Reaction Dynamics of CN Radicals in Acetonitrile Solutions

Article abstract of DOI:10.1021/acs.jpca.5b10720

The bimolecular reactions that follow 267 nm ultraviolet photolysis of ICN in acetonitrile solution have been studied using transient absorption spectroscopy on the picosecond time scale. Time-resolved electronic absorption spectroscopy (TEAS) in the ultraviolet and visible spectral regions observes rapid production and loss (with a decay time constant of 0.6 ± 0.1 ps) of the photolytically generated free CN radicals. Some of these radicals convert to a solvated form which decays with a lifetime of 8.5 ± 2.1 ps. Time-resolved vibrational absorption spectroscopy (TVAS) reveals that the free and solvated CN-radicals undergo geminate recombination with I atoms to make ICN and INC, H atom abstraction reactions, and addition reactions to solvent molecules to make C₃H₃N₂ radical species. These radical products have a characteristic absorption band at 2036 cm^-1 that shifts to 2010 cm^-1 when ICN is photolyzed in CD₃CN. The HCN yield is low, suggesting the addition pathway competes effectively with H atom abstraction from CH₃CN, but the delayed growth of the C₃H₃N₂ radical band is best described by reaction of solvated CN radicals through an unobserved intermediate species. Addition of methanol or tetrahydrofuran as a cosolute promotes H atom abstraction reactions that produce vibrationally hot HCN. The combination of TEAS and TVAS measurements shows that the rate-limiting process for production of ground-state HCN is vibrational cooling, the rate of which is accelerated by the presence of methanol or tetrahydrofuran.

Full text of DOI:10.1021/acs.jpca.5b10720

Article
pubs.acs.org/JPCA
Reaction Dynamics of CN Radicals in Acetonitrile Solutions
Daisuke Koyama,† Philip Coulter,† Michael P. Grubb,† Gregory M. Greetham,‡ Ian P. Clark,‡
and Andrew J. Orr-Ewing*,†
†School of Chemistry, University of Bristol, Cantock’s Close, Bristol BS8 1TS, U.K.
‡Central Laser Facility, Research Complex at Harwell, Science and Technology Facilities Council, Rutherford Appleton Laboratory,
Harwell Oxford, Didcot, Oxfordshire OX11 0QX, U.K.
S
* Supporting Information
ABSTRACT: The bimolecular reactions that follow 267 nm ultraviolet
photolysis of ICN in acetonitrile solution have been studied using transient
absorption spectroscopy on the picosecond time scale. Time-resolved electronic
absorption spectroscopy (TEAS) in the ultraviolet and visible spectral regions
observes rapid production and loss (with a decay time constant of 0.6 0.1 ps)
of the photolytically generated free CN radicals. Some of these radicals convert to
a solvated form which decays with a lifetime of 8.5
2.1 ps. Time-resolved
vibrational absorption spectroscopy (TVAS) reveals that the free and solvated
CN-radicals undergo geminate recombination with I atoms to make ICN and
INC, H atom abstraction reactions, and addition reactions to solvent molecules
to make C3H3N2 radical species. These radical products have a characteristic
absorption band at 2036 cm−1 that shifts to 2010 cm−1 when ICN is photolyzed
in CD3CN. The HCN yield is low, suggesting the addition pathway competes effectively with H atom abstraction from CH3CN,
but the delayed growth of the C3H3N2 radical band is best described by reaction of solvated CN radicals through an unobserved
intermediate species. Addition of methanol or tetrahydrofuran as a cosolute promotes H atom abstraction reactions that produce
vibrationally hot HCN. The combination of TEAS and TVAS measurements shows that the rate-limiting process for production
of ground-state HCN is vibrational cooling, the rate of which is accelerated by the presence of methanol or tetrahydrofuran.
employed as a source of CN radicals.4,5 These reactive radicals
typically abstract hydrogen atoms from organic molecules as
1. INTRODUCTION
Elementary bimolecular chemical reactions are the basic steps
of many more-complex chemical processes; consequently, their
shown by eq 1, and the reactions with alkanes have been
extensively studied in the gas phase.9−14
rates, mechanisms and reaction dynamics have been extensively
studied by both experimental and theoretical methods.1,2 Many
CN + RH → HCN(v1, v2, v3) + R
(1)
of these studies have concentrated on radical reactions in the
gas phase. Under gas-phase conditions dominated by isolated
collisions, competing reaction pathways can be distinguished, as
can the release of excess energy from exothermic reactions into
specific translational, rotational, vibrational or electronic
degrees of freedom of the reaction products. These types of
observation reveal detailed information about the structures of
key intermediates along the reaction path.
Reactive collisions cannot be considered to occur in isolation
in the liquid phase, where many important synthetic and
biochemical processes take place. Instead, the surrounding
solvent interacts continuously with the reacting species and can
modify energy barriers, reaction pathways, product yields and
the flow of any energy released.3 These interactions occur on
ultrafast (femtosecond to picosecond) time scales, but
nonequilibrium reaction dynamics reminiscent of the gas-
phase can persist if reactive events are fast enough to compete
with the response of the surrounding solvent.4−8
Here, (v1,v2,v3) denotes the number of quanta of the three
vibrational modes of HCN, the C−N stretch (v1), the bend
(v2), and the C−H stretch (v3). These reactions typically
release more than 100 kJ mol−1 of energy, and a significant
fraction flows into the internal vibrational modes of the
products. The potential energy surfaces (PESs) for these
reactions have early and low energy barriers, with a flat angular
dependence at the transition state.15,16 Consequently, the
energy released by the reaction excites the HCN in the C−H
stretching and bending vibrations.
Reactions of the type illustrated by eq 1 have also been
investigated in solution in common chlorinated organic
solvents. For example, Orr-Ewing and co-workers explored
the reaction dynamics of CN radicals with various organic
molecules (cyclohexane, d12-cyclohexane, tetramethylsilane,
tetrahydrofuran (THF), acetone) in chloroform and dichloro-
To investigate some of the effects of solvent on the dynamics
and pathways of bimolecular reactions, ultraviolet (UV)
photolysis of dissolved cyanogen iodide (ICN) has been
Received: November 2, 2015
Revised: December 3, 2015
Published: December 4, 2015
© 2015 American Chemical Society
12924
DOI: 10.1021/acs.jpca.5b10720
J. Phys. Chem. A 2015, 119, 12924−12934

The Journal of Physical Chemistry A
Article
methane using time-resolved vibrational absorption spectros-
copy (TVAS).17−21 The experimental measurements demon-
strated that the degree of vibrational excitation of the nascent
products is reduced in the liquid phase, but signatures of the
gas-phase dynamics remain. Complete quenching of nascent
vibrational excitation by interaction with the solvent bath can
take tens to hundreds of picoseconds in chloroform and
dichloromethane solutions.4,5,17−19,22 Modifications of the
reaction mechanism by interactions with the solvent bath
include formation of complexes between CN radicals and
solvent molecules.23,24
The current study extends these prior investigations to
reactions of CN radicals in acetonitrile (CH3CN or CD3CN),
which is chosen because it is a more strongly interacting solvent
than chloroform and dichloromethane. The response of
acetonitrile molecules to changes in polarity in solutes is
known to be fast,25,26 with subpicosecond restructuring of the
solvent shell. We consider reactions of the CN radical with both
the acetonitrile solvent and with THF or methanol (CH3OH)
present as cosolutes. We observe the production and loss of
CN radicals by transient electronic absorption spectroscopy
(TEAS), and use TVAS to monitor the outcomes of competing
chemical reactions. Abstraction reactions of the type
represented by eq 1 occur, but we also report evidence of a
pathway involving addition of a CN radical to the nitrile group
of the acetonitrile solvent.
ascertained by comparison with steady state FT-IR spectrum of
stable compounds such as CH3CN, CD3CN, and HCN.
