The Adiabatic Ionization Energy and Triplet T1 Energy of Jet-Cooled

Nov 19, 2012 - Gas-phase cytosine exists in five different tautomer/rotamer forms 1, 2a, 2b, 3a, and 3b. We determine the threshold ionization energy ...
0 downloads 0 Views 922KB Size
Letter pubs.acs.org/JPCL

The Adiabatic Ionization Energy and Triplet T1 Energy of Jet-Cooled Keto-Amino Cytosine Simon Lobsiger and Samuel Leutwyler* Department of Chemistry and Biochemistry, University of Bern, Freiestrasse 3, CH-3012 Bern, Switzerland ABSTRACT: Gas-phase cytosine exists in five different tautomer/rotamer forms 1, 2a, 2b, 3a, and 3b. We determine the threshold ionization energy (IE) of the keto-amino tautomer 1 as 8.73 ± 0.02 eV, using resonant two-photon ionization mass spectrometry in a supersonic molecular beam via the 1ππ* excited state. This is the first IE threshold measurement for the biologically relevant tautomer 1. The IE of the thermal gas-phase mixture of cytosine has been measured as 8.60 ± 0.05 eV by Kostko et al. using singlephoton VUV photoionization [Phys. Chem. Chem. Phys., 2010, 12, 2860]. Given the tautomer distribution and ionization energies calculated in that work, our determination of the keto-amino tautomer IE implies that the IE measured by Kostko et al. is dominated by the enol-amino tautomers 2a and 2b. Upon excitation of keto-amino cytosine to its 1ππ* state, relaxation occurs to a lower-lying long-lived state. The IE threshold measured via this state places its energy about 0.69 eV below the 1ππ* state, in good agreement with the triplet T1 energy of keto-amino cytosine calculated by several high-level ab initio methods. The identification of keto-amino cytosine T1 is the basis for characterizing the intersystem crossing rates into and the photochemical reactions of this long-lived state. SECTION: Spectroscopy, Photochemistry, and Excited States

G

species is calculated to be the amino-enol tautomer 2b, followed by its rotamer 2a, the amino-keto tautomer 1, and the imino-keto tautomer 3a.3,8−13 An overview of tautomer concentrations at different temperatures is given in Table 1 of ref 3. Due to this broad tautomer distribution, the adiabatic ionization energies and photoelectron spectra reported for single-photon experiments1−7 are superpositions of the different tautomers. In an XPS study of cytosine, Feyer et al.7 have suggested that three tautomers are populated at 450 K, with the enol-amino tautomers 2a and 2b being the dominant species. Kostko et al. have calculated the cytosine tautomer distribution at their experimental temperature of T = 582 K and reported a single-photon IE threshold of 8.60 ± 0.05 eV for cytosine.3 However, they did not assign this value to any specific tautomer or set of tautomers.3 In contrast to the single-photon PI-MS, UPS, and XPS methods, resonant two-photon ionization (R2PI) spectroscopy in many cases allows one to perform tautomer-selective measurements by exciting a given tautomer to an intermediate excited state, followed by ionization of that state.15−20 De Vries and co-workers have performed pioneering work on the R2PI spectra of cytosine.16−18 On the basis of the IR-UV double resonance spectrum of cytosine near 32000 cm−1 and supported by the close similarity of the UV spectra of cytosine and 1-methylcytosine (which exists dominantly as the keto-

iven the importance of the ionization of nucleobases both in the gas phase and in solution, there have been a number of studies of the ionization energy (IE) of gas-phase cytosine, using photoionization mass spectrometry (PI-MS),1−3 valence photoelectron spectroscopy (UPS),4−6 or core-level photoelectron spectroscopy (XPS).7 Most recently, Kostko et al.3 have studied the ionization of cytosine monomers and dimers using single-photon vacuum-ultraviolet (VUV) photoionization in combination with high-level electronic structure calculations.3 For cytosine (and guanine), however, the interpretation of the single-photon ionization data is difficult, because a number of different gas-phase tautomers and rotamers are produced under typical thermal vaporization conditions. The relative stabilities and free energies of the cytosine gas-phase tautomers have been extensively investigated by theoretical calculations ranging from density functional to CCSD(T) methods.3,8−13 Scheme 1 shows the structures of the five most populated tautomers of gas-phase cytosine.3,8−13 The tautomer numbering follows that of ref 14. At the temperatures T = 500 − 580 K typically employed for vaporization of cytosine in most experimental studies (510 K in this work), the dominant Scheme 1

