AmericanChemicalSociety.com

(This work was conducted by graduate student Tamika Madison, as part of her dissertation work with Prof. Peter Siska - the author of the proposal.  Pete passed away and JJG stepped in and advised Tamika as she completed this part of her dissertation.) The first carefully measured mass spectra obtained by Penning Ionization (PI) of formamide with helium, neon, or argon metastable atoms have been obtained at kinetic energies near 1 kcal/mol.  This target organic molecule was selected as a model compound for investigating selective fragmention of polypeptides in a desire to find a method for sequencing of polypeptides by mass spectrometry.  Data were reproducibly collected on the Siska group’s apparatus that employs a supersonic excited noble gas beam crossing an effusive beam of formamide vapor.  Product ions are extracted perpendicular to the plane of the crossed beams, analyzed by a quadrupole mass filter, and counted by a scintillation-type ion counter.  Noble gas ions and noble gas Rydberg atoms were removed by electric deflection/quenching fields.   These data, which have been averaged over a number of independent experiments, have also been corrected for background reactions and for any possible mass discrimination effects (due to the differing m/z values being monitored). 

For the He* reaction, the average yields of the various product ions (with a one-standard-deviation error bar) are:  m/z 16 - 7.5(1.5)%, m/z 17 - 11.5(1.9)%, m/z 18 – 1.9(0.5)%, m/z 29 – 15.7(1.3)%, m/z 42 – 0.8(0.7)%, m/z 43 – 13.0(1.0)%, m/z 44 – 18.0(1.2%), and m/z 45 – 31.6(1.4)%. For the Ne* reaction (with less interaction energy than that for helium), the average yields of the product ions (with a one-standard-deviation error bar) are:  m/z 16 – 2.9(1.6)%, m/z 17 – 21.4(0.7)%, m/z 28 – 1.9(0.9)%, m/z 29 – 24.5(1.9)%, m/z 43 – 0.6(0.4)%, m/z 44 – 25.0(1.1)%, and m/z 45 – 23.7(1.3)%. For the Ar* reaction (with the least interaction energy of the three systems), the average yields of the product ions (with a one-standard-deviation error bar) are:  m/z  17 - 5.4(1.5)%, m/z 29 – 0.7(0.3)%, m/z 44 – 6.0(0.6)%, and m/z 45 – 87.9(0.6)%. High level ab initio calculations for thirty unique fragmentation channels have also been completed, these calculations provide thermodynamic data and more.  The three energetically favored channels are formation of the ammonia radical cation (m/z 17, NH₃⁺), the formamide radical cation (m/z 45, HCONH₂⁺) and OCNH₂⁺ (m/z 44).

Ab initio calculations were also used to determine energies and structures of relevant transition states connecting the initially formed molecular radical cation to observed ionic (and assumed neutral) products. In order to compute a ‘theoretical mass spectrum’ to compare to that obtained in the lab, RRKM calculations were carried out using a custom developed code.  This was completed for the Ar* PI reaction of formamide, and a plot created that show the k(E) for the two allowed fragmentation channels (loss of H and formation of OCNH₂⁺; loss of CO and formation of NH₃⁺), and the deposition function. The PI spectra show different ion yields when compared to the standard 70 eV EI spectrum for formamide, which we demonstrated can be accounted for by analyzing the dynamics of the PI reaction.  The number and identity of observed ions in these spectra depend on the excitation energy of Ng* as well as the ΔE₀ values for the fragmentation channels that include them.  We have focused on the Ar* + HCONH₂ spectrum.  The yields calculated are in agreement with our experimental yields.  RRKM theory also allowed us to determine the fraction of our ion yields that can be attributed to tunneling.  Future measurements of these spectra with formamide – d₃ would allow us to experimentally determine how much of the ion yield comes from tunneling.  Our plans for future work had included collecting the spectra for five other amides, systematically growing alkyl chains from the carbonyl carbon and nitrogen atoms and doing similar RRKM calculations for these systems.  We also had hoped to collect PIES spectra for these systems in order to have more accurate energy deposition functions for ion yield calculations.  The calculation of ion yields using unimolecular rate theory is often a difficult task because of the difficulty in obtaining the internal energy distribution for the molecular ion in question.  The ease in obtaining the energy deposition function for a molecular ion produced by PI makes this task less daunting.  PI mass spectrometry then provides a great venue for the application of RRKM-QET theory to the calculation of ion yields.

Unfortunately, with the passing of Pete Siska, this project has ended and his lab closed.  Tamika published the above results in J. Chem. Phys. 2009, 131, 134309 and will include them in her dissertation.  She has now moved to another research group to complete her dissertation work in a new research area.