Cytosine Radical Cations: A Gas-Phase Study Combining IRMPD Spectroscopy, UVPD Spectroscopy, Ion–Molecule Reactions, and Theoretical Calculations

Article abstract of DOI:10.1002/cphc.201700281

The radical cation of cytosine (Cyt^.+) is generated by dissociative oxidation from a ternary Cu^II complex in the gas phase. The radical cation is characterized by infrared multiple photon dissociation (IRMPD) spectroscopy in the fingerprint region, UV/Vis photodissociation (UVPD) spectroscopy, ion–molecule reactions, and theoretical calculations (density functional theory and ab initio). The experimental IRMPD spectrum features diagnostic bands for two enol-amino and two keto-amino tautomers of Cyt^.+ that are calculated to be among the lowest energy isomers, in agreement with a previous study. Although the UVPD action spectrum can also be matched to a combination of the four lowest energy tautomers, the presence of a nonclassical distonic radical cation cannot be ruled out. Its formation is, however, unlikely due to the high energy of this isomer and the respective ternary Cu^II complex. Gas-phase ion–molecule reactions showed that Cyt^.+ undergoes hydrogen-atom abstraction from 1-propanethiol, radical recombination reactions with nitric oxide, and electron transfer from dimethyl disulfide.

Full text of DOI:10.1002/cphc.201700281

Accepted Article
Title: Cytosine radical cation: a gas-phase study combining IRMPD
spectroscopy, UV-PD spectroscopy, ion-molecule reactions, and
theoretical calculations
Authors: Michael Lesslie, John T Lawler, Andy Dang, Joseph A Korn,
Daniel Bim, Vincent Steinmetz, Philippe Maitre, Frantisek
Turecek, and Victor Ryzhov
This manuscript has been accepted after peer review and appears as an
Accepted Article online prior to editing, proofing, and formal publication
of the final Version of Record (VoR). This work is currently citable by
using the Digital Object Identifier (DOI) given below. The VoR will be
published online in Early View as soon as possible and may be different
to this Accepted Article as a result of editing. Readers should obtain
the VoR from the journal website shown below when it is published
to ensure accuracy of information. The authors are responsible for the
content of this Accepted Article.
To be cited as: ChemPhysChem 10.1002/cphc.201700281
Link to VoR: http://dx.doi.org/10.1002/cphc.201700281
A Journal of
www.chemphyschem.org

10.1002/cphc.201700281
ChemPhysChem
FULL PAPER
Cytosine radical cation: a gas-phase study combining IRMPD
spectroscopy, UV-PD spectroscopy, ion-molecule reactions, and
theoretical calculations
Michael Lesslie,[a] John T. Lawler,[a] Andy Dang,[b] Joseph A. Korn,[b] Daniel Bím,[c] Vincent Steinmetz,[d]
Philippe Maître,[d] Frantisek Tureček*[b] and Victor Ryzhov*[a]
Abstract: The radical cation of cytosine (Cyt•+) was generated via
dissociative oxidation from a ternary Cu(II) complex in the gas phase. cancer.[2] Whether direct, (radiation induced) or indirect (via
consequences such as strand breaks, mutagenesis, and
The radical cation was characterized by infrared multiple photon
dissociation (IRMPD) spectroscopy in the fingerprint region, UV-VIS
photodissociation (UV-PD), ion-molecule reactions, and theoretical
calculations (density functional theory and ab initio). The IRMPD
experimental spectrum has diagnostic bands for two enol/amino and
two keto/amino tautomers of Cyt•+, that were calculated to be among
the lowest-energy isomers in concert with a previous work (J.
Wolken, et al., Int. J. Mass Spectrom. 2007). While UV-PD action
spectrum can also be matched to a combination of the four lowest-
energy tautomers, it does not rule out the presence of a non-
classical distonic radical cation. Its formation is, however, unlikely
due to the high energy of this isomer and the respective ternary
Cu(II) complex. Gas-phase ion-molecule reactions showed that Cyt•+
undergoes hydrogen atom abstraction from n-propyl thiol, radical
recombination reactions with nitric oxide, and electron transfer from
dimethyl disulfide.
chemical reaction), oxidation often results in the formation of a
nucleobase radical cation within the DNA strand.[3] This species
undergoes a variety of reactions.[4] Most notably, charge
transfer via electron hole-migration is followed by proton transfer,
which disrupts the hydrogen bonding network leading to
eventual strand scission.[4]
It is notable that the local environment experienced by
nucleobases within DNA is different from model systems in
water. In that regard, gas-phase data can be useful in providing
insight into intrinsic chemistry of the often-transient nucleobase
radicals. Such solvent-free analysis is unencumbered by multi-
conformational averaging and counterion effects, allowing for
direct experimental observation of the specific radical species.
Multiple gas-phase studies have been devoted to the
charged even-electron species of nucleobases and nucleosides
in an attempt to uncover the intrinsic structure and properties of
the DNA building blocks. Early mass spectrometry based
characterization methods, such as kinetic methods, lacked the
ability to accurately discriminate between the multiple low-
energy structures which are possible for the nucleobases and
nucleosides. However, they were successful in characterization
of properties such as proton affinity,[5] and metal-binding
energetics.[6] Likewise, ion-mobility mass spectrometry (IM-MS)
was successful in the characterization of the conformational
Introduction
The building blocks of DNA are susceptible to undesired
chemical modifications that may result in loss or change of
function.[1] For nucleobases, radical formation via oxidative
chemical reactions or ionizing radiation is of particular concern
as such modifications can have particularly severe
shapes
of
oligonucleotides,[7]
dinucleotides,[8]
and
mononucleotides.[9] Such IM-MS studies, which focus on large
flexible species, are unfortunately unable to resolve the
nucleobase conformers due to their similar cross sectional area.
The development of infrared multiple photon dissociation
(IRMPD) spectroscopy, in combination with computational
chemistry, presented a novel step forward for the direct
structural observation of gas phase ions, and has been
leveraged extensively for the structural elucidation of the DNA
[a]
[b]
M. Lesslie, J. T. Lawler, V. Ryzhov*
Department of Chemistry and Biochemistry
Northern Illinois University
DeKalb, IL 60115, USA
Email: ryzhov@niu.edu
A. Dang, J. A. Korn, F. Tureček*
Department of Chemistry
constituents.
Maître, Salpin, and co-workers originally
University of Washington
Bagley Hall, Box 351700
determined the protonated structure of the gas-phase pyrimidine
nucleobases to be predominantly the lowest energy enol/amino
form with only minor contributions of the keto/amino tautomer.[10]
This work was quickly followed by the analysis of monohydrated
Seattle, Washington 98195, USA
Email: turecek@chem.washington.edu
D. Bím
Institute of Organic Chemistry and Biochemistry
Academy of Sciences of the Czech Republic
166 10 Prague 6, Czech Republic
V. Steinmetz, P. Maître
Laboratoire de Chimie Physique
Université Paris-Sud
UMR8000 CNRS
[c]
[d]
nucleobases.[11]
Since those initial studies, IRMPD
spectroscopy has been utilized to examine a multitude of
species related to DNA constituents, including, nucleosides,[12]
nucleotides,[13] nucleobase dimers,[14] and metal-bound
species.[15]
Significantly less work has been done, however, on the
examination of one-electron oxidation products of DNA and its
91405 Orsay, France
Supporting information for this article is given via a link at the end of
the document.
This article is protected by copyright. All rights reserved.

