Direct Detection of the Open-Shell Singlet Phenyloxenium Ion: An Atom-Centered Diradical Reacts as an Electrophile

Article abstract of DOI:10.1021/jacs.7b07512

A new photoprecursor to the phenyloxenium ion, 4-methoxyphenoxypyridinium tetrafluoroborate, was investigated using trapping studies, product analysis, computational investigations, and laser flash photolysis experiments ranging from the femtosecond to the millisecond time scale. These experiments allowed us to trace the complete arc of the photophysics and photochemistry of this photoprecursor beginning with the initially populated excited states to its sequential formation of transient intermediates and ultimate formation of stable photoproducts. We find that the excited state of the photoprecursor undergoes heterolysis to generate the phenyloxenium ion in ～2 ps but surprisingly generates the ion in its open-shell singlet diradical configuration (¹A₂), permitting an unexpected look at the reactivity of an atom-centered open-shell singlet diradical. The open-shell phenyloxenium ion (¹A₂) has a much shorter lifetime (τ ～ 0.2 ns) in acetonitrile than the previously observed closed-shell singlet (¹A₁) phenyloxenium ion (τ ～ 5 ns). Remarkably, despite possessing no empty valence orbitals, this open-shell singlet oxenium ion behaves as an even more powerful electrophile than the closed-shell singlet oxenium ion, undergoing solvent trapping by weakly nucleophilic solvents such as water and acetonitrile or externally added nucleophiles (e.g., azide) rather than engaging in typical diradical chemistry, such as H atom abstraction, which we have previously observed for a triplet oxenium ion. In acetonitrile, the open-shell singlet oxenium ion is trapped to generate ortho and para Ritter intermediates, one of which (para) is directly observed as a longer-lived species (τ ～ 0.1 ms) in time-resolved resonance Raman experiments. The Ritter intermediates are ultimately trapped by either the 4-methoxypyridine leaving group (in the case of para addition) or trapped internally via an essentially barrierless rearrangement (in the case of ortho addition) to generate a cyclized product. The expectation that singlet diradicals react similarly to triplet or uncoupled diradicals needs to be reconsidered, as a recent study by Perrin and Reyes-Rodríguez (J. Am. Chem. Soc. 2014, 136, 15263) suggested the unsettling possibility that singlet p-benzyne could suffer nucleophilic attack to generate a naked phenyl anion. Now, this study provides direct spectroscopic observation of this phenomenon, with an atom-centered open-shell singlet diradical reacting as a powerful electrophile. To the question of whether a nucleophile can attack a singly occupied molecular orbital, the answer is apparently yes, at least if another partially occupied orbital is available to avoid violation of the rules of valence.

Full text of DOI:10.1021/jacs.7b07512

Subscriber access provided by University of Birmingham
Article
Direct Detection of the Open-Shell Singlet Phenyloxenium
ion: An Atom-Centered Diradical Reacts as an Electrophile
Lili Du, Yunfan Qiu, Xin Lan, Ruixue Zhu, David Lee
Phillips, Ming-De Li, Andrew S. Dutton, and Arthur H. Winter
J. Am. Chem. Soc., Just Accepted Manuscript • DOI: 10.1021/jacs.7b07512 • Publication Date (Web): 25 Sep 2017
Downloaded from http://pubs.acs.org on September 25, 2017
Just Accepted
“Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted
online prior to technical editing, formatting for publication and author proofing. The American Chemical
Society provides “Just Accepted” as a free service to the research community to expedite the
dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts
appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been
fully peer reviewed, but should not be considered the official version of record. They are accessible to all
readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered
to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published
in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just
Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor
changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers
and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors
or consequences arising from the use of information contained in these “Just Accepted” manuscripts.
Journal of the American Chemical Society is published by the American Chemical
Society. 1155 Sixteenth Street N.W., Washington, DC 20036
Published by American Chemical Society. Copyright © American Chemical Society.
However, no copyright claim is made to original U.S. Government works, or works
produced by employees of any Commonwealth realm Crown government in the course
of their duties.

Page 1 of 8
Journal of the American Chemical Society
1
2
3
4
5
6
Direct Detection of the OpenꢀShell Singlet Phenyloxenium ion: An
AtomꢀCentered Diradical Reacts as an Electrophile
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
a,†
a,‡
†
†
†
,†┴
‡
Lili Du, Yunfan Qiu, Xin Lan, Ruixue Zhu, David Lee Phillips,* MingꢀDe Li, * Andrew S. Dutton and Arthur H.
‡
Winter*
†Department of Chemistry, The University of Hong Kong, Pokfulam Road, Hong Kong, People’s Republic of China
‡
Department of Chemistry, Iowa State University, 2101d Hach Hall, Ames, Iowa 50011, United States
Department of Chemistry, Shantou University, Guangdong 515063, P. R. China.
