2,6-Diazasemibullvalenes: Synthesis, structural characterization, reaction chemistry, and theoretical analysis

Article abstract of DOI:10.1021/ja305581f

A series of 2,6-diazasemibullvalenes (NSBVs) were synthesized and isolated from the reaction of 1,4-dilithio-1,3-dienes with nitriles via oxidant-induced C-N bond formation. For the first time, the activation barrier and an X-ray crystal structure of a su

Full text of DOI:10.1021/ja305581f

Communication
pubs.acs.org/JACS
2,6-Diazasemibullvalenes: Synthesis, Structural Characterization,
Reaction Chemistry, and Theoretical Analysis
Shaoguang Zhang,† Junnian Wei,† Ming Zhan,† Qian Luo,† Chao Wang,† Wen-Xiong Zhang,†
and Zhenfeng Xi*,†,‡
†Beijing National Laboratory for Molecular Sciences (BNLMS), Key Laboratory of Bioorganic Chemistry and Molecular Engineering
of the Ministry of Education, College of Chemistry, Peking University, Beijing 100871, China
‡State Key Laboratory of Organometallic Chemistry, Shanghai Institute of Organic Chemistry, Chinese Academy of Sciences,
Shanghai 200032, China
S
* Supporting Information
Scheme 1. Semibullvalenes and 2,6-Diazasemibullvalenes
ABSTRACT: A series of 2,6-diazasemibullvalenes
(NSBVs) were synthesized and isolated from the reaction
of 1,4-dilithio-1,3-dienes with nitriles via oxidant-induced
C−N bond formation. For the first time, the activation
barrier and an X-ray crystal structure of a substituted 2,6-
diazasemibullvalene were determined. All NSBVs show
extremely rapid aza-Cope rearrangement in solution, but
the rapid aza-Cope rearrangement is “frozen” in the solid
state, as shown by solid-state NMR measurements and X-
ray single-crystal structural analysis. Insertion of unsatu-
rated compounds or a low-valent metal center into the
NSBV C−N bond gave diverse and interesting ring-
expansion products. Theoretical analysis showed that the
localized structure is predominant and that the homoar-
omatic delocalized structure exists as a minor component
in the equilibrium.
During the past 30 years, no further report have followed in the
literature, leaving the structure and reaction chemistry of
NSBVs almost totally unknown.5c In this paper, we report (1)
the synthesis and isolation of a series of NSBVs, (2) the first
single-crystal structure of an NSBV (1a), (3) the insertion
reaction chemistry of NSBVs, and (4) theoretical/computa-
tional calculations and analysis.
2,6-Diazasemibullvalene (NSBV) and its all-carbon analogue
semibullvalene (SBV) have long been of fundamental interest
both theoretically and experimentally1−5 because of their
unique strained ring systems, their intramolecular skeletal
rearrangements, their rapid degenerate Cope rearrangement,6
and the predicted existence of a homoaromatic delocalized
structure.1d,7 Synthesis and structural studies of these highly
strained ring systems have been a great challenge in organic
chemistry. Reaction studies and synthetic applications of their
nonclassical bonding have long been attractive.
For SBVs, there have been many reports and remarkable
achievements on the synthesis, structures, and reaction
chemistry2 as well as theoretical studies3 since Zimmerman
and co-workers reported the first synthesis in 1966.2a It has
been predicted theoretically that NSBVs should undergo a
more rapid aza-Cope rearrangement with a lower activation
barrier than the all-carbon SBV analogues and that potentially
they should approach a delocalized homoaromatic structure.
However, little is known experimentally.4,5 In fact, as the only
experimental in situ NMR identification of an NSBV to date,
1,5-dimethyl-3,7-diphenyl-2,6-diazasemibullvalene (2) was re-
We have developed two preparative methods (A and B) for
the efficient synthesis of NSBV derivatives 1 from the reaction
of dilithio reagents 4 with nitriles (Scheme 2).8 Both methods
involve lithiation and oxidant-induced intramolecular C−N
bond formation.9 Method A is a one-pot synthesis of 1. In the
type-I synthesis, 4a was generated in situ from its
corresponding 1,4-diiodo compound and t-BuLi. Reaction of
4a with 2.4 equiv of trimethylacetonitrile (t-BuCN) readily
afforded dianion 5a.8 Addition of di-tert-butyl peroxide [(t-
BuO)2, 4.0 equiv] as oxidant led to 1,5-bridged NSBV
derivative 1a via intramolecular C−N bond formation.