10.1002/cphc.201700281
ChemPhysChem
FULL PAPER
constituents (i.e., nucleobase radical cations) in the gas phase.
This first instances revolved around the use of the high-energy
electron impact (EI) ionization method to study the fragmentation
pathways of the nucleobase radical cations.[16] Neutralization-
reionization mass spectrometry (NR-MS) has been used in
combination with high-level theoretical calculations to model,
study the energetics, and examine the probable structures of the
nucleobases and their corresponding radical cations.[17] Most
recently, a mixture of cytosine radical cation isomers has been
produced by synchrotron-based VUV photoionization of gas-
phase cytosine produced in a molecular beam and characterized
by ionization energy measurements that were in accord with
high-level ab initio calculations.[18]
formed via CID inside the ion trap. To avoid such complication,
we complement the IRMPD spectroscopy of the cytosine radical
cation with the higher-energy single-photon technique, UV-PD
action spectroscopy. Gas-phase spectroscopic experiments in
this range have previously been utilized for the protonated and
monohydrated forms of thymine and uracil.[23] Ion-molecule
reactions (IMR) are carried out with various neutrals in order to
further examine the chemistry of the cytosine radical cation.
Computational chemistry methods are utilized to evaluate low-
energy conformations and estimate theoretical absorbance
bands.
An alternative nucleobase radical cation generation
method was required to be compatible with the mass
spectrometers used for gas-phase spectroscopic methods.
Dissociation of ternary copper complexes under collision-
induced dissociation conditions in ion traps was originally shown
to generate singly-charged radical cations of peptides (Eq. 1).[19]
Subsequently, this method was extended to the generation of
Results and Discussion
Cytosine radical cation isomers
When the nucleobases are bound within a DNA strand, the
hydrogen bonding network of the Watson-Crick pairing results in
highly-favored tautomeric conformations of each base. For
cytosine, the keto/amino form is preferred. However, the free
nucleobases can possess a number of different tautomeric
nucleobase and nucleoside radical cations.[20]
Using this
technique, O’Hair and co-workers were able to successfully
determine the tautomeric form of the substituted nucleobase 9-
methyl guanine radical cation,[21] as well as the nucleoside
deoxyguanosine radical cation, via IRMPD spectroscopy.[22]
forms.
The neutral cytosine molecule has seven formal
tautomers (Scheme 2) from which one-electron oxidation would
result in independent radical cations of the nucleobase. When
considering these seven tautomers computationally, rotational
isomers must also be taken into account, resulting in fifteen
possible radical cation species of cytosine (Figure S1).
[CuII(terpy)(M)]•2+  [CuI (terpy)]+ + M•+
(Eq. 1)
While these studies demonstrate the applicability of gas-phase
spectroscopic techniques in defining the native structure of the
DNA constituents, the structures of the unsubstituted
nucleobase radical cations in the gas phase formed via oxidative
dissociation are yet to be determined.
The current work focuses on defining the gas-phase
conformational composition of the radical cation of cytosine.
Protonated cytosine was found to be a mixture of the two lowest
energy tautomers, the enol/amino and keto/amino species B and
C (Scheme 1), both formed via protonation of the most stable
neutral species A (conventional numbering used throughout this
work is shown in red).[10, 11b]
Scheme 2. The seven tautomers of neutral Cyt.
The relative energetics of the cytosine tautomeric forms
have previously been determined computationally, both as a
17g]
neutral species and as radical cations.[11b,
Herein, the
Scheme 1. Low energy tautomers of Cyt (A) and protonated Cyt (B & C).
cytosine radical cation was initially explored at the
B3LYP/6311++G(d,p) level of theory. This approach was found
to reproduce the relative energy rankings of the low-energy
tautomers at the highest level of theory previously used,
CCSD(T)/6-311++G(3df,2p), with only minor exceptions (notably
species 3, corresponding to the keto amino N1H tautomer, is
found ~ 5 kJ mol-1 higher in energy at both the CCSD(T)/6-
311++G(3df,2p) and CCSD(T)/aug-cc-pVTZ level of theory).[17g]
In an effort to define the specific tautomeric mixture
protonated cytosine ions were hydrated in the gas phase. The
water molecule attachment significantly lowered the dissociation
barrier and allowed for observation of the species in the N-H/O-
H stretch region.[11b] Such an approach, however, is not easily
implemented for the radical cation of cytosine, which has to be
This article is protected by copyright. All rights reserved.

10.1002/cphc.201700281
ChemPhysChem
FULL PAPER
Figure 1. Optimized structures of the cytosine radical cation at the B3LYP/6-
311++G(d,p) level of theory. Relative energies calculated at that level are
displayed including zero-point vibrational energies and referring to 0 K in kJ
mol-1 (CCSD(T)/6-311++G(3f,2p)[17g] and CCSD(T)/aug-cc-pVTZ are given in
parentheses for comparison). The spin electron density on heteroatoms is
shown in italics.
Figure 2. Experimental (black/gray) and theoretically calculated (red) IR
spectra of the four lowest energy cytosine radical cations (1-4). The linear
combination (LC 1-4) is based on equal parts of each isomer. Theoretical
spectra were calculated at the B3LYP/6-311++G(d,p) level of theory with the
FWHM set to 30 cm-1 and scaled using a factor of 0.98.
In fact, the same seven species are predicted within 15 kJ mol-1
of the lowest energy conformer by both computational methods
(Figure 1, 1-7). Of these species, all isomers possessing the
amino form (1-4) and isomer 7 were found within 10 kJ mol-1 of
the reference isomer 4. All other structures were found to be
higher in energy (see Figure S1).
absent from the experimental spectrum (Figure S3). The
enol/imino with a proton at the N3 position (5) also displays a
poor match to the experimental spectrum (Figure S3).
Combined with the tautomer existing at over 10 kJ mol-1 higher
energy than the lowest energy isomer at both levels of theory, it
is unlikely that this species is present in any substantial
population under equilibrium conditions. The four remaining
amino-type isomers (1-4) were found to be the lowest in energy
when calculated using the B3LYP/6311++G(d,p) level of theory
(within 3 kJ mol-1, Figure 1). This energy difference is small
enough to consider that these isomers will be populated. The
experimentally observed spectrum (Figure 2) clearly indicates a
mixture of isomers and has diagnostic bands for both enol/amino
(1 and 2) and keto/amino species (3 and 4) as discussed below.
A linear combination (LC) of the four isomers was considered
and displayed a suitable match for the experimental spectra (see
LC 1-4 in Figure 2).
Gas-phase action spectroscopy – IRMPD
The infrared multiple photon dissociation (IRMPD) action
spectrum was collected in the “fingerprint region” (1000 – 1900
cm-1) for the cytosine radical cation. The experimental spectrum
for the cytosine radical cation reveals eight notable peaks
(Figure 2 and Table 1). Based on the number of observed
bands alone, it is unlikely that the profile corresponds to a single
isomeric species in the gas phase.
Comparison to the
theoretically calculated absorbance bands for each of the seven
low-energy isomers confirms the lack of a mono-isomeric match
(Figure 2, see Figure S3 for the simulated IR of all remaining
This situation is similar to that of the protonated cytosine
previously studied by Maître, Salpin, and co-workers via IRMPD
action spectroscopy.[10] The two lowest energy isomers of
[Cyt+H]+, shown as species B and C (Scheme 1), within 0.3 kJ
cytosine radical structures).
In particular, the calculated
keto/imino tautomers (6 and 7) have a strong absorbance near
1850 cm-1, corresponding to the C=O stretching motion, which is
This article is protected by copyright. All rights reserved.