3. RESULTS AND DISCUSSION
Unravelling the chemistry that occurs following UV photolysis
of ICN in acetonitrile solution requires information from both
TEAS and TVAS experiments. The TEAS measurements
identify the production, solvation and loss of CN radicals and
are reported first because they establish the time scales for
reactive removal of CN. The TVAS measurements examine the
products of CN radical reactions, as well as geminate
recombination of CN with I atoms. Comparison of the
product-formation time scales with the CN-loss rates obtained
by TEAS reveals reactive pathways and provides evidence for
intermediate species that are not directly observed in our
experiments. We concentrate first on the ICN/CH3CN and
ICN/CD3CN solutions to establish the processes that occur
when CN is produced photolytically in acetonitrile, and then
consider the changes that result from addition of THF or
methanol.
3.1. TEAS of Solution of ICN in CH3CN Following 267
nm Excitation. Figure 1 shows TEA spectra of 290 mM ICN
in CH3CN following 267 nm UV excitation. The spectra can be
decomposed into three contributing features, in accord with
prior analysis by Rivera et al. and illustrated for a TEA spectrum
obtained at a time delay of 0.6 ps in Figure 1b.30 We performed
this decomposition in the KOALA program.31 Figure 1c shows
exponential fits to the time-dependence of the intensities of
these three spectral features, and the corresponding time
constants are summarized in Table 1.
A sharp band centered at 389 nm at early time delays grows
within our instrument response limit (∼200 fs) and decays with
a time constant of 0.6 0.1 ps (Figure 1c). The location of this
peak matches the B ← X electronic transition of the CN radical
in the gas phase, and in agreement with Rivera et al. and
Crowther et al., we assign it to nonsolvated (“free”) CN
radicals.23,24,30,32 In response to the decay of the nonsolvated
CN radicals, we see evolution of a broad new component
peaking to the shorter wavelength side of the nonsolvated CN
radicals with a time constant for growth of 0.6 0.1 ps. Given
this correspondence between time constants, we assign this
broad feature to solvated CN radicals, although a contribution
from I atoms must also be considered. Similar correspondences
between decay of free CN radicals and growth of solvated CN
radicals are also observed in other solvent systems.21 Following
its rapid growth, the solvated CN radical absorption decays, and
as Figure 1c shows, fitting this decay to a biexponential function
gives time constants of 8.5 2.1 and 108 26 ps. The faster
decay component of the solvated CN radical is attributed
predominantly to reaction with CH3CN and, to a lesser extent,
geminate recombination to ICN and INC, on the basis of
TVAS results presented in section 3.2.
The third component of the spectra at short times is a broad
band peaking around 580 nm and extending across much of the
visible region. Similar bands were observed previously by Rivera
et al. following ICN photolysis in water and ethanol and
assigned to a charge-transfer (CT) band of spin−orbit excited
and solvated I*(2P1/2) atoms.30 The corresponding band of
ground-state I(2P3/2) atoms lies to shorter wavelength and
partially overlaps the solvated CN absorption band. The I* CT
band decays rapidly because of spin−orbit relaxation, as shown
in Figure 1c, with concomitant growth of the I(2P3/2)-solvent
CT feature expected. The evidence presented here shows that
2. EXPERIMENTAL DETAILS
Transient absorption spectroscopy experiments were per-
formed using the ULTRA laser system at the Central Laser
Facility of the Rutherford Appleton Laboratory27 or an ultrafast
laser system at the University of Bristol. The details of each
system and our experimental methods have been described
previously;19,28 therefore, only a brief description relating
directly to the current study is provided here.
ICN (98%; Acros Organics) was recrystallized from toluene
(Sigma-Aldrich, analytical grade) solution before use. The ICN
samples were prepared as 290 mM solutions in CH3CN,
CD3CN, THF in CH3CN, neat THF, or methanol in CH3CN
(all solvents from Sigma-Aldrich, analytical grade; CD3CN-d3
99.5 atom %). Measurements were carried out in a Harrick cell
with a 380-μm thick PTFE spacer and CaF2 windows and the
5−10 mL sample solutions were circulated by a peristaltic
pump. The reactions were initiated by a 267 nm UV pulse
(duration ∼50 fs; 1 μJ per pulse) and probed by transient
infrared (IR) absorption using a ∼500 cm−1 (∼300 cm−1 for
the Bristol laser system) bandwidth IR laser pulse for TVAS
experiments or by transient UV/vis absorption using a
broadband white light continuum (340−620 nm) for TEAS
experiments. The chosen ICN concentration gave an
absorbance of 1.0 at the 267 nm excitation wavelength.
Density functional theory (DFT) calculations of harmonic
vibrational frequencies were performed using the Gaussian 09
package and the B3LYP density functional with the 6-311+
+G(3df,3pd) basis set.29 This method was chosen because of its
good balance between computational efficiency and reliability,
as well as the availability of scaling factors for infrared frequency
calculations. DFT calculations using the BPW91 functional
were also conducted to confirm the results obtained using the
B3LYP method. Various types of methods for calculation of
harmonic vibrational frequencies were tested, but calculated
frequencies obtained with the BPW91 functional provided the
best results for our system without using a scaling factor, as
12925
DOI: 10.1021/acs.jpca.5b10720
J. Phys. Chem. A 2015, 119, 12924−12934

The Journal of Physical Chemistry A
Article
recombination with other radical or atomic species, for example
forming I2.
At the longest time delays in our measurements, a spectral
feature remains that is centered at 400 nm. A similar feature
was reported in Dunning et al.’s recent study of ICN photolysis
in acetone solutions, and was assigned to a stable CN-acetone
complex.20 One possible assignment of the 400 nm band is
therefore to a CN-acetonitrile complex, but this appears to be
only a partial explanation. In our ICN/acetonitrile solutions we
−
see evidence for development of an I3 band centered at 365
nm even without any laser irradiation. Therefore, we interpret
the 400 nm feature as a combination of a CN-acetonitrile
−
complex absorption band and a bleach feature derived from I3 .
−
The interference from the I3 bleach becomes more significant
when THF is added and valuable information is lost in
CH3CN/THF solutions. Therefore, TEA spectra for solutions
containing THF are not shown and discussed.
3.2. TVAS of Solutions of ICN in CH3CN Following 267
nm Excitation. One pathway for reaction of CN radicals with
CH3CN might involve the H atom abstraction of eq 2 in which
HCN is a product. TVAS observations of these HCN products
were restricted to the C−N stretching region at wavenumbers
around 2050 cm−1 because of strong interferences in the C−H
stretching region from CH3CN modes.
CN + CH3CN → HCN + CH2CN
(2)
Examples of TVA spectra obtained by photolysis of 290 mM
ICN in CH3CN and CD3CN are shown in Figure 2, panels a
and b. None of the time-dependent spectral features are seen in
experiments conducted on CH3CN or CD3CN samples
without ICN solute. A bleach of the ICN absorption band
induced by the pump laser is not shown in the figure, but it is
observed as a negative-going signal at 2167 cm−1 in both
solvents and shows almost no recovery. The band at 2118 cm−1
in the CD3CN solution is attributed to the solvent by
comparison with steady state FT-IR spectra. This negative-
going band increases in depth with greater delay times, which
suggests reactive loss of solvent molecules because the transient
feature is not observed in the absence of ICN. The time-
dependence of the integrated band intensity is shown in Figure
2(c). Following a rapid onset of the bleach of the CD3CN
band, its further development is fitted well by a biexponential
function with time constants of 8.7 1.8 ps and 82 30 ps.