Decomposition of 1a occurred when the normal workup
procedure and column chromatography using silica gel or
alumina were used to purify the product. Bulb-to-bulb
distillation (220 °C, 0.01 kPa) was found to be an efficient
purification procedure, and pure 1a was obtained in 66%
isolated yield as light-yellow crystals. The reactions of 4a with
different nitriles followed by treatment with (t-BuO)2 afforded
moderate yields of 1,5-bridged NSBVs 1b−e with different
ported by Mullen and co-workers as a breakthrough in 1982
̈
(Scheme 1).5a In 1985, the Mullen group reported the thermal
̈
rearrangement of 2 to give a 1,5-diazocine, which is the only
Received: June 8, 2012
Published: July 11, 2012
recorded example of the reaction chemistry of NSBVs.5b
© 2012 American Chemical Society
11964
dx.doi.org/10.1021/ja305581f | J. Am. Chem. Soc. 2012, 134, 11964−11967

Journal of the American Chemical Society
Communication
Scheme 2. Synthetic Strategies for NSBVs 1
Scheme 3. Thermolysis of NSBVs 1 to 1,5-Diazocines 6
°C and did not undergo the transformation. These results show
that the substituents at the 1- and 5-positions of 1 play an
important role in their thermal stability.2c Notably, thermal
conversion of the all-carbon SBV analogues to cyclooctate-
traenes (COTs) occurs at higher temperatures in most cases.2g
1,5-Diazocines are interesting aza analogues of COTs, but
efficient methods for synthesizing such compounds are very
rare.5b,10
The solution NMR spectra of the isolated NSBVs 1a−h
showed a rapid equilibrium between two localized structures, 1
and 1′. In 1a, for example, the aziridinyl H4 and vinyl H8
displayed only one singlet at 4.79 ppm in the 1H NMR
spectrum in THF-d8 and C1/C5, C3/C7, and C4/C8 displayed
singlets at 79.2, 162.9, and 99.1 ppm, respectively, in the 13C
NMR spectrum in THF-d8 as a result of the rapid degenerate
aza-Cope rearrangement. The chemical shift of C4/C8 in 1a is
comparable to those found in NBSV 2 (99.4 ppm for C4/
C8).5a Low-temperature 1H and 13C NMR data for 1a in THF-
d8 recorded on a 600 MHz spectrometer unambiguously
showed that 1a continued to undergo rapid aza-Cope
rearrangement even down to −100 °C. However, at −110
°C, with addition of CS2 in the solvent, line broadening of the
singlet peak for C4/C8 was observed (width at half-height W1/2
= 41.9 Hz), while no obvious line broadening of the peak for
the CH3 carbon on the t-Bu group took place (W1/2 = 8.8 Hz),
suggesting that the aza-Cope rearrangement was slowed. These
experimental observations indicate that 1a is not a static
homoaromatic form but instead is dynamically balanced by the
rapid degenerate aza-Cope rearrangement, in good agreement
a
Conditions for obtaining 3 in method B: HMPA (2.0 equiv), rt, 0.5 h;
b
R′CN (2.4 equiv), reflux, 3h; NaHCO3(aq). [O] = (t-BuO)2 (4.0
equiv). [O] = PhI(OAc)2 (1.0 equiv). [O] = t-BuOCl (1.0 equiv).
c
d
substituents at the 3- and 7-positions. In the type-II synthesis,
dilithio reagent 4b was successfully applied for the one-pot
formation of NSBV derivative 1f in 51% isolated yield.
Method B is a stepwise synthesis of NSBVs 1. Dianions 5
could be readily generated in situ via dilithiation of Δ1-
bipyrrolines 3. Sequential addition of phenyliodine diacetate
[PhI(OAc)2] as oxidant afforded the corresponding NSBVs 1
in good isolated yields. The use of (t-BuO)2 led to a slightly
lower yield. With method B, 1a−e could all be obtained in
higher isolated yields. The 1,5-dialkyl-substituted Δ1-bipyrro-
lines 3 obtained from type-III reagent 4c could also be
converted to the corresponding non-bridged NSBVs 1g and 1h
in 72 and 73% isolated yield, respectively. For the synthesis of
type-I NSBV derivatives, method B was found to be more
efficient than method A. All of the NSBV derivatives are stable
in an inert atmosphere at room temperature.