Received: October 24, 2012 Accepted: November 19, 2012 Published: November 19, 2012 © 2012 American Chemical Society

3576

dx.doi.org/10.1021/jz301719h | J. Phys. Chem. Lett. 2012, 3, 3576−3580

The Journal of Physical Chemistry Letters

Letter

amino tautomer), they assigned the spectrum of cytosine near 32000 cm−1 to the keto-amino tautomer 1. On the basis of excitation to the 000 band in the R2PI spectrum of the keto-amino tautomer at 31835 cm−1 (3.947 eV), we have measured the IE threshold of the amino-keto tautomer. We have recently measured the rotational contour of the 000 band and from its polarization have determined that it indeed leads to a 1ππ* excited state.21 In contrast to the 1−2 ps electronic lifetimes reported for cytosine at higher excitation energies, the lifetime of the 1ππ* v = 0 level is τ ∼ 40 ps as measured by line broadening,21 which is long enough to perform a two-step photoionization measurement even with nanosecond lasers. Similar lifetimes (τ = 30 − 50 ps) have recently been observed for 5-methylcytosine.22 Neon carrier gas (Linde, ≥99.995%) at ∼1.8 bar backing pressure was passed through a pulsed nozzle (0.4 mm diameter) containing cytosine (Sigma, >99% purity) heated to 508 K. Photoionization efficiency (PIE) curves were measured by crossing the skimmed supersonic jet with the unfocused UV excitation and ionization laser beams in the source of a linear time-of-flight mass spectrometer (TOF-MS). Excitation was performed with 100−200 μJ UV pulses from a frequency-doubled Radiant Dyes NarrowScan D-R dye laser. The excited states were ionized using ∼150−200 μJ pulses of ∼10 cm−1 bandwidth from an Ekspla NT342B optical parametric oscillator (UV-OPO), which is tunable over the relevant spectral range 215−270 nm. The molecular beam/ mass spectrometer experimental setup is similar to that described in refs 23 and 24. The jet-cooled cytosine was excited to the 000 band of the 1 ππ* state, as indicated in the inset of Figure 1. Figure 1b shows the PIE curve when the two laser pulses are optimally overlapped for maximum signal at zero time-delay. The frequency of the ionization laser was scanned in 0.05 nm steps. The electric field of the ion source of the TOF-MS was switched on 50 ns after ionization to ensure field-free photoionization conditions. In these two-photon ionization measurements, the ionization laser can itself excite optical transitions to discrete vibronic levels of other tautomers (presumably the enol-amino tautomers 2a, 2b) in the same spectral region, which give rise to one-color two-photon signals that appear below the ionization threshold. To eliminate these, the PIE curve was measured with the ionization laser always on, but the excitation laser alternated on/off on successive laser pulses. The profile shown in Figure 1b corresponds to the PIE curve with both lasers on minus that with only the ionization laser on. The dip in the PIE curve at 41200 cm−1 is due to an imperfection in the nonlinear doubling scheme employed by the Ekspla NT342B OPO. The ionization threshold of the S1 state, obtained by extrapolation of the linear rise to the zerosignal line is at 38600 ± 80 cm−1 (4.786 ± 0.010 eV). The sum of the S1 → S0 excitation and the ionization energies corresponds to IE = 8.73 ± 0.01 eV. Figure 3a compares the time-dependent B3LYP minimum geometries of the 1ππ* excited state and of the ion ground state D0. These geometries are similar, differing only in the amino group angle. This argues that the vertical transition in ionization is not far from adiabatic. Note that the 1ππ* excitation also mainly involves a change of the -NH2 angle. The strongest band in the corresponding R2PI spectrum (inset in Figure 1) lies only 110 cm−1 above the 000 electronic origin. We therefore believe that we are not missing the adiabatic threshold

Figure 1. PIE curve for keto-amino cytosine 1 following excitation to its lowest 1ππ* electronic origin at 31836 cm−1 (3.947 eV). The IE threshold obtained by extrapolation of the linear rise to the zero-signal line is at 38600 ± 80 cm−1 (4.786 ± 0.01 eV).