The subpicosecond initial growth of the CD3CN bleach
suggests some reactive removal by the free CN radicals
produced by ICN photolysis. The subsequent 8.7 ps time
constant for the development of the CD3CN bleach feature
matches that obtained from TEAS data for decay of solvated
CN radicals in CH3CN (Table 1). This correspondence
supports the reaction of acetonitrile with solvated CN radicals
as a contributing cause of the time-dependence of both the
TEAS and TVAS bands. The cause of the longer-time loss of
CD3CN (with τ2 = 82 30 ps) is uncertain, but the relative
amplitudes of the different components of the time-evolution of
this band show that this is a minor pathway compared to the
reactive removal by CN radicals.
Figure 1. Transient electronic absorption spectra of 290 mM ICN in
CH3CN following 267 nm excitation and the spectral decomposition
into component absorptions. (a) TEA spectra for selected time delays
from 0−1300 ps, with a color code provided as an inset key. (b)
Example of decomposition of the spectrum obtained at a time delay of
0.6 ps into its constituent parts. (c) Time-dependence of the free and
solvated CN radical absorption bands and the I* charge-transfer band
obtained from decomposition of TEA spectra. The solid lines are
exponential fits with time constants reported in Table 1. The long-time
decay of intensity in the region spanned by the solvated CN band is a
consequence of an overlapping solvent-I atom charge transfer band.
Table 1. Time Constants Obtained from Exponential Fits to
Time-Dependent Intensity Data from TEA Spectra for
a
Bands Labelled as Free CN, Solvated CN and I*.
τ1/ps
τ2/ps
τ3/ps
free CN
0.6 0.1
0.6 0.1
1.5 0.1
b
c
solvated CN
8.5 2.1
108 26
I*(2P1/2
)
a
b
Uncertainties are 2 SD from the fits. This time constant corresponds
c
to growth of the spectral feature. Overlapping contribution from
I(2P3/2) atoms.
the early time temporal behavior of the broad band extending
from the near-UV to the visible region is more consistent with
assignment to loss of solvated CN radicals than to ground-state
I atoms. However, the slower (108 26 ps) decay component
of this feature is not observed in experimental studies of BrCN
photolysis in acetonitrile,21 and is therefore attributed to the
I(2P3/2) atoms. These ground-state I atoms are lost by diffusive
The band centered at 2068 cm−1 is assigned to the CN
stretching band of INC(v = 0) formed by geminate
recombination on the basis of our and others’ prior
observations.17−21,33−36 This assignment is further supported
by the fact there is no isotope shift when CH3CN is replaced by
CD3CN. The INC(v = 0) band grows with a time constant of
7.7 0.7 ps in CH3CN that is commensurate with the loss of
12926
DOI: 10.1021/acs.jpca.5b10720
J. Phys. Chem. A 2015, 119, 12924−12934

The Journal of Physical Chemistry A
Article
Figure 2. Transient vibrational absorption spectra obtained following 267 nm UV excitation of 290 mM ICN in (a) CH3CN and (b) CD3CN
obtained in the 1980−2130 cm−1 range. Inset color keys identify spectra obtained at selected time delays. Band assignments are indicated and
discussed further in the main text. Time-dependences of integrated band intensities are shown for (c) the CD3CN band at 2118 cm−1 and (d) the
weak HCN band in CH3CN at 2092 cm−1. Exponential or biexponential fits to these time-dependent intensities give the reported time constants.
Table 2. Observed and Calculated Fundamental Infrared Frequencies and Intensities of the H2CCNCHN and D2CC
NCDN Radicals formed by CN Addition to Acetonitrile
observed in TVAS
frequency/cm−1
B3LYP/6-311++G(3df,3pd)
BPW91/6-311++G(3df,3pd)
frequency/cm−1 intensity/km mol−1
a
frequency/cm−1
intensity/km mol−1
H2CCNCHN
D2CCNCDN
2036
2010
2016
1992
567
481
2043
2021
408
345
Frequencies calculated at the B3LYP level were multiplied by a factor of 0.9604.39
a
CN seen by TEAS with τ = 8.5 2.1 ps, followed by a more
constants indicates that these peaks derive from corresponding
reactions in CH3CN and CD3CN. The isotope shift of 26 cm−1
demonstrates that the responsible species contains H or D
atoms and this species is derived from CH3CN or CD3CN
because these peaks are not observed in other solvents such as
chloroform or dichloromethane.17−21 Assignment to CH2CN
(or CD2CN) products of reaction 2 was discounted because we
conducted experiments on the UV photolysis of ICH2CN in
acetonitrile and did not observe the same spectral features.
DFT calculations also fail to predict CH2CN/CD2CN bands in
proximity to the observed features and with the correct isotope
shifts. These measurements and calculations are reported in the
Supporting Information. Consequently, we assign these peaks
to the radical products of addition of CN to acetonitrile:
gradual increase with a time constant of 341
21 ps. In
addition to the fundamental C−N stretching band of INC, we
assign the peaks at 2043 and 2018 cm−1 to absorption from the
v = 1 and 2 vibrational levels of the CN stretching mode of
INC.21 These bands are shifted to lower wavenumber than the
fundamental band by intervals corresponding to the 12.5 cm−1
diagonal anharmonicity of this mode deduced from electronic
structure calculations.21 The value is also consistent with the
anharmonicity of the CN radical vibration.37 These two hot
bands grow with the same time constants as the fundamental
INC band, and decay with time constants of 282 67 ps for
the INC(v = 2) band and 630 76 ps for the INC(v = 1) band
that are in a ratio consistent with expectations from Landau−
Teller theory.38 The geminate recombination is therefore
concluded to produce some vibrationally excited INC
molecules, and competes with ICN, but is a minor
recombination channel (accounting for only a few percent of
the dissociated I atoms and CN radicals) observable only
because of the large IR transition dipole moment of the CN
stretching mode.33 We estimate that approximately 30% of the
initially formed INC is vibrationally excited in the CN
stretching mode by comparing IR band intensities at early time
delay (∼15 ps), with account taken for differences in the
transition dipole moments for the bands.
•
•
CN + CH3CN(CD3CN) → C3H3N2 (C3D3N2 )
(3)
•
Our DFT calculations identified numerous C3H3N2 (or
•
C3D3N2 ) adducts lower in energy than separated CN and
CH3CN (see Supporting Information for further details), but
on the basis of computed vibrational frequencies, isotope shifts
and band intensities, our preferred assignment is to H2CC
NCHN in CH3CN and D2CCNCDN in CD3CN, with
the structures shown in Figure 2, parts a and b. The computed
wavenumbers and intensities of the bands of these two
structures in this spectral region are reported in Table 2; the
large transition dipole moments facilitate observation of these
bands in our TVA spectra. The observed vibrational mode
corresponds to stretching of the central CN bond. Addition
of CN to the C atom of the nitrile group in acetonitrile is also
The bands at 2036 cm−1 in CH3CN and 2010 cm−1 in
CD3CN grow with time constants of 50 3 ps and 50 2 ps
respectively if their time-dependent intensities are fitted to
single exponential functions. The agreement between time
12927
DOI: 10.1021/acs.jpca.5b10720
J. Phys. Chem. A 2015, 119, 12924−12934

The Journal of Physical Chemistry A
Article
Figure 3. (a) Transient vibrational absorption spectra of UV-photoexcited 290 mM ICN in CH3CN in the 1675−1850 cm−1 range. A small bleach
feature observed at around 1825 cm−1 is attributed to a CH3CN solvent band. (b) Time dependence of the bands at 1733 and 1756 cm −1. The
lower two panels show integrated band intensities and fits to the sequential reaction model of eq 4 for (c) the H2CCNCHN band at 2036 cm−1
in CH3CN and (d) the D2CCNCDN band at 2010 cm−1 in CD3CN both of which are evident in Figure 2.
plausible, but the computed vibrational transition dipole
moments of the resulting CH3C(N•)CN radicals are 1 or 2
orders of magnitude smaller in our spectroscopic window (see
Supporting Information). These possible alternative reaction
products may therefore be overlooked.