However, when 2,3-diphenyl-1,4-dilithio-1,3-butadiene (4d)
was applied using method A (Scheme 3), 1,5-diazocine 6a was
obtained in 53% isolated yield and structurally characterized.
We assumed that the expected NSBV derivative 1i might be
unstable at room temperature and readily transformed into the
thermodynamically more stable 6a, which was also obtained via
method B [see the Supporting Information (SI)].5c The
nonbridged NSBV derivatives 1g and 1h could be quantitively
converted to their corresponding 1,5-diazocines 6b and 6c, but
a higher temperature was required.2g,5b On the contrary, 1,5-
bridged NSBVs 1a−f showed good thermal stability under 200
with Mullen’s report. Through line-shape analysis of the low-
̈
temperature 13C NMR spectra as reported by Quast and co-
workers,2f the upper limit of the activation barrier of the aza-
Cope rearrangement at 163 K (ΔG⧧163K) was found to be 4.4
kcal/mol, which is indeed lower than those of their
corresponding all-carbon analogues.2b,f
The solid-state 13C NMR spectrum of 1a at room
temperature showed a “frozen” unsymmetrical structure. C4
and C8 showed two broad singlets at 74.7 and 125.3 ppm,
respectively. Other peaks of 1a (e.g., C1/C5 and C3/C7) all
showed different chemical shifts from one another, indicating
that the degenerate aza-Cope rearrangement was “frozen” in
the solid state.
A single crystal of 1a suitable for X-ray structural
determination, obtained at −20 °C in hexane/diethyl ether
solution, provided the first example of a single-crystal structure
of an NSBV. In sharp contrast to its solution behavior, the
single-crystal structure shows a localized structure with a
strained aziridine ring (Figure 1a). This is in good agreement
with the solid-state 13C NMR data. The 1,5-bridge exists as a
distorted boatlike cyclohexane ring. The C4−N6 bond (1.628
Å) is much longer than that in simple aziridine compounds
11965
dx.doi.org/10.1021/ja305581f | J. Am. Chem. Soc. 2012, 134, 11964−11967

Journal of the American Chemical Society
Communication
respectively. Thus, both 1adeloc and 1a* could be homoaromatic
on the basis of their NICS values.
The reaction chemistry of NSBVs is unknown except for the
thermolysis of 2 to give 1,5-diazocine as reported by Mullen
̈
and co-workers.5b Although all of the isolated NSBVs 1 are
stable in an inert atmosphere, they are sensitive to acid, base,
and silica gel and decompose slowly when exposed to moisture.
These observations indicated their highly reactive nature and
suggested that the reaction chemistry of such fluxional
molecules should be very interesting.
The insertion of unsaturated compounds or low-valent
transition metal centers into the weakened C−N bonds
interrupt the rapid Cope rearrangement and lead to the
diversified ring-expansion products (Scheme 4).14,15 Regiospe-
Figure 1. (a) X-ray structure of 1a (30% thermal ellipsoids; H atoms
except H4 and H8 omitted for clarity). (b) B3LYP/6-31G*-optimized
localized structure of 1a. Selected bond lengths (Å) and angles (deg)
in the two structures are shown.
