Design and performance of an atmospheric pressure ion mobility Fourier transform ion cyclotron resonance mass spectrometer



This manuscript presents our initial results and development of a novel hybrid instrument that combines atmospheric pressure ion mobility spectrometry (AP-IMS) with Fourier transform ion cyclotron resonance mass spectrometry (FTICR-MS). Our preliminary results obtained from atmospheric pressure mobility separation of peptide mixtures combined with high-resolution FTICR mass analysis are demonstrated. The custom IMS system was constructed in-house and was coupled to the commercial FTICR-MS instrument through a flared inlet capillary interface. Dual-gate ion filtration was adapted to allow concurrent measurement of both mobility and m/z values. The feasibility of mobility separation was demonstrated with baseline separation of the peptides bradykinin and angiotensin II and their measured reduced mobility constants which were consistent with those previously reported. Furthermore, the unique size-to-charge separation mechanism of IMS that allows isomer separation was explored and demonstrated with the partial separation of two isomeric phosphopeptides. We feel the combination of IMS and FTICR-MS holds great potential for accurate mass analysis of mobility-selected ions and these results are the first to demonstrate the feasibility of coupling these two techniques. Copyright © 2007 John Wiley & Sons, Ltd.

Since its inception in 1970,1, 2 ion mobility spectrometry (IMS) has been combined with a variety of other analytical technologies and employed for detection of a wide range of analytes including both small organic compounds3–6 and large biological molecules.7–13 IMS separates ions in the gas phase on the basis of their differential mobility under a uniform weak electric field. It is of great interest in many analytical fields due to the high speed with which this separation technique can be achieved. In its early developments, IMS was termed “plasma chromatography”1, 2 because of its gas-phase separation properties that are analogous to conventional separation techniques such as gas chromatography (GC) and liquid chromatography (LC). In fact, IMS shares similarities to both chromatography and mass spectrometry (MS) but is based on principles different from both, thus it can be coupled to either GC14 and LC15 as a detector or MS as a supplementary separation device.

While standalone IMS with a Faraday plate detector offers rapid and sensitive detection in real-time with low-cost and field-deployable benefits, coupling IMS with MS has recently attracted increased interest due to additional qualitative and specific fragmentation information that can be gained with MS and MS/MS. One important feature that distinguishes IMS from MS is that IMS separates ions based on their size-to-charge ratio (Ω/z, where Ω is cross section) whereas MS measures ions based on their mass-to-charge ratio (m/z). IMS is not only a function of size but also shape and ion-neutral potential interactions. This Ω/z-based separation mechanism of IMS enables separation of isomers since they can have different conformation in the gas phase even though they have the same m/z. Thus combining IMS with MS produces more comprehensive information than is possible with either technology alone. In addition, as a separation technique, IMS is more advantageous for analysis of biological samples owing to the orders of magnitude faster separation (on ms scale) than typical condensed-phase chromatography such as LC, capillary electrophoresis, or gel electrophoresis.