10−2 ps−1), as well as the 8.5 2.1 ps decay of solvated CN
radicals and the 8.7 1.8 ps growth of the CD3CN bleach. A
common time scale is thus observed for CN radical
consumption by these competing reaction and recombination
pathways. The isomerization or quenching step to H2CC
NCHN or D2CCNCDN radicals is estimated to occur
with a τ2 = 1/k2 ≈ 33 ps time constant. The band at 1756 cm−1
(Figure 3, parts a and b) grows with a time constant of 18.4
2.4 ps when fitted by a single exponential function and its
growth is slower than the expected reaction time scale. The
carrier of this band might therefore be formed through other
sequential process associated with the addition reaction 3.
The weak TVAS band at 2092 cm−1 in CH3CN (Figure 2a)
is assigned to the fundamental CN stretching band of the
ground vibrational state of HCN (denoted by HCN(0))
formed by reaction 2. This assignment is made on the basis that
the peak position agrees with a steady state FT-IR spectrum of
HCN obtained by photolysis of ICN in CH3CN, and is not
observed for experiments conducted in CD3CN. The time-
dependent integrated band intensities plotted in Figure 2d
suggest positive but declining intensities at the earliest time
delays (the first 5 ps) followed by a growth with a time constant
of 73 17 ps when fitted by a single exponential function. The
resulting integrated absorbance remains constant from
approximately 150 ps until the limit of our experimental time
delay (3 ns). The early time positive intensity may be a
consequence of baseline noise affecting the low HCN band
intensity, or could be a signature of vibrationally hot products
formed on this time scale.
The time constants for formation of the radical adduct
products (∼50 ps) are significantly slower than the expected
reaction time scale of ∼8.5 ps for solvated CN (Table 1). TVAS
data obtained in other wavenumber regions indicate that there
are some product bands that grow with faster time constants in
the 9−20 ps range. Two such examples are bands centered at
1733 and 1756 cm−1, as shown in Figure 3 for ICN photolysis
in CH3CN. The slower buildup of intensity on the bands at
2036 cm−1 (in CH3CN) and 2010 cm−1 (in CD3CN) can be
understood if the proposed H2CCNCHN or D2CC
NCDN products are not directly produced by addition of CN
to the acetonitrile molecule, but are instead the result of
isomerization or quenching of an initially formed intermediate.
The postulated intermediate is not necessarily the same species
as is responsible for the 1733 or 1756 cm−1 features. The
sequential kinetic model of eq 4 (shown for the CH3CN case,
and with CN denoting a solvated form of the radical) provides
a good account of the observed band intensities, with fitted rate
coefficients of k1 = (13.8 4.7) × 10−2 ps−1 and k2 = (3.0
0.4) × 10−2 ps−1 for CH3CN and k1 = (9.2 2.8) × 10−2 ps−1
and k2 = (3.1 0.5) × 10−2 ps−1 for CD3CN solutions.
k1
CN + CH3CN → CN−CH3CN adduct
k2
The intensity of the HCN fundamental band at late times for
measurements conducted in CH3CN is significantly weaker
than we observed for experiments conducted in other organic
solvents including CHCl3, CH2Cl2 and THF (see section
3.4).17−19 We attribute the apparently low yield of HCN from
the abstraction reaction 2 to the competitive addition reaction
→ H2CCNCHN
(4)
Fits to this model are shown in Figure 3, parts c and d. The k1
values agree within the 2 SD uncertainties with the rate
coefficient obtained from growth of the band at 1733 cm−1 seen
in the TVA spectra in Figure 3a (k1 = 1/τ1 = (10.4 1.2) ×
12928
DOI: 10.1021/acs.jpca.5b10720
J. Phys. Chem. A 2015, 119, 12924−12934

The Journal of Physical Chemistry A
Article
Figure 4. (a) Transient vibrational absorption spectra of 290 mM ICN in 2.0 M CH3OH/CH3CN solutions. The feature at 2090 cm−1 is assigned to
HCN(0). The features at 2068, 2043, and 2018 cm−1 are assigned to the fundamental band and vibrational hot bands of INC.21 The band at 2035
cm−1 is assigned to H2CCNCHN. (b) Transient vibrational absorption spectra of 290 mM ICN in 2.0 M CH3OH/CD3CN solutions. The
features at 2010 and 2118 cm−1 are assigned to D2CCNCDN and CD3CN respectively. (c) Time dependence of the HCN(0) band at 2090
cm−1 in a 2.0 M CH3OH/CH3CN solution. The solid line is a fit to a sequential kinetic model described in the text. The inset shows an expanded
view of the first 40 ps. (d) Time dependence of the HCN(0) band at 2090 cm−1 (black) and the CD3CN band at 2118 cm−1 (blue) in a 2.0 M
CH3OH/CD3CN solution. The intensities for the HCN(0) band are multiplied by 3 for clarity. The solid lines are fits to a single exponential
function for the HCN(0) band and to a biexponential function for the CD3CN band.
shown in eqs 3 and 4. The time constant for growth of the
HCN fundamental band intensity of 73 17 ps is much larger
than the ∼8.5 ps time constant for the loss of solvated CN
radicals observed in TEAS and CD3CN in TVAS, indicating
that the bimolecular reaction is not the rate limiting process in
HCN(0) formation. Instead, we attribute the slower rate of
buildup of HCN(0) to relaxation of HCN molecules formed in
vibrationally excited levels by the exothermic H atom
abstraction reaction. Further justification for this interpretation
is presented in the following sections.
3.3. Effect of Adding Methanol on HCN Formation in
Acetonitrile. The data presented in section 3.2 showed that
the peak intensity of the HCN fundamental band in CH3CN
solution is weak because of competition between H atom
abstraction and addition reaction pathways. We explored this
competition further by adding organic cosolutes to the ICN
solutions in acetonitrile, the first of which was chosen to be
methanol.
Measurements of TVA spectra of ICN in CH3OH/CH3CN
solutions were restricted to relatively low CH3OH concen-
trations (≤2.0 M) because unidentified deposits developed on
the CaF2 windows of the Harrick cell during laser irradiation of
samples with higher CH3OH concentrations. Figure 4
illustrates TVA spectra of 290 mM ICN in a 2.0 M solution
of CH3OH in CH3CN (a) and CH3OH in CD3CN (b) with
indications of peak assignments. The shapes and the peak
positions for components of the spectra are all similar to those
observed for ICN photolysis in neat CH3CN or CD3CN. The
most obvious change is the increase of the HCN(0) band
intensity at 2090 cm−1, for which the kinetics in CH3OH/
CH3CN solution are shown in Figure 4c following spectral
decomposition in KOALA,31 suggesting H atom abstraction
from CH3OH is faster than that from CH3CN and can compete
with the addition reaction 4.
