Axial Chirality about Boron-Carbon Bond: Atropisomeric Azaborines

Article abstract of DOI:10.1021/acs.orglett.6b01159

The preparation of atropisomeric 2,1-borazaronaphthalenes is described. Resolution of the atropisomeric pair was achieved by preparative Chiral Stationary Phase HPLC (CSP-HPLC). The absolute configuration of the stereogenic axis was derived from Time-Depe

Full text of DOI:10.1021/acs.orglett.6b01159

Letter
pubs.acs.org/OrgLett
Axial Chirality about Boron−Carbon Bond: Atropisomeric Azaborines
Andrea Mazzanti, Elia Mercanti, and Michele Mancinelli*
Department of Industrial Chemistry “Toso Montanari”, University of Bologna, Viale del Risorgimento 4, I-40136 Bologna, Italy
S
* Supporting Information
ABSTRACT: The preparation of atropisomeric 2,1-borazaro-
naphthalenes is described. Resolution of the atropisomeric pair
was achieved by preparative Chiral Stationary Phase HPLC
(CSP-HPLC). The absolute configuration of the stereogenic
axis was derived from Time-Dependent DFT (TD-DFT)
simulation of the Electronic Circular Dichroism spectra
(ECD). X-ray diffraction and Dynamic NMR data allowed
structural and dynamic comparison with the analogue isosteric
carbon compounds.
he discovery in the 20th century of biologically active
Taxially chiral compounds useful as pharmaceutical drugs
has led chemists to develop new atropisomeric compounds and
to address their importance in drug discovery.1,2
Dewar and co-worker proposed the first preparation of
azaboraphenanthrene3 and 2,1-borazaronaphthalene4 where
they realized the isosteric replacement of CC with B−N
within the context of an aromatic system. The forefather 1,2-
azaborine was prepared only recently.5 This new class of
compounds has attracted much attention over the past few
years6 because of their importance in both medicinal and
materials science. The replacement of two sp2 carbon atoms by
one boron and one nitrogen atom produces systems that are
isoelectronic with their all-carbon analogues,7 yet they possess
at least local dipole moments, as the N−B bond contains a
significant ionic component. This polarity can significantly alter
both molecular and solid-state electronic and optical properties
of the system by modifying the character of the frontier
molecular orbitals, and the intermolecular interactions present
in solid phases.
Figure 1.
structure of a B−N fused polycyclic aromatic structure where
the azaborine substructure was twisted by steric interaction,
reporting a B−N bond distance of 1.426 Å, typical of a BN
double bond. Most of the reported X-ray structures of
compounds containing the 2,1-borazaronaphthalene ring11 are
Cr(CO)3 complexes or contain boron−oxygen bonds. To the
best of our knowledge this is the first crystal structure of a 2,1-
borazaronaphthalene with both N−C and B−C bonds.12
A detailed investigation of the bond lengths can therefore
cast new light on the electronic features of these compounds
(Figure 2). The 2,1-borazaronaphthalene ring is perfectly
planar and the B−N bond length is 1.427 Å, shorter than the
C4−C5 bond length. This confirms the strong contribution of
the nitrogen lone pair to develop a double bond between
nitrogen and boron. In contrast the B−C3 bond length (1.515
Å) is more consistent with a B−C single bond, and C3−C4
length is typical of an isolated CC double bond. The
structure in the solid state has the m-tolyl group displaced out
of the azaborine plane by 72°, and the p-nitrophenyl group is
almost perpendicular to the same plane (B−N−CH2−Cq
dihedral angle = 104°). Due to the twisted conformation, the
m-tolyl ring cannot develop conjugation with boron, and the
B−Cq bond length is 1.581 Å.
In the recent past, we have been interested in the preparation
of a new class of atropisomeric compounds bearing N−C
stereogenic axes, and newly synthetic procedures of 2,1-
borazaronaphthalenes8 stimulated our interest in the oppor-
tunity for a boron−carbon stereogenic axis able to produce
thermally stable atropisomers.
In order to detect the existence of a stereogenic axis, a
chirality probe is needed within the molecular scaffold. The
benzylic group is well suitable to this purpose because of the
isolated CH2 that develops a simple AB-system when
conformational enantiomers are formed. In this case a p-
nitrobenzyl moiety9 was linked to the nitrogen of 2,1-
borazaronaphthalene ring, and aromatic rings with different
steric hindrances were bound to boron (compounds 1−3,
Figure 1).