┴
KEYWORDS : Oxenium Ion, Ritter Reaction, Resonance Raman, Time-resolved Spectroscopy, Open-shell Singlet, Diradicals
ABSTRACT: A new photoprecursor to the phenyloxenium ion, 4ꢀmethoxyphenoxypyridinium tetrafluoroborate, was investigated
using trapping studies, product analysis, computational investigations, and laser flash photolysis experiments ranging from the
femtosecond to the millisecond timescale. These experiments allowed us to trace the complete arc of the photophysics and photoꢀ
chemistry of this photoprecursor beginning with the initially populated excited states to its sequential formation of transient interꢀ
mediates and ultimate formation of stable photoproducts. We find that the excited state of the photoprecursor undergoes heterolysis
1
to generate the phenyloxenium ion in ~ 2 ps, but surprisingly generates the ion in its openꢀshell singlet diradical configuration ( A ),
2
permitting an unexpected look at the reactivity of an atomꢀcentered openꢀshell singlet diradical. The openꢀshell phenyloxenium ion
1
1
(
A ) has a much shorter lifetime (τ ~ 0.2 ns) in acetonitrile than the previouslyꢀobserved closedꢀshell singlet ( A ) phenyloxenium
2
1
ion (τ ~ 5 ns). Remarkably, despite possessing no empty valence orbitals, this openꢀshell singlet oxenium ion behaves as an even
more powerful electrophile than the closedꢀshell singlet oxenium ion, undergoing solvent trapping by weaklyꢀnucleophilic solvents
such as water and acetonitrile or externally added nucleophiles (e.g. azide) rather than engaging in typical diradical chemistry such
as H atom abstraction, which we have previously observed for a triplet oxenium ion. In acetonitrile, the openꢀshell singlet oxenium
ion is trapped to generate orthoꢀ and paraꢀRitter intermediates, one of which (para) is directly observed as a longerꢀlived species (τ
~
0.1 ms) in timeꢀresolved Resonance Raman experiments. The Ritter intermediates are ultimately trapped by either the 4ꢀ
methoxypyridine leaving group (in the case of para addition) or trapped internally via an essentially barrierless rearrangement (in
the case of ortho addition) to generate a cyclized product. The expectation that singlet diradicals react similarly to triplet or uncouꢀ
pled diradicals needs to be reconsidered, as a recent study by Perrin and coworkers suggested the unsettling possibility that singlet
pꢀbenzyne could suffer nucleophilic attack to generate a naked phenyl anion. Now, this study provides direct spectroscopic obserꢀ
vation of this phenomenon with an atomꢀcentered openꢀshell singlet diradical reacting as a powerful electrophile. To the question
of whether a nucleophile can attack a SOMO, the answer is apparently yes, at least if another partially occupied orbital is available
to avoid violation of the rules of valence.
INTRODUCTION
derivatives and observed the 4’ꢀmethylꢀ4ꢀbiphenylyloxenium
ion directly, which demonstrated that oxenium ions are
discrete intermediates as well as the feasibility of studying
oxenium ions with timeꢀresolved spectroscopic methods.
However, further studies on the design of suitable precursors
for the phenyloxenium ion have been difficult due to the
limitations on the requirements for the para substituent of the
precursors in efficiently generating the oxenum ion.
Oxenium ions are highly reactive intermediates with the
+
formula RꢀO . These species are isoelectronic to the betterꢀ
understood nitrene intermediates, but bear a formally positive
charge on a hypovalent oxygen atom. Oxenium ions have been
proposed as intermediates in important umpolung synthetic
1ꢀ6
7ꢀ9
reactions, in the generation of industrial thermoplastics, in
1
0
forming stable complexes with transition metals, and in the
enzymatic oxidative halophenol dehalogenation process by the
dehaloperoxidase in biology. Oxenium ions have also been
implicated in astrochemical studies
persist in interstellar space.
Following Novak and Platz’s initial reports, we observed the
1
1
cs
closedꢀshell ground state parent phenyloxenium ion 2 from
1
2ꢀ14
and are thought to
Despite oxenium ions taking
photolysis of a protonated phenylhydroxylamine (see Scheme
15ꢀ16
20
1). The design of positively charged precursors have several
part in a variety of important processes, studies of the
reactivities and properties of oxenium ions have been limited
by the lack of general photoprecursors to these reactive
advantages, including the ability to (in principle) generate the
oxenium ion in any solvent system, rather than requiring
ionizing solvents. However, like the photoprecursors in the
work of Novak and Platz, photolysis of the protonated
hydroxylamine photoprecursors led to competing generation
17
intermediates.
In 2007, Novak and Platz
of an oxenium ion in solution from laser flash photolysis of
the 4ꢀ(4ꢀmethylphenyl)ꢀ4ꢀacetoxy cyclohexadienonyl
ACS Paragon Plus Environment
18ꢀ19
reported the first direct detection
21
of phenoxy radicals, complicating the subsequent analysis.
22
Inspired by previous studies for generating arylnitrenium

Journal of the American Chemical Society
ions from photolysis of Nꢀpyridinium ions, we synthesized
RESULTS AND DISCUSSION
Page 2 of 8
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
precursor 1 as a potentially improved photoprecursor for the
phenyloxenium ion 2 and anticipated this photoprecursor
could reduce the concomitant radical generation resulting from
competing homolysis. (Scheme 1). Indeed, photolysis of this
photoprecursor relatively cleanly generates the phenyloxenium,
Transient Absorption Study of 1. Photolysis of 1 by fsꢀTA
was investigated in MeCN solution after excitation by 267 nm
and the resulting transient absorption spectra are shown in
Figure 1a and 1b. The initial spectrum at 1 ps (Figure 1a)
shows a broad band at 510 nm and a band tail from 325 nm to
350 nm, which can be assigned to the singlet excited state of
the precursor 1. Both bands at 510 nm and 325 nm
simultaneously decay over a few picoseconds giving rise to a
new peak having maxima at 383 nm and a small broad tail ~
but surprisingly generates the oxenium ion in its excited openꢀ
1
shell singlet diradical ( A ) electronic configuration, allowing
2
us to directly observe the lifetime and reactivity of an atomꢀ
centered openꢀshell singlet diradical.