10.1002/cphc.201700281
ChemPhysChem
FULL PAPER
Table 1. Experimental and theoretical vibrational frequencies for the cytosine radical cation.
Exp.[a]
Theoretical cytosine radical cation
Vibrational modes[c]
1[b]
2[b]
3[b]
4[b]
1650 (0.68)
1560 (0.87)
1525 (0.94)
1632 (98)
1630 (65)
1645 (494)
1667 (218)
1641 (133)
sNH2
C=O + sNH2
1561 (380)

asN3C2O + sNH2 + OH
C5C6 + sNH2, highly coupled
asN3C2O + asNH2 + OH
1576 (132)
1486 (91)
1365 (63)
1526 (253)
1516 (404)

N3C4 + C2O + N1C6
N3C4 + asNH2 + N1C6
1516 (128)
1485 (39)
1505 (1.0)
1491 (71)

asN8C4C5 + C5H
N1H + C4N8 + sNH2
C4C5, highly coupled
N1C2 + OH + C6H
C5C6 + C6H
1491 (94)
1403 (34)
1415 (0.72)
1380 (0.67)
1240 (0.30)
1100 (0.28)
1429 (72)
1367 (72)
1218 (160)
1410 (397)
1367 (59)
1220 (110)
C6H, highly coupled
N3C4, highly coupled
OH
N1H + NH2 + C5H + C6H
NH2 + N1C6 + C=O + N1H + C6H
1231 (42)
1152 (134)
1116 (66)
1082 (47)

asN1C2N3 + C5C6 + C5H + C6H
NH2+ C2N3 + C5H
[a] Experimental frequency in cm-1 (relative intensity in parenthesis). [b] Theoretical frequency in cm-1 (km mol-1 intensity in parenthesis). Scaled by
a factor of 0.98. [c] See numbering system in Scheme 1.
mol-1 of one another, were found to be potentially contributing to
the experimentally observed IRMPD spectrum in the fingerprint
region. Additional isomers were calculated to be much higher in
energy and did not match the experimental IR spectrum.[11b]
Analysis of the individual absorbance bands and the
corresponding theoretically predicted vibrational modes (Table
1) reveals that each isomer contributes substantially to the
observed experimental spectrum. The peak at ~ 1650 cm-1
correlates primarily to the strong absorbance from the scissoring
motion of the NH2 group (sNH2) which was most intense in the
keto/amino forms 3 and 4. Combined carbonyl stretching motion
(C=O) and NH2 scissoring was found with a relatively high
intensity in 4, also contributing to the 1650 cm-1 peak. The
enol/amino species 1 and 2 have less intense sNH2 modes that
are slightly red-shifted from their keto/amino counterparts and
may cause the peak shoulder observed at ~1630 cm-1. A sharp
peak at ~1560 cm-1 in the IRMPD spectrum may result from the
coupled asymmetric stretching of the C2-N3-O (asN3C2O), the
asymmetric NH2 scissoring (sNH2), and the OH bending (OH)
vibrational mode calculated to be an intense mode of species 2.
Assigning the split peaks at ~1505 cm-1 and ~1525 cm-1 is
challenging. However, species 1 has two strong vibrational
modes in this range at 1526 cm-1 (the analogous mode to that
found in species 2 at ~1560 cm-1) and 1516 cm-1, which is the
most intense mode from the coupled N3-C4 (N3C4), C2-OH
(C2O), and N1C6 (N1C6) stretching. Species 2 possess a
similar mode at 1516 cm-1 that likely also contributes to the large
observed peak in this region. All four of the low energy
structures calculated have less intense and highly-coupled
vibrational modes in the 1485-1500 cm-1 region, likely resulting
in the broadening and tailing of the larger peak towards the
lower frequencies. The observed peak at 1415 cm-1 aligns well
with the intense mode calculated for species 2 for the N1-C2
(N1C2) stretch coupled with the OH (OH) and C6H (C6H)
bending motions. An analogous vibrational mode is also present,
although less intense, in species 1. The neighboring peak at
~1380 cm-1 is closely approximated by the highly coupled C6H
motion of species 1 & 2 at 1367 cm-1 and the N3C4 mode in
species 3 at 1365 cm-1. In the lower-frequency region of the
experimental spectrum, several low intensity peaks are
observed and correspond to highly coupled vibrational modes.
Notably, the O-H bending motion in the enol/amino species 1
and 2 is calculated to be reasonably intense at ~1220 cm-1,
possibly contributing substantially to the broad experimental
peak found at ~1240 cm-1. The final peak near 1100 cm-1 is
matched most reasonably by two vibrational modes calculated
for species 4 (at 1116 cm-1 and 1082 cm-1) which bracket the
experimentally observed peak.
Gas-phase action spectroscopy – UV-PD
Single-photon UV-PD action spectroscopy was also
acquired for the cytosine (M) radical cation generated from the
[CuII(terpy)(M)]•2+ complex. The experimental UV-PD spectra in
the 210 – 700 nm range for the cytosine radical cations were
generated from the combination of two fragmentation channels:
the loss of 28 and 42 Da (Figure 3). These fragments were
consistently seen in both IRMPD and CID experiments and are
assigned to the neutral losses of CO and NCO.
Such
eliminations were found previously and explored theoretically.[17g]
As the barrier toward such eliminations was found higher in
energy than the transition states leading to isomerization of the
cytosine radical cation,[17g] the fragments cannot be diagnostic in
the determination of the species present in the gas phase. The
action spectrum showed several bands that were represented by
the neutral losses mentioned above. Weak bands were
observed at 600, 530, and 455 nm, whereas stronger bands
appeared at 400, 350, 280 (shoulder), and 260 nm. The short
This article is protected by copyright. All rights reserved.