The HCN band rises with a time constant of 39
3 ps
following an initial induction of ∼10 ps and shows an almost
constant absorbance after ∼120 ps. This kinetic behavior, with
a delayed onset to the growth of the band, indicates that higher
vibrational levels of the HCN are formed from the reaction and
that vibrational relaxation processes from the higher levels to
the ground state take place on this time scale, as previously was
argued by Rose et al.17,18 The HCN(0) band intensities are well
fitted by a stepwise kinetic model describing this behavior
(Figure 4c). Figure 4d shows the kinetics of the HCN(0) and
CD3CN solvent bands in a 2.0 M CH3OH/CD3CN solution.
As is the case for neat CD3CN, the CD3CN band shows a
negative-going feature with a time-dependence that is well fitted
by a biexponential function, but with time constants of 5.9
2.0 and 58 17 ps. The faster time constant agrees with that
for formation of INC (τ = 6.4
0.9 ps), which is to be
expected for two competing processes involving consumption
of solvated-CN radicals. The poorer signal-to-noise ratios on
HCN band intensities for measurements in the CH3OH/
CD3CN solution prevent analysis using the stepwise kinetic
model applied to HCN production in the CH3OH/CH3CN
solution.
The key time constants deduced from analysis of UV-excited
ICN/acetonitrile and ICN/2.0 M CH3OH/acetonitrile sol-
utions are reported in Table 3. These time constants can be
inverted to give pseudo-first order rate coefficients because
both acetonitrile and CH3OH are in considerable excess over
CN radicals.
Figure 5 shows the reaction rate coefficients deduced in this
way by fitting HCN(0) rise times to single-exponential
functions for solutions of different CH3OH concentration in
12929
DOI: 10.1021/acs.jpca.5b10720
J. Phys. Chem. A 2015, 119, 12924−12934

The Journal of Physical Chemistry A
Article
relaxation. The bimolecular reactions can be with either
CH3CN or CH3OH molecules in the solution to produce
HCN* which is then quenched by two mechanisms:
Table 3. Summary of the Time Constants for Solvated CN
Radicals, CD3CN and HCN(0) in Acetonitrile and 2.0 M
CH3OH/Acetonitrile Solutions
k5a
τ1/ps
τ2/ps
HCN* + CH3CN → HCN(0) + CH3CN
(5a)
a
decay of solvated CN radicals in neat CH3CN
formation of HCN(0) in neat CH3CN
8.5 2.1
73 17
108 26
k5b
HCN* + CH3OH → HCN(0) + CH3OH
(5b)
formation of HCN(0) in 2.0 M CH3OH/
CH3CN
39
3
If these relaxation steps are indeed rate limiting, the rate of
HCN(0) formation is given by
decay of CD3CN in neat CD3CN
8.7 1.8
5.9 2.0
82 30
58 17
decay of CD3CN in 2.0 M CH3OH/CD3CN
formation of HCN(0) in 2.0 M CH3OH/
CD3CN
53
9
d[HCN(0)]
′
= k5[HCN*]
(6)
dt
a
Attributed to an overlapping solvent-I atom CT band.
With the pseudo first order rate coefficient:
′
k5 = k5a[CH3CN] + k5b[CH3OH]
(7)
A linear fit to a plot of k′5 against [CH3OH] is shown in Figure 5
and gives a gradient of (5.2 0.4) × 10−3 M−1 ps−1 and an
intercept of (1.5 0.05) × 10−2 ps−1. The pseudo first order
analysis therefore indicates bimolecular rate coefficients k5a
=
(7.8 0.3) × 108 M−1 s−1 for quenching by acetonitrile (using
a molarity of neat CH3CN of 19.1 M) and k5b = (5.2 0.4) ×
109 M−1 s−1 for the quenching by CH3OH. The signal-to-noise
levels in TVAS spectra prevented a more complete
interpretation of the vibrational cooling of the HCN*.
However, for studies in which THF was added to the ICN/
acetonitrile solutions, a more detailed analysis was possible, as
is discussed in section 3.4.
3.4. Effect of Adding Tetrahydrofuran on HCN
Formation in Acetonitrile. TVAS data were obtained for
ICN in CH3CN/THF mixtures with THF concentrations in
the range 0−12.3 M, the upper limit corresponding to neat
THF. Figure 6 illustrates TVA spectra obtained for a solution of
290 mM ICN in neat THF, integrated band intensities, and
kinetic fits. The spectra resemble those reported previously by
Rose et al.18 and the two prominent bands at 2079 and 2059
cm−1 are assigned to the fundamental CN stretching v1 = 1
← v1 = 0 absorption of HCN, and the fundamental CN
stretching band of INC which is probably overlapped by
underlying bands of nascent vibrationally hot HCN.21
However, another unidentified band can also be seen at 2045
cm−1. Although we always used fresh ICN and THF samples,
this band was not consistently observed in every data set. Since
our main interest in this study is in the effect of solvent on the
dynamics of CN radicals and HCN, we do not consider this
additional band further.
Figure 5. Pseudo first order rate coefficients for HCN(0) formation as
a function of CH3OH concentration. The rate coefficients are obtained
as reciprocals of exponential time constants for the HCN(0) growth.
CH3CN. The first-order rate coefficients depend linearly on
CH3OH concentration. The HCN band intensities also
increase as the concentration of CH3OH rises, indicating
successful competition of the CN + CH3OH reaction with the
addition of CN radicals to acetonitrile molecules. The time
constant for removal of CD3CN in 2.0 M CH3OH/CD3CN
solutions is a measure of the rate at which CN radicals are
consumed by all competing pathways including H atom
abstraction from CH3OH. This time constant is significantly
smaller than that for HCN(0) growth under the same
conditions. The abstraction reaction is therefore not the rate-
determining step; instead, production of HCN(0) is controlled
by the vibrational relaxation of nascent, internally hot HCN*
molecules. The pseudo first order rate coefficients therefore
describe how addition of CH3OH accelerates this vibrational
Figure 6. (a) Transient vibrational absorption spectra of a 290 mM solution of ICN in neat THF, obtained in the 1990−2125 cm−1 range. (b) Time
dependence of the integrated intensities of the HCN fundamental CN stretching band at 2079 cm−1. The solid line is a fit to the sequential kinetic
model of eqs 8−10). The inset shows an expanded view of the first 20 ps.
12930
DOI: 10.1021/acs.jpca.5b10720
J. Phys. Chem. A 2015, 119, 12924−12934

The Journal of Physical Chemistry A
Article
′
The HCN band is considerably stronger relative to the INC
band than we observe for reactions in acetonitrile. It rises with a
time constant of 15.4 0.7 ps following an initial induction of
∼5 ps (Figure 6b inset) and shows an almost constant
absorbance after ∼50 ps, suggesting HCN is also formed via
vibrationally excited HCN*.
k10 = k10a[CH3CN] + k10b[THF]
(15)
Fits of the rise of the HCN(0) band intensity to the model of
eq 13 return two rate coefficient values but do not distinguish
k9′ and k1′0. We therefore assign the larger of the two rate
coefficient values to the first vibrational relaxation step (9), on
the basis that the Landau−Teller model predicts vibrational
quenching rates in solution that increase with the vibrational
quantum number. The fits were performed using TVAS data
obtained at several THF concentrations in acetonitrile. Figure 7
A sequential kinetic model with vibrationally excited HCN as
an intermediate species was employed in our initial analysis of
the time dependence of the HCN fundamental band intensity
for THF/CH3CN solutions of various concentrations. In all
cases, the time constants for growth of the HCN(0) absorption
band were significantly larger than those for the loss of solvated
CN radicals in an ICN/CH3CN solution determined using
TEAS. As was the case for the ICN/methanol/CH3CN
experiments reported in section 3.3, we concluded that the
rate-determining step for HCN(0) growth is vibrational
cooling. Moreover, we found it necessary to adopt a two-
stage vibrational relaxation model to account successfully for
our observations. The analysis model we adopt is therefore
summarized by eqs 8−10, which incorporate reactive
production of highly vibrationally excited HCN** products
which undergo stepwise relaxation to HCN* and finally to
ground state HCN products by interaction with either CH3CN
or THF molecules in the solution.
