Photo- and thermal-induced multistructural transformation of 2-phenylazolyl chelate boron compounds

Article abstract of DOI:10.1021/ja400917r

The new N,C-chelate boron compounds B(2-phenylazolyl)Mes₂ [Mes = mesityl; azolyl = benzothiazolyl (1a), 4-methylthiazolyl (2a), benzoxazolyl (3a), benzimidazolyl (4a)] undergo an unprecedented multistructural transformation upon light irradiation or heating, sequentially producing isomers b, c, d, and e. The dark isomers b generated by photoisomerization of a undergo a rare thermal intramolecular H-atom transfer (HAT), reducing the azole ring and generating new isomers c, which are further transformed into isomers d. Remarkably, isomers d can be converted to their diastereomers e quantitatively by heating, and e can be converted back to d by irradiation at 300 nm. The structures of isomers 1d and 1e were established by X-ray diffraction. The unusual HAT reactivity can be attributed to the geometry of the highly energetic isomers b and the relatively low aromaticity of the azole rings. The boryl unit plays a key role in the reversible interconversion of d and e, as shown by mechanistic pathways established through DFT and TD-DFT calculations.

Full text of DOI:10.1021/ja400917r

Communication
pubs.acs.org/JACS
Photo- and Thermal-Induced Multistructural Transformation of
2‑Phenylazolyl Chelate Boron Compounds
Ying-Li Rao, Hazem Amarne, Leanne D. Chen, Matthew L. Brown, Nicholas J. Mosey, and Suning Wang*
Department of Chemistry, Queen’s University, Kingston, Ontario, Canada K7L 3N6
S
* Supporting Information
Scheme 1
ABSTRACT: The new N,C-chelate boron compounds
B(2-phenylazolyl)Mes2 [Mes = mesityl; azolyl = benzo-
thiazolyl (1a), 4-methylthiazolyl (2a), benzoxazolyl (3a),
benzimidazolyl (4a)] undergo an unprecedented multi-
structural transformation upon light irradiation or heating,
sequentially producing isomers b, c, d, and e. The dark
isomers b generated by photoisomerization of a undergo a
rare thermal intramolecular H-atom transfer (HAT),
reducing the azole ring and generating new isomers c,
which are further transformed into isomers d. Remarkably,
isomers d can be converted to their diastereomers e
quantitatively by heating, and e can be converted back to d
phenylpyridine).5 However, unlike B(ppy)Mes2, which under-
by irradiation at 300 nm. The structures of isomers 1d and
1e were established by X-ray diffraction. The unusual HAT
reactivity can be attributed to the geometry of the highly
energetic isomers b and the relatively low aromaticity of
the azole rings. The boryl unit plays a key role in the
reversible interconversion of d and e, as shown by
mechanistic pathways established through DFT and TD-
DFT calculations.
goes facile and efficient photothermal isomerization between the
colorless isomer a and the deep-blue isomer b only, 1a−4a
undergo a sequential multistructural transformation in response
to light irradiation or heating. This unusual transformation
involves (i) photoisomerization (PI) of a to b; (ii) thermal
intramolecular H-atom transfer (HAT) in b to give the new
isomer c; (iii) a 1,3-shift of the boryl group in c to give another
new isomer, d; and (iv) diastereomer interconversion via ring
opening/closing, converting d to the new isomer e (Scheme 2).
olecular isomerism is a common phenomenon in the
Mchemical world. It plays an important role in biology,
medicine, and the development of advanced materials.1−4 The
most commonly observed molecular isomerism involves a
structural change between two different forms of a molecule,
such as the cis/trans isomerization,2 the interconversion of mer
and fac isomers of octahedral complexes,3 and the transformation
between the ring-opened and ring-closed isomers of dithienyl-
ethenes and spiropyrans.4 Because molecular isomerization can
often be controlled by external stimuli such as light, heat, or
pressure and the different isomers usually have distinct chemical
and physical properties, this phenomenon has been exploited
extensively in switching/controlling the chemical reactivity and
optical, electronic, and physical properties of molecules and
materials.1−4 Although molecular isomerism is common,
molecular systems displaying multistructural transformations
involving more than two isomers in response to heat or light
remain rare. Here we disclose an unprecedented multistructural
transformation of organoboron molecules.