10.1002/cphc.201700281
ChemPhysChem
FULL PAPER
wavelength band at <210 nm appeared only in the m/z 69 (loss
of NCO) channel. The action spectrum of cytosine cation radical
substantially differs from the UV spectrum of neutral cytosine
which shows the longest wavelength absorption maximum at
270 nm.[24]
Interestingly, none of the four low-energy isomers of the
cytosine radical cation had the same calculated absorption band
pattern as that found in the experimental UV-PD spectrum
(Figure 3a, see Figure S4 for additional theoretical spectra).
The long wavelength band at 600 nm is represented by an
absorption band in the spectrum of 1 (Figure 3b) whereas the
530 and 455 nm bands appear in 1 and 4 (Figure 3d). The 350
nm band appears in 1 and 3 (Figure 3c) whereas the 400 nm
band appears in 4. Isomer 2 UV-PD spectrum (Figure S4) is
very similar to that of 1. All these isomers also show strong
absorption bands at 230-260 nm approximately matching the
260 nm experimental band. Thus, concurring with the IRMPD
data, the UV-PD action spectra indicate the presence of several
cytosine cation radical isomers that are formed by intramolecular
electron transfer in the gas-phase [CuII(terpy)(M)]•2+ complex.
Interestingly, a reasonable match to the 250-450 nm region of
the action spectrum was found for the calculated absorption
spectrum of the high-energy isomer 8 (Figure 3e). This feature,
not revealed by IRMPD, does not rule out the potential presence
of the formally distonic radical cation (with the charge at the
amine and the unpaired spin on the oxygen atom, species 8).
However, this ion is more than 120 kJ mol-1 higher in energy
than the keto/amino and enol/amino tautomers (Figure 1 and
Table S1), and so its formation appears unlikely and is
discussed further below.
Comparison of UV and IR spectroscopy results
Given that both IR and UV action spectroscopy data point to a
mixture of several Cyt●+ isomers present in the gas phase during
laser activation, their origin needs to be discussed. As the
radical cation of cytosine is formed via CID of the
[CuII(terpy)(M)]•2+ complex, the structure of this complex, both in
solution and in the gas phase, may affect the structure of the
resulting Cyt●+ ions. The relative energies of selected
[CuII(terpy)(M)]•2+ complexes were obtained for fully optimized
structures (Figure S5), corresponding to ligation of Cu(terpy) by
different cytosine tautomers (we considered two binding modes
of 1, and one each for 4 and 8). Consistent with the order of
relative energies for Cyt●+, the N1-ligated [Cu(terpy)(4-N1)]●2+
complex was found to be lowest in energy, closely followed by
the isomer 1 also bound via N1. Since the complexes are
formed in solution, solvation effects were considered by running
Polarizable Continuum Model[25] calculations in water and
methanol. Although there are some energy differences between
the complexes due to different solvation, the relative differences
between isomers 4 and 1 bound to Cu(terpy) in solution are still
minor. Therefore, it is reasonable to suggest that the mixture of
Cyt●+ isomers results from the isomeric Cu complexes found in
solution.
Figure 3. (a) Experimental UV-PD action spectra of the cytosine radical cation
displaying individual fragmentation channels, total fragments, and the
smoothed pattern as indicated in the legend. The spectrum is compiled from
the average of two individual runs on separate days. Calculated TD-DFT M06-
2X/6-311++G(2d,p) (black lines) of (b) 1, (c) 3, (d) 4, and (e) 8. The green
error bars are from Newton-X calculations of vibronic transitions at 300 K.
Red bars represent the EOM-CCSD/6-31+G(d,p) absorption lines. The red
curves depict wavelength-benchmarked vibronic spectra based on a linear
correlation of M06-2X and EOM-CCSD calculated transition energies.
Formation of the Cyt●+ isomeric mixture is also possible
during CID of the [CuII(terpy)(M)]•2+ complexes when Cyt●+ ions
This article is protected by copyright. All rights reserved.