by more than ±160 cm−1. This increases the above error estimate to ±0.02 eV. Kostko et al. have previously measured the ionization threshold of the thermally populated mixture of cytosine tautomers as 8.60 eV, using single-photon photoionization.3 Their threshold is 0.13 eV lower than our experimental IE for the keto-amino tautomer 1. Given the tautomer distribution calculated by Kostko et al., with 2b being the most abundant gas-phase species followed by 2a,3 we surmise that their threshold is dominated by the IE of the enol-amino tautomers 2a and 2b.3 Using the EOM-IP-CCSD method, Kostko et al. calculated the adiabatic IE of 1 as 8.67 eV,3 which is 0.06 eV lower than our experimental IE for 1. The EOM-IP-CCSD calculated adiabatic IEs for 2a and 2b are 8.52 and 8.54 eV, respectively; these are also 0.06−0.08 eV lower than their measured threshold.3 We conclude that the EOM-IP-CCSD adiabatic IEs of 1, 2a, and 2b are consistently shifted by about −0.07 eV relative to their respective experimental values. As noted by de Vries and co-workers,17 a part of the ketoamino tautomer population excited at the 000 band relaxes to a lower-lying state. This state in turn decays with a lifetime that is long enough to be measured by temporally delaying the 5 ns ionization pulse. De Vries and co-workers used 193 nm pulses from an ArF excimer laser for photoionization and measured a lifetime of 290 ns, on the basis of which they tentatively assigned it as a triplet state.17 In our experiment, the 5 ns pulses from the UV OPO can be tuned from the S1 ionization 3577

dx.doi.org/10.1021/jz301719h | J. Phys. Chem. Lett. 2012, 3, 3576−3580

The Journal of Physical Chemistry Letters

Letter

the ionization threshold. In this interpretation, the PI onset at 43200 cm−1 (5.356 eV) is a lower limit of the adiabatic IE. Comparison to the IE measured via the 1ππ* state implies that the long-lived state lies less than 3.37 eV above the ground state. On the other hand, we estimate an approximate upper limit of the adiabatic IE as 44200 cm−1 (5.480 eV) by backextrapolation of the straight part of the PIE curve to zero signal, as shown by the black arrow in Figure 2b. Comparison of this value with the IE measured via the 1ππ* state implies that the long-lived state lies at least 3.26 eV above the ground state. The range 3.26−3.37 eV estimated for the energy of the long-lived state compares nicely to the results of timedependent B3LYP calculations, which predict the adiabatic T1 3 ππ* state 3.18 eV above the S0 ground state, or 0.76 eV below the 1ππ* state (see Table 1). Merchan et al. have previously calculated the energies of S1 to S3 singlet excited states and the T1 to T4 triplet states using CASSCF/CASPT2 calculations, also given in Table 1.26 Their CASPT2//CASSCF calculations also identify the T1 state as a 3ππ* state, 3.04 eV above the ground state. Abouaf et al. have previously measured the T1 and T2 states by electron energy loss spectroscopy and observe the T1 state at 3.5 ± 0.1 eV.25 However, their measurement was performed on an effusive thermal beam at 423 K, in which the dominant species are the tautomers 2a and 2b (see above). Abouaf et al. also calculated the T1 energy of keto-amino cytosine 1 using different density functional methods as well as CCSD and CCSD(T) correlated wave function methods. In Table 1, we give their unrestricted CCSD(T) T1 energy calculated with the largest basis set, which is 3.13 eV.25 Note that the values of the T1 energy of keto-amino cytosine in Table 1 were calculated with very different methods, but lie in the narrow range of 3.04−3.18 eV. Having identified the long-lived state as T1, we return to the question of adiabatic versus vertical ionization from T1. Figure 3b shows the superposed equilibrium structures of the T1(3ππ*) and 1ππ* states, calculated by the time-dependent B3LYP method with the TZVP basis set. The T1 structure is distinctly nonplanar, with the C5−C6 bond strongly twisted, in agreement with Merchan et al.,26 who also reported a C5−C6 twisted 3ππ* minimum. For such a large geometry change, one predicts a very small ionization Franck−Condon factor from the v = 0 level of T1 to the ion v = 0 level, precluding adiabatic ionization from the cold T1 state. This means that the adiabatic IE from T1 must be smaller than the IE extrapolated in Figure 2b; hence the T1 energy of 3.26 eV given above is a lower limit. Figure 4 compares our experimental and the TD-B3LYP calculated adiabatic energies of keto-amino cytosine 1 (see also Table 1). The experimental 1ππ* origin band lies at 3.947 eV, which is in excellent agreement with the adiabatic 1ππ* excitation energy of 3.94 eV. The energy of the T1 3ππ* state is

