytochrome C for a gas-filled digitally operated ion guide optimized for TIFS isolation. (c) Enhancement of the signal intensity observed by increasing the collection time by a factor of 2.5.

before applying the isolation waveform at frequency F2 for another 200 RF periods. The ions were then promptly ejected 12 ACS Paragon Plus Environment

Page 13 of 21

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


with the 7.5/85/7.5 duty cycle. The frequency difference between F1 and F2 was slowly increased until the resolution stopped changing and the signal intensity remained high. As above, a spectrum was generated stepping F1 and F2 through a frequency list generating consistent m/z steps. The spectrum with a 200 ms collection time is shown in Figure 4 (b). This ion isolation process has been dubbed “Trap, Isolate and Frequency Shift” or TIFS by our group. Each point in the spectrum is an individual separation from the broad band of ions trapped. It reveals a resolving power of approximately 400 (R = m/∆m). Under our instrumental conditions, there is approximately 100 Hz between m/z = 1000 and 1001. Because the frequency can be defined with microhertz precision, the mass window can be frequency adjusted at the 10 parts per billion level within the one unit window. These high resolution mass windows cannot be achieved with the current setup because of contributions of other errors in the system such as ripple in the DC power supplies in the part per thousand range and the variation in the value of quadrupole radius, r0, along the length of the ion guide. However, these variances can be significantly reduced. For example, a quadrupole that has been designed and tested as a mass filter could be used instead of our home built ion guide. Low ripple DC power are also available on the market. Consequently, the 400 resolving power achieved could be greatly improved. Figures 4 (b) and (c) show that the intensity of our spectra is defined by the trapping time and the rate of ion generation. Figure 4 (c) reveals a significant improvement in the signal-to-noise ratio just by increasing the trapping time to 13 ACS Paragon Plus Environment


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 14 of 21

500 ms without loss of resolution. Each of these points in the mass spectrum is a separate isolation inside of the linear ion guide with a mass window of about 2.5 Da at m/z 1000 (R = 400). Consequently, our isolation process will increase the concentration of the ions of interest in the quadrupole just by increasing the trapping time. However, there is a limited number of charges that an ion trap will hold. The trick to ensuring that most of them are the ions of interest in a very narrow range is to rapidly execute multiple brief TIFS operations before ejection. To demonstrate this concept, the integrated detector signal was measured as a function of the number of 10 ms trapping periods followed by the described isolation cycle

Figure 5: Integrated detector response as a function of iterations of TIFS isolation of the z = 9 charge state of cytochrome C before ejection. Data shows that it is possible to fill the quad to saturation (greater than 10 million charges estimated at saturation) with a single charge state.

(Figure 5). The isolated ion signal continually increases and then levels out indicating that the quadrupole has reached saturation after about 400 ms. Subsequent to this investigation, the instrument was equipped with a Faraday detector that revealed that 10 million charges are promptly ejected from the charge-saturated quadrupole in a 200 µs pulse. The trap can hold much larger quantities of charge, but it can only hold about 10 million charges near the exit where the fields from the end cap electrode penetrate far enough into the trap to promptly eject the ions when the duty cycle is switched. Overfilling the trap is represented by the same intense flux of ions


Page 15 of 21

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


when the waveform is switched followed by a lower intensity tail made up of the ions that have to move from deeper in the quadrupole to the point where the endcap potential from the switched waveform catches the ions and pulls them out of the trap. Increasing the number of ions injected into any mass analyzer has a marked effect on the signal to noise. For example, the scattering based noise in time-of-flight is generally a single ion event that can occur at any time whereas the TOF ion signal is coherent. When tens of thousands of ions make up the coherent signal whose amplitude is defined by an eight bit digitizer, the incoherent single ion noise does not register even at the one bit level. Therefore, trapping and concentrating the ions vastly improves the signal to noise of TOF spectra. Comparison to Existing Technology At this point it is reasonable to ask why sinusoidal waveform technology cannot rapidly isolate large ions in a narrow mass range without significant ion loss. First, the majority of ion isolation techniques in traps currently in use are based on resonant excitation. Resonance based isolation techniques using either sinusoidal or digital waveform technology can be used for unit ion isolation in the m/z range where the massive intact ions would be.12,13 However, they do not yield efficient ion isolation because resonance techniques are perturbative. Low voltage waveforms are used to excite ions to eject out of high voltage trapping potentials. The low voltage isolation waveform has to be applied for a sufficiently long time to radially excite the unwanted ions out of the trap. In doing so, they 15 ACS Paragon Plus Environment