IMS can be operated at either atmospheric pressure or low vacuum pressure (1–10 Torr) and both types of IMS have been coupled with various types of mass spectrometers,16 such as quadrupole MS,3, 17–21 time-of-flight (TOF)-MS,7, 8, 22, 23 and ion trap MS.12, 24–26 In addition, low-pressure IMS has been coupled to Fourier transform ion cyclotron resonance (FTICR)-MS.27 To synchronize the acquisition cycle of IMS and MS, most two-dimensional IMS-MS measurements are achieved by either acquiring IMS spectra for a single m/z value or scanning the m/z window for a selected drift time window, with the exception of TOF-MS which is able to measure a wide window of m/z values simultaneously on a time scale compatible with on-line IMS. The IMS-TOF-MS combination allows hundreds of MS spectra to be recorded for elution of a selected IMS peak, which was called a ‘nested’ technique.7, 28 However, given the relatively slow data acquisition nature of most other MS instruments, Clemmer and coworkers24 and Creaser et al.25 have employed a reversed sequence in the hybrid IMS-MS design. In their setup, a quadrupole ion trap was used as an ion storage and concentration device to collect ions from a continuous ionization source (electrospray ionization (ESI)) and then pulses of ions were injected to drift cell for mobility measurement. Russell and coworkers27 also employed a reversed sequence in their hybrid FTICR-MS-IMS instrument with the ICR cell placed prior to the drift cell. In their novel design, the ICR cell was used as a device to study ion-molecule reactions and TOF-MS was the actual mass analyzer for IMS. In all injection types of MS-IMS hybrids reported thus far, low-pressure IMS was used in part to limit the dramatic pressure difference between the MS region and IMS region. This type of hybrid construction typically requires major instrumental modifications that impose significant technological challenges, such as differential pumping and aperture alignment.27 Conversely, atmospheric pressure (AP)-IMS can be readily coupled with an atmospheric pressure ionization source and interfaced in front of a mass spectrometer with minimal or no modifications to the instrument. Recently Clowers and Hill26 have implemented a union of an atmospheric pressure IMS instrument that uses dual-gate selection with quadrupole ion trap MS without any modifications to the mass spectrometer. The dual-gate design reduces the speed of mobility measurement but allows synchronization of the IMS and MS acquisition cycles. In this manuscript, a novel hybrid instrument of IM-FTICR-MS is reported that adapts the dual-gate IMS design and further includes a flared inlet capillary reported by Bruce and coworkers29 to enhance ion transfer efficiency at the interface. This report will highlight the construction of this hybrid instrument and present the preliminary results that demonstrate high-resolution mass analysis and AP-IMS separation of standard peptide mixtures and isomeric phosphopeptides.


A schematic diagram of the hybrid nano-ESI-AP-IM-FTICR-MS instrument is presented in Fig. 1. The basic components of this hybrid instrument consist of (i) a nano-ESI source; (ii) an AP-IMS drift tube; (iii) a flared inlet capillary interface; and (iv) a FTICR mass spectrometer. Each module is described in detail below.

Figure 1.

Schematic of the electrospray ionization atmospheric pressure ion mobility-Fourier transform ion cyclotron resonance mass spectrometry (ESI-AP-IM-FTICR-MS) instrument.

Nano-ESI source

ESI nanospray emitters used for this research were made by etching fused-silica capillary (360 µm o.d. × 20 µm i.d.; Polymicro Technologies, Phoenix, AZ, USA) in 49% HF solution as previously described.30 The emitter was then trimmed to a length of ∼5 cm. The electrical potential used to produce ESI was applied through a stainless steel zero dead volume union (Upchurch Scientific, Oak Harbor, WA, USA) that connected the emitter with the fused-silica capillary transfer line (360 µm o.d. × 75 µm i.d., ∼30 cm long).

Atmospheric pressure ion mobility spectrometer

The atmospheric pressure ion mobility spectrometer used for this research was constructed at Washington State University, Pullman, WA, USA. The drift tube was built upon a standard stacked ring configuration by assembling repeating units. Each repeating unit, composed of a conductive stainless steel ring with a dimension of 50 mm (o.d.) × 48 mm (i.d.) × 3 mm (width) and an insulating ceramic ring with a dimension of 60 mm (o.d.) × 50 mm (i.d.) × 4.5 mm (width), was stacked sequentially in an alumina tube. Stainless steel rings, also called guard rings, were connected via a series of half (desolvation region) and one (drift region) megaohm resistors (Caddock Electronics Inc., Riverside, CA, USA). The ceramic ring served to isolate the guard rings from the alumina tube as well as from each other. The current instrument consists of 42 repeating units with a total length of 34 cm, but the length of the drift tube can be easily shortened or lengthened by removing or adding a number of rings. Two Bradbury-Nielsen gate rings divided the drift tube into three regions: 7.5-cm long desolvation region, 26.5-cm long drift region and ∼1-cm long interface region. 10 kV was normally applied to the first ring electrode and the last ring electrode voltage was adjusted by variable resistor to be ∼200 V referenced to ground. The drift voltage was dropped gradually across the drift tube via the resistor chain to form an electric field of 157 V/cm in the desolvation region and 333 V/cm in the drift region. The lower electric field in the desolvation region allowed solvated ions to spend more time in the heated drift gas to promote more efficient desolvation prior to injection to drift region. The two Bradbury-Nielsen gates were made of electrically isolated alternating parallel Alloy 46 wires (76 µm in diameter) (California Fine Wire Co., Grover Beach, CA, USA) spaced 0.65 mm apart. When the potentials, applied on alternating wires, were the same as the reference potential, the gate was ‘open’ to allow ions to pass through; while the potentials on the adjacent wires were offset ±50 V with the respect to the reference potential, an electric field of ∼1500 V/cm was created orthogonal to the drift field and the gate was ‘closed’ to shut off ion transmission.