6
00 nm, assignable to a new species after cleavage of the OꢀN
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
bond. Fitting the kinetics at 383 nm to a biꢀexponential
function, two time constants (2 ps and 214 ps) were
determined in Figure 1c. The first (2 ps) is attributed to bond
cleavage of the excited state to yield the openꢀshell oxenium
ion, while the second is assigned to the decay of the openꢀshell
oxenium ion into downstream intermediates.
1
.0 ps
1.2 ps
2.0 ps
6.7 ps
1.5 ps
3.1 ps
10.1 ps
11.3 ps
16.8 ps
70.2 ps
195 ps
6
4
2
0
1.7 ps
2
1
0
4.9 ps
1
1.3 ps
587 ps
1.56 ns
2.79 ns
b
a
3
50 400 450 500 550 600 650
Wavelength (nm)
350 400 450 500 550 600 650
Wavelength (nm)
Scheme 1. Overview of prior work and this work.
4
2
0
2
508 nm
4
Here, we are pleased to report a timeꢀresolved spectroscopic
study of the photochemistry of precursor 1 osand an
investigation of the reactivity of phenyloxenium 2 using
femtosecond transient absorption (fsꢀTA), nanosecond
transient absorption (nsꢀTA) and nanosecond timeꢀresolved
τ
= 2.3 ps
τ1= 2.0 ps
2
0
-
10
20
30
40
50
Time delay (ps)
383 nm
τ2= 214 ps
3
c
Resonance Raman (nsꢀTR ) spectroscopy methods. This study
0
500 1000 1500 2000 2500 3000
Time delay (ps)
demonstrates a remarkable effect of the photoleaving group, as
2
0
a
related protonated hydroxylamine photoprecursor
1
Figure 1. FsꢀTA of the precursor 1 after 267 nm excitation in
MeCN (a, b), the kinetics at 383 nm are shown in c and the kinetꢀ
ics at 508 nm are shown in the inset. The solid red lines indicate
the fitting to the experimental data.
generates the closedꢀshell singlet phenyloxenium ion ( A ) in
1
combination with the phenoxy radical, while this new
photoprecursor with a pyridine leaving group intriguingly
produces the openꢀshell singlet parent phenyloxenium ion
1
(
A ). The quenching study of the phenyloxenium 2 with
2
Assignment of the openꢀshell oxenium ion. A recent highꢀ
level (CASPT2/ANOꢀL) computational study of the
phenyloxenium ion indicated that the phenyloxenium ion
possesses a closedꢀshell singlet ground state ( A , S C ) with
a singletꢀtriplet gap gap of ~20 kcal/mol ( A ꢀ A ). An openꢀ
shell n,π* singlet state ( A ) is computed as its first singlet
excited state (S ), with an adiabatic energy gap of ~31
kcal/mol. Other π, π* singlet excited states are higher in
energy. Surprisingly, the new species following the excited
state of the precursor 1 is not the expected closedꢀshell singlet
oxenium ion, which was previously detected at 524 nm in
MeCN following decay of the excited state (the computed
closed shell oxenium ion band has a maxima at 520 nm, in
weakly nucleophilic solvents such as water, as well as
externally added azide, gives evidence of the electrophilic
character of this openꢀshell singlet oxenium ion species.
Furthermore, in MeCN, we also directly observe an efficient
nucleophilic addition reaction between the openꢀshell singlet
phenyloxenium 2 and the solvent by laser flash photolysis and
photochemical product studies. The consequent nitrilium ion
adducts 3 and 4 in Scheme 2 are known to be significant
1
1
0, 2v
1
3
1
2
1
2
1
27
2
3
intermediates in Ritter reactions. Nucleophilic additions to
24
diradicals or diradicaloids has been proposed by Perrin and
2
5
Myers for the addition of nucleophiles to pꢀbenzyne and
dehydrotoluene. In the latter case, Myers proposed
26
2
0
zwitterionic character for the intermediate. Squires and coꢀ
workers have also shown that benzynes can be formed from
collisionꢀinduced dissociation (CID) of haloanionions, which
can be thought of as this process in reverse. Here we provide
direct spectroscopic support for nucleophilic additions to an
atom centered diradical 2 acting as an electrophile. Here, the
direct observation of openꢀshell singlet parent phenyloxenium
ion 2 and the Ritter intermediate formation provide new
insight into the structure and chemical reactivity of openꢀshell
singlet oxenium ions, and more broadly provide insight into
the reactivity patterns of singlet diradicals.
good agreement with experiment; Figure 2, red line). Instead,
it is assigned to the openꢀshell singlet phenyloxenium ion
from the following observations. As shown in Figure 2,
comparison of the spectrum at 11.3 ps in MeCN solution with
os
2
1
the MSꢀCASPT2ꢀsimulated UVꢀVis spectrum of the
A2
singlet openꢀshell phenyloxenium ion indicates the bands at
83 nm and 600 nm are in reasonable agreement
experimentally, the longꢀwavelength shoulder band apears to
3
(
be solvatochromic, explaining some disagreement with the
gasꢀphase computed value for this peak). We also considered
that, in principle, the excited state of the photoprecursor can
2
ACS Paragon Plus Environment

Page 3 of 8
Journal of the American Chemical Society
connect to any electronic state of the product 2. MSꢀCASPT2ꢀ
state electronic distribution may lead to the unique adiabatic
heterolysis for 1.
