High-resolution photoelectron spectra of the pyrimidine-type nucleobases

Abstract

High-resolution photoelectron spectra of the gas phase pyrimidine-type nucleobases, thymine, uracil, and cytosine, were collected using synchrotron radiation over the photon energy range 17 ≤ hν ≤ 150 eV. These data provide the highest resolution photoelectron spectra of thymine, uracil, and cytosine published to date. By comparing integrated regions of the energy dependent photoelectron spectra of thymine, the ionization potentials of the first four ionic states of thymine were estimated to be 8.8, 9.8, 10.3, and 10.8 eV. The thymine data also show evidence for low energy shape resonances in three of the outermost valence electronic states. Comparing the uracil spectrum with the thymine spectrum, the four outermost valence electronic states of uracil likely begin at binding energies 9.3, 9.9, 10.5, and 11.0 eV. High-resolution spectra indicate only one tautomeric form of cytosine contributes significantly to the spectrum with the four outermost valence electronic states beginning at binding energies 8.9, 9.9, 10.4, and 10.85 eV.

Computing the Ultrafast and Radiationless Electronic Excited State Decay of Cytosine and 5‐methyl‐cytosine Cations: Uncovering the Role of Dynamic Electron Correlation

“…By averaging the values for each electronic cationic state within the different formulations of the CASPT2 method (see Table ) we predict the first four vertical ionisation potentials relative to the , , and states (see Figure ) to be placed at 8.74, 9.44, 9.59 and 9.94 eV, respectively, while the first adiabatic ionisation potential is situated at 8.56 eV, which is in good agreement with the available experimental evidence and previous theoretical estimates . A standard deviation of ∼0.2 eV for the CASPT2 average is obtained for all states.…”

Ultrafast and radiationless electronic excited state decay of uracil and thymine cations: computing the effects of dynamic electron correlation

“…We start by analysing and comparing the results obtained for the two well-defined experimental magnitudes often employed to benchmark the cationic manifold computationally, that is, the vertical and adiabatic ionisation potentials. Averaging values from Table 1 for each state in turn, the different formulations of the CASPT2 method predict the first three vertical ionisation potentials to be placed at 9.53, 10.00 and 10.59 eV, while the first adiabatic ionisation potential is placed at 9.35 eV, in agreement with the available experimental evidence [69][70][71][72] , as well as previous theoretical estimates. 6,15,74 As has been reported by Roca-Sanjuán et al, 15 the introduction of the IPEA shift appears to blue-shift the estimates obtained at the CASPT2 level, bringing them closer to the experimental values.…”

“…Averaged CASPT2 estimates of the first three vertical and first adiabatic ionisation potential are 9.22, 9.93, 10.51 and 9.05 eV, respectively, which are placed roughly within a tenth of an eV from the experimental evidence (see Table 2). [69][70][71][72] We observe a 0.2 (D 2 ) to 0.3 (D 0 ) eV red-shift on vertical and adiabatic ionisation potentials of 2 π + character related to 5methylation, i.e. moving from uracil + to thymine + .…”

“…We reproduce the trend registered experimentally whereby 5methylation induces a stabilisation on the 2 π + cationic manifold. [69][70][71][72] Within the cationic manifold, CASSCF predicts 1 eV D 0 − D 1 and 0.08 eV D 1 − D 2 energy gaps at the FC region, in contrast with the XMS-CASPT2 estimates of 0.71 and 0.57 eV, respectively, leading to differences of ∼0.3 eV for the former and ∼0.5 eV for the latter. The red-shift on D 2 and the negligible effect on D 1 upon methylation leads to a near degeneracy between these states, which is however broken when including dynamic correlation.…”

Assessment of Electron Propagator Methods for the Simulation of Vibrationally Resolved Valence and Core Photoionization Spectra

“…The VIEs have been computed using both diagonal and non-diagonal approaches, whereas the gradients have been computed in all cases at the OVGF level since, as shown above for furane, the choice of the propagator method used in the simulation has a minor impact in the reproduction of the gradients, which can therefore be computed using cheaper methods. The reliability of the various propagator approaches has been checked comparing the theoretical results with the experimental photoelectron spectrum taken from ref 86. that is, up to our knowledge, the most accurate spectrum available in the literature, displaying an higher resolution with other spectra available in the literature 87,88.…”