Scheme 2
Organoboron compounds that display unusual and unique
photo/thermal isomerization phenomena have recently been
reported.5,6 This work further illustrates the exceptional ability of
N,C-chelate boron compounds to transform and adapt in
response to light and heat.
Compounds 1a−4a were synthesized in good yields by
lithiation of the appropriate bromophenylazole or phenylazole
This discovery was made during our investigation of a new
class of N,C-chelate organoboron compounds, B(2-
phenylazolyl)Mes2 [Mes = mesityl; azolyl = benzothiazolyl
(1a), 4-methylthiazolyl (2a), benzoxazolyl (3a), benzimidazolyl
(4a)] (Scheme 1), which are analogues of the previously
reported N,C-chelate compound B(ppy)Mes2 (ppy = 2-
Received: January 31, 2013
Published: February 20, 2013
© 2013 American Chemical Society
3407
dx.doi.org/10.1021/ja400917r | J. Am. Chem. Soc. 2013, 135, 3407−3410

Journal of the American Chemical Society
Communication
starting materials followed by the addition of BMes2F and fully
characterized by NMR and elemental analyses. Single-crystal X-
ray diffraction (scXRD) showed that the structures of 2a and 3a7
are similar to those of B(ppy)Mes2 and its derivatives.5 Synthetic
details, full characterization data for 1a−4a, and crystal data for
2a and 3a can be found in the Supporting Information (SI). All
four compounds are either colorless or light yellow and are
fluorescent with moderate quantum efficiencies (see the SI).
The absorption spectra (Figure 1) and time-dependent
density functional theory (TD-DFT) calculations (see the SI)
singlet peaks from the olefinic protons of the cyclohexadienyl
(CHD) ring at 5.5−5.8 ppm and the two singlet peaks
characteristic of a vinyl group at 4.5−5.0 ppm in C6D6. The
structure of 1c was established using 2D NMR spectra and
confirmed by DFT calculations and NMR spectra (see the SI).
The HAT from a methyl group to C2 of the thiazole ring,
transforming the thiazole into a thiazoline, is significant and
unusual since the reduction/hydrogenation of azole compounds
usually requires a catalyst or the use of a highly reactive hydride.8
NMR data further showed that 1c undergoes a subsequent
transformation to give 1d upon prolonged heating; 1d is stable
under ambient air and was isolated and fully characterized by
NMR, elemental, and scXRD analyses. The 1H NMR spectrum
of 1d contains distinct methylene peaks with an AB pattern and
2JH−H = 11 Hz (Figure 2), indicating that 1d has an asymmetric
nature. The 11B NMR chemical shift of 1d (46.0 ppm) is shifted
downfield by >50 ppm relative to that of 1b, similar to those
observed in Mes2BNR2 and R2BNR′2 compounds,9a−e thus
supporting the possible presence of a BN bond.
The crystal structure of 1d (Figure 3) confirms the NMR
findings unequivocally. C25 of the thiazole ring has been
Figure 1. Absorption spectra of (left) 1a−4a in toluene, (middle) their
dark isomers 1b−4b in toluene, and (right) 1d and 1e in benzene. Inset:
Photographs showing the colors of 1a−4a and 1b−4b in C6D6.