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
simulated UVꢀvis spectra (Figure S1) of the other lowꢀlying
1
singlet excited states of the phenyloxenium ion (e.g. B and
1
1
B ) are in poorer agreement with the experimental spectrum
2
1
1
1.3 ps
than the A state. Furthermore, these alternative states at their
optimized geometries are close in energy with the A state
2
1
2
and would be expected to rapidly undergo internal conversion
1
to the A2 state if they were initially populated upon
heterolytic cleavage of 1. Although in appearance the sharp
3
50
os
400
400
450
500
550
600
650
peaks at 380 nm and 400 nm resemble the absorption of the
closed-shell singlet
20
2
phenoxy radical, several experiments show it is not mainly
due to the radical. First, most of this transient signal is
trapped by azide and by water to generate water adducts, and
the species has a much shorter lifetime in water (Figure S2)
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
350
450
500
550
600
650
than MeCN (14 ps in H O vs. 214 ps in MeCN). In pure
2
Wavelength (nm)
MeCN, most of this transient signal is followed by solventꢀ
trapping Ritter intermediates, which are clearly characterized
Figure 2. The fsꢀTA spectrum of 1 at 11.3 ps in MeCN solution
(Top). Computed UVꢀvis spectra of the openꢀshell
phenyloxenium ion 2 (Bottom, Black) at CASPT2(8,8)/ ANOꢀLꢀ
VTZP level and closedꢀshell phenyloxenium (Bottom, Red) at
3
by the nsꢀTR experiments described later. These behaviors
are indicative of a cationic intermediate, not a radical.
Furthermore, the previouslyꢀobserved phenoxy radical has a
20
TDꢀB3LYP/6ꢀ311G (2d,p) level. The computed 380 nm band of
lifetime of 150 ns in MeCN, while most of this transient
signal has a lifetime of ~0.2 ns in MeCN. We conclude that
because the openꢀshell singlet oxenium ion has identical pi
orbital occupation as the phenoxy radical, it has a relatively
similar absorption profile. While it is possible a small amount
of radical is generated concomittantly with the oxenium ion, i3t
is present in too small amounts to be seen in the nsꢀTR
experiments, described later. This is in contrast to our prior
os
2
, which has a low computed oscillator strength, is artificially
magnified by 100ꢀfold relative to the nearꢀIR shoulder bands to
make it observable.
Azide trapping studies. As further confirmation of the
os
existence of the phenyloxenium ion 2 and its electrophilic
character, we performed a quenching study by adding an azide
trap to the solution in Figure 3. As shown in Figure 3a, there
are two transient absorption bands around 300 nm and 370 nm
cs
study of the 2 where the radical could be clearly seen in the
3
nsꢀTR experiments. The possibility that the transient is the
triplet oxenium ion can be ruled out because it is generated too
quickly to allow for intersystem crossing, which for π,π*
absorptions typically occurs with rate constants on the order of
at just after the pulse in the 1:4 H O:MeCN solution. At a
2
later delay time, the 370 nm transient absorption band
gradually disappears and the 300 nm transient band shifts
down to 280 nm. The band at 370 nm has a lifetime of around
6
8
ꢀ1
1
0 – 10 sec . Furthermore, a previously observed triplet
28
100 µs in the 1:4 H O:MeCN solution. However, in the
oxenium ion underwent H atom abstraction reactions, not
nucleophilic additions.
2
presence of 12.5 mM NaN (Figure 3b), no band at 370 nm
3
shows up and a more noticeable sharp band at 400 nm is
observed. The comparison of the spectra just after the pulse is
also displayed in Figure 3c. In the presence of the quencher,
the band at 370 nm is lower than the one without a quencher.
The initial band at 370 nm is assigned to Ritter intermediate 4
produced from the efficient reaction between the
Though 2 is generated in the form of the openꢀshell singlet
cs
excited state, the internal conversion to its ground state ( 2) is
not detected within 200 ps lifetime before reaction occurs
based on the absence of the closedꢀshell singlet transient peak
(Figure 2, red line). Indeed, a prior theoretical investigation
has demonstrated that phenyloxenium ion has a closedꢀshell
3
1
phenyloxenium ion 2 and MeCN, based on the nsꢀTR
singlet ground state ( A , S ) and its openꢀshell singlet state
(
3
1
0
1
spectrum with resonance enhancement of this band, discussed
A ) is computed as its first singlet excited state (S ) with a
2 1
27
later. These trapping results are consistent with
a
0.8 kcal/mol adiabatic energy difference. This large energy
1 1
photogenerated phenyloxenium ion 2 that can undergo the
nucleophilic addition reaction with MeCN to form Ritter
intermediates 3 or 4 (after deprotonation), while NaN3
performs as a superior competing trapping agent. Though the
gap between the A and A is anticipated to cause a slow
1
2
internal conversion over 200 ps. In addition, the symmetry
restriction could also prevent the adequate mixing of the
orbitals, preventing the internal conversion from occurring
efficiently. This relatively slow internal conversion permits
the direct spectroscopic observation of the reactivity of the
openꢀshell singlet oxenium ion.
os
openꢀshell singlet 2, as an atom centered diradical, would be
anticipated
to react similarily to triplet or uncoupled
diradicals, both the formation of the Ritter intermedates and
the quenching study indicate that the phenyloxenium ion 2
reacts as a powerful electrophile.
An intriguing question is why the protonated hydroxylamine
affords the closedꢀshell singlet phenyloxenium upon
photolysis while 1 generates the openꢀshell singlet state.