(1.520 Å), indicating enhanced strain and through-bond
coupling in the NSBV molecule.11 The other bond lengths in
the NSBV core are all in the normal range, comparable to those
in the calculated localized structure of unsubstituted NSBV.4b
The structures of both localized NSBV 1a and delocalized
NSBV 1adeloc were optimized using DFT calculations. At the
B3LYP/6-31G* level,12 the ground states of 1a and 1adeloc were
found to be energy minima, as confirmed by frequency
calculations (Figure 1b). The (GIAO)B3LYP/6-311+G**-
calculated 13C NMR spectrum of the localized structure was
similar to the solid-state 13C NMR spectrum of 1a. The Gibbs
free energy of 1adeloc at 163 K was 1.8 kcal/mol higher than that
of the calculated localized 1a. This trend is consistent with the
calculations on unsubstituted NSBV at the MP2/cc-pVDZ level
by Greve.4b In addition, the transition state 1a* for the aza-
Cope rearrangement was optimized. The calculations indicated
that 1a* and 1adeloc are very close in energy, and the potential
energy surface has a broad, flat transition-state region (Figure
2). The value of ΔG⧧163K was calculated to be only 2.1 kcal/mol,
Scheme 4. Insertions into the Weakened C−N Bonds
cific cycloaddition of 1a with the activated alkynes dimethyl and
diethyl acetylenedicarboxylate in toluene at 90 °C readily
afforded the 1,5-diazatriquinacenes 7a and 7b in 60 and 53%
isolated yield, respectively (Scheme 4a).16 To the best of our
knowledge, the syntheses of triquinacenes and azatriquinacenes
generally require multistep procedures; thus, this straightfor-
ward synthesis of 1,5-diazatriquinacenes should be useful for
the construction of structurally interesting yet otherwise
unavailable “bowl-shaped” polycyclic frameworks.15,17 The
cycloaddition of 1a with isocyanates without any catalyst led
to tetracyclic imidazolidinone derivatives 8a−c in moderate
yields (Scheme 4b). It is noteworthy that all reported reactions
of simple aziridines with isocyanates require a catalyst such as a
transition-metal salt,18 indicating that the C−N bonds in
NSBVs might be more reactive than those in simple aziridines
because of the enhanced ring strain caused by the rigid ring
system as well as through-bond coupling. The structures of 7a
and 8a were determined by X-ray diffraction (see the SI).
Carbonylation of 1a using Co2(CO)8 at room temperature
gave the tetracyclic β-lactam 9 in 61% yield (Scheme 4c). In
contrast, carbonylation reactions of simple aziridines all occur
only at elevated temperatures, at high CO pressures, or in the
Figure 2. Calculated gas-phase relative energies, Gibbs free energies,
and enthalpies (all in kcal/mol) at 163 K, 1 atm.
which is comparable with the experimental result. Because of
the small activation barrier from 1adeloc to 1a, the rearrange-
ment of 1adeloc to 1a should be extremely fast. These results
show that the 1a is the predominant form in solution and the
gas phase, with the homoaromatic delocalized 1adeloc existing as
a minor component in the equilibrium. For details of the DFT
calculations, including optimized structures and selected bond
lengths and angles in 1adeloc and 1a*, see the SI.
The B3LYP/6-311+G**-calculated nucleus-independent
chemical shift (NICS) values for 1adeloc were NICS(0) =
−19.0 ppm and NICS(−1) = −14.2 ppm, whose magnitudes
are larger than those of 1a [NICS(0) = −8.4 ppm, NICS(−1) =
−7.8 ppm].4b,13 In addition, the NICS(0) and NICS(−1)
values of the transition-state 1a* were −17.6 and −13.3 ppm,
11966
dx.doi.org/10.1021/ja305581f | J. Am. Chem. Soc. 2012, 134, 11964−11967

Journal of the American Chemical Society
Communication
presence of promoters.19 Insertion of a low-valent transition
metal into the weakened C−N bond was demonstrated by the
reaction of 1a with an N-heterocyclic carbene-ligated Ni(0)
complex (Scheme 4d). Addition of 1a to a 1:1 mixture of
bis(1,5-cyclooctadiene)nickel(0) [Ni(cod)2] and 1,3-bis(2,6-
diisopropylphenyl)imidazol-2-ylidene (IPr) in THF resulted in
a rapid color change from dark brown to red. The three-
coordinated, four-membered azanickelacycle 10 was isolated in
80% yield, and its structure is shown in Scheme 4. Only one IPr
ligand coordinates to the Ni(II) center, probably because of
steric hindrance. The angles N2−Ni1−C19 (159.7°) and C5−
Ni1−C19 (109.9°) reveal a distorted T-shaped Ni coordination
environment.20 Clearly, the enhanced ring strain and through-
bond coupling in NSBV molecules in solution weakens the C−
N bonds, leading to reactivities different from those of simple
aziridine analogues.
In summary, we have successfully established experimental
models for structurally and theoretically interesting 2,6-
diazasemibullvalenes (NSBVs). The efficient one-pot synthesis
and isolation of a series of NSBVs was performed by oxidant-
induced C−N bond formation. For the first time, the single-
crystal structure of an NSBV (1a) was determined, revealing a
localized structure. The C2-symmetric structure of 1a in
solution along with line broadening of the NMR signal at
−110 °C indicates an extremely low barrier for the rapid
degenerate aza-Cope rearrangement. Theoretical analysis
showed that the localized structure is predominant, and the
homoaromatic delocalized structure exists as a minor
component in the equilibrium. Insertion reactions of
unsaturated compounds and a low-valent metal center into
the NSBV C−N bond generated diverse ring-expansion
products, demonstrating the unusual reactivity of NSBVs.