for 1a−4a indicated that the lowest excited states of these
molecules (S1 and T1, which are responsible for PI of N,C-chelate
BMes2 compounds) have features similar to those of B(ppy)-
Mes2 and its derivatives.5c Thus, not surprisingly, 1a−4a, like
B(ppy)Mes2, undergo PI upon irradiation at 350 nm, with a color
change to blue, deep blue, or purple (Figure 1), forming 1b−4b
quantitatively with moderate PI quantum efficiencies (0.46, 0.35,
1
0.25, and 0.38, respectively). H and 11B NMR data and DFT
calculations confirmed that the structures of 1b−4b are similar to
Figure 3. Crystal structures of (left) 1d and (right) 1e. Important bond
lengths (Å) an angles (deg) for 1d: B−N, 1.408(7); B−C1, 1.595(9);
B−C10, 1.570(7); C1−C2, 1.492(7); C25−C26, 1.509(8); S−C25,
1.822(6); N−C25, 1.508(6); C1−B−C10, 117.8(5); C1−B−N,
121.2(4); C10−B−N, 120.9(5); B−C1−C2, 102.4(5). For 1e: B−N,
1.424(3); B−C1, 1.597(3); B−C10, 1.579(3); C1−C2, 1.513(3); C25−
C26, 1.519(3); S−C25, 1.831(2); N−C25, 1.499(3); C1−B−C10,
113.90(8); C1−B−N, 123.5(2); C10−B−N, 122.50(19); B−C1−C2,
124.17(18).
that of the dark isomer of B(ppy)Mes2.
When heated, 1b−4b undergo further isomerization to form
1d−4d, respectively, with 1c−4c as the corresponding
intermediates. To illustrate this unusual and complex process,
1
we focus on the transformation of 1b. The H NMR spectral
changes of 1a upon irradiation and the subsequent changes upon
heating are shown in Figure 2. The structures of the isomers and
the key transition state involved are shown in Scheme 2. The dark
isomer 1b is transformed into the new species 1c in solution; the
process is slow at ambient temperature and fast upon heating.
transformed from an sp2 carbon to a chiral sp3 carbon, while the
C1 methyl group bound to the aliphatic C atom in 1b has
become a CH2 group, forming a σ bond with the B atom. The B
atom has a trigonal-planar geometry and a BN bond with a
bond length of 1.408(7) Å, which is comparable to some
previously reported BN bond lengths.9a,d−f The thiazoline ring
is considerably puckered, with a mean deviation of 0.19 Å from
the best least-squares plane through the five atoms. H25 is
oriented syn with respect to the adjacent phenyl C26−C27 bond.
The formation of 1d arises from a 1,3-sigmatropic shift of the B
atom in 1c driven by rearomatization of the CHD ring (Scheme
2). This leads to an expansion of the six-membered ring in 1c to
an eight-membered ring in 1d. Although 1,2-sigmatropic
migration of organoboron groups is well-known,10 1,3-migration
as in the 1c → 1d transformation is not common.
1
The most distinct H NMR spectral features of 1c are the two
Most exciting is the quantitative isomerization of 1d to its
diastereoisomer 1e upon heating at 90 °C, as evidenced by NMR
spectra (Figure 2 top). The 1H NMR spectrum of 1e is distinctly
different from that of 1d. For example, the H25 peak shifts
downfield in going from 1d to 1e. The geminal 2JH−H coupling
constant of the CH2 protons becomes much bigger (16 Hz in 1e
vs 11 Hz in 1d). The crystal structure of 1e was determined by
Figure 2. 1H NMR monitoring of (bottom) the sequential conversions
1a (red) → 1b (blue) → 1c (green) → 1d (purple) and (top) the
interconversion of 1d and 1e in C6D6.