While this question can really only be addressed by excited
state PES mapping and conical intersection searches, a TDꢀ
DFT groundꢀstate—excited state difference density plot
a
5
4
3
2
1
0
b
0 ns
6
4
2
0
0 ns
6
0 ns
5
1
20
ꢁ
0
s
ꢁ
ꢁ
ꢁ
120 ns
s
s
s
ꢁ
ꢁ
1
80 ns
240 ns
300 ns
6
0
(
Figure S3) indicates that the nature of the first singlet excited
180
s
s
3
00
states between the two photoprecursors are quite different.
The protonated hydroxylamine photoprecursor has a π,π*
singlet excited state necessarily located on the phenyl ring. In
contrast, 1 has an excited state featuring significant charge
redistribution to the pyridinium ring. The differing excited
3
00
400
500
600
700
800
300
400
500
600
700
800
Wavelength ( nm )
Wavelength ( nm )
3
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 4 of 8
c
0 ns without NaN3
MeCN. In Figure 4a, the most intense Raman band is ascribed
4
3
2
1
0
0
ns with NaN3
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
to the aromatic C=C stretches of the phenyl group near 1629
O
N
ꢀ1
cm . The second most intense Raman band in this region is
ꢀ
1
the 2124 cm band due to the stretching mode of the
isonitrilium CꢀN triple bond. The changes with time for the
ꢀ
1
1
629 cm feature integrated area is shown in Figure S6 and
the fitted lifetime is 134 µs, which is assigned to the lifetime
of the Ritter intermediate 4, about the same as the longerꢀlived
3
00
400
500
600
700
800
Wavelength ( nm )
3
80 nm transient observed in the nsꢀTA experiments, also
Figure 3. NsꢀTA of the precursor 1 after 266 nm excitation (a)
attributed to this intermediate. Our product studies indicate
that this longꢀlived species is ultimately trapped by the pꢀ
methoxypyridine leaving group to generate an unusual imino
pyridinium salt.
1:4 H O:MeCN, (b) 1:4 H O:MeCN with the presence of 12.5
2
2
mM NaN , (c) comparison of spectra immediately after the pulse
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
3
in both conditions.
*
a
*
b
Scheme 2. The possible reaction pathways for the phenyloxenium
ion reactions with the solvents and the quencher are shown.
8
00
1000 1200 1400 1600 1800 2000 2200
-1
Raman shift (cm )
3
NsꢀTR study of 1. In the nsꢀTA spectra (Figure 3), the longꢀ
3
Figure 4. (a) nsꢀTR spectra at 10 ns of the precursor 1 after 266
nm excitation in pure MeCN; b) computed Raman spectrum of
Ritter intermediate 4 (B3PW91/6ꢀ31+G(d,p)). Asterisks represent
solvent subtraction artifacts.
lived intermediate gives a band ranging from 325 nm to 3803
nm. In order to unveil the nature of this intermediate, nsꢀTR
experiments of the new photoprecursor were performed by
using both 355 nm and 369 nm as the probe wavelengths in
pure MeCN and this data is shown in Figures S4 and S5. The
same Raman spectra obtained in Figure S4 and S5 indicate the
same species is detected by both of these probe wavelengths,
which is consistent with the nsꢀTA results. Therefore, we only
discuss the spectra we observed using the 355 nm probe
wavelength here. As shown in Figure S4, only one species is
obtained and it decays from 10 ns to 300 µs. An unusual band
Photochemical Product Study of 1. Stable end photoproduct
analysis was performed both in the presence and absence of
trapping agents. Upon formation of the openꢀshell singlet
phenyloxenium ion 2 in pure acetonitrile, both the solvent
(acetonitrile) and the leaving group (4ꢀmethoxypyridine) are
found to attack either the para or ortho position of the ring to
produce the ultimate pyridineꢀadducts and two nitrilium ions,
3
intermediates in Ritter reactions where a nitrile undergoes
nucleophilic addition. Compound 6 is immediately generated
after a barrierless cyclization reaction of 3 according to the
computational geometry optimization as discussed above.
Another nitrilium ion 4 is detected by fsꢀTA and nsꢀTR and
reacts with 4ꢀmethoxypridine to generate a pyridineꢀstablized
nitrilium ion. This type of reaction has been investigated
recently by van Dijk and coꢀworkers. Notably, reaction
between a closedꢀshell singlet phenyloxenium ion and the
solvent acetonitrile was not observed in our previous study,
while the existence of 3 and 4 demonstrates that the openꢀshell
singlet phenyloxenium ion 2 can react with acetonitrile to
produce para and ortho ring adducts.
ꢀ1
at 2124 cm is typical of a triple bond such as an isonitrile. As
discussed above, both Ritter intermediates 3 and 4 are possible
reaction intermediates in MeCN. In an attempt to identify
these intermediates, DFT calculations were used to predict the
Raman spectra for intermediates 3 and 4 at the B3PW91/6ꢀ
31+G(d,p) level of theory. However, the deprotonated version
of Ritter intermediate 3 was not found to be a minimum, and
underwent cyclization apparently without a barrier to produce
compound 6, a compound isolated in our product studies.
Thus, it is perhaps not surprising that we only detect the para
Ritter intermediate, as it does not have access to this internal
trapping decay channel.
and 4. These particular types of nitrilium ions are important
3
29
2
0
Comparison of the computed normal Raman spectra of 4 with
3
the experimental nsꢀTR spectra are displayed in Figure 4.