The drift tube oven was constructed from two pieces of 20-cm long aluminum cylinder with three heating cartridges (Heatcon, Seattle, WA, USA) embedded inside each cylinder. The drift tube was normally heated at 150°C. All mobility experiments were carried out at atmospheric pressure which typically ranged from 690 to 710 Torr at Washington State University.

Flared inlet capillary interface

Bruce and coworkers29 recently reported that a 2–5-fold signal gain can be achieved by replacing the original glass capillary with a flared inlet metal capillary in an FTICR-MS instrument. To improve ion transmission from the IMS tube to the mass spectrometer, this flared tube interface was adapted for the current hybrid instrument without any further modification. In particular, the flared inlet tube used in this research had a dimension of 0.015″ i.d. and 0.03″ o.d. at the unflared end and was flared with a 45° angle from the center line to ∼2.2 mm o.d. at other end. The length of the tube is 10″. The inlet tube was housed in a ¼″ stainless steel tube which served as an interchangeable adaptor with the original glass capillary tube.


The mass spectrometer used for the hybrid instrumentation is a Bruker Apex-Q 7T FTICR mass spectrometer (Bruker Daltonics, Billerica, MA, USA). This instrument consists of a Q-interface: a front hexapole (h1), a mass-selective quadrupole filter (Q), and a second hexapole (h2), and other associated ion transfer optics prior to the ICR cell. h2 can be effectively used as both an ion storage trap and a collision cell. Ions formed by the nano-ESI source entered the instrument through the flared inlet capillary and then passed through h1 and Q followed by accumulation in h2 for a duration of 1–2 s. Ions were then injected into the ICR cell through an electrostatic ion guide. Xmass 7.0.6 software program was used to acquire all mass spectra. All datasets were acquired with 256K points.

Instrument operating modes

Continuous mode

In this mode, both ion gates were open and ions formed at the ESI source were continuously transferred to the MS instrument. This mode was normally used for tuning the instrument for optimal signal.

Selected mobility monitoring

In this mode of operation, the first ion gate was set at a certain pulse width, e.g., 0.5 ms, while the second gate was open only in the user-defined scan window at selected delay times, e.g., 25–35 ms, which means only those ions with drift time between 25 and 35 ms would pass gate 2 and enter the mass spectrometer. Therefore, this mode of operation can be used to narrow down the mobility scanning window for the analyte ions of interest in a short period of time.

Mobility scanning dual-gate mode

Mobility scanning mode sets both gates at a certain pulse width (e.g., 0.5 ms) with a delay applied to gate 2. The scanning of this delay time with defined step increment allows acquisition of the mobility spectrum for the analyte ions. The delay time scan window (start of the delay – end of the delay) can be approximately determined using selected mobility monitoring mode as described above. For example, if the scan window is 25–30 ms and the step increment is 0.1 ms, 50 data points will be obtained for the mobility spectrum.

Acquisition timing sequence

To allow concurrent mobility and m/z measurement for the hybrid IM-FTICR-MS instrument, IMS dual-gate selection events will be ‘nested’ under each MS acquisition cycle. Figure 2 shows an example of a typical acquisition timing sequence. The rising edge of the h2 TTL pulse triggers the timing sequence of dual-gate pulsing with defined delays. We normally set h2 accumulation time at 2 s. Thus, for a delay time of 40 ms, we are able to repeat IMS injections 50 times for each MS spectrum acquisition. It should be noted that the maximum of IMS injection pulses per MS cycle is governed by the end time of the scan window. To enhance signal-to-noise (S/N) ratio, we typically set data averaging of 50–100 for each data point.

Figure 2.

Illustration of timing sequence for IM-FTICR-MS data acquisition. Each MS injection cycle includes a number of repeated IMS dual-gate pulsing cycles. For example, if h2 accumulation time is 2000 ms, gate 2 delay is 40 ms, then the dual-gate filtration events can be repeated 50 times. The maximum number of IMS pulsing repeats is determined by the end of the delay time in the mobility scanning dual-gate mode.