3408
dx.doi.org/10.1021/ja400917r | J. Am. Chem. Soc. 2013, 135, 3407−3410

Journal of the American Chemical Society
Communication
scXRD (Figure 3). One key difference between 1d and 1e is the
inversion of the configuration of the chiral C atom, C25. As a
consequence, H25 becomes anti and the S atom syn to the C26−
C27 bond in 1e. H27 forms an intramolecular hydrogen bond
with the S atom (H27···S = 2.89 Å), in agreement with the >1
ppm downfield shift of the H27 peak in the NMR spectrum of 1e
relative to that of 1d (Figure 2). The C2−C1−B bond angle in 1e
[124.17(18)°] is much greater than that in 1d [102.4(5)°]. The
H−C1−H angle in 1e (106.3°) is smaller than that in 1d
(111.2°), consistent with the bigger 2JH−H coupling constant of
the CH2 protons. The other key difference between 1d and 1e is
that the thiazoline ring in 1e is coplanar with the B, C1, and C10
atoms in 1e (the mean deviation from the plane defined by the
benzothiazoline ring, B, C1, and C10 is 0.052 Å), indicating
greater π conjugation of the benzothiazoline and the B atom in 1e
than in 1d. This is corroborated by the appearance of a distinct
low-energy absorption band at 300−350 nm in the UV−vis
absorption spectrum of 1e, which is much weaker and at a slightly
higher energy in the spectrum of 1d (Figure 1). The activation
energy for the 1d → 1e transformation was determined to be 180
10 kJ mol−1 in C6D6. Most noteworthy is that fact that the two
diastereomers 1d and 1e can be interconverted quantitatively
(see the SI). 1d is stable toward irradiation and can be converted
to 1e only thermally. In contrast, 1e is thermally stable but,
remarkably, can be fully converted back to 1d upon irradiation at
300 nm. This reversible diastereomer interconversion is a rare
phenomenon, and it illustrates the possibility of controlling the
stereochemistry of azo-heterocyclic molecules with a B unit.
The dark isomers 2b−4b undergo a multistage isomerization
process similar to that for 1b. However, to achieve appreciable
thermal transformation of b to d, higher temperatures are needed
for 2b (90 °C) and 4b (130 °C). Hence, not surprisingly, the
intermediates 2c and 4c were not observed in the NMR spectra
because of their rapid conversion to 2d and 4d, respectively, at
the higher temperatures. 3b can isomerize to 3d at 50 °C, but the
heating time of 30 h needed to achieve the maximum conversion
(∼100%) is much longer than for 1b, which reached the
maximum conversion to 1d (∼82%) in 9 h at 50 °C. In addition,
the 1b → 1d isomerization occurs at ambient temperature, while
the 3b → 3d isomerization was not observed at this temperature.
Qualitatively, the b → d isomerization reactivity follows the order
1b > 3b > 2b ≫ 4b. The formation of c dearomatizes the azole
ring in b. Thus, the much lower reactivity of 4b relative to 1b−3b
may be due to the aromaticity of the imidazolyl ring in 4b, which
is much greater than that of the azole rings in 1b−3b (see the SI).
The fact that the dark isomer of B(ppy)Mes2 does not undergo
intramolecular HAT can also be explained by the well-known
high aromaticity of the pyridyl ring. Another important
observation is that upon heating, besides isomerizing to d,
isomers b can also thermally revert to the original isomers 1a
(∼18%), 2a (∼13%), 3a (∼0%), and 4a (∼82%).
To gain a better understanding of the multistage isomerization
process and the difference in the isomerization behaviors of 1a−
4a, the mechanistic pathways were investigated using DFT.11
The results for 1 are shown in Figure 4; those for 2−4, which
Figure 4. Relative ground-state energies and structures of the isomers,
intermediates, and transition states involved in the 1a → 1e
transformation. The involvement of excited states (S1 and Sx) in the
transformation is also illustrated.
follow the same mechanistic pathway, are given in the SI. The 1a
→ 1b transformation in the ground state follows the same
isomerization pathway as established previously for B(ppy)Mes2
and related compounds.6c In the optimized structure of 1b, the H
atom of the methyl group on the BC2 triangle is ∼2.46 Å from the
thiazole C2 atom. In the transition state (TS3), the H atom is
midway between the methylene and the thiazole C atoms and is
transferred directly to the thiazole, forming the syn diaster-
eoisomer 1c exclusively. The calculated barrier for 1b → 1a
thermal reversal is ∼30 kJ mol−1 higher than that for the 1b → 1c
isomerization, which explains why thermal treatment transforms
1b predominantly to 1c instead of 1a. The activation barrier for
the 1c → 1d reaction is smaller than that for 1d → 1c. Thus, once
1c is formed, it readily converts to 1d via TS4, in which the B−
CCHD bond is broken and the B atom is roughly equidistant from
CCHD and the two vinyl C atoms, supporting a concerted
sigmatropic migration of the B group.