Figure 4 indicates that both the Raman vibrational frequencies
and relative intensities computed for Ritter intermediate 4 are
Photochemical products in water also shows the electrophilic
property of the openꢀshell singlet phenyloxenium ion, forming
the water adducts hydroquinone and catechol. Sodium azide
was used for quench study in nsꢀTA studies and the
3
reasonably consistent with the TR spectrum. The transient
3
species observed by nsꢀTR at an early time can be tentatively
assigned to the reaction product of Ritter intermediate 4 which
is produced from the reaction of the phenyloxenium ion 2 and
1
corresponding H NMR (Figure S17) shows that there are less
4
ACS Paragon Plus Environment

Page 5 of 8
Journal of the American Chemical Society
The nanosecond (nsꢀTA) experiments were performed using
adducts between 2 and methoxypyridine or acetonitrile
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
produced, indicating the azide can successfully trap 2.
the fourth harmonic output of an Nd:YAG laser supplied the
266 nm laser pump pulse. The probe light came from a 450 W
xenon lamp. The 266 nm pump laser beam photoexcited the
sample, and at a right angle the probe light from the xenon
lamp traversed the sample held in a 1 cm flowing quartz cell.
The transmitted probe light was then detected by a single deꢀ
tector (for kinetic analysis) or by an array detector (for specꢀ
tral analysis).
Scheme 3. Photochemical product studies on Rayonet
photolysis (254 nm) of 1. Relative percentages were
1
determined by H NMR integration.
OH
OH
O
OMe
OH
OH
OH
OH
N
OH
BF4-
N
N
O
N
N
OMe
N
OH
3
OMe
OMe
The nanosecond timeꢀresolved Resonance Raman (nsꢀTR )
hv
2
0%
27%
21%
22%
--
--
--
10%
7%
experiments were performed using the fourth harmonic of an
Nd:YAG nanosecond pulsed laser supplied the 266 nm pump
wavelength. The pump pulse photoexcited the sample to start
the photochemical processes and the probe pulse monitored
the sample and the intermediate species formed. The laser
beams were lightly focused and lined up so that they merged
together onto a flowing argon purged sample solution. A pulse
delay generator was utilized to electronically set the time deꢀ
lay between the pump and probe laser pulses. The Raman
scattered signal was collected using a backscattering geometry
and observed by a liquid nitrogenꢀcooled chargeꢀcoupled deꢀ
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
MeCN
hv
9
%
24%
--
43%
17%
2
H O
Conclusion. In conclusion, the new photoprecursor for the
generation of the phenyloxenium ion, 4ꢀ
methoxyphenoxypyridinium tetrafluoroborate salt, was
investigated using timeꢀresolved spectroscopic experiments
3
(
fsꢀTA, nsꢀTA and nsꢀTR ) along with the necessary
computational calculations and photochemical product studies.
These experiments allow us to make several conclusions.
First, the pyridinium leaving group appears to provide a better
photoprecursor to oxenium ions than the previouslyꢀ
investigated ammonia leaving groups, as we observe only
oxenium generation, rather than a mixture of oxenium and
radical generation (a small amount of generated radical, not
3
vice (CCD) detector. The nsꢀTR spectra shown here were
found from subtraction of an appropriately scaled probeꢀ
beforeꢀpump spectrum from the correlated pump−probe resoꢀ
nance Raman spectrum to mostly get rid of nontransient
bands. The Raman bands of MeCN were used to calibrate the
− 1
Raman shifts with an estimated uncertainty of 5 cm .
3
B. Computational methods. Density functional theory (DFT)
detectable by the nsꢀTR experiments cannot be ruled out).
computations were performed using the Gaussian09 software
Second, this study demonstrates the importance of excited
state dynamics on photochemical reaction outcomes, as this
precursor gives birth to the oxenium ion in its first excited
state rather than its ground state. Of more general importance,
this study provides direct spectroscopic evidence for a concept
proposed by Perrin from deuterium isotope studies on pꢀ
32
suite. In all cases, optimized geometries were found to have
zero imaginary frequencies and corrections for the zero point
vibrational energy were added unscaled. Geometries of the
closedꢀshell singlet phenyloxenium ion and the phenoxy radiꢀ
cal molecular were optimized at the B3LYP/ccꢀpVTZ level of
theory. The computed UVꢀvis spectrum of Ritter intermediate
2
4
benzyne: the unsettling possibility that a nucleophile can
attack a singlyꢀoccupied orbital. Furthermore, not only does
the openꢀshell oxenium ion show surprising electrophilic
behavior, but it appears to be a much more powerful
electrophile than the closedꢀshell species, as born out by a
much shorter lifetime (0.2 ns vs. 5 ns in MeCN) and its ability
to trap recalcitrant nucleophiles such as MeCN to form Ritter
intermediates, which is not seen with the closedꢀshell oxenium
ion. A theoretical basis for understanding nucleophilic
additions to singlet diradicals is clearly needed.
4
was calculated at the TDꢀ(U)B3LYP/6ꢀ31+(d,p) level with
the optimized geometry at (U)B3LYP/6ꢀ31+(d,p) level emꢀ
ploying a SMD (acetonitrile) solvent model. To predict the
3
TR spectra of the Ritter intermediate 4, B3PW91/6ꢀ
3
1+G(d,p) was employed to optimize the structure and predict
the Raman spectra. A simulated band width for the spectra
ꢀ
1
was determined using a Lorentzian of 10 cm band width for
the vibrational band frequencies. A frequency scaling factor of
0.960 was used in the comparison of the calculated results
with the experimental spectra. CASPT2(8,8)/ ANOꢀL calculaꢀ
tion was utilized to optimize the geometry of the openꢀshell
1
Experimental and Computational Methods
singlet phenyloxenium ion ( A ) using the Molcas 8.0 softꢀ
2
3
33
A. fsꢀTA, nsꢀTA and nsꢀTR Experiments. The experimental
ware. To predict the UVꢀvis absorption spectrum, single
setups and methods used for these experiments have been
point energies at the MSꢀCASPT2 level were calculated using
the final geometry from the previous calculation for five states
and calculated the transition dipole for the transitions between
the states. The band shapes were simulated using the Gaussꢀ
Sum method.
