Effects of modulation defects on hadamard transform time-of-flight mass spectrometry (HT-TOFMS)

Joel R. Kimmel, Facundo M. Fernández, Richard N. Zare

In any Hadamard multiplexing technique, discrepancies between the intended and the applied encoding sequences may reduce the intensity of real spectral features and create discrete, artificial signals. In our implementation of Hadamard transform time-of-flight mass spectrometry (HT-TOFMS), the encoding sequence is applied to the ion beam by means of an interleaved comb of wires (Bradbury-Nielson gate), which shutters the ion beam on and off. By isolating and exaggerating individual skewing effects in simulating the HT-TOFMS process, we determined the nature of errors that arise from various defects. In particular, we find that the most damaging defects are: mismatched voltages between the wire sets and the acceleration voltage of the instrument, which cause positive and negative peaks throughout mass spectra; insufficient deflection voltage, which reduces the intensity of real peaks and causes negative peaks that are spread across the entire mass range; and voltage errors as the wire sets return from their deflection voltage to their transmission value, which yield significant reductions in peak intensities, create artificial peaks throughout mass spectra, and broaden real peaks by causing positive peaks to grow in the bins adjacent to them. Because the magnitude of the modulation defects grows as the applied modulation voltage is increased, Bradbury-Nielson gates with finer wire spacing, and hence stronger effective fields for a given applied voltage, were produced and installed. Operating at 10 to 15 V where errors in the electronics are essentially absent, the most finely spaced gate (100 microm) yielded signal-to-noise ratios that were more than two times higher than those achieved with more widely spaced gates. As an alternative method for minimizing skewing effects, HT-TOFMS data were post processed using an exact knowledge of the modulation defects. Nonbinary matrices that mimic the actual encoding process were built by measuring voltage versus time traces and then translating these traces to transmission versus time. Use of these matrices in the deconvolution step led to marked improvements in spectral resolution but require full knowledge of the encoding defects.