10.1002/cphc.201700281
ChemPhysChem
FULL PAPER
dissociating from the complex carry excess energy sufficient for
nitric oxide even when the spin density at the oxygen atom is as
low as 0.17.[27]
isomerization.
It
has
previously
been
established
computationally by Wolken et al. that isomerization between the
low energy isomers of cytosine radical cation would take less
energy than their fragmentation.[17g] The most energetically
favorable elimination channel (loss of CO) was found to have a
critical transition state energy of 180 kJ mol-1, relative to the
lowest energy isomer (4), at the CCSD(T)/6-311++G(3df,2p)
level of theory.[17g] Isomerization pathways of species 1-4, were
found to proceed through transition states with energies of 145 -
174 kJ mol-1 relative to ion 4 at the same level of theory.[17g]
Therefore, is not inconceivable that the radical cation of cytosine
formed during the CID of the Cu complex will be able to
rearrange before the spectroscopy experiments. This would
result in a mixture of low energy isomers as the best match to
the experimental IRMPD and UV-PD spectroscopy data.
We investigated the possibility that the distonic ion (8) was
co-formed via dissociation of the copper complex (Eq. 1, where
M = Cyt, and Scheme S1). The properties of this highly
energetic isomer and its ability to rearrange into more stable
species is discussed in detail in the supporting information
(Discussion of isomer 8, Schemes S2-S3, Figures S6-S7).
However, the O-ligated [Cu(terpy)(8-O)]●2+ complex was found
higher in energy than the 4 counterpart by 97 kJ mol-1 and
higher than the other isomers in the gas phase (Figure S5 and
Table S3). Although the energy differences between the
complexes somewhat decreased due to different solvation,
complex [Cu(terpy)(8-O)]●2+ was still >90 kJ mol-1 less stable
than [Cu(terpy)(4-N1)]●2+ and also substantially less stable than
the other solvated complexes in both water and methanol. Thus,
is seems unlikely that [Cu(terpy)(8-O)]●2+ was present under
equilibrium conditions in bulk solution.
Electron transfer reaction (Eq. 3) dominated the reactivity
with dimethyl disulfide (IP  8.2 eV[28]), as demonstrated by the
appearance of the charged product at m/z 94. Such reactions
often occur when a more stable radical ion can be formed (of the
two neutral species in Eq. 3, the one with the lower IP will
become the cation). We found such reactions typical for a highly
delocalized -radical of tryptophan.[26] For Cyt•+, the ionization
energies were calculated previously to be in the range of 8.2 –
8.9 eV depending on the specific tautomer and level of
theory,[17g] in agreement with the experimental values[29] that
have recently been narrowed to 8.695-8.738 eV.[29]
The hydrogen abstraction reaction was found to be the
main process for the cytosine radical cation in the presence of n-
propyl thiol (Eq. 4). This reaction leads to the protonated
cytosine and can also occur via the O- or N-based radical site. In
effect, that will result in a new O-H or N-H bond and a thiyl
radical (RS•). This type of hydrogen atom abstraction was
theoretically predicted by Tehrani et al. as a typical reaction for
the cytosine radical cation.[30] They calculated the reaction to be
exothermic by ~75 kJ mol-1 and noted that only a small activation
barrier was required.[30]
While this reactivity cannot be used to discern the isomeric
composition of Cyt•+, it is consistent with the reactivity trends
predicted for nucleobase radicals. These trends suggest that if
an electron is stripped from the Cyt site, resulting in a transient
radical cation species, a facile hydrogen atom or electron
transfer reaction may follow. This will lead to propagation of the
radical site either intramolecularly within the DNA chain or
intermolecularly through reactions with neighboring species.
Conclusions
Ion-molecule reactions (IMR)
The gas-phase reactivity of the cytosine radical cation
(Cyt•+) was explored toward a variety of neutral reagents
including nitric oxide, dimethyl disulfide, and n-propyl thiol. This
resulted in the following reactivity (where M = Cyt):
This combined experimental and computational study allows us
to arrive at the following conclusions. Cytosine radical cations
are formed as a mixture of isomers upon collision-induced
oxidation of the cytosine ligand via dissociation from the copper
(II) ternary complex, [CuII(terpy)(M)]•2+. A mixture of several low-
energy isomers was also found in a recent investigation of
cytosine radical cations formed via VUV photodissociation.[18]
This is consistent with the existence of multiple Cyt•+ isomers
within 10 kJ mol-1 of one another as found by multiple level of
theory. In this work, the Cyt•+ isomeric mixture may originate
from different [CuII(terpy)(M)]•2+ isomers present in solution or
from the gas-phase rearrangement of Cyt•+ upon CID of the Cu
complex. IRMPD action spectroscopy points to a mixture of low
energy amino-type Cyt•+ tautomers (1-4) that contribute to the
vibrational signature in the fingerprint region. The UVPD action
spectroscopy spectrum similarly contains features of the low
energy Cyt•+ ions. The presence of a high-energy distonic isomer
8 which displayed a reasonable UV-PD match was considered.
Radical recombination: M•+ + •NO  M-NO+
(Eq. 2)
•+
Electron transfer: M•+ + CH3SSCH3  M + CH3SSCH3 (Eq. 3)
Hydrogen abstraction: M•+ + RSH  MH+ + RS•
(Eq. 4)
The corresponding mass spectra are given in Figure S8.
•
Exposing Cyt•+ to nitric oxide produced the NO addition
reaction (Eq. 2) characteristic of -radicals.[26] It is notable that
this reaction can potentially occur at both the N- and O-based
radical locations in cytosine radical cation. In the enol/amino
structures (1 and 2) there is a substantial spin density (0.33) at
N1 atom and only minimal unpaired spin at the hydroxyl O7.
The keto/amino structures (3 and 4) possess significant spin
density at the carbonyl O7 (0.49 and 0.27, respectively), as well
as considerable spin density at the unprotonated N (0.29 at N3
and 0.40 at N1, respectively). We have previously shown that
highly delocalized phenoxy radicals display reactivity towards
Analysis of energy data for gas-phase Cyt•+
and
[CuII(terpy)(M)]•2+ complexes in solution and the gas phase
indicates that the formation of isomer 8 is unlikely. As all
isomers possess highly delocalized unpaired spin, ion-molecule
This article is protected by copyright. All rights reserved.

