Supporting Information
Experimental Characterization of Secular Frequency Scanning in Ion Trap Mass Spectrometers
Dalton T. Snyder; Christopher J. Pulliam; Wiley, J. S.; Duncan, J.; R. Graham Cooks*
*Email:
Figure S1: Typical scan diagrams (voltage amplitudes) for obtaining a mass spectrum by resonance ejection with an rf amplitude scan and proposed ac frequency sweep (“AC scan” or “secular frequency scan”), enabling acquisition of a mass spectrum with a constant rf amplitude and frequency. During the mass scan, the frequency of the supplementary ac waveform is swept, ejecting ions in a mass selective manner.
Figure S2: Instrumental setup for secular frequency scans on a miniature mass spectrometer.
Figure S3: Secular (AC) frequency scanning in an ion trap: (a) forward secular frequency scan (reverse mass sweep) of tetraalkylammonium salts (m/z 284, 360, and 382) recorded using the Mini 12 (300 ms, 10-500 kHz, 1.5 Vpp, LMCO = 100 Th), (b) simulated secular frequency scan (4 mTorr He, 3 ms scan from 10 to 300 kHz, 10 V0-p), (c) resonance ejection using the Mini 12, (d) equivalent secular frequency scan performed using a benchtop LTQ (same conditions as (a), but 1 Vpp), (e) scan in (d) converted to the mass domain and (f) comparison to resonance ejection using the LTQ.
Figure S4: Secular frequency scan mass spectrum of 1,3-dibromopropane on a cylindrical ion trap with an internal electron ionization source showing resolved bromine isotope peaks. The lower mass cutoff was ~70 Da, and the ac amplitude was ~0.7 V0-p. The peak at m/z 152 is methyl salicylate background from a previous experiment.
Figure S5: Effect of scan direction on the secular frequency scan mass spectra of tetraalkylammonium salts where ions are ejected at different pressures in each scan: (a) Mini 12 and (b) LTQ. Scan time was 300 ms from 10 to 500 kHz (or vice versa) with an ac amplitude of 4 Vpp for (a) and 1 Vpp for (b). Here resolution and sensitivity appear to be superior in the forward sweep, but as we show later in the paper, this is caused by ion ejection later in the scan in the reverse frequency sweep, which drastically reduces resolution and signal intensity, whereas in the forward frequency sweep the ions are ejected at higher pressure near the beginning of the mass scan.
Figure S6: Resonance absorption curves for excitation of n-butylbenzene (m/z 134). Of considerable interest is (i) the shift of the resonance condition to higher frequencies at higher amplitudes (when ions are far from the center of the trap), and (ii) the asymmetry in the curves, broader at low frequency and sharper at high frequency, particularly for low excitation amplitudes. Reprinted with permission from Williams, J. D.; Cox, K. A.; Cooks, R. G.; Mcluckey, S. A.; Hart, K. J.; Goeringer, D. E. Anal. Chem. 1994, 66, 725. Copyright 1994 American Chemical Society.
Figure S7: Effect of scan rate on the secular frequency scan of tetraalkylammonium salts (m/z 285, 360, and 383) demonstrated using (a) Mini 12 and (b) LTQ. Scan rate was altered by changing the scan time (inset) on the function generator while keeping the frequency range the same. Scan range was 10-500 kHz and amplitude was (a) 3 Vpp and (b) 1 Vpp. The LMCO was set at 100 Th.
Figure S8: Forward frequency scan of a calibration solution of Ultramark (m/z 1022-1922, every 100 m/z), MRFA (m/z 525), and caffeine (m/z 195) showing effect of rf frequency on the spectra. Rf amplitude was held constant at 6,000 DAC steps, ac amplitude was 7 Vpp, and scan time was 800 ms.
Figure S9: Secular frequency scan mass spectrum showing higher order parametric resonances on a benchtop LTQ XL. Ions from an Ultramark 1621 calibration solution were analyzed with a secular frequency scan 10-580 kHz with a 1 Vpp amplitude over 800 ms during an Ultrazoom scan starting at 1300 Th. The intense peaks on the right side of the spectrum (>350 ms) are ions ejected at their secular frequency, whereas the same distribution is observed in the 150-350 ms range, indicative of ejection at the K = 4 parametric resonance. See Table S1 for calculations.
Ejection Time (ms) / Ejection Frequency (kHz) / Parametric Frequency / Ejection Frequency / K0 / 10 / *calibration parameters
800 / 580 / *calibration parameters
675.19 / 491.07 / 2 / 2 or dipolar
582.59 / 425.10 / 2 / 2 or dipolar
516.30 / 377.86 / 2 / 2 or dipolar
464.81 / 341.18 / 2 / 2 or dipolar
422.96 / 311.36 / 2 / 2 or dipolar
387.41 / 286.03 / 2 / 2 or dipolar
357.41 / 264.65 / 2 / 2 or dipolar
339.26 / 251.72 / 3.90 / 4
293.33 / 219.00 / 3.88 / 4
256.30 / 192.61 / 3.92 / 4
231.11 / 174.67 / 3.91 / 4
205.19 / 156.19 / 3.99 / 4
183.33 / 140.62 / 4.07 / 4
173.70 / 133.76 / 3.96 / 4
Table S1: Calculated times and frequencies of ejection for the scan in Figure S9. The top two rows of data show the parameters for calculating ejection frequency from ejection time. The data in blue show ion ejection times and frequencies corresponding to those times. These are their experimental secular frequencies. The ions in purple correspond to the ions in blue (first blue ion = first purple ion and so on), but were ejected at a higher order parametric resonance corresponding to K = 4, as calculated in column 3. For the calculation in column 3, it was assumed that the parametric resonance frequency was twice the frequency in blue for each respective ion. Hence, the ratio of the parametric frequency to each blue frequency is exactly 2. The K value corresponds to the order of the parametric resonance.
Figure S10: Mathieu stability diagram illustrating parametric K = 1 to K = 6 resonances. Shaded areas represent regions of instability. Here Q is directly proportional to the excitation voltage, indicating that the higher order resonances require higher excitation amplitudes. Reprinted from Collings, B.A., Sudakov, M., Londry, F.A.: Resonance shifts in the excitation of the n = 0, K = 1 to 6 quadrupolar resonances for ions confined in a linear ion trap. J. Am. Soc. Mass Spectrom. 13, 577-586 (2002).
Figure S11: Mass range extension on the Mini 12. Secular frequency scans of a solution of caffeine, MRFA, and Ultramark 1621 showing mass range extension at low rf amplitudes: (a) full scan on Mini 12 showing caffeine, MRFA, and Ultramark peaks (3.3 Vpp, 10-500 kHz, LMCO = 98 Th, 800 ms scan), (b) zoomed in image showing resolved Ultramark peaks, (c) resonance ejection performed on an LTQ, and (d) secular frequency scan of the Ultramark solution on an LTQ performed over 800 ms, 10-500 kHz, with a 1 Vpp amplitude and a lower mass cutoff of 1,000 Th.
Figure S12: Resonance ejection mass spectrum at 345 kHz of a calibration solution of caffeine (m/z 195), MRFA (m/z 525), and Ultramark 1621 at 345 kHz (not within mass range). Spectrum was acquired on the Mini 12. Sample number is linearly correlated with m/z.