Figure 7. Pseudo-first order rate coefficients for HCN* (filled circles)
and HCN(0) (open circles) formation as a function of THF
concentration in acetonitrile solutions. The rate coefficients are
obtained by fitting the HCN(0) kinetics to a sequential model
described in the text.
k8
CN + CH3CN/c‐C4H8O →
HCN** + CH2CN/c‐C4H7O
(8)
k9a
HCN** + CH3CN → HCN* + CH3CN
(9a)
shows the derived pseudo first-order rate coefficients (k′9 and
k1′0). Both rate coefficients depend linearly on THF
concentration. The gradients of the best fit lines in Figure 7
are (8.5 0.6) × 10−3 M−1 ps−1 and (7.2 0.3) × 10−3 M−1
ps−1 for vibrational relaxation steps 9b and 10b, respectively.
The corresponding intercepts are (3.7 0.2) × 10−2 ps−1 and
k9b
HCN** + c‐C4H8O → HCN* + c‐C4H8O
(9b)
k10a
HCN* + CH3CN ⎯⎯→ HCN(0) + CH3CN
(10a)
(1.2
0.1) × 10−2 ps−1. From these values and the
k10b
HCN* + c‐C4H8O ⎯⎯→ HCN(0) + c‐C4H8O
(10b)
concentration of neat CH3CN, the bimolecular relaxation rate
coefficients are deduced to be k9a = (1.9 0.1) × 109 M−1 s−1,
On the basis of the above kinetic model, the HCN* and
HCN(0) formation rates are given by
k
9b = (8.5 0.6) × 109 M−1 s−1, k10a = (6.2 0.3) × 108 M−1
s−1 and k10b = (7.2 0.3) × 109 M−1 s−1. The exact nature of
HCN** is uncertain, but it corresponds to HCN vibrationally
excited either with two (or more) quanta of the same mode, or
with quanta in each of two or more modes.
d[HCN*]
= {k9a[CH3CN] + k9b[THF]}[HCN**]
dt
(11)
We also fitted the HCN kinetics in the THF/CH3CN
solution systems with a single exponential function to compare
directly with the bimolecular rate coefficient obtained in the
CH3OH/CH3CN solution systems. Bimolecular rate coeffi-
cients for HCN(0) production of k10a = (6.7 0.4) × 108 M−1
s−1 and k10b = (4.5 0.3) × 109 M−1 s−1 are deduced. The
vibrational relaxation rate coefficients for each solvent are
summarized in Table 4.
The bimolecular rate coefficient values for quenching to
HCN(0) by CH3CN deduced from the THF/CH3CN and the
MeOH/CH3CN measurements are in reasonable agreement
and we prefer the value of k10a = (6.2 0.3) × 108 M−1 s−1
obtained from the sequential kinetic model as a representative
rate coefficient. The deduced rate coefficients for quenching of
internally excited HCN by THF and CH3OH are also similar,
and almost an order of magnitude larger than the rate
coefficient for quenching by CH3CN. These differences arise
from the couplings of the vibrational modes of HCN to the
solvent modes, and hence depend on the structures and
vibrational frequencies of the solvent molecules: THF and
d[HCN(0)]
= {k10a[CH3CN] + k10b[THF]}[HCN*]
dt
(12)
We observe the buildup of ground-state HCN(0) by TVAS,
regardless of whether it forms from CN radical reactions with
CH3CN or THF, so these reactions are not distinguished in 8
but the experimental data demonstrate that the reaction with
THF dominates. The reactive step is significantly faster than
the vibrational relaxation processes, and a reduced model that
describes our observations is therefore
′
′
k10
k9
fast
CN ⎯⎯→ HCN** → HCN* → HCN(0)
(13)
in which k9′ and k1′0 are pseudo-first order rate coefficients for
vibrational relaxation. These two first-order rate coefficients
relate to the bimolecular rate coefficients for reactions 9 and 10
via:
′
k9 = k9a[CH3CN] + k9b[THF]
(14)
12931
DOI: 10.1021/acs.jpca.5b10720
J. Phys. Chem. A 2015, 119, 12924−12934

The Journal of Physical Chemistry A
Article
or more quanta of vibrational excitation and undergoes
stepwise vibrational relaxation. This relaxation is significantly
accelerated by the presence of the THF.
Table 4. Summary of Bimolecular Relaxation Rate
Coefficients for Vibrationally Excited HCN with CH3CN,
and with CH3OH and THF in Acetonitrile Solutions
The use of acetonitrile in place of other common solvents
such as chloroform or dichloromethane opens up new pathways
for CN radical reactions in solution. The competition between
abstraction and addition pathways is reminiscent of the
dynamics of CN radical reactions with alkenes in the gas
phase,40−43 but with the difference that the radical adduct can
be stabilized by the surrounding solvent bath. Photoinduced
radical chemistry in solution is known to lead to complicated
reaction processes, and as the current study illustrates, some of
these processes are highly dependent on the choice of solvent.
Moreover, the rates of growth of absorption bands of reaction
products can be controlled by either the reactive step or the
cooling of nascent, internally excited product molecules.
Nevertheless, the rate coefficients of several of the early time
steps in these competing or sequential pathways can be
unraveled by a combination of time-resolved vibrational and
electronic absorption spectroscopies.
HCN** → HCN* relaxation HCN* → HCN(0) relaxation
rate coefficient/M−1 s−1
rate coefficient/M−1 s−1
a
CH3CN
(7.3 0.7) × 108
b
b
(1.9 0.1) × 109
(6.2 0.3) × 108
c
CH3OH
THF
(5.2 0.4) × 109
c
(4.5 0.3) × 109
b
b
(8.5 0.6) × 109
(7.2 0.3) × 109
a
From fitting with a single exponential function and averaging values
for the CH3OH/CH3CN and THF/CH3CN solutions, with an
b
uncertainty that encompasses both values. Fitting to the sequential
c
kinetic model. Fitting with a single exponential function.
CH3OH solutes appear to offer acceptor modes for the HCN
internal energy that are closer to resonant than those of
acetonitrile. Much faster quenching by THF than by CH3CN
means that the differences in the rates of the HCN**→HCN*
and HCN*→HCN(0) steps are not as pronounced.
ASSOCIATED CONTENT
* Supporting Information
■
4. CONCLUSIONS
S
The reactions that occur in photoexcited solutions of ICN in
CH3CN have been investigated by time-resolved absorption
spectroscopy. Following irradiation at 267 nm, nonsolvated CN
radicals form within the instrument response and become
The Supporting Information is available free of charge on the
ACS Publications website at DOI: 10.1021/acs.jpca.5b10720.
All experimental data are archived in the University of Bristol’s
Research Data Storage Facility (DOI: 10.5523/bris.