C. Product Studies. Photochemical product studies in deuterꢀ
ated solvents were performed with 2.5 mg of precursor 1 disꢀ
solved in 1 ml of CD CN and 1 ꢁL internal standard decane.
3
1
3
0ꢀ31
previously described.
Briefly, the fsꢀTA measurements
were accomplished using a femtosecond regenerative ampliꢀ
fied Ti:sapphire laser system in which the amplifier was seedꢀ
ed with the 120 fs laser pulses from an oscillator laser system.
The laser probe pulse was produced by utilizing ∼ 5% of the
amplified 800 nm laser pulses to generate a whiteꢀlight conꢀ
tinuum (325ꢀ650 nm) in a CaF crystal and then this probe
2
beam was split into two parts before traversing the sample.
The flowing sample was excited by a 267 nm pump laser
beam. An absorbance of 1 at 267 nm was used for the sample
solutions for the fsꢀTA experiments in order to maintain the
same number of photons being absorbed for the same irradiatꢀ
ing conditions for the samples.
The mixture was filtered and an initial H NMR was taken
with a 90° angle and a relaxation delay of 60 seconds in a
quartz NMR tube. The sample was then irradiated with 254
nm UV light from a mercury vapor lamp for different desired
time durations in a Rayonet photoreactor. Spectra were taken
1
with the previous parameters. After the photolysis and H
NMR characterization were completed, the sample was subꢀ
mitted to LCꢀMS for further product confirmation. Compariꢀ
5
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 6 of 8
1
son between the H NMR spectra of the sample and the standꢀ
ard chemicals also clarified the peak assignment and the prodꢀ
ucts generated (See Figure S7ꢀS18, Supporting Information
for more details.)
15. Leyva, E.; Platz, M. S.; Persy, G.; Wirz, J. J. Am. Chem.
Soc. 1986, 108, 3783.
16. Smith, D. Chem. Rev. 1992, 92, 1473.
17. Qiu, Y.; Winter, A. H. Org. Biomol. Chem. 2017, 15,
2666.
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
ASSOCIATED CONTENT
Supporting Information.
Description of the synthesis of 1 and its characterization data
18. Wang, Y.ꢀT.; Wang, J.; Platz, M. S.; Novak, M. J. Am.
Chem. Soc. 2007, 129, 14566.
1
9. Wang, Y.ꢀT.; Jin, K. J.; Leopold, S. H.; Wang, J.; Peng,
3
H.ꢀL.; Platz, M. S.; Xue, J.; Phillips, D. L.; Glover, S. A.;
Novak, M. J. Am. Chem. Soc. 2008, 130, 16021.
(NMR and Mass spectra), fsꢀTA, nsꢀTR , product studies and
computation details described in this study (PDF).
2
0. Hanway, P. J.; Xue, J. D.; Bhattacharjee, U.; Milot, M. J.;
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Zhu, R. X.; Phillips, D. L.; Winter, A. H. J. Am. Chem. Soc.
The Supporting Information is available free of charge on the
ACS Publications website at DOI:
2
013, 135, 9078.
2
1. Li, M. D.; Hanway, P. J.; Albright, T. R.; Winter, A. H.;
AUTHOR INFORMATION
Corresponding Authors
Phillips, D. L. J. Am. Chem. Soc. 2014, 136, 12364.
2. Chiapperino, D.; McIlroy, S.; Falvey, D. E. J. Am. Chem.
2
Soc. 2002, 124, 3567.
*Eꢀmail: phillips@hku.hk
*Eꢀmail: mdli@stu.edu.cn
*Eꢀmail: winter@iastate.edu
23. Ritter, J. J.; Kalish, J. J. Am. Chem. Soc. 1948, 70, 4048.
24. Perrin, C. L.; ReyesꢀRodríguez, G. J. J. Am. Chem. Soc.
2014, 136, 15263.
25.Myers, A. G.; Dragovich, P. S.; Kuo, E. Y., J. Am. Chem.
Soc. 1992, 114, 9369.
Author Contributions
a
These authors contributed equally to this work.
2
6.Wenthold, P. G.; Wierschke, S. G.; Nash, J. J.; Squires, R.
R., J. Am. Chem. Soc. 1994, 116, 7378.
Funding Sources
2
5
7.Hanway, P. J.; Winter, A. H., J. Am. Chem. Soc. 2011, 133
086.
Financial support from NSF (CHEꢀ1464956) and ACS PRF
(PRF55820ꢀND4) for AHW are gratefully acknowledged.
28. Li, M. D.; Albright, T. R.; Hanway, P. J.; Liu, M. Y.; Lan,
X.; Li, S. B.; Peterson, J.; Winter, A. H.; Phillips, D. L. J. Am.
Chem. Soc. 2015, 137, 10391.
ACKNOWLEDGMENT
2
9. van Dijk, T.; Bakker, M. S.; Holtrop, F.; Nieger, M.;
Slootweg, J. C.; Lammertsma, K. Org. Lett. 2015, 17, 1461.
0. Du, L.; Li, M.ꢀD.; Zhang, Y.; Xue, J.; Zhang, X.; Zhu, R.;
AHW thanks the Iowa State chemistry instrumentation facilities.