Data analysis

IMS control software was programmed by LabVIEW 6.1 (National Instruments, Austin, TX, USA) and communicated with the gate controller and TTL trigger through an interface board (PCI-6601; National Instruments) installed in a Dell PC computer. More details were described previously.26 FTICR-MS was operated in chromatography mode during mobility scanning dual-gate mode and the recorded chromatogram was reconstructed to obtain the selected-ion chromatogram using the program ICR-2LS31 (version 2.30).

Chemicals and reagents

Acetonitrile used in this study was HPLC grade and obtained from J.T. Baker (Phillipsburgh, NJ, USA). Formic acid and hydrofluoric acid (HF) were obtained from Sigma (St. Louis, MO, USA). Water used in this research was deionized to a resistivity of 18.2 MΩ-cm (prepared by Barnstead Nanopure Water Systems). Standard peptides were purchased from Sigma and used without further purification. Phosphopeptides were made in the Peptide Synthesis Facility at Washington State University. The sequence and m/z information of all the peptides used in this research is presented in Table 1. Typically, a 100 µM solution prepared in 50% acetonitrile and 0.1% formic acid was used to acquire mobility spectra.

Table 1. Sequences, charge states, and reduced mobility constants for five peptides used in this research
Peptide namePeptide sequencem/z [M+H]+m/z [M+2H]2+m/z [M+3H]3+K0 (cm2/V s)d (measured)K0 (cm2/V s)e (literature)
  • a

    The N-terminus of the neurotensin sequence has a pyroglutamate modification.

  • b

    S3P peptide has a phosphorylation modification at serine 3.

  • c

    S5P peptide has a phosphorylation modification at serine 5.

  • d

    The measured K0 is for the ions with m/z underlined.

  • e

    The literature value for K0 was for the ions with m/z underlined. ‘-’ means no K0 values could be found in previous literature.

Angiotensin IIDRVYIHPF1046.54523.77262.391.081.10


Optimizing ion signal with continuous mode

Most common interfaces used with hybrid AP-IMS-MS instruments consist of an unmodified MS inlet, which is either a small pinhole aperture (50–200 um i.d.) or an inlet capillary (0.5 mm i.d. × 15–25 cm long). Ion loss at the interface during IMS-MS can be more than 99%,32 which tremendously limits the sensitivity and has been a bottleneck for wider application of this technique. Smith and coworkers32 recently reported using an ion funnel interface at both front ESI-IMS and rear IMS-QTOF regions to achieve nearly 100% ion transmission efficiency. However, the radio-frequency (rf) ion focusing capability of the ion funnel works only at a relatively low pressure (1–10 Torr) regime.33 Therefore, ion funnel focusing is not suitable for AP-IMS-MS. However, a recently implemented flared inlet capillary for the FTICR-MS instrument demonstrated that the increased size of the entrance aperture offered by flared inlet improves ion transfer efficiency.29 We adapted this flared inlet interface for coupling AP-IMS with FTICR-MS.

With the IMS instrument operating in the continuous mode, we were able to optimize the position of the flared tube entrance relative to the exit of the ion mobility mass spectrometer. The optimal signal was normally achieved when the distance between the second gate and the flared tube entrance was ∼1 cm. With 10 kV applied at the first ring electrode in the IMS tube, the voltage at gate 2 was ∼660 V and the voltage on the last ring electrode located right after gate 2 was ∼200 V. The voltage on the flared inlet capillary could be varied from 40 to 150 V without dramatically changing signal intensity. In addition, we found that ion signals could be further increased ∼20% if we applied 800–1000 V to the IMS exit cover, which was made of aluminum and placed right after the last ring electrode.