1bvvhkmn0dhd61ol5zmtgyv2dk).
solvated in 0.6
0.1 ps. The short time constant of the
solvation compared to other solvents studied previously
indicates rapid reorientation dynamics of the acetonitrile
molecules surrounding the I + CN photofragments, perhaps
driven by strong interactions between the CN radical and the
CN functional group in CH3CN. The solvated CN radicals
decay with a time constant of 8.5 2.1 ps, which is mostly
determined by competing chemical processes that we have
characterized by transient electronic and vibrational absorption
spectroscopy to include geminate recombination (to ICN and
INC) and reaction with solvent molecules. The contribution of
reactions with the solvent is further supported by the observed
loss of CD3CN when experiments are conducted in deuterated
acetonitrile. Most of the reactions appear to take place with the
solvated CN radicals, although we do observe evidence for
some reaction of CN radicals before they have equilibrated into
solvated forms.
The yield of HCN from the reactions of CN radicals with the
solvent is low in comparison to studies in other common
solvents,17−19 because a second pathway involving CN addition
to the nitrile group of acetonitrile competes effectively. This
addition can lead to numerous thermodynamically favored
product radicals, but the evidence from our time-resolved
spectra indicates production of H2CCNCHN following
formation of an initial, unidentified radical adduct. Other radical
products may form, but not be observed because they lack
strong IR bands in the spectral region covered.
TVA spectra of UV-excited ICH2CN in CH3CN, and
computed energies, vibrational frequencies and IR band
intensities of possible products of the addition of CN to
CH3CN and CD3CN (PDF)
AUTHOR INFORMATION
Corresponding Author
*(A.J.O.-E.) Telephone: +44 (0)117 9287672. E-mail: A.Orr-
Ewing@bristol.ac.uk.
■
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
The Bristol group thanks the European Research Council
(ERC, Advanced Grant 290966 CAPRI) for financial support.
M.P.G. acknowledges award of a Marie Curie International
Incoming Fellowship (PIIF-GA-2012-326988). The ULTRA
Laser Facility is supported by the Science and Technology
Facilities Council (STFC, Facility Grant ST/501784). We
thank Igor V. Sazanovich and Mike Towrie (STFC Rutherford
Appleton Laboratory) for assistance with the experimental
measurements.
REFERENCES
■
(1) Levine, R. D. Molecular Reaction Dynamics. Cambridge University
Press: Cambridge, U.K., 2005.
When THF or methanol is added to the ICN/CH3CN
solutions and the ternary mixture is photoexcited, the faster
abstraction of an H atom from these cosolutes competes with
the addition reaction of CN to acetonitrile. The TVAS
measurements provide evidence of formation of vibrationally
hot HCN, as was previously reported for CN reactions with
other organic molecules in chlorinated solvents.17−19 Our data
for CN radical reactions with THF in acetonitrile are best
explained by a mechanism in which the HCN forms with two
(2) Tutorials in Molecular Reaction Dynamics; Brouard, M., Vallance,
C., Eds.; Royal Society of Chemistry: Cambridge, U.K., 2010.
(3) Elles, C. G.; Crim, F. F. Connecting Chemical Dynamics in Gases
and Liquids. Annu. Rev. Phys. Chem. 2006, 57, 273−302.
(4) Orr-Ewing, A. J. Dynamics of Bimolecular Reactions in Solution.
Annu. Rev. Phys. Chem. 2015, 66, 119−141.
(5) Orr-Ewing, A. J. Bimolecular Chemical Reaction Dynamics in
Solution. J. Chem. Phys. 2014, 140, 090901.
12932
DOI: 10.1021/acs.jpca.5b10720
J. Phys. Chem. A 2015, 119, 12924−12934

The Journal of Physical Chemistry A
Article
(6) Glowacki, D. R.; Orr-Ewing, A. J.; Harvey, J. N. Non-Equilibrium
Reaction and Relaxation Dynamics in a Strongly Interacting Explicit
Solvent: F + CD3DN Modelled with a Parallel Multi-State EVB Model.
J. Chem. Phys. 2015, 143, 044120.
Alkanes, Alcohols, and Chloroalkanes. J. Phys. Chem. A 2009, 113,
3758−3764.
(24) Crowther, A. C.; Carrier, S. L.; Preston, T. J.; Crim, F. F. Time-
Resolved Studies of CN Radical Reactions and the Role of Complexes
in Solution. J. Phys. Chem. A 2008, 112, 12081−12089.
(25) Rosenthal, S. J.; Xie, X. L.; Du, M.; Fleming, G. R. Femtosecond
Solvation Dynamics in Acetonitrile: Observation of the Inertial
Contribution to the Solvent Response. J. Chem. Phys. 1991, 95,
4715−4718.
(7) Dunning, G. T.; Murdock, D.; Greetham, G. M.; Clark, I. P.; Orr-
Ewing, A. J. Solvent Response to Fluorine-Atom Reaction Dynamics in
Liquid Acetonitrile. Phys. Chem. Chem. Phys. 2015, 17, 9465−9470.
(8) Dunning, G. T.; Glowacki, D. R.; Preston, T. J.; Greaves, S. J.;
Greetham, G. M.; Clark, I. P.; Towrie, M.; Harvey, J. N.; Orr-Ewing, A.
J. Vibrational Relaxation and Microsolvation of DF after F-Atom
Reactions in Polar Solvents. Science 2015, 347, 530−533.
(9) Morris, V. R.; Mohammad, F.; Valdry, L.; Jackson, W. M. Steric
Effects on Nascent Vibrational Distributions of the HCN Product
Produced in CN Radical Reactions with Ethane, Propane and
Chloroform. Chem. Phys. Lett. 1994, 220, 448−454.
(26) Bagchi, B.; Jana, B. Solvation Dynamics in Dipolar Liquids.
Chem. Soc. Rev. 2010, 39, 1936−1954.
(27) Greetham, G. M.; Burgos, P.; Cao, Q.; Clark, I. P.; Codd, P. S.;
Farrow, R. C.; George, M. W.; Kogimtzis, M.; Matousek, P.; Parker, A.
W.; et al. ULTRA: A Unique Instrument for Time-Resolved
Spectroscopy. Appl. Spectrosc. 2010, 64, 1311−1319.
(28) Roberts, G. M.; Marroux, H. J. B.; Grubb, M. P.; Ashfold, M. N.
R.; Orr-Ewing, A. J. On the Participation of Photoinduced N−H Bond
Fission in Aqueous Adenine at 266 and 220 nm: A Combined
Ultrafast Transient Electronic and Vibrational Absorption Spectros-
copy Study. J. Phys. Chem. A 2014, 118, 11211−11225.
(29) Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.;
Robb, M. A.; Cheeseman, J. R.; Scalmani, G.; Barone, V.; Mennucci,
B.; Petersson, G. A.; et al. Gaussian 09; Gaussian Inc.: Wallingford, CT,
2009.
(30) Rivera, C. A.; Winter, N.; Harper, R. V.; Benjamin, I.; Bradforth,
S. E. The Dynamical Role of Solvent on the ICN Photodissociation
Reaction: Connecting Experimental Observables Directly with
Molecular Dynamics Simulations. Phys. Chem. Chem. Phys. 2011, 13,
8269−8283.
(31) Grubb, M. P.; Orr-Ewing, A. J.; Ashfold, M. N. R. Koala: A New
Program for the Processing and Decomposition of Transient Spectra.
Rev. Sci. Instrum. 2014, 85, 064104.
(32) Moskun, A. C.; Bradforth, S. E. Photodissociation of ICN in
Polar Solvents: Evidence for Long Lived Rotational Excitation in
Room Temperature Liquids. J. Chem. Phys. 2003, 119, 4500−4515.
(33) Samuni, U.; Kahana, S.; Fraenkel, R.; Haas, Y.; Danovich, D.;
Shaik, S. The ICN-INC System - Experiment and Quantum-Chemical
Calculations. Chem. Phys. Lett. 1994, 225, 391−397.