threshold at 259 nm down to 215 nm energy. When delaying the ionizing laser by 50 ns, we can observe the PIE curve of the long-lived state, which is shown in Figure 2b.

Figure 2. Comparison of the PIE curves for keto-amino cytosine following excitation at the S1 1ππ* origin (a) with prompt ionization (0 ns delay) and (b) with the ionization laser pulse delayed by 50 ns. The 1ππ* excited cytosine relaxes rapidly to a lower-lying long-lived state. The upper limit to the ionization threshold of the long-lived state is obtained by back-extrapolation (black arrow) of the linear rise to zero signal as ∼44′200 cm−1. The estimated uncertainty of the IE is indicated as a red horizontal bar (see the text).

The onset of the PIE curve for delayed ionization at ∼43200 cm−1 is offset by ∼4600 cm−1 to higher energy, with a slow initial rise above the background. If we crudely assume that 4600 cm−1 is the adiabatic energy difference of the 1ππ* and the lower state, this energy is randomized over the 33 vibrational modes of Cyt during the 50 ns delay time. Photoionization out of these vibrationally “hot” states could then be interpreted as giving rise to the sluggish onset of the delayed-ionization PIE curve in Figure 2b, thereby obscuring

Table 1. Adiabatic Energies (in eV) for Keto-Amino Cytosine (1), Calculated by the UCCSD(T), CASPT2 and TimeDependent B3LYP Methods, Compared to Experiment ππ* nOπ* T1(3ππ*) T2(3nOπ*) T3(3nNπ*)

UCCSD(T)a/6-311++G(3df,2p)

CASPT2//CASSCFb (12,9)/6-31G(d,p)

TD-B3LYPc/TZVP

experimentc

3.94 3.74 3.18

3.95 (±0.02)

3.13

3.64 3.87 3.04 3.88 3.89

1 1

a

3.26−3.37

Abouaf et al., ref 25. bMerchan et al., ref 26. cThis work. 3578

dx.doi.org/10.1021/jz301719h | J. Phys. Chem. Lett. 2012, 3, 3576−3580

The Journal of Physical Chemistry Letters

Letter

of the thermal tautomer distribution performed by singlephoton VUV ionization and high-level EOM-IP-CCSD calculations by Kostko et al. allows one to attribute their ionization threshold3 as that of the nonbiological enol-amino tautomers 2a and 2b. Our measurement of the IE of the longlived lower state to which the optically pumped 1ππ* v = 0 level relaxes17 identifies it as the T1(3ππ*) state of cytosine. The experimental limits to its IE places the T1 state at 3.26−3.37 eV above the ground state.



AUTHOR INFORMATION

Corresponding Author

Figure 3. Time-dependent B3LYP and unrestricted B3LYP (UB3LYP) calculated and overlaid equilibrium structures of (a) the optically excited S1(1ππ*) state and the ion ground state, and (b) the T1 3ππ* state and the ion ground state. The 1ππ* and 3ππ* geometries are drawn in gray. The ion ground-state geometry (in color) is planar.

*E-mail: [email protected]. Notes

The authors declare no competing financial interest.



ACKNOWLEDGMENTS Financial support by the Schweiz. Nationalfonds (Project No. 200020-132540) is gratefully acknowledged.