Single crystals of compound 1 were obtained by slow
evaporation of a dichloromethane solution. Crystal structures of
azaborines are rare. Seki and Hatakeyama10 reported the crystal
Received: April 21, 2016
© XXXX American Chemical Society
A
DOI: 10.1021/acs.orglett.6b01159
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
Figure 2. X-ray structure of compound 1 (crystals from dichloro-
methane). On the left are reported some relevant bond lengths. On
the right is reported a side view showing the out-of-plane disposition
of boron and nitrogen substituents. All the hydrogens are omitted for
clarity.
Density Functional Theory (DFT) calculations were run13 at
the B3LYP/6-31G(d) level of theory to explore the conforma-
tional space due to the different dispositions allowed to p-
nitrobenzyl and m-tolyl rings. Four stable conformations were
found, as from Figure S1 of Supporting Information (SI). They
differ because of the relative disposition of the 3-methyl moiety
and the benzyl moiety that can be on the same side (syn) or on
the opposite side (anti) with respect to the azaborine plane. For
each of the two conformations, two additional conformers are
generated by a different dihedral angle of the m-tolyl ring (|61°|
and |122°|; see Figure S1). DFT data well reproduced the
ground state geometries in terms of bond distances (B−N =
1.443 Å, N−Cq = 1.404 Å, B−C3 = 1.524 Å). Between the two
available transition states for the m-tolyl rotation, the lowest
energy one (Figure S2, top) corresponds to a rotational barrier
of 6.5 kcal/mol. Albeit rather small, this barrier should be
observable by the Dynamic NMR technique.14 On the other
hand, two transitions states were modeled and optimized also
for the p-nitrobenzyl rotation (Figure S2, bottom). The more
favorable transition state puts the two aryl rings perfectly
stacked, and its energy was calculated as 10.5 kcal/mol; thus,
also this conformational change could be detected by low
temperature NMR.
When a sample of 1 in CD2Cl2 was cooled down, the CH2
signal started to broaden below −10 °C, reached coalescence at
−37 °C and eventually split into an AB-system at −89 °C
(Figure 3). Line-shape simulations provided the rate constants
for the rotational process, and hence the free energy of the
process (10.7 0.15 kcal/mol). This values should therefore be
related to the N-(p-nitrobenzyl) rotation. On further lowering
the temperature of a CDFCl2 sample of 1, the methyl signal of
the m-tolyl ring broadened below −120 °C and eventually split
at −150 °C into two lines with a 78:22 ratio (ΔG° = 0.31 kcal/
mol, Figure S3), thus confirming the formation of two
conformational diastereoisomers where both rotations are
frozen in the NMR time scale. Line-shape simulation provided
a value of 6.6 kcal/mol for this lower barrier, in excellent
agreement with the value suggested by DFT for the B-Aryl
rotation (Table S1).
Figure 3. VT spectra for compound 1 (1H NMR 600 MHz in
CD2Cl2). On the left is shown the evolution of the CH2 signal on
lowering the temperature. On the right are reported the line-shape
simulations with the corresponding rate constants.
calculations suggest a barrier of 16.5 kcal/mol for the B-Aryl
rotation and 10.0 kcal/mol for the p-nitrobenzyl rotation
(Figure S4).
At ambient temperature, the spectrum of compound 2 in
C2D2Cl4 exhibits a well resolved AB system for the benzylic
CH2, albeit with a much smaller chemical shift separation with
respect to compound 1 (36 Hz for 2 and 267 Hz for 1,
respectively). On raising the temperature, the coalescence point
is reached at +111 °C and a single peak is observed above +130
°C (Figure S5). The corresponding energy barrier derived from
line-shape simulations is 19.1 kcal/mol. In this compound this
barrier can be safely assigned to the B-Aryl rotation due to the
relevant steric hindrance of the ortho-methyl.
When a sample of 2 in CDFCl2 was cooled, the two doublets
of the AB system broaden and split at −90 °C into two pairs of
signals with a 75:25 ratio, thus confirming the formation of two
conformational diastereoisomers generated by the two stereo-
genic axes (Figure S6). The experimental energy barrier
determined by line-shape simulation was 10.0 kcal/mol, very
similar to the higher barrier of compound 1 and identical to
that suggested by calculations for N-(p-nitrobenzyl) rotation.