(34) Larsen, J.; Madsen, D.; Poulsen, J.; Poulsen, T. D.; Keiding, S.
R.; Thøgersen, J. The Photoisomerization of Aqueous ICN Studied by
Subpicosecond Transient Absorption Spectroscopy. J. Chem. Phys.
2002, 116, 7997−8005.
(35) Helbing, J.; Chergui, M.; Fernandez-Alberti, S.; Echave, J.;
Halberstadt, N.; Alberto Beswick, J. A. Caging and Excited State
Emission of ICN Trapped in Cryogenic Matrices: Experiment and
Theory. Phys. Chem. Chem. Phys. 2000, 2, 4131−4138.
(36) Helbing, J.; Chergui, M. Spectroscopy and Photoinduced
Dynamics of ICN and Its Photoproducts in Solid Argon. J. Phys. Chem.
A 2000, 104, 10293−10303.
(37) NIST Chemistry Webbook, NIST Standard Reference Database
Number 69. http://webbook.nist.gov.
(38) Owrutsky, J. C.; Raftery, D.; Hochstrasser, R. M. Vibrational
Relaxation Dynamics in Solutions. Annu. Rev. Phys. Chem. 1994, 45,
519−555.
(39) Andersson, M. P.; Uvdal, P. New Scale Factors for Harmonic
Vibrational Frequencies Using the B3LYP Density Functional Method
with the Triple-ζ Basis Set 6-311+G(d,p). J. Phys. Chem. A 2005, 109,
2937−2941.
(40) Trevitt, A. J.; Soorkia, S.; Savee, J. D.; Selby, T. S.; Osborn, D.
L.; Taatjes, C. A.; Leone, S. R. Branching Fractions of the CN+C3H6
Reaction Using Synchrotron Photoionization Mass Spectrometry:
Evidence for the 3-Cyanopropene Product. J. Phys. Chem. A 2011, 115,
13467−13473.
(10) Few, J. FTIR Studies of Chemical Processes. D.Phil. Thesis,
University of Oxford: 2013.
(11) Copeland, L. R.; Mohammad, F.; Zahedi, M.; Volman, D. H.;
Jackson, W. M. Rate Constants for CN Reactions with Hydrocarbons
and the Product HCN Vibrational Populations: Examples of Heavy-
Light-Heavy Abstraction Reactions. J. Chem. Phys. 1992, 96, 5817−
5826.
(12) Bethardy, G. A.; Northrup, F. J.; Macdonald, R. G. The Initial
Vibrational Level Distribution and Relaxation of HCN[X1Σ+(v10v3)]
in the CN(X2Σ+)+CH4→HCN+CH3 Reaction System. J. Chem. Phys.
1996, 105, 4533−4549.
(13) Bethardy, G. A.; Northrup, F. J.; Macdonald, R. G. The Initial
Vibrational-State Distribution of HCN X1Σ+(v10v3) from the Reaction
CN(X2Σ+)+C2H6→HCN+C2H5. J. Chem. Phys. 1995, 102, 7966−
7982.
(14) Bethardy, G. A.; Northrup, F. J.; He, G.; Tokue, I.; Macdonald,
R. G. Initial Vibrational Level Distribution of HCN[X1Σ+(v10v3)] from
the CN(X2Σ+)+H2→HCN+H Reaction. J. Chem. Phys. 1998, 109,
4224−4236.
(15) Preston, T. J.; Hornung, B.; Pandit, S.; Harvey, J. N.; Orr-Ewing,
A. J. Dynamical Effects and Product Distributions in Simulated CN +
Methane Reactions. J. Phys. Chem. A 2015, submitted for publication.
(16) Glowacki, D. R.; Orr-Ewing, A. J.; Harvey, J. N. Product Energy
Deposition of CN + Alkane H Abstraction Reactions in Gas and
Solution Phases. J. Chem. Phys. 2011, 134, 214508.
(17) Rose, R. A.; Greaves, S. J.; Oliver, T. A. A.; Clark, I. P.;
Greetham, G. M.; Parker, A. W.; Towrie, M.; Orr-Ewing, A. J.
Vibrationally Quantum-State-Specific Dynamics of the Reactions of
CN Radicals with Organic Molecules in Solution. J. Chem. Phys. 2011,
134, 244503.
(18) Rose, R. A.; Greaves, S. J.; Abou-Chahine, F.; Glowacki, D. R.;
Oliver, T. A. A.; Ashfold, M. N. R.; Clark, I. P.; Greetham, G. M.;
Towrie, M.; Orr-Ewing, A. J. Reaction Dynamics of CN Radicals with
Tetrahydrofuran in Liquid Solutions. Phys. Chem. Chem. Phys. 2012,
14, 10424−10437.
(19) Greaves, S. J.; Rose, R. A.; Oliver, T. A. A.; Glowacki, D. R.;
Ashfold, M. N. R.; Harvey, J. N.; Clark, I. P.; Greetham, G. M.; Parker,
A. W.; Towrie, M.; Orr-Ewing, A. J. Vibrationally Quantum-State-
Specific Reaction Dynamics of H Atom Abstraction by CN Radical in
Solution. Science 2011, 331, 1423−1426.
(20) Dunning, G. T.; Preston, T. J.; Greaves, S. J.; Greetham, G. M.;
Clark, I. P.; Orr-Ewing, A. J. Vibrational Excitation of Both Products of
the Reaction of CN Radicals with Acetone in Solution. J. Phys. Chem. A
2015, DOI: 10.1021/acs.jpca.5b05624.
(21) Coulter, P. C.; Grubb, M. P.; Koyama, D.; Orr-Ewing, A. J.
Dynamics of Solvation, Reaction and Recombination of the Photo-
products of ICN in Solution. J. Phys. Chem. A 2015, DOI: 10.1021/
acs.jpca.5b10716.
(22) Raftery, D.; Gooding, E.; Romanovsky, A.; Hochstrasser, R. M.
Vibrational Product State Dynamics in Solution-Phase Bimolecular
Reactions - Transient Infrared Study of CN Radical Reactions. J. Chem.
Phys. 1994, 101, 8572−8579.
(23) Crowther, A. C.; Carrier, S. L.; Preston, T. J.; Crim, F. F. Time-
Resolved Studies of the Reactions of CN Radical Complexes with
(41) Trevitt, A. J.; Goulay, F.; Meloni, G.; Osborn, D. L.; Taatjes, C.
A.; Leone, S. R. Isomer-Specific Product Detection of CN Radical
Reactions with Ethene and Propene by Tunable VUV Photoionization
Mass Spectrometry. Int. J. Mass Spectrom. 2009, 280, 113−118.
(42) Gannon, K. L.; Glowacki, D. R.; Blitz, M. A.; Hughes, K. J.;
Pilling, M. J.; Seakins, P. W. H Atom Yields from the Reactions of CN
12933
DOI: 10.1021/acs.jpca.5b10720
J. Phys. Chem. A 2015, 119, 12924−12934

The Journal of Physical Chemistry A
Article
Radicals with C2H2, C2H4, C3H6, Trans-2-C4H8, and Iso-C4H8. J. Phys.
Chem. A 2007, 111, 6679−6692.
(43) Estillore, A. D.; Visger, L. M.; Kaiser, R. I.; Suits, A. G. Crossed-
Beam Imaging of the H Abstraction Channel in the Reaction of CN
with 1-Pentene. J. Phys. Chem. Lett. 2010, 1, 2417−2421.
12934
DOI: 10.1021/acs.jpca.5b10720
J. Phys. Chem. A 2015, 119, 12924−12934