Further studies of the chemical and physical properties of these
otherwise unavailable azasemibullvalenes are in progress.
(d) Quast, H.; Carlsen, J.; Janiak, R.; Peters, E.-M.; Peters, K.; von
Schnering, H. G. Chem. Ber. 1992, 125, 955. (e) Williams, R. V.;
Gadgil, V. R.; Chauhan, K.; van der Helm, D.; Hossain, M. B.;
Jackman, L. M.; Fernandes, E. J. Am. Chem. Soc. 1996, 118, 4208.
(f) Jackman, L. M.; Fernandes, E.; Heubes, M.; Quast, H. Eur. J. Org.
Chem. 1998, 2209. (g) Quast, H.; Heubes, M.; Dietz, T.; Witzel, A.;
Boenke, M.; Roth, W. R. Eur. J. Org. Chem. 1999, 813. (h) Seefelder,
M.; Heubes, M.; Quast, H.; Edwards, W. D.; Armantrout, J. R.;
Williams, R. V.; Cramer, C. J.; Goren, A. C.; Hrovat, D. A.; Borden, W.
T. J. Org. Chem. 2005, 70, 3437. (i) Wang, C.; Yuan, J.; Li, G.; Wang,
Z.; Zhang, S.; Xi, Z. J. Am. Chem. Soc. 2006, 128, 4564. (j) Griffiths, P.
R.; Pivonka, D. E.; Williams, R. V. Chem.Eur. J. 2011, 17, 9193.
(3) (a) Jiao, H.; Schleyer, P. v. R. Angew. Chem., Int. Ed. Engl. 1993,
32, 1760. (b) Jiao, H.; Nagelkerke, R.; Kurtz, H. A.; Williams, R. V.;
Borden, W. T.; Schleyer, P. v. R. J. Am. Chem. Soc. 1997, 119, 5921.
(c) Goren, A. C.; Hrovat, D. A.; Seefelder, M.; Quast, H.; Borden, W.
T. J. Am. Chem. Soc. 2002, 124, 3469. (d) Wang, S. C.; Tantillo, D. J. J.
Phys. Chem. A 2007, 111, 7149.
(4) Theoretical studies of NSBVs: (a) Dewar, M. J. S.; Nah
́ ́
lovska, Z.;
Nahlovsky, B. D. Chem. Commun. 1971, 1377. (b) Greve, D. R. J. Phys.
́
́
Org. Chem. 2011, 24, 222. Other heterosemibullvalenes (c) Wu, H.-S.;
Jiao, H.; Wang, Z.-X.; Schleyer, P. v. R. J. Am. Chem. Soc. 2003, 125,
10524.
(5) Experimental studies of NSBVs: (a) Schnieders, C.; Altenbach,
H. J.; Mullen, K. Angew. Chem., Int. Ed. Engl. 1982, 21, 637.
̈
(b) Schnieders, C.; Huber, W.; Lex, J.; Mullen, K. Angew. Chem., Int.
̈
Ed. Engl. 1985, 24, 576. (c) Dull, B.; Mullen, K. Tetrahedron Lett.
1992, 33, 8047.
̈
̈
(6) (a) Houk, K. N.; Gonzalez, J.; Li, Y. Acc. Chem. Res. 1995, 28, 81.
(b) Graulich, N. Wiley Interdiscip. Rev.: Comput. Mol. Sci. 2011, 1, 172.
Metal-promoted Cope rearrangements: (c) Siebert, M. R.; Tantillo, D.
J. J. Am. Chem. Soc. 2007, 129, 8686 and references therein.
(7) Winstein, S. J. Am. Chem. Soc. 1959, 81, 6524.
(8) Yu, N.; Wang, C.; Zhao, F.; Liu, L.; Zhang, W.-X.; Xi, Z. Chem.
Eur. J. 2008, 14, 5670.
(9) West, S. P.; Bisai, A.; Lim, A. D.; Narayan, R. R.; Sarpong, R. J.
Am. Chem. Soc. 2009, 131, 11187.