In order to compare the experimental barriers determined for
B-Aryl rotation in compounds 1 and 2 with that of standard
Aryl−Aryl rotation, we prepared the two naphthyl compounds
4 and 5 (Figure 1) that are the carbon isosteric analogues of 1
and 2 (see the synthetic procedure in SI). Compounds 4 and 5
were analyzed with dynamic NMR and kinetic racemization to
determine the two barriers for m-tolyl and 2,3-dimethylphenyl
rotation, which were found to be 10.7 and 25.4 kcal/mol,
respectively. Rotational barriers of 8.0 and 7.6 kcal/mol were
measured for N-(p-nitrobenzyl) rotation (see Figures S7, S8,
S9, and S10).
To unambiguously confirm which conformational process is
responsible for the higher and lower barrier of compound 1, we
prepared compound 2 where the m-tolyl is replaced by a more
hindered 2,3-dimethylphenyl ring. The presence of a methyl in
the ortho position raises the barrier of the B-Aryl rotation, while
the rotational barrier of the p-nitrobenzyl group should not be
modified at a great extent because the 2,3-dimethylphenyl ring
is parallel to the p-nitrophenyl ring in the transition state. DFT
The experimental data (Table 1) underline striking differ-
ences in the rotational barriers of a standard biaryl compound
2
with respect to that of a B−Csp axis. The m-tolyl rotational
barrier of 4 is 4.1 kcal/mol higher with respect to 1, and that of
B
DOI: 10.1021/acs.orglett.6b01159
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
Table 1. Experimental Energy Barriers for B-Aryl and N-(p-
Nitrobenzyl) Rotations (Energies in kcal/mol)
compd
B-Aryl
N-Benzyl
compd
C-Aryl
C-Benzyl
1
2
3
6.6
10.7
10.0
−
4
5
6
10.7
25.4
>40
8.0
7.6
−
19.1
33.0
5 is 6.3 kcal/mol higher than that of 2. In contrast, the N-(p-
2
nitrobenzyl) rotation has a higher barrier than Csp −benzyl
rotation. In the case of the B-Aryl axis this outcome could be
explained by taking into account the bond distances. The
experimental B-Aryl bond length is 1.58 Å while that of a
standard biaryl is 1.49 Å.15 The longer distance allows better
accommodation of the rotating aryl ring in the transition state.
A confirmation of this hypothesis is that the difference raises
when the energy barrier increases. If a conjugative stabilizing
effect were present in the transition state, the rotational barrier
should have the same difference upon raising the steric
hindrance of the rotating ring. The same approach based on
bond lengths can rationalize the different barriers observed for
N-(p-nitrobenzyl) rotation. The N−CH2 bond distance (1.46
Å) is shorter than C−CH2 in similar compounds (1.51 Å);16
thus, compounds 1−2 exhibit N-(p-nitrobenzyl) rotational
barriers higher than compounds 4−5.
Albeit rather high, the B-Aryl rotational barrier of compound
2 does not allow for a physical resolution of the atropisomeric
pair. To this aim we prepared compound 3 bearing the 2-
methylnaphthalene moiety. DFT calculations suggest an
enantiomerization barrier of about 31 kcal/mol for this
compound (Figure S11). Taking into account the errors of
calculations observed for compounds 1 and 2, the experimental
barrier should be in any case larger than 25−26 kcal/mol, thus
making feasible the resolution of the atropisomeric pair.
Due to the steric hindrance, compound 3 had to be prepared
by means of Dewar’s procedure4 followed by the addition of 2-
methylnaphthylmagnesium bromide (Scheme 1). This inter-
Figure 4. (Top) CSP-HPLC chromatogram of compound 3.
(Bottom) ECD spectra of the two atropisomers of 3, recorded in
acetonitrile.
and +140 °C in C2D2Cl4 and monitoring racemization by
HPLC (see Figure S18).
