We have measured the IR and UV spectra of cytosine in a low-temperature argon matrix. An attempt was made to determine the tautomeric ratios existing in the matrix, making use of the matrix-isolation IR spectrum and computed IR intensities of the tautomers in a least squares fitting procedure. The mole fractions are about 0.22 for oxo(-amino) form, 0.26 and 0.44 for the two rotamers, respectively, of the hydroxy(-amino) form and 0.08 for the (oxo-)imino tautomer. These ratios were then used to simulate the matrix-isolation UV spectrum as a composite of the individual spectra, the latter calculated ab initio at high levels of electron correlation theory. The agreement between simulated and experimental UV spectra seems satisfactory. This indicates that, in contrast to the solid state and solution spectra described up to now by the oxo(-amino) form alone, the reproduction of the matrix-isolation UV spectrum needs at least the hydroxy(-amino) and oxo(-amino) forms, and probably also the (oxo-)imino form.

Conformer ttc/VIp of glycine and glycine-N,N,O-d(3) has been prepared in low-temperature Ar, Kr, Xe, and N(2) matrices by near-infrared (NIR) laser irradiation of the first OH stretching overtone of conformer ttt/Ip. Glycine (and glycine-N,N,O-d(3)) ttc/VIp was found to convert back to ttt/Ip in the dark by hydrogen-atom tunneling. The observed half-lives of ttc/VIp in Ar, Kr, and Xe matrices at 12 K were 4.4 ± 1 s (50.0 ± 1 h), 4.0 ± 1 s (48.0 ± 1 h), and 2.8 ± 1 s (99.3 ± 2 h), respectively. In correspondence with the observation for the cis-to-trans conversion of formic and acetic acid, the tunneling half-life of glycine ttc/VIp in a N(2) matrix is more than 3 orders of magnitude longer (6.69 × 10(3) and 1.38 × 10(4) s for two different sites) than in noble gas matrices due to complex formation with the host molecules. The present results are important to understand the lack of experimental observation of some computationally predicted conformers of glycine and other amino acids.

No abstract

The near- and mid-IR spectra of α-alanine isolated in low-temperature Ar, Kr, and N2 matrixes were measured. Production of the short-lived conformer VI at the expense of the predominant conformer I was observed upon short irradiation with NIR laser light at the first O-H stretching overtone band of conformer I. Conformer VI decays by H-atom tunneling at 12 K with half-lives of 5.7 ± 1 s, 2.8 ± 1 s in Ar (two different sites), 7.0 ± 1 s in Kr, and 2.8 × 10(3) ± 1.2 × 10(3) s in N2. Upon prolonged irradiation, conformer I slowly transformed into conformer IIa. On the basis of these irradiation experiments, the unambiguous vibrational assignments of conformers I, IIa, and VI are given. In contrast to similar experiments for glycine, the irradiation experiments did not lead to the formation of conformer IIIb. This is explained by a very low IIIb → I barrier height computed for alanine, which results in a very fast depletion of conformer IIIb even in low-temperature matrixes.

Former assignments of the matrix-isolation infrared (MI-IR) spectrum of 2-chloropropionic acid are revised with the help of near-infrared (NIR) laser irradiation induced change in conformer ratios. This method allows not only the unambiguous assignment of each band in the MI-IR spectrum to the two trans (Z) and the cis (E) conformers but also the assignment of the spectral bands to different matrix sites. The tunneling decay of the higher-energy cis conformer prepared from both trans conformers in different sites is also investigated. It is shown that the tunneling decay time is very sensitive to the matrix site, especially if the in situ prepared high-energy conformer has a strained geometry in the matrix cage. The analysis shows that the kinetics of some cis → trans back conversion processes cannot be fitted by a single exponential decay. The possible reasons of this observation are examined and discussed. The present and former results clearly show that, in addition to tunneling processes, the decay rates strongly depend on solid-state effects. Therefore, simple theoretical predictions of decay rates, which do not take into account the solid-state effects, can only be compared to experimental observations only if experimentally proven that these effects do not significantly affect the experimentally measured tunneling rates.