(10) (a) Robins, L. I.; Carpenter, R. D.; Fettinger, J. C.; Haddadin,
M. J.; Tinti, D. S.; Kurth, M. J. J. Org. Chem. 2006, 71, 2480.
(b) Lodewyk, M. W.; Kurth, M. J.; Tantillo, D. J. J. Org. Chem. 2009,
74, 4804.
(11) Sasaki, M.; Yudin, A. K. J. Am. Chem. Soc. 2003, 125, 14242.
(12) (a) Becke, A. D. J. Chem. Phys. 1993, 98, 5648. (b) Lee, C.;
Yang, W.; Parr, R. G. Phys. Rev. B 1988, 37, 785. (c) Hehre, W. J.;
Radom, L.; Schleyer, P. v. R.; Pople, J. A. Ab Initio Molecular Orbital
Theory; Wiley: New York, 1986.
(13) (a) Schleyer, P. v. R.; Maerker, C.; Dransfeld, A.; Jiao, H.;
Hommes, N. J. R. v. E. J. Am. Chem. Soc. 1996, 118, 6317. (b) Chen,
Z.; Wannere, C. S.; Corminboeuf, C.; Puchta, R.; Schleyer, P. v. R.
Chem. Rev. 2005, 105, 3842.
ASSOCIATED CONTENT
* Supporting Information
■
S
Experimental and computational procedures, NMR spectra, and
crystallographic data (CIF). This material is available free of
charge via the Internet at http://pubs.acs.org.
AUTHOR INFORMATION
Corresponding Author
zfxi@pku.edu.cn
■
Notes
(14) Dauban, P.; Malik, G. Angew. Chem., Int. Ed. 2009, 48, 9026 and
references therein.
The authors declare no competing financial interest.
(15) Askani, R.; Kirsten, R.; Dugall, B. Tetrahedron 1981, 37, 4437.
(16) (a) Dalili, S.; Yudin, A. K. Org. Lett. 2005, 7, 1161.
(b) Baktharaman, S.; Afagh, A.; Vandersteen, A.; Yudin, A. K. Org.
Lett. 2010, 12, 240.
ACKNOWLEDGMENTS
This work was supported by the NSFC and the 973 Program
(2011CB808700). We thank Prof. Zhi-Xiang Yu of Peking
■
(17) (a) Mascal, M.; Lera, M.; Blake, A. J. J. Org. Chem. 2000, 65,
7253. (b) Cadieux, J. A.; Buller, D. J.; Wilson, P. D. Org. Lett. 2003, 5,
3983 and references therein.
University and Prof. Todd B. Marder of Universitat Wurzburg
for insightful discussions and comments.
̈
̈
(18) Munegumi, T.; Azumaya, I.; Kato, T.; Masu, H.; Saito, S. Org.
Lett. 2006, 8, 379 and references therein.
REFERENCES
■
(1) Reviews of semibullvalenes: (a) Williams, R. V. Adv. Theor.
Interesting Mol. 1998, 4, 157. (b) Hopf, H. Classics in Hydrocarbon
Chemistry; Wiley-VCH: Weinheim, Germany, 2000; Chapter 10, p
209. (c) Williams, R. V. Eur. J. Org. Chem. 2001, 227. (d) Williams, R.
V. Chem. Rev. 2001, 101, 1185.
(2) (a) Zimmerman, H. E.; Grunewald, G. L. J. Am. Chem. Soc. 1966,
88, 183. (b) Cheng, A. K.; Anet, F. A. L.; Mioduski, J.; Meinwald, J. J.
Am. Chem. Soc. 1974, 96, 2887. (c) Quast, H.; Mayer, A.; Peters, E.-
M.; Peters, K.; von Schnering, H. G. Chem. Ber. 1989, 122, 1291.
(19) Piotti, M. E.; Alper, H. J. Am. Chem. Soc. 1996, 118, 111.
(20) (a) Lin, B. L.; Clough, C. R.; Hillhouse, G. L. J. Am. Chem. Soc.
2002, 124, 2890. Related T-shaped Ni−NHC complexes: (b) Wang,
S. C.; Troast, D. M.; Conda-Sheridan, M.; Zuo, G.; LaGarde, D.;
Louie, J.; Tantillo, D. J. J. Org. Chem. 2009, 74, 7822.
11967
dx.doi.org/10.1021/ja305581f | J. Am. Chem. Soc. 2012, 134, 11964−11967