Having in hand the two separated atropisomers of compound
3, we pursued the determination of their absolute configuration
(AC) by time-dependent DFT (TD-DFT) simulation of the
electronic circular dichroism spectra (ECD).18,19 Due to the
hindered rotation of the N-(p-nitrobenzyl) group, two
conformations have to be considered in the simulation of the
ECD spectra. However, the exact ratio of the two
conformations (50:50) can be easily determined from low-
temperature NMR spectra (Figure S19) and used in the
weighting of their two simulated ECD spectra (Figure 5; see SI
Scheme 1
Figure 5. Comparison of the experimental ECD spectrum (black
trace) of the second eluted atropisomer of 3 with the simulated
spectrum (red trace) calculated at the TD-DFT CAM-B3LYP/6-311+
+G(2d,p) level of theory and assuming the P absolute configuration.
mediate was lithiated with n-BuLi/TMEDA and reacted with p-
nitrobenzaldehyde to obtain compound 7 (see Figures S12−
S16 in the SI for further details). The latter was effectively
reduced to compound 3 with iodotrimethylsilane.17
The atropisomeric pair was resolved by means of
enantioselective HPLC on a Chiralpak AS-H column, with a
large separation between the two atropisomers (Figure 4, top).
Good amounts of the two atropisomers were obtained by
means of a semipreparative approach, and the Electronic
Circular Dichroism (ECD) spectra recorded in acetonitrile
showed the expected opposite traces for the two atropisomers
(Figure 4, bottom). The racemization barrier was determined
as 33.0 kcal/mol by keeping an enantiopure sample at +130 °C
for further details). The theoretical ECD spectra were obtained
using the 6-311++G(2d,p) basis set20 and four different
functionals (CAM-B3LYP,21 BH&HLYP,22 M06-2X,23 and
ωB97-XD24) to allow for redundancy (see SI for further
details). All the simulations assigned the P absolute
configuration to the second eluted atropisomer of 3. Using
the same approach, the absolute configuration of the
atropisomers of compounds 5, 6, and 7 was determined (see
SI for details).25
C
DOI: 10.1021/acs.orglett.6b01159
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
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, O.; Foresman, J. B.; Ortiz, J. V.;
ASSOCIATED CONTENT
* Supporting Information
■
S
̈
The Supporting Information is available free of charge on the
ACS Publications website at DOI: 10.1021/acs.or-
glett.6b01159.
Cioslowski, J.; Fox, D. J. Gaussian, Inc.: Wallingford, CT, 2009.
(14) Casarini, D.; Lunazzi, L.; Mazzanti, A. Eur. J. Org. Chem. 2010,
2010, 2035.
(15) Debeaux, M.; Brandhorst, K.; Jones, P. G.; Hopf, H.;
Grunenberg, V.; Kowalsky, W.; Johannes, H. H. Beilstein J. Org.
Chem. 2009, 5, 31.
Additional Figures S1−S19; detailed procedure for the
absolute configuration determination; experimental de-
tails and spectroscopic data for compounds 1−7 and
intermediates; X-ray data for 1; computational data for
(16) Irngartinger, H.; Fettel, P. W.; Siemund, V. Eur. J. Org. Chem.
1998, 1998, 2079.
13
1
(17) Cain, G. A.; Holler, E. R. Chem. Commun. 2001, 1168.
(18) The reference method to assign the AC relies on the X-ray
anomalous scattering that in the present case would require the use of
a Cu Kα X-ray source. See: Hooft, R. W. W.; Straver, L. H.; Spek, A. L.
J. Appl. Crystallogr. 2008, 41, 96−103.
(19) For reviews, see: (a) Mazzanti, A.; Casarini, D. WIREs Comput.
Mol. Sci. 2012, 2, 613. (b) Pescitelli, G.; Di Bari, L.; Berova, N. Chem.
Soc. Rev. 2011, 40, 4603. (c) Bringmann, G.; Bruhn, T.; Maksimenka,
K.; Hemberger, Y. Eur. J. Org. Chem. 2009, 2009, 2717.
1−7; copies of H, C NMR spectra (PDF)
Crystallographic data for 1 (CIF)
AUTHOR INFORMATION
Corresponding Author
■
*E-mail: michele.mancinelli@unibo.it.
Notes
́
(20) (a) Gunasekaran, P.; Perumal, S.; Carlos Menendez, J.;
The authors declare no competing financial interest.
Mancinelli, M.; Ranieri, S.; Mazzanti, A. J. Org. Chem. 2014, 79,
11039. (b) Ambrogi, M.; Ciogli, A.; Mancinelli, M.; Ranieri, S.;
Mazzanti, A. J. Org. Chem. 2013, 78, 3709.