10.1002/cphc.201700281
ChemPhysChem
FULL PAPER
directed into a PG142C unit (Altos Photonics, Bozeman, MT, USA). The
reactions are unable to shed light on the isomeric composition of
the Cyt radical cation population. They do, however, reveal that
the cytosine radical cations are highly reactive and undergo
facile hydrogen atom tranfer, electron transfer, and radical
recombination reactions.
PG142C unit consists of
a third harmonic generator and optical
parametric oscillator coupled with an optional second harmonic generator
(SH), and provides wavelength tuning between 210-700 nm at 0.52-
12.69 mJ/pulse peak. The power of the PG142C output beam was
measured at each wavelength using an EnergyMax-USB J-10MB energy
sensor (Coherent Inc., Santa Clara, CA, USA). After recording the power
readings, the energy sensor was removed from the beam path, and the
PG142C output beam (6-mm diameter) was focused through a small
aperture drilled in the CI source and into the linear ion trap. In this
particular case, the typical experimental setup consisted of electro-
spraying our copper ternary complex solution into the ion trap, performing
CID on the doubly-charged complex to produce the desired singly-
charged nucleobase cation radical, and then isolating the nucleobase
cation radical for photo-activation from 210-700 nm. The intensities of the
resulting UVPD-MS3 photo-fragments were monitored as a function of
wavelength, and their final intensities were normalized to the laser output
power to plot the action spectra. The number of laser pulses used during
each isolation depended on the degree of photo-dissociation at each
given wavelength, and ranged from 1 to 19 pulses (i.e. 100 to 1000 ms
activation time, respectively, with each successive pulse spaced by 50
ms).
More specifically, the prepared cytosine-copper ternary complex
solution was introduced into the ESI source of the mass spectrometer at
a flow rate of 1 uL/min in the positive ion mode. The spray voltage,
capillary temperature, and capillary voltage were adjusted to 2.7 kV,
275 °C, and 15 V, respectively. The doubly-charged cytosine-copper
ternary complex (m/z 203.5) was isolated using a 2 m/z isolation window
and subjected to CID at 35 NCE for 30 ms. The resulting singly-charged
cytosine cation radical (m/z 111) produced via oxidative dissociation from
the ternary complex was then isolated with a 5 m/z isolation window and
subjected to action spectroscopy via UVPD. The number of laser pulses
used throughout the SSH/FSH (210-354 nm), ESH (355-409 nm), and
PG region (410-700 nm) ranged from 3, 15, and 1 pulse(s), respectively
(i.e. 200, 750, and 100 ms activation times). “Blank” background scans
were also recorded to avoid “false” photo-fragment peak assignments
due to spectral noise produced by the laser.
Experimental Section
Materials. All chemicals and reagents were used as received without any
additional purification. Cytosine, copper(II) nitrate trihydrate, 2,2’:6’,2”-
terpyridine, n-propyl thiol, and dimethyl disulfide were purchased from
Sigma-Aldrich (Milwaukee, WI). Methanol (HPLC grade) was purchased
from Fisher Scientific (Pittsburg, PA). Premixed gas containing 1% nitric
oxide in laser grade helium was purchased from Airgas (Chicago, IL).
Water was purified (18 MΩ) in house.
Sample preparation. Stock solutions (1 mg/mL) of cytosine, and
2,2’:6’,2”-terpyridine (terpy) were each prepared in methanol, and
Cu(NO3)2 (1 mg/mL) was prepared in 50:50 methanol/water. The stock
solutions of terpy and Cu(NO3)2 were combined 1:1 and diluted 10-fold in
methanol. An aliquot of this solution was then combined 1:1 with the
stock solution cytosine, vortexed and allowed to react for 10 min. at room
temperature. The resulting mixture was diluted with methanol to an
appropriate concentration for direct infusion into the mass spectrometer.
Cytosine radical generation was accomplished via successive isolation of
the ternary metal complex, [CuII(terpy)(M)]2+, where M= cytosine, CID of
the complex, and isolation of the dissociation product M•+.
Ion-molecule reactions. The isolated radical cations were subjected to
IMRs with the neutral species nitric oxide, n-propyl thiol, and dimethyl
disulfide via a modified Bruker Esquire 3000 quadrupole ion trap mass
spectrometer (Bruker Daltronics, Bremen, Germany), as previously
described.[31] Reactions with nitric oxide were achieved by introducing
nitric oxide into the trap via the helium line from a 1% nitric oxide in
helium premixed cylinder. The internal helium regulator was unmodified
from normal usage. The dimethyl disulfide and n-propyl thiol were
introduced to the trap through a leak valve. In all cases, a scan delay of
500 ms was employed as the reaction time.
Calculations. All calculations were performed using the Gaussian 09
quantum chemical program.[34] Geometries were optimized within the
framework of Becke’s three-parameter DFT hybrid functional (B3LYP)
and the 6-311++G(d,p) basis set.
All optimized structures were
IRMPD action spectroscopy. Gas-phase infrared spectra of the cytosine
radical cation were acquired using the FEL beamline at the Centre Laser
Infrarouge d'Orsay (CLIO) facility. Solutions were prepared as previously
described. Radical formation was achieved by using CID in the modified
ion trap of the Bruker Esquire 3000+ (Bruker Daltonics, Bremen,
Germany) quadrupole ion trap mass spectrometer, and were
subsequently mass selected. Details of the instrumental setup are
described elsewhere.[32] IRMPD was performed on the mass-selected
ions, and mass spectra were recorded over eight averages with a
wavenumber step of 4.5 cm-1, an irradiation time of 1 s, and laser power
of 700-1500 mW. To ensure reproducibility, each spectrum was acquired
twice. The IRMPD spectra were produced based on -ln (fragment
subjected to vibrational frequency analyses to ensure that they
corresponded to minima (no imaginary frequencies) and transition states
(1 imaginary frequency). For comparison of theoretical IR absorbance
bands and the experimental IRMPD spectroscopy, a scaling factor of
0.98 was applied to the predicted IR spectra as calculated at the
B3LYP/6311++G(d,p) level. This scaling factor is known to be
appropriate for DFT comparison with IRMPD spectra.[35] Predicted IR
peaks were convoluted using Gaussian profiles with a full width at half-
maximum of 30 cm-1.
Another set of optimized geometries and relative energies of all
cytosine cation-radical tautomers were obtained with the ωB97X-D[36]
and M06-2X[37] hybrid functionals using the 6-311+G(2d,p) basis set. At
our highest level of theory, select structures were optimized with
UMP2(full)/6-31G(d,p) and used for single-point energy coupled-cluster
calculations[38] with single, double, and disconnected triple excitations,[39]
CCSD(T), with the aug-cc-pVTZ basis set,[40] providing benchmark
relative energies. All these calculations were performed for doublet spin
states within the spin-unrestricted formalism.Vertical excitation energies
and transition intensities (oscillator strengths) were calculated for lowest
25 excited states with time-dependent DFT calculations[41] using the
ωB97X-D and M06-2X functionals and the 6-311+G(2d,p) basis set.
Another set of excitation energies were obtained with equation-of-motion
ion/fragment+parent ion) signal ratio as
wavenumber.
a function of excitation
UV-PD action spectroscopy. Tandem MS (i.e. UVPD-MSn and CID-MSn)
and action spectroscopy were performed on an LTQ-XL-ETD linear ion
trap (ThermoElectron Fisher, San Jose, CA) modified with an external
EKSPLA NL301G (Altos Photonics, Bozeman, MT, USA) Nd-YAG laser
source operating at 20 Hz frequency with a 3-6 ns pulse width.[33] The
pump laser is interfaced to the LTQ via LabView software (National
Instruments, Austin, TX, USA), and generates photons which are
This article is protected by copyright. All rights reserved.