The flared inlet tube was heated to 200°C for all experiments. No significant change in ion signals was observed whether the flared inlet tube was heated or not, suggesting that desolvation critical to mass spectral analysis is efficiently achieved in the IMS region prior to the entrance of the mass spectrometer. We further investigated the influence of the drift tube temperature using a peptide mixture of bradykinin, neurotensin, and angiotensin II. The heating temperature of the IMS tube appeared to correlate with observed charge states in the mass spectra. As shown in Fig. 3, the charge states of all peptides consistently shifted to lower values as the temperature rose. In particular, the singly charged neurotensin ions were observed at only 250°C but not the other three lower temperature settings. Li and Cole34 reported that charge states can be selectively controlled by varying the orifice diameter of the nanospray tip, i.e., smaller tip orifice favors highly charged ion formation. For a given precursor molecule, ions of higher charge states are more prone to fragmentation than those of lower charge state.35, 36 For the current purpose of mobility measurement, we chose a heating temperature of 150°C for further experiments since the ions of higher charge states have higher mobility and separate with improved IMS resolution.

Figure 3.

Charge state selection observed by varying the drift tube heating temperature. Data were collected using continuous mode ESI-AP-IM-FTICR-MS analysis of a peptide mixture containing angiotensin II, bradykinin, and neurotensin with the drift tube heating temperature set at (a) 100°C, (b) 150°C, (c) 200°C, and (d) 250°C. Lower charge states of the ions were favored at higher heating temperature for all three peptides. The inset figure in (d) shows that, at 250°C, a weak signal of singly charged neurotensin was observed.

Nano-ESI has become the electrospray technique of choice in most biological mass spectrometry laboratories because of its improved ionization efficiency, ion transfer efficiency, and sensitivity as compared to conventional ESI. We recently reported a discovery that with standalone IMS the observed ion signal increases with increased flow rate and this result was contrary to previous observations in nano-ESI-MS.30 Here, with the AP-IMS system placed in between the ESI source and the MS, we further investigated the flow rate effects with use of a nanospray tip. Different from both ESI-MS and ESI-IMS, the flow rate of 1–2 µL/min seemed to offer the optimal signal for ESI-IMS-MS (Fig. 4). Therefore, the following experiments were performed with a flow rate of 1 µL/min.

Figure 4.

Measured peak intensities for doubly charged bradykinin at m/z 530, triply charged neurotensin at m/z 558, and doubly charged angiotensin II at m/z 523 as a function of infusion flow rates. Each FTICR-MS spectrum was collected with the use of continuous IMS mode, 256K data points and 20 data averages. Each data point is the average of three measurements of peak intensities extracted from FTICR-MS spectra. Error bars indicate the standard deviation calculated from three replicate measurements.

Separation of peptide mixtures

After optimizing the instrument conditions, we evaluated the utility of this hybrid instrument with the analysis of peptide mixtures. The initial attempt was to separate three standard peptide ions, i.e., doubly charged bradykinin (m/z 530), doubly charged angiotensin II (m/z 523), and triply charged neurotensin (m/z 558), with dual-gate filtration experiments. The selected mobility scanning mode was utilized to predict the scan window, which was determined to be 46–51 ms. Then the mobility separation experiments were set up using the mobility scanning dual-gate mode with the following parameters: 2 s h2 accumulation time, 39 dual-gate pulses per h2 accumulation cycle, 100 data averages, 0.2 ms step increment, 12.5 kV nanospray voltage, 10 kV applied to the first ring electrode, 150°C drift tube heating temperature (measured drift space temperature was 125°C), and air used as drift gas. Figure 5 shows the reconstructed mobility spectra of these three ions with gate pulse width at 0.5 ms (Fig. 5(a)) and 0.4 ms (Fig. 5(b)). It should be noted the same pulse width was applied to both gates. As shown in Fig. 5, bradykinin was baseline-resolved from neurotensin and angiotensin II; neurotensin and angiotensin II were partially resolved. A pulse width of 0.4 ms showed slightly better resolution, but further decrease in gate pulse width resulted in significantly reduced ion signals. Given the nature of dual-gate mobility experiments, the effective drift length (L) and drift voltage (V) are the distance and potential difference between the first gate and the second gate, respectively, since once ions exit the second gate, the drift time measurement is completed. We calculated the reduced mobility constants (K0) for these three ions using the equation:

equation image

with L ∼26.5 cm, V 8.3 kV, T (temperature) 398 K, and P (pressure) 700 Torr. The measured K0 values for all three ions are listed in Table 1. The measured K0 values for doubly charged bradykinin and angiotensin II ions in this study were 1.12 and 1.08, respectively, which were within 98% of reported literature values.15

Figure 5.