ACKNOWLEDGMENTS
■
The University of Bologna (RFO Funds 2013 and 2014) is
gratefully acknowledged for financial support. ALCHEMY Fine
Chemicals & Research (Bologna, www.alchemy.it) is acknowl-
edged for a generous gift of chemicals.
(21) Yanai, T.; Tew, D.; Handy, N. Chem. Phys. Lett. 2004, 393, 51.
(22) In Gaussian 09 the BH&HLYP functional has the following
HF
LSDA
Becke88
LYP
form: 0.5*EX + 0.5*EX
+ 0.5*ΔEX
+ EC
.
(23) Zhao, Y.; Truhlar, D. G. Theor. Chem. Acc. 2008, 120, 215.
(24) Chai, J.-D.; Head-Gordon, M. Phys. Chem. Chem. Phys. 2008, 10,
6615.
REFERENCES
■
(25) The reduction of a single stereoisomer of 7b (R,P or S,M)
yielded a racemic mixture of compound 3, suggesting a fast rotation of
the 2-methylnaphthyl group in the radical transition states. See:
Stoner, E. J.; Cothron, D. A.; Balmer, M. K.; Roden, B. A. Tetrahedron
1995, 51, 11043. See also Figure S17 for further details.
(1) (a) Clayden, J.; La Plante, S. Angew. Chem., Int. Ed. 2009, 48,
6398. (b) La Plante, S. ChemMedChem 2011, 6, 505. (c) La Plante, S.
J. Med. Chem. 2011, 54, 7005.
(2) Zask, A.; Murphy, J.; Ellestad, G. A. Chirality 2013, 25, 265.
(3) Dewar, M. J. S.; Kubba, V. P.; Pettit, R. J. Chem. Soc. 1958, 3073.
(4) Dewar, M. J. S.; Dietz, R. J. Chem. Soc. 1959, 2728.
(5) Edel, K.; Brough, S. A.; Lamm, A. N.; Liu, S.-Y.; Bettinger, H. F.
Angew. Chem., Int. Ed. 2015, 54, 7819.
(6) For reviews, see: (a) Campbell, P. G.; Marwitz, A. J.V.; Liu, S.-Y.
Angew. Chem., Int. Ed. 2012, 51, 6074. (b) Bosdet, M. J. D.; Piers, W.
E. Can. J. Chem. 2009, 87, 8. (c) Liu, Z.; Marder, T. B. Angew. Chem.,
Int. Ed. 2008, 47, 242.
(7) Marwitz, A. J. V.; Matus, M. H.; Zakharov, L. N.; Dixon, D. A.;
Liu, S.-Y. Angew. Chem., Int. Ed. 2009, 48, 973.
(8) (a) Wisniewski, S. R.; Guenther, C. L.; Argintaru, O. A.;
Molander, G. A. J. Org. Chem. 2014, 79, 365. (b) Molander, G. A.;
Wisniewski, S. R. J. Org. Chem. 2014, 79, 6663. (c) Molander, G. A.;
Wisniewski, S. R. J. Org. Chem. 2014, 79, 8339. (d) Molander, G. A.;
Wisniewski, S. R.; Davan, E. E. J. Org. Chem. 2014, 79, 11199.
(9) The p-nitrophenyl moiety was used to reduce the crowding of the
1
aromatic region of the H NMR spectrum.
(10) Hatakeyama, T.; Hashimoto, S.; Seki, S.; Nakamura, M. J. Am.
Chem. Soc. 2011, 133, 18614.
(11) CSD search reported 23 hits using the 1,2-azaborine
substructure as the input search.
(12) Braunschweig, H.; Celik, M. A.; Hupp, F.; Krummenacher, I.;
Mailander, L. Angew. Chem., Int. Ed. 2015, 54, 6347.
̈
(13) Gaussian 09 Rev. D.01. 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, J. A., Jr.; Peralta, J. E.; Ogliaro, F.;
Bearpark, M.; Heyd, J. J.; Brothers, E.; Kudin, K. N.; Staroverov, V. N.;
Kobayashi, R.; Normand, J.; Raghavachari, K.; Rendell, A.; Burant, J.
C.; Iyengar, S. S.; Tomasi, J.; Cossi, M.; Rega, N.; Millam, J. M.; 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.;
D
DOI: 10.1021/acs.orglett.6b01159
Org. Lett. XXXX, XXX, XXX−XXX