REFERENCES

(1) Orlov, V. M.; Smirnov, A. N.; Varshafsky, Y. M. Ionization Potentials and Electron-Donor Ability of Nucleic Acid Bases and their Analogs. Tetrahedron Lett. 1976, 17, 4377−4378. (2) Belau, L.; Wilson, K. R.; Leone, S. R.; Ahmed, M. VacuumUltraviolet Studies of the Microhydration of DNA Bases (Guanine, Cytosine, Adenine and Thymine). J. Phys. Chem. A 2007, 111, 7562. (3) Kostko, O.; Bravaya, K.; Krylov, A.; Ahmed, M. Ionization of Cytosine Monomer and Dimer Studied by VUV Photoionization and Electronic Structure Calculations. Phys. Chem. Chem. Phys. 2010, 12, 2860. (4) Hush, N. S.; Cheung, A. S. Ionization Potentials and Donor Properties of Nucleic-Acid Bases and Related Compounds. Chem. Phys. Lett. 1975, 34, 11−13. (5) Yu, C.; Peng, S.; Akiyima, I.; Lin, J.; LeBreton, P. R. Ultraviolet Photoelectron Studies of Biological Pyrimidines - Valence Electronic Structure of Cytosine. J. Am. Chem. Soc. 1978, 100, 2303−2307. (6) Trofimov, A. B.; Schirmer, J.; Kobychev, V. B.; Potts, A. W.; Holland, D. M. P.; Karlsson, L. Photoelectron Spectra of the Nucleobases Cytosine, Thymine and Adenine. J. Phys. B: At., Mol., Opt. Phys. 2006, 39, 305−329. (7) Feyer, V.; Plekan, O.; Richter, R.; Coreno, M.; Vall-Ilosera, B.; Prince, K. C.; Trofimov, A. B.; Zaytseva, I. L.; Moskovskaya, T. E.; Gromov, E. V.; Schirmer, J. Tautomerism in Cytosine and Uracil: An Experimental and Theoretical Core Level Spectroscopic Study. J. Phys. Chem. A 2009, 113, 5736. (8) Fogarasi, G. Relative Stabilities of Three Low-Energy Tautomers of Cytosine: A Coupled Cluster Electron Correlation Study. J. Phys. Chem. A 2002, 106, 1381. (9) Trygubenko, S. A.; Bogdan, T. V.; Rueda, M.; Orozco, M.; Luque, F. J.; Sponer, J.; Slavicek, P.; Hobza, P. Correlated Ab Initio Study of Cytosine and Its Tautomers in the Gas Phase, in a Microhydrated Environment and in Aqueous Solution. Phys. Chem. Chem. Phys. 2002, 4, 4192. (10) Piacenza, M.; Grimme, S. Systematic Quantum Chemical Study of DNA-Base Tautomers. J. Comput. Chem. 2004, 25, 83. (11) Yang, Z.; Rodgers, M. T. Theoretical Studies of the Unimolecular and Bimolecular Tautomerization of Cytosine. Phys. Chem. Chem. Phys. 2004, 6, 2749. (12) Wolken, J. K.; Yao, C.; Turecek, F.; Polce, M. J.; Wesdemiotis, C. Cytosine Neutral Molecules and Cation-Radicals in the Gas-Phase: Structures, Energetics, Ion Chemistry, and Neutralization-Reionization Mass Spectrometry. Int. J. Mass.Spec. 2007, 267, 30−42. (13) Kobayashi, R. A CCSD(T) Study of the Relative Stabilities of Cytosine Tautomers. J. Phys. Chem. A 1998, 102, 10813.

Figure 4. Experimental (red) and time-dependent B3LYP/TZVP calculated adiabatic energies (black) of the low-lying singlet and triplet states of keto-amino cytosine (tautomer 1). The EOM-IP-CCSD calculated adiabatic IE is taken from ref 3.

calculated at 3.18 eV, which is 0.07 eV lower than the 3.26 eV estimated from the extrapolation in Figure 2b. Note that the UCCSD(T) and CASPT2 T1 triplet energies in Table 1 are even lower than the TD-B3LYP value. Given (1) the reasonable agreement of the observed and calculated energies and (2) the long gas-phase lifetime of this state, we consider the assignment of the dark state as the T1 state to be firm. This result also supports the conjecture of de Vries and co-workers.17 Given that the T1 state is efficiently produced from the v = 0 level of the lowest 1ππ* state and that the electronic lifetime of this level is τ ∼ 40 ps, the intersystem crossing (ISC) rate is of the order of kISC ∼ 1010 s−1. This value is similar to that measured for 5-methyl-2-hydroxypyrimidine, for which ISC is the dominant nonradiative process out of the S1 state, with kISC ∼ 1/τfl ∼ 2 × 1010 s−1.24 We are currently investigating the ISC kinetics of cold gas-phase cytosine.21 In summary, the two-color R2PI technique yields an IE for the biologically relevant keto-amino tautomer of cytosine as IE = 8.73 ± 0.02 eV. Combination with the previous measurement 3579