10.1002/cphc.201700281
ChemPhysChem
FULL PAPER
[13]
[14]
[15]
Y. w. Nei, K. T. Crampton, G. Berden, J. Oomens, M. T. Rodgers,
J. Phys. Chem. A 2013, 117, 10634-10649.
B. Yang, R. R. Wu, G. Berden, J. Oomens, M. T. Rodgers, J. Phys.
Chem. B 2013, 117, 14191-14201.
a) B. Yang, R. Wu, N. Polfer, G. Berden, J. Oomens, M. Rodgers,
J. Am. Soc. Mass Spectrom. 2013, 24, 1523-1533; b) E. A. Gillis,
K. Rajabi, T. D. Fridgen, J. Phys. Chem. A 2008, 113, 824-832; c)
E. A. Gillis, T. D. Fridgen, Int. J. Mass Spectrom. 2010, 297, 2-8; d)
K. Rajabi, E. A. Gillis, T. D. Fridgen, J. Phys. Chem. A 2010, 114,
3449-3456.
a) J. M. Rice, G. O. Dudek, M. Barber, J. Am. Chem. Soc. 1965,
87, 4569-4576; b) J. M. Rice, G. O. Dudek, J. Am. Chem. Soc.
1967, 89, 2719-2725.
a) F. Tureček, J. Mass Spectrom. 1998, 33, 779-795; b) J. K.
Wolken, F. Turecek, J. Phys. Chem. A 1999, 103, 6268-6281; c) J.
K. Wolken, F. Tureček, J. Phys. Chem. A 2001, 105, 8352-8360;
d) X. Chen, E. A. Syrstad, M. T. Nguyen, P. Gerbaux, F. Tureček,
J. Phys. Chem. A 2005, 109, 8121-8132; e) C. Yao, M. L.
Cuadrado‐Peinado, M. Polášek, F. Tureček, Angew. Chem. Int. Ed.
2005, 44, 6708-6711; f) F. Tureček, Adv. Quantum Chem. 2007,
52, 89-120; g) J. K. Wolken, C. Yao, F. Tureček, M. J. Polce, C.
Wesdemiotis, Int. J. Mass Spectrom. 2007, 267, 30-42.
Z. Chen, K.-C. Lau, G. A. Garcia, L. Nahon, D. K. Božanić, L.
Poisson, M. M. Al-Mogren, M. Schwell, J. S. Francisco, A. Bellili, M.
Hochlaf, J. Am. Chem. Soc. 2016, 138, 16596-16599.
a) I. K. Chu, C. F. Rodriquez, T.-C. Lau, A. C. Hopkinson, K. W. M.
Siu, J. Phys. Chem. B 2000, 104, 3393-3397; b) I. K. Chu, C. F.
Rodriguez, A. C. Hopkinson, K. W. M. Siu, T.-C. Lau, J. Am. Soc.
Mass Spectrom. 2001, 12, 1114-1119; c) E. Bagheri-Majdi, Y. Ke,
G. Orlova, I. K. Chu, A. C. Hopkinson, K. W. M. Siu, J. Phys. Chem.
B 2004, 108, 11170-11181.
a) S. Wee, R. A. J. O'Hair, W. D. McFadyen, Rapid Commun.
Mass Spectrom. 2005, 19, 1797-1805; b) A. K. Lam, B. F.
Abrahams, M. J. Grannas, W. D. McFadyen, R. A. J. O'Hair, Dalt.
Trans. 2006, 5051-5061.
L. Feketeová, G. N. Khairallah, B. Chan, V. Steinmetz, P. Maître, L.
Radom, R. A. J. O'Hair, Chem. Comm. 2013, 49, 7343-7345.
L. Feketeová, B. Chan, G. N. Khairallah, V. Steinmetz, P. Maître, L.
Radom, A. Richard, Phys. Chem. Chem. Phys. 2015, 17, 25837-
25844.
calculations with coupled clusters and single and double excitations
(EOM-CCSD)[42] using the 6-31+G(d,p) basis set and, for structures 4
and 8 also using the 6-311+G(2d,p) basis set, to benchmark the TD-DFT
energies.[43]
The
EOM-CCSD/6-31+G(d,p)
and
6-311+G(2d,p)
calculations of 4 and 8 showed nearly identical results (Table S2,
Supporting Information) with root mean square deviations of absorption
wavelentgths of rmsd < 2.3 nm. The smaller 6-31+G(d,p) basis set was
then used in EOM-CCSD calculations of all cytosine cation radicals.
Comparison of the TD-DFT and EOM-CCSD data showed a closer
agreement between the M06-2X and EOM-CCSD benchmark
computations for most of the cytosine structures (Table S2, Supporting
Information), and this functional was therefore used to further compute
the vibronically-broadened absorption spectra at 300 K.[44]
Vibronically-broadened spectra were generated with Newton-X
(version 1.4, www.newtonx.org),[45] PuTTY 0.67 SSH suite (Simon
Tatham, 1997-2016), Xming (Colin Harrison, 2005-2007), and WinSCP
(Martin Prikryl, 2000-2016) programs. Optimized Cartesian atomic
coordinates and harmonic frequencies obtained with M06-2X/6-
31+G(d,p) were used to generate random configurations that were
weighted according to their Boltzmann factors at 300 K. A total of 12
excited electronic states and 500 configurations were used to produce
each spectrum.
[16]
[17]
[18]
[19]
Acknowledgements
[20]
This study was supported by Northern Illinois University. M.L.
thanks the Graduate School of NIU for the Dissertation
Completion Fellowship. F.T. thanks the Chemistry Division of the
National Science Foundation for funding (Grants CHE-1359810
and CHE-1543805).
[21]
[22]
[23]
[24]
S. Ø. Pedersen, C. S. Byskov, F. Turecek, S. B. Nielsen, J. Phys.
Chem. A 2014, 118, 4256-4265.
a) F. Zaloudek, J. S. Novros, L. B. Clark, J. Am. Chem. Soc. 1985,
107, 7344-7351; b) H. Du, R. C. A. Fuh, J. Li, L. A. Corkan, J. S.
Lindsey, Photochem. Photobiol. 1998, 68, 141-142; c) C. T.
Middleton, K. de La Harpe, C. Su, Y. K. Law, C. E. Crespo-
Hernández, B. Kohler, Annu. Rev. Phys. Chem. 2009, 60, 217-239.
J. Tomasi, B. Mennucci, R. Cammi, Chem. Rev. 2005, 105, 2999-
3094.
A. Piatkivskyi, S. Osburn, K. Jaderberg, J. Grzetic, J. D. Steill, J.
Oomens, J. Zhao, J. K.-C. Lau, U. H. Verkerk, A. C. Hopkinson, V.
Ryzhov, J. Am. Soc. Mass Spectrom. 2013, 24, 513-523.
M. Lesslie, A. Piatkivskyi, J. Lawler, T. R. Helgren, S. Osburn, R. A.
J. O'Hair, V. Ryzhov, Int. J. Mass Spectrom. 2015, 390, 124-131.
D. T. Leeck, H. I. Kenttämaa, Org. Mass. Spectrom. 1994, 29, 106-
107.
a) D. Dougherty, E. Younathan, R. Voll, S. Abdulnur, S. McGlynn,
J. Electron. Spectrosc. Relat. Phenom. 1978, 13, 379-393; b) N.
Hush, A. S. Cheung, Chem. Phys. Lett. 1975, 34, 11-13.
Z. A. Tehrani, M. J. Javan, A. Fattahi, M. M. Hashemi, J. Theor.
Comput. Chem. 2012, 11, 313-327.
Y. Pyatkivskyy, V. Ryzhov, Rapid Commun. Mass Spectrom. 2008,
22, 1288-1294.
L. Mac Aleese, A. Simon, T. B. McMahon, J.-M. Ortega, D. Scuderi,
J. Lemaire, P. Maître, Int. J. Mass Spectrom. 2006, 249, 14-20.
a) C. J. Shaffer, R. Pepin, F. Tureček, J. Mass Spectrom. 2015, 50,
1438-1442; b) H. T. Nguyen, C. J. Shaffer, R. Pepin, F. Tureček, J.
Phys. Chem. Lett. 2015, 6, 4722-4727; c) E. Viglino, C. J. Shaffer,
F. Tureček, Angew. Chem. Int. Ed. 2016.
M. J. Frisch, G. W. Trucks, H. B. Schlegel, G. E. Scuseria, M. A.
Robb, J. R. Cheeseman, G. Scalmani, V. Barone, B. Mennucci, G.
A. Petersson, H. Nakatsuji, M. Caricato, X. Li, H. P. Hratchian, A. F.
Izmaylov, J. Bloino, G. Zheng, J. L. Sonnenberg, M. Hada, M.
Ehara, K. Toyota, R. Fukuda, J. Hasegawa, M. Ishida, T. Nakajima,
Y. Honda, O. Kitao, H. Nakai, T. Vreven, J. A. Montgomery Jr., J. E.
Peralta, F. Ogliaro, M. J. Bearpark, J. Heyd, E. N. Brothers, K. N.
Kudin, V. N. Staroverov, R. Kobayashi, J. Normand, K.
Raghavachari, A. P. Rendell, J. C. Burant, S. S. Iyengar, J. Tomasi,
M. Cossi, N. Rega, N. J. Millam, M. Klene, J. E. Knox, J. B. Cross,
V. Bakken, C. Adamo, J. Jaramillo, R. Gomperts, R. E. Stratmann,
Keywords: nucleobases • radical ions • UV-PD spectroscopy •
IRMPD spectroscopy • ion-molecule reactions
[1]
[2]
[3]
a) R. P. Rastogi, A. Kumar, M. B. Tyagi, R. P. Sinha, J. Nucleic
Acids 2010, 2010; b) A. Kumar, M. D. Sevilla, Chem. Rev. 2010,
110, 7002-7023.
a) J. Cadet, T. Douki, J.-L. Ravanat, Free Radical Biol. Med. 2010,
49, 9-21; b) C. J. Burrows, J. G. Muller, Chem. Rev. 1998, 98,
1109-1152.
W. A. Glass, M. N. Varma, Physical and chemical mechanisms in
molecular radiation biology, Vol. 58, Springer Science & Business
Media, 2012.
[25]
[26]
[27]
[28]
[29]
[4]
[5]
S. Steenken, Chem. Rev. 1989, 89, 503-520.
a) F. Greco, A. Liguori, G. Sindona, N. Uccella, J. Am. Chem. Soc.
1990, 112, 9092-9096; b) A. Liguori, A. Napoli, G. Sindona, R.
Cooks, Rapid Commun. Mass Spectrom. 1994, 8, 89-93.
M. Rodgers, P. B. Armentrout, J. Am. Chem. Soc. 2000, 122,
8548-8558.
[30]
[31]
[32]
[33]
[6]
[7]
J. Gidden, J. E. Bushnell, M. T. Bowers, J. Am. Chem. Soc. 2001,
123, 5610-5611.
[8]
[9]
J. Gidden, M. T. Bowers, Eur. Phys. J. D 2002, 20, 409-419.
a) D. Liu, T. Wyttenbach, M. T. Bowers, J. Am. Chem. Soc. 2006,
128, 15155-15163; b) J. Gidden, M. T. Bowers, J. Phys. Chem. B
2003, 107, 12829-12837.
[10]
[11]
J. Y. Salpin, S. Guillaumont, J. Tortajada, L. MacAleese, J.
Lemaire, P. Maitre, ChemPhysChem 2007, 8, 2235-2244.
a) J. M. Bakker, R. K. Sinha, T. Besson, M. Brugnara, P. Tosi, J.-Y.
Salpin, P. Maître, J. Phys. Chem. A 2008, 112, 12393-12400; b) J.
M. Bakker, J.-Y. Salpin, P. Maître, Int. J. Mass Spectrom. 2009,
283, 214-221.
[34]
[12]
a) J. Y. Salpin, D. Scuderi, Rapid Commun. Mass Spectrom. 2015,
29, 1898-1904; b) A. Filippi, C. Fraschetti, F. Rondino, S. Piccirillo,
V. Steinmetz, L. Guidoni, M. Speranza, Int. J. Mass Spectrom.
2013, 354, 54-61; c) R. Wu, B. Yang, C. Frieler, G. Berden, J.
Oomens, M. Rodgers, Phys. Chem. Chem. Phys. 2015, 17, 25978-
25988; d) R. Wu, B. Yang, C. Frieler, G. Berden, J. Oomens, M.
Rodgers, J. Phys. Chem. B 2015, 119, 5773-5784.
This article is protected by copyright. All rights reserved.