The most important finding from the computational work is
that the 1d → 1e conversion involves the transition state TS5
(Scheme 2 and Figure 4), in which the C−S bond dissociates, a
CN bond forms, and the BN bond becomes a B−N bond.
The structure of TS5 is reminiscent of the ring-opened isomer of
spiropyrans and related compounds.4a,12 The C−S bond
dissociation is the key for the configuration inversion of the
chiral carbon atom and the interconversion of 1d and 1e. The
ability of the B center to accommodate both B−N and BN
bonds likely also plays an important role in this reversible
isomerization. The barrier for the 1d → 1e reaction is much
higher than that for the 1b → 1c (1d) process, in agreement with
the higher temperature needed for the 1d → 1e conversion.
Isomer 1e is ∼27 kJ mol−1 more stable than 1d, which may
explain the inability of 1e to undergo thermal reversion to 1d.
The S1 state of 1e is below that of 1d, and the HOMO → LUMO
transition dominates the S1 state for both compounds. However,
the oscillator strength for 1e is ∼5 times greater than that for 1d,
in agreement with the UV−vis spectrum (Figure 1). Thus, 1e
must be excited above the S1 state for it to be converted
effectively to 1d in the excited state. It is conceivable that a
transition state similar to TS5 is involved in the excited-state
Isomers 2d−4d can be quantitatively converted to 2e−4e at
110, 90, and 130 °C, respectively, in the same manner as 1d
(because of the high temperature used, 4e was observed in the 4b
→ 4d conversion). However, unlike 1e, which can be
quantitatively converted back to 1d by irradiation at 300 nm,
2e−4e can be only partially converted back to 2d−4d (max.
∼60% for 2d, ∼20% for 3d, and ∼10% for 4d). Prolonged
irradiation of 2e−4e did not increase the conversion yield but
instead formed unidentified products that may be caused by
either alternative PI pathways of 2e−4e or the poor photo-
stability of 2d−4d (see the SI).
3409
dx.doi.org/10.1021/ja400917r | J. Am. Chem. Soc. 2013, 135, 3407−3410

Journal of the American Chemical Society
Communication
conversion of 1e to 1d, requiring the BN bond to change into a
B−N bond and the C−S bond to break. Thus, to achieve effective
PI of 1e to 1d, the HOMO should be localized on the BN
bond and the thiazoline unit, which is indeed the case for 1e. In
contrast, for 1d, there is a large contribution to the HOMO from
the dimethylbenzene ring (the C2 ring in Figure 3), which may
explain the inability of 1d to photoisomerize to 1e. The
calculations also showed that among 1a−1e, the thiazoline
isomers 1d and 1e have the lowest energy. Thus, the unusual
multistructural transformation is thermodynamically favored.
The calculated barriers for the b → c conversions follow the
order 1b < 3b ≈ 2b < 4b, in agreement with the experimentally
observed reactivity trend. The relative stability of c versus b (i.e.,
the energy difference of the two isomers) follows the order 1c
(−20 kJ mol−1) > 3c (−6 kJ mol−1) > 2c (5 kJ mol−1) > 4c (25 kJ
mol−1). The poor stability of 4c and the high 4b → 4c activation
barrier are consistent with the relatively high aromaticity of the
imidazolyl ring and are responsible for the low activity of the 4b
→ 4d isomerization. The greater activity of the 1b → 1d
conversion compared with the 3b → 3d conversion, despite the
somewhat lower aromaticity of the benzoxazole ring in 3b, may
arise because weaker oxazole N donor in 3b stabilizes TS3 less
effectively than the thiazole N donor in 1b. The observed
reactivity trend for b → d conversion likely results from both
aromaticity and the azole N donor strength. The activation
barriers of the b → a thermal reversal follow the order 1b < 2b <
3b ≈ 4b, and the relative stability of a versus b follows the order
4a > 1a ≈ 2a > 3a, which, along with the b → c reactivity trend, is
responsible for the observed relative distribution of a and d/e in
the thermal isomerization of b.
facilities. S.W. thanks the Canada Council for the Arts for the
Killam Research Fellowship. Y.-L.R. thanks the Government of
Canada for the Vanier Scholarship.