DLP thanks support from the Hong Kong Research Grants Counꢀ
cil grants HKU 7035/13P, AoE/Pꢀ03/08, SEG HKU/07 and The
University of Hong Kong Development Fund 2013ꢀ2014 project
3
Cheng, S. C.; Li, X.; Phillips, D. L. J. Org. Chem. 2015, 80,
7340.
31. Du, L.; Zhang, X.; Xue, J.; Tang, W.; Li, M.ꢀD.; Lan, X.;
Zhu, J.; Zhu, R.; Weng, Y.; Li, Y.; Phillips, D. L. J. Phys.
Chem. B 2016, 120, 11132.
“New Ultrafast Spectroscopy Experiments for Shared Facilities.”
ABBREVIATIONS
REFERENCES
1. Swenton, J. S.; Carpenter, K.; Chen, Y.; Kerns, M. L.;
Morrow, G. W. J. Org. Chem. 1993, 58, 3308.
32. 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.; Nakatsuji, H.; Caricato, M.;
Li, X.; Hratchian, H. P.; Izmaylov, A. F.; Bloino, J.; Zheng,
G.; Sonnenberg, J. L.; Hada, M.; Ehara, M.; Toyota, K.;
Fukuda, R.; Hasegawa, J.; Ishida, M.; Nakajima, T.; Honda,
Y.; Kitao, O.; Nakai, H.; Vreven, T.; Montgomery Jr., J. A.;
Peralta, J. E.; Ogliaro, F.; Bearpark, M. J.; Heyd, J.; Brothers,
E. N.; Kudin,
K. N.; Staroverov, V. N.; Kobayashi, R.; Normand, J.;
Raghavachari, K.; Rendell, A. P.; Burant, J. C.; Iyengar, S. S.;
Tomasi, J.; Cossi, M.; Rega, N.; Millam, N. J.; Klene, M.;
Knox, J. E.; Cross, J. B.; Bakken, V.; Adamo, C.; Jaramillo,
J.; Gomperts, R.; Stratmann, R. E.; Yazyev, O.; Austin, A. J.;
Cammi, R.; Pomelli, C.; Ochterski, J. W.; Martin, R. L.;
Morokuma, K.; Zakrzewski, V. G.; Voth, G. A.; Salvador, P.;
Dannenberg, J. J.; Dapprich, S.; Daniels, A. D.; Farkas, Ö.;
Foresman, J. B.; Ortiz, J. V.; Cioslowski, J.; Fox, D. J.
Gaussian 09, Gaussian, Inc.: Wallingford, CT, USA, 2009.
2
3
4
1
5
2
6
6
7
. Omura, K. J. Org. Chem. 1996, 61, 2006.
. Omura, K. J. Org. Chem. 1996, 61, 7156.
. Williams, L. L.; Webster, R. D. J. Am. Chem. Soc. 2004,
26, 12441.
. Peng, H. M.; Webster, R. D. J. Org. Chem. 2008, 73, 2169ꢀ
175.
. Troisi, F.; Pierro, T.; Gaeta, C.; Neri, P. Org. Lett. 2009, 11,
97ꢀ700.
. Baesjou, P. J.; Driessen, W. L.; Challa, G.; Reedijk, J. J.
Am. Chem. Soc. 1997, 119, 12590.
. Driessen, W. L.; Baesjou, P. J.; Bol, J. E.; Kooijman, H.;
Spek, A. L.; Reedijk, J. Inorg. Chim. Acta. 2001, 324, 16.
. Kobayashi, S.; Higashimura, H. Prog. Polym. Sci. 2003, 28,
8
9
1015.
10. Vigalok, A.; Rybtchinski, B.; Gozin, Y.; Koblenz, T. S.;
BenꢀDavid, Y.; Rozenberg, H.; Milstein, D. J. Am. Chem. Soc.
3
3. Aquilante, F.; Autschbach, J.; Carlson, R. K.; Chibotaru,
2
003, 125, 15692.
L. F.; Delcey, M. G.; De Vico, L.; Fdez. Galván, I.; Ferré, N.;
Frutos, L. M.; Gagliardi, L.; Garavelli, M.; Giussani, A.;
Hoyer, C. E.; Li Manni, G.; Lischka, H.; Ma, D.; Malmqvist,
P. Å.; Müller, T.; Nenov, A.; Olivucci, M.; Pedersen, T. B.;
Peng, D.; Plasser, F.; Pritchard, B.; Reiher, M.; Rivalta, I.;
Schapiro, I.; SegarraꢀMartí, J.; Stenrup, M.; Truhlar, D. G.;
Ungur, L.; Valentini, A.; Vancoillie, S.; Veryazov, V.;
1
1. Osborne, R. L.; Coggins, M. K.; Raner, G. M.; Walla, M.;
Dawson, J. H. Biochemistry-Us 2009, 48, 4231.
2. Reiner, E. J.; Harrison, A. G. Int. J. Mass. Spectrom. Ion.
Process. 1984, 58, 97.
1
1
1
1
3. Siuzdak, G.; Belbruno, J. J. Laser. Chem. 1991, 11, 83.
4. Siuzdak, G.; North, S.; Belbruno, J. J. J. Phys. Chem-Us.
991, 95, 5186.
6
ACS Paragon Plus Environment

Page 7 of 8
Journal of the American Chemical Society
Vysotskiy, V. P.; Weingart, O.; Zapata, F.; Lindh, R. J.
Comput. Chem. 2016, 37, 506.
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
7
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 8 of 8
Graphical TOC
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
8
ACS Paragon Plus Environment