10.1002/cphc.201700281
ChemPhysChem
FULL PAPER
O. Yazyev, A. J. Austin, R. Cammi, C. Pomelli, J. W. Ochterski, R.
[39]
[40]
[41]
[42]
G. D. Purvis, R. J. Bartlett, J. Chem. Phys. 1982, 76, 1910-1918.
T. H. Dunning Jr, J. Chem. Phys. 1989, 90, 1007-1023.
F. Furche, R. Ahlrichs, J. Chem. Phys. 2002, 117, 7433-7447.
D. C. Comeau, R. J. Bartlett, Chem. Phys. Lett. 1993, 207, 414-
423.
L. Martin, K. Morokuma, V. G. Zakrzewski, G. A. Voth, P. Salvador,
J. J. Dannenberg, S. Dapprich, A. D. Daniels, Ö. Farkas, J. B.
Foresman, J. V. Ortiz, J. Cioslowski, D. J. Fox, Gaussian, Inc.,
Wallingford, CT, USA, 2009.
[35]
a) J. Steill, J. Zhao, C. K. Siu, Y. Ke, U. H. Verkerk, J. Oomens, R.
C. Dunbar, A. C. Hopkinson, K. W. M. Siu, Angew. Chem. 2008,
120, 9812-9814; b) J. Zhao, J. K.-C. Lau, J. Grzetic, U. H. Verkerk,
J. Oomens, K. W. M. Siu, A. C. Hopkinson, J. Am. Soc. Mass.
Spectrom. 2013, 24, 1957-1968; c) O. Hernandez, B. Paizs, P.
Maître, Int. J. Mass Spectrom. 2015, 377, 172-178.
J.-D. Chai, M. Head-Gordon, Phys. Chem. Chem. Phys. 2008, 10,
6615-6620.
[43]
a) J. A. Sonk, H. B. Schlegel, J. Phys. Chem. A 2011, 115, 11832-
11840; b) V. Riffet, D. Jacquemin, E. Cauet, G. Frison, J. Chem.
Theory Comp. 2014, 10, 3308-3318; c) F. Turecek, J. Phys. Chem.
A 2015, 119, 10101-10111.
M. Barbatti, A. J. Aquino, H. Lischka, Phys. Chem. Chem. Phys.
2010, 12, 4959-4967.
M. Barbatti, M. Ruckenbauer, F. Plasser, J. Pittner, G. Granucci, M.
Persico, H. Lischka, Wiley Interdisciplinary Reviews:
Computational Molecular Science 2014, 4, 26-33.
[44]
[45]
[36]
[37]
[38]
Y. Zhao, D. G. Truhlar, Theor. Chem. Acc. 2008, 120, 215-241.
J. Čížek, J. Paldus, L. Šroubková, Int. J. Quantum Chem. 1969, 3,
149-167.
This article is protected by copyright. All rights reserved.

10.1002/cphc.201700281
ChemPhysChem
FULL PAPER
Entry for the Table of Contents (Please choose one layout)
Layout 1:
FULL PAPER
All mixed up: The cytosine radical
cation (Cyt•+) is present as a mixture
of several low-energy amino-type
tautomers in the gas phase when
formed via dissociation from the
Michael Lesslie, John T. Lawler, Andy
Dang, Joseph A. Korn, Daniel Bím,
Vincent Steinmetz, Philippe Maitre,
Frantisek Tureček* and Victor Ryzhov*
Page No. – Page No.
ternary Cu(II) complex.
isomers are distinguished
inspection of IR and UV action
spectroscopy techniques, ion-
molecule reactions, and theoretical
calculations.
Specific
via
Cytosine radical cation: a gas-phase
study combining IRMPD
spectroscopy, UV-PD spectroscopy,
ion-molecule reactions, and
theoretical calculations
This article is protected by copyright. All rights reserved.