REFERENCES
■
(1) (a) Woolley, G. A. Nat. Chem. 2012, 4, 75. (b) Beharry, A. A.;
Woolley, G. A. Chem. Soc. Rev. 2011, 40, 4422. (c) Balzani, V.; Credi, A.;
Venturi, M. Molecular Devices and Machines; Wiley-VCH: Weinheim,
Germany, 2003. (d) Feringa, B. L. Molecular Switches; Wiley-VCH:
Weinheim, Germany, 2001.
(2) (a) Natansohn, A.; Rochon, P. Adv. Mater. 1999, 11, 1387.
(b) Natansohn, A.; Rochon, P. Chem. Rev. 2002, 102, 4139. (c) Kume,
S.; Nishihara, H. Dalton Trans. 2008, 3260. (d) Yokoyama, Y.; Saito, M.
In Chiral Photochemistry; Inoue, Y., Ramamurthy, Y., Eds.; Marcel
Dekker: New York, 2004; Chapter 6.
(3) (a) Colle, M.; Dinnebier, R. E.; Brutting, W. Chem. Commun. 2002,
̈
̈
23, 2908. (b) Colle, M.; Gmeiner, J.; Milius, W.; Hillebrecht, H.;
̈
Brutting, W. Adv. Funct. Mater. 2003, 13, 108. (c) Tamayo, A. B.;
Alleyne, B. D.; Djurovich, P. I.; Lamansky, S.; Tsyba, I.; Ho, N. N.; Bau,
R.; Thompson, M. E. J. Am. Chem. Soc. 2003, 125, 7377.
̈
(4) (a) Berkovic, G.; Krongauz, V.; Weiss, V. Chem. Rev. 2000, 100,
1741 and references therein. (b) Irie, M. Chem. Rev. 2000, 100, 1685.
(c) Ko, C.-C.; Yam, V. W.-W. J. Mater. Chem. 2010, 20, 2063.
(5) (a) Rao, Y. L.; Amarne, H.; Zhao, S. B.; McCormick, T. M.; Martic,
S.; Sun, Y.; Wang, R. Y.; Wang, S. J. Am. Chem. Soc. 2008, 130, 12898.
(b) Amarne, H.; Baik, C.; Murphy, S. K.; Wang, S. Chem.Eur. J. 2010,
16, 4750. (c) Rao, Y. L.; Amarne, H.; Wang, S. Coord. Chem. Rev. 2012,
256, 759. (d) Rao, Y. L.; Amarne, H.; Lu, J. S.; Wang, S. Dalton Trans.
2013, 42, 638.
(6) (a) Ansorg, K.; Braunschweig, H.; Chiu, C. W.; Engels, B.; Gamon,
D.; Hugel, M.; Kupfer, T.; Radacki, K. Angew. Chem., Int. Ed. 2011, 50,
̈
2833. (b) Wilkey, J. D.; Schuster, G. B. J. Am. Chem. Soc. 1988, 110,
7569. (c) Rao, Y.-L.; Chen, L. D.; Mosey, N. J.; Wang, S. J. Am. Chem.
In summary, an unprecedented multistructural transformation
of B(2-phenylazolyl)Mes2 compounds has been demonstrated,
and mechanistic pathways for this remarkable phenomenon have
been established. Because the formation of the dark isomer b is
the key for the subsequent transformation, this work
demonstrates that light can be used as a convenient trigger or
switch to turn on unusual and complex structural/chemical
transformations of organoboron compounds. The intramolecu-
lar HAT observed in this system is facilitated by the relatively low
aromaticity of the azole rings. The ability of a BRR′ unit to
accommodate both BN and B−N bonds plays a key role in
promoting the reversible isomerization of d and e, illustrating the
potential use of organoboron chromophores in controlling
stereoselective ring-opening/closing processes of azole hetero-
cycles.
Soc. 2012, 134, 11026. (d) Nagura, K.; Saito, S.; Frohlich, R.; Glorius, F.;
̈
Yamaguchi, S. Angew. Chem., Int. Ed. 2012, 51, 7762. (e) Fukazawa, A.;
Yamaguchi, E.; Ito, E.; Yamada, H.; Wang, J.; Irle, S.; Yamaguchi, S.