dx.doi.org/10.1021/jz301719h | J. Phys. Chem. Lett. 2012, 3, 3576−3580

The Journal of Physical Chemistry Letters

Letter

(14) Fogarasi, G. High-Level Electron Correlation Calculations on Some Tautomers of Cytosine. J. Mol. Struct. 1997, 413, 271−278. (15) Plützer, C.; Hünig, I.; Kleinermanns, K. Tautomers and Electronic States of Jet-Cooled Adenine Investigated by Double Resonance Spectroscopy. Phys. Chem. Chem. Phys. 2003, 4, 4877. (16) Nir, E.; Janzen, C.; Imhof, P.; Kleinermanns, K.; de Vries, M. S. Pairing of the Nucleobases Guanine and Cytosine in the Gas Phase Studied by IR-UV Double-Resonance Spectroscopy and Ab Initio Calculations. Phys. Chem. Chem. Phys. 2002, 4, 732. (17) Nir, E.; Müller, M.; Grace, L. I.; de Vries, M. S. REMPI Spectroscopy of Cytosine. Chem. Phys. Lett. 2002, 355, 59−64. (18) Nir, E.; Hünig, I.; Kleinermanns, K.; de Vries, M. S. The Nucleobase Cytosine and the Cytosine Dimer Investigated by DoubleResonance Laser Spectroscopy and Ab initio Calculations. Phys. Chem. Chem. Phys. 2003, 5, 4780−4785. (19) Lee, Y.; Schmitt, M.; Kleinermanns, K.; Kim, B. Observation of Ultraviolet Rotational Band Contours of the DNA Base Adenine: Determination of the Transition Moment. J. Phys. Chem. A 2006, 110, 11819. (20) Lobsiger, S.; Sinha, R. K.; Trachsel, M.; Leutwyler, S. Low-Lying Excited States and Nonradiative Processes of the Adenine Analogues 7H-and 9H-2-Aminopurine. J. Chem. Phys. 2011, 134, 114307−11420. (21) Lobsiger, S.; Trachsel, M. A.; Frey, H. M.; Leutwyler, S. ExcitedState Dynamics of Cytosine: The Decay of the Keto-Amino Tautomer is Not Ultrafast. Submitted for publication, 2012. (22) Trachsel, M. A.; Lobsiger, S.; Leutwyler, S. Out-of-Plane LowFrequency Vibrations and Nonradiative Decay in the 1ππ* State of JetCooled 5-Methylcytosine. J. Phys. Chem. B 2012, 116, 11081−11091. (23) Lobsiger, S.; Frey, H. M.; Leutwyler, S. Supersonic Jet UV Spectrum and Nonradiative Processes of the Thymine Analogue 5Methyl-2-hydroxypyrimidine. Phys. Chem. Chem. Phys. 2010, 12, 5032−5040. (24) Lobsiger, S.; Frey, H. M.; Leutwyler, S.; Morgan, P.; Pratt, D. S(0) and S(1) State Structure, Methyl Torsional Barrier Heights, and Fast Intersystem Crossing Dynamics of 5-Methyl-2-hydroxypyrimidine. J. Phys. Chem. A 2011, 115, 13281−13290. (25) Abouaf, R.; Pommier, J.; Dunet, H.; Quan, P.; Nam, P.-C.; Nguyen, M. T. The Triplet State of Cytosine and Its Derivatives: Electron Impact and Quantum Chemical Study. J. Chem. Phys. 2004, 121, 11668−11674. (26) Merchan, M.; Serrano-Andrés, L.; Robb, M. A.; Blancafort, L. Triplet-State Formation along the Ultrafast Decay of Excited Singlet Cytosine. J. Am. Chem. Soc. 2005, 127, 1820.

3580

dx.doi.org/10.1021/jz301719h | J. Phys. Chem. Lett. 2012, 3, 3576−3580