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 16 of 21

also radially excite the bound ions in the vicinity of the ejected ions. Radial excitation of the bound ions reduces their binding in the trap (well depth). It also increases the spread of the secular frequency bandwidth of the bound ions making it easier for them to escape the trap. Moreover, the collisional damping experienced by large ions in a gas-filled trap broadens the bandwidth of each ion’s secular frequency. Decreasing the well depth exacerbates this phenomenon and makes ion isolation exceedingly inefficient with increasing molecular mass. The net effect is a lowering of the charge capacity for the isolated ions, making it difficult to fill the trap to its normal capacity. The techniques presented here use the stability boundary to perform the isolation. Although it is not typically used anymore in commercial instruments, sinusoidal technology is also capable of performing isolation using the boundary by similarly cooling the ions and then translating the ions to be isolated to the boundary of the stability diagram by applying a DC potential and thereby performing boundary ejection. Digital waveform technology does something similar. It moves the boundary and changes the frequency to situate the wanted ions just inside the boundary. Physically both techniques are the same, but they operate on dramatically different timescales. Digital isolation waveforms can be applied for as little as one RF period. Consequently, the duration of the isolation waveform can be optimized so that the unwanted ions are ejected without significantly reducing the population of wanted ions. This agility permits this new technology to take advantage of the different rates of excitation. Ions pushed beyond the boundary rapidly gain kinetic energy and eject from the trap within a


Page 17 of 21

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


few RF periods while ions inside the stable region gain energy relatively slowly. The criterion for efficient ion isolation in this case is that the band of secular frequencies of the ions to be isolated cannot significantly overlap with the ions to be ejected while in the deeper trapping well. The sudden switch to the isolation waveform allows the ions to be excited at dramatically different rates. The ions to be ejected can be separately placed beyond the boundary by precise duty cycle and frequency manipulation as previously described and illustrated while the ions to be kept remain within the boundary where they radially excite at a much slower rate. Consequently, any ions that can be baseline separated from the unwanted ions within a mass spectrum can be efficiently separated. These different rates of excitation apply at any value of m or m/z because the rate of excitation for an ion moving from a bound state to an unbound state will always be much greater than an ion moving from a bound state to another bound state. The agility of this isolation methods will always be able to take advantage of this difference; therefore, this method of isolation is essentially mass independent. Digital waveform technology can instantaneously jump the ions back to the broadband trapping conditions where the pseudopotential well is deep, the ion secular frequencies are better defined and the ion capacity is large before the isolated ions can sufficiently excite thereby preventing excess population from seeping out of the trap. Because this technique efficiently isolates large ions in very narrow mass windows with 400 resolving power, it should be possible to efficiently isolate and preconcentrate individual charge states of proteins and complexes with molecular masses beyond 200 kDa.



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 18 of 21

Make no mistake, when the isolation waveform with a 66 mass unit wide window is applied to even cytochrome c without switching the trapping conditions back to broad band trapping, the isolated ion population diminished because the isolated ions were radially excite as well. In fact a loss of almost the entirety of the isolated ion population within 100 ms was observed. Since sinusoidal technology can also perform boundary based isolation, it conceivably could yield the same efficient isolation if the isolation could be applied for approximately ~1 ms as we do with our technique. Unfortunately, current sinusoidal technology is not that agile. It requires a DC power supply to create a potential between the electrodes. The typical rise time of a high voltage DC power supply is on the order of 100 ms. Moreover, settling time (the time it takes to stabilize within some fixed percentage of the specified value) can be even longer. Sinusoidal technology in its current state cannot precisely and rapidly apply the DC potential necessary to create a 1 ms isolation. Reduced ion populations are inevitable using the DC/RF sinusoidal excitation. Thus far, only digital waveform technology has sufficient agility to isolate large ions in a narrow mass window without significant loss of ion population. Conclusion: An in house constructed linear quadrupole ion guide used digital waveform manipulation to demonstrate an isolation resolving power of 400 (m/∆m) in the presence of a buffer gas at 5 mTorr. The ability to precisely control the frequency and duration of the isolation waveform permits precise isolation while minimizing the ion loss that plagues mass filter and resonant dipolar isolation techniques. Ions 18 ACS Paragon Plus Environment