ESI-AP-IM-FTICR-MS analysis of standard peptide mixtures using mobility scanning dual-gate mode with (a) 0.5 ms and (b) 0.4 ms gate pulse width. Dotted line represents the reconstructed selected ion mobility spectrum of [bradykinin+2H]2+ at m/z 530.8; solid line represents the reconstructed mobility spectrum of [neurotensin+3H]3+ at m/z 558.7; and dashed line represents the reconstructed mobility spectrum of [angiotensin II+2H]2+ at m/z 523.8. Inset figures show examples of dual-gate filtered FTICR-MS spectra for bradykinin, neurotensin, and angiotensin II.

One primary advantage of IMS is its capability to separate isomeric compounds, which is unattainable for FTICR-MS in spite of its superior resolving power. Tandem mass spectrometry (MS/MS) fragmentation experiments are often necessary for distinguishing these isomeric compounds. However, LC/MS/MS experiments will not be able to resolve the differences and make correct assignments of fragment ions if two isomers coelute, which is unfortunately true in many cases because of high structural similarities of isomers. Thus, we took further efforts to look into the isomer separation capability of this hybrid instrument. The isomers we chose for this experiment were two peptides with the same sequence but different phosphorylation sites (see Table 1). Protein and peptide phosphorylation is a type of posttranslational modification with great significance in many biological systems, since site-specific phosphorylation can have dramatic effects on system functionality.37 The mobility scanning dual-gate mode was applied to acquire the mobility spectra for each peptide individually and mixtures of these two peptides. The applied parameters for this set of experiments included 2 s h2 accumulation time, 47 dual-gate pulses per h2 cycle, 100 data averages, 37–42 ms scan window, 0.5 ms gate pulse width, 13 kV nanospray voltage, 11.8 kV first ring electrode voltage, drift temperature 125°C, and air as drift gas. The effective drift voltage was 9.7 kV. The resulting mobility spectra for each individual peptide and mixtures of two peptides are presented in Fig. 6, which shows partial separation of the two isomeric phosphopeptides. The calculated K0 values for these two peptides are provided in Table 1. Due to the limited instrument sensitivity, further decrease in gate pulse width resulted in no mobility signal. Therefore, additional improvement of ion transfer efficiency and duty cycle will be essential for achieving baseline separation of these two isomeric phosphopeptides with the use of 0.2 ms or lower gate pulse width.

Figure 6.

ESI-AP-IM-FTICR-MS analysis of isomeric phosphopeptides with gate pulse width at 0.5 ms. Shown is the reconstructed selected ion mobility spectrum of doubly charged ion at m/z 459.7. Dashed line represents the mobility spectrum collected with S3P alone; dotted line represents the spectrum collected with S5P alone; and solid line represents the mobility spectrum collected with a mixture of S3P and S5P. Inset figure shows an example FTICR-MS spectrum of doubly charged peptide ion at m/z 459.7 after dual-gate selection.


In this report, we presented the initial results from the combination of AP-IMS with FTICR-MS which allowed two-dimensional mobility and high mass resolution m/z measurements without major instrument modification. We found the charge states of peptide ions were tunable with the control of drift tube heating temperature, which could be beneficial when selection of a specific charge state is needed. A nanospray emitter was employed; however, in contrast to nano-ESI-MS which favors low flow rate, i.e., 50–200 nL/min, the optimal signal was obtained with a relatively high flow rate, i.e., 1 µL/min. The feasibility of combined mobility separation and high-resolution mass measurements by FTICR-MS with the use of this hybrid instrument was demonstrated with standard peptide mixtures. In addition, the capability of separating isomeric phosphopeptides with IMS was demonstrated for the first time. Due to the limited instrument sensitivity, a relatively large gate pulse width (0.5 ms) was used. Further improvement of the instrument sensitivity will allow use of smaller gate pulse width and thus increase the IMS resolution. Finally, though a flared inlet capillary was employed as interface for signal improvement, further enhancement in ion transfer and duty cycle will be critical for applying this technique to the analysis of biological samples in limited quantity.


This research was supported by grants from NIH (Grant No. R21 DK070274); National Science Foundation (Grant No. DBI-0352451); US Department of Energy, the Office of Science (BER) (Grant No. DE-FG02-04ER63924); and the Murdock Charitable Trust.