Organometallics 2011, 30, 3870. (f) Fukazawa, A.; Yamada, H.;
Yamaguchi, S. Angew. Chem., Int. Ed. 2008, 47, 5582. (g) Kano, N.;
Yoshino, J.; Kawashima, T. Org. Lett. 2005, 7, 3909. (h) Yoshino, J.;
Kano, N.; Kawashima, T. Tetrahedron 2008, 64, 7774. (i) Wakamiya, A.;
Taniguchi, R.; Yamaguchi, S. Angew. Chem., Int. Ed. 2006, 45, 3170.
(7) The crystal structure data files for 2a, 3a, 1d, and 1e have been
deposited at the Cambridge Crystallographic Data Centre (CCDC
913143, 913142, 913141, and 913144, respectively).
(8) (a) Thiazole and Its Derivatives; Metzger, J. V., Ed.; The Chemistry
of Heterocyclic Compounds, Vol. 34; Wiley: New York, 1979; Part I, pp
132−133. (b) Hofmann, K. Imidazole and Its Derivatives, Part I; The
Chemistry of Heterocyclic Compounds, Vol. 6; Interscience: New York,
1953. (c) Clarke, G. M.; Sykes, P. J. Chem. Soc. C 1967, 1411.
(d) Katritzky, A. R.; Ramsden, C. A.; Joule, J. A.; Zhdankin, V. V.
Handbook of Heterocyclic Chemistry, 3rd ed.; Elsevier: Oxford, U.K.,
2010; Chapter 3.4, p 535.
(9) (a) Cui, Y.; Li, F. H.; Lu, Z. H.; Wang, S. Dalton Trans. 2007, 2634.
(b) Robertson, A. P. M.; Leitao, E. M.; Manners, I. J. Am. Chem. Soc.
2011, 133, 19322. (c) Brown, N. M. D.; Davidson, F.; Wilson, J. W. J.
Organomet. Chem. 1980, 192, 133. (d) Pestana, D. C.; Power, P. P. Inorg.
Chem. 1991, 30, 528. (e) Lichtblau, A.; Hausen, H. D.; Schwarz, W.;
Kaim, W. Inorg. Chem. 1993, 32, 73. (f) Okada, K.; Suzuki, R.; Oda, M. J.
Chem. Soc., Chem. Commun. 1995, 2069.
(10) (a) Bromm, L. O.; Laaziri, H.; Lhermitte, F.; Harms, K.; Knochel,
P. J. Am. Chem. Soc. 2000, 122, 10218. (b) Varela, J. A.; Pena, D.;
Goldfuss, B.; Polborn, K.; Knochel, P. Org. Lett. 2001, 3, 2395.
(c) Wood, S. E.; Rickborn, B. J. Org. Chem. 1983, 48, 555.
ASSOCIATED CONTENT
* Supporting Information
■
S
Synthesis and computational details, additional results, and
complete ref 11 (as ref 6S). This material is available free of
charge via the Internet at http://pubs.acs.org.
AUTHOR INFORMATION
Corresponding Author
wangs@chem.queensu.ca
■
Notes
The authors declare no competing financial interest.
(11) Frisch, M. J.; et al. Gaussian 09, revision B.01; Gaussian, Inc.:
Wallingford, CT, 2010.
ACKNOWLEDGMENTS
■
(12) (a) Toman, P.; Bartkowiak, W.; Nespurek, S.; Sworakowski, J.;
̌
̊
We thank the Natural Sciences and Engineering Research
Council of Canada for financial support and the High
Performance Computing Virtual Laboratory for computing
́
Zalesny, R. Chem. Phys. 2005, 316, 267. (b) Paramonov, S. V.; Lokshin,
V.; Fedorova, O. A. J. Photochem. Photobiol., C 2011, 12, 209.
3410
dx.doi.org/10.1021/ja400917r | J. Am. Chem. Soc. 2013, 135, 3407−3410