Page 19 of 21

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


in the 9+ charge state of cytochrome C were isolated with 400 resolving power and concentrated to charge saturation in the ion guide. Subsequent work by our group with digitally operated ion guides and Faraday detection revealed a saturation charge capacity of 10 million. This isolation technique requires the agility of digital waveform technology because it takes advantage of the different rates of excitation that occur when the system is instantaneously switched to an isolation waveform. The ions in the unstable regions of a,q space rapidly eject while those in the stable region slowly excite radially. The agility allows the waveform to be switched back to a broadband trapping waveform with a deep potential well before the isolated ions can significantly excite. This method will greatly extend the range of molecular masses that can be efficiently isolated because those different rates of excitation will exist at any value of mass or m/z and digital waveform technology will always be able to take advantage of them. Individual charge states of any protein and most complexes should now be amenable to efficient isolation and preconcentration to the capacity of an external trap for detection by any downstream mass analyzer. Acknowledgements: This work was supported by the Defense Threat Reduction Agency, Basic Research Award # HDTRA1-12-0015, to Washington State University and the National Science Foundation Award No. 1352780. Supporting Information Available:



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 20 of 21

The effects of changing the duty cycle on the stability diagrams is illustrated in the supplementary information section as is the definition of the nomenclature that defines the waveforms.

References: (1) Brabeck, G. F.; Reilly, P. T. A. J. Am. Soc. Mass Spectrom.2016, 27, 1122-1127. (2) Zhao, X. Z.; Douglas, D. J. J. Mass Spectrom. 2008, 275, 91-103. (3) Xu, W.; Song, Q.; Smith, S. A.; Chappell, W. J.; Ouyang, Z. J. Am. Soc. Mass Spectrom. 2009, 20, 21442153. (4) Wilcox, B. E.; Hendrickson, C. L.; Marshall, A. G. J. Am. Soc. Mass Spectrom. 2002, 13, 1304-1312. (5) Wang, X.; Chen, H.; Lee, J.; Reilly, P. T. A. Int. J. Mass Spectrom. 2012, 328, 28-35. (6) Koizumi, H.; Jatko, B.; Andrews, W. H.; Whitten, W. B.; Reilly, P. T. A. Int. J. Mass Spectrom. 2010, 292, 23-31. (7) Brabeck, G. F.; Reilly P. T. A.; Wang, L. A. Proceedings of the 63rd ASMS Conference on Mass Spectrometry and Allied Topics, St. Louis, MO, May 31 - June 4, 2015; American Society for Mass Spectrometry; 271004

(8) Fenn, J. B.; Mann, M.; Meng, C. K.; Wong, S. F.; Whitehouse, C. M. Mass Spectrom. Rev. 1990, 9, 3770. (9) Lee, J.; Marino, M. A.; Koizumi, H.; Reilly, P. T. A. Int. J. Mass Spectrom. 2011, 304, 36-40. (10) Brabeck, G. F.; Reilly, P. T. A. Int. J. Mass Spectrom. 2014, 364, 1-8. (11) Brancia, F. L.; McCullough, B.; Entwistle, A.; Grossmann, J. G.; Ding, L. J. Am. Soc. Mass Spectrom. 2010, 21, 1530-1533. (12) Ding, L.; Sudakov, M.; Brancia, F. L.; Giles, R.; Kumashiro, S. J. Mass Spectrom. 2004, 39, 471-484. (13) Schwartz, J. C.; Senko, M. W.; Syka, J. E. P. J. Am. Soc. Mass Spectrom. 2002, 13, 659-669.


Page 21 of 21

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


319x183mm (150 x 150 DPI)

ACS Paragon Plus Environment

Methodology and Characterization of Isolation and Preconcentration

Recommend Documents