Fine-tuning of boron complexes with cage-shaped ligand geometry: Rational design of triphenolic ligand as a template for structure control

Article abstract of DOI:10.1021/ol7030944

Boron complexes surrounded with organic cages were controlled precisely by a remote atom placed at the bottom of the cage. A replacement of the bottom tether atom (carbon or silicon) changed the characteristics (kinetic and thermodynamic factors) of boron complexes by geometric effects. A theoretical study shows that the bottom atoms also control eigenvalues of MO. This cage complex will provide a systematic template for fine-tuning of metal complexes to create various properties.

Full text of DOI:10.1021/ol7030944

ORGANIC
LETTERS
2008
Vol. 10, No. 5
929-932
Fine-Tuning of Boron Complexes with
Cage-Shaped Ligand Geometry:
Rational Design of Triphenolic Ligand
as a Template for Structure Control
Makoto Yasuda,*,† Sachiko Yoshioka,† Hideto Nakajima,† Kouji Chiba,‡ and
Akio Baba*,†
Department of Applied Chemistry and Center for Atomic and Molecular Technologies,
Graduate School of Engineering, Osaka UniVersity, 2-1 Yamadaoka, Suita,
Osaka 565-0871, Japan, and Science and Technology System DiVision,
Ryoka Systems Inc., 1-28-38 Shinkawa, Chuo-ku, Tokyo 104-0033, Japan
yasuda@chem.eng.osaka-u.ac.jp; baba@chem.eng.osaka-u.ac.jp
Received December 23, 2007
ABSTRACT
Boron complexes surrounded with organic cages were controlled precisely by a remote atom placed at the bottom of the cage. A replacement
of the bottom tether atom (carbon or silicon) changed the characteristics (kinetic and thermodynamic factors) of boron complexes by geometric
effects. A theoretical study shows that the bottom atoms also control eigenvalues of MO. This cage complex will provide a systematic template
for fine-tuning of metal complexes to create various properties.
The properties of metal complexes undoubtedly depend on
the metal centers and ligands. Numerous studies have been
performed to alter various properties of the complexes by
varying the metal centers and ligands. However, replacing a
metal center and/or its ligands often causes changes in the
properties of the metal-ligand complex that are too large.
Changing the geometry around the metal can alter the
properties of metal complexes, but this approach may cause
relatively large alterations in the metal complex properties.1
Character changes of metal complexes have been extensively
studied when the metals or ligands are replaced; however,
no attention has been focused on systematic study of fine-
tuning of metal complexes in spite of the importance of finely
tuned catalysts or materials. One reason for the undeveloped
situation is the lack of suitable templates to investigate. In
this context, the purpose of this study was to create a new
ligand system that changes neither chemical connectivity nor
geometry around the metal but instead alters the steric
structure of the ligand apart from the metal. Recently, our
research group reported cage-shaped borate esters, such as
B(OC6H4)3CH (1B), that have higher Lewis acidity and
catalytic activity than the ordinal open-shaped borate ester
B(OPh)3 (2).2,3 The cage shape forces a larger dihedral angle
(C-O-B-O) than the open-shaped borate ester.2 The
characters of open- and cage-shaped borates still have large
† Osaka University.
‡ Ryoka Systems Inc.
(1) Some examples are shown: (a) Denmark, S. E.; Griedel, B. D.; Coe,
D. M.; Schnute, M. E. J. Am. Chem. Soc. 1994, 116, 7026-7043. (b) Nelson,
S. G.; Kim, B.-K.; Peelen, T. J. J. Am. Chem. Soc. 2000, 122, 9318-9319.
(c) Kobayashi, J.; Kawaguchi, K.; Kawashima, T. J. Am. Chem. Soc. 2004,
126, 16318-16319. (d) Wagner, C. E.; Kim, J.-S.; Shea, K. J. J. Am. Chem.
Soc. 2003, 125, 12179-12195.
(2) Although we previously proposed the concept to explain the different
characters between open- and cage-shaped borates, theoretical discussion
had not been employed: Yasuda, M.; Yoshioka, S.; Yamasaki, S.; Somyo,
T.; Chiba, K.; Baba, A. Org. Lett. 2006, 8, 761-764.
10.1021/ol7030944 CCC: $40.75
© 2008 American Chemical Society
Published on Web 01/26/2008

differences between them and are not finely tuned despite
bearing the same planar geometries around the boron centers.
However, the cage-shaped borates led to us to consider the
cage-ligand as a suitable template for investigating a
systematic fine-tuning system. Accordingly, a sila derivative
of the cage-shaped borate, B(OC6H4)3SiR (3B), with a larger
tether introduced into the cage, was thought to be worthwhile
to examine. We report here on a systematic fine-tuning
protocol of metal complexes with precisely designed ligands
that does not directly affect on the boron center but control
the character by slight change of the ligand structure. A
gradual and useful character change has been accomplished
using the triphenolic cage-ligand system with different
tethers.
Table 1. Theoretical Calculation of 3B, 1B, and 2
∆E in
pyridine
com-
dihedral
angle θ
(C-O-B-O) eigenvalue plexation
entry
borate
(deg)
(kcal/mol) (kcal/mol)
1a
2
B(OC6H4)3CH
1B
48.4
45.4
43.6
2.0
-18.26
-16.81
-16.46
-12.46
-19.2
-13.2
-13.1
-5.0
B(OC6H4)3SiMe 3aB
3
B(OC6H4)3SiiBu 3bB
B(OPh)3
4a
2
a Reference 2.
result, the energy levels of the lowest unoccupied MO4
concerning Lewis acids also were intermediate to the values
obtained for 1B and 2, as follows: 1B < 3aB < 3bB < 2.
The MO diagrams show accessible lobes on the borons,
which are suitable for accepting a Lewis base (Scheme 2).
The pyridine-complexation energies, ∆E, also showed the
order 1B < 3aB < 3bB < 2. The theoretically calculated
results prompted the comparison of silicon-tethered cage-
shaped borate esters with carbon-tethered esters because fine-
tuning of the Lewis acidity of the boron complexes may be
possible by ligand structure control.
The structural change of the cage-ligand is related to
dihedral angle of (C-O-B-O) that determines the character
of metal complexes.2 Therefore, B(OH)3 was chosen as a
model compound to perceive any change of its character:
The relationship between the dihedral angle (H-O-B-O)
θ and the lowest unoccupied MO contributing to the Lewis
acid was theoretically calculated,4 keeping connectivity and
geometry around the boron center (planar structure in sp2
hybridization) constant as shown in Scheme 1. The angle (θ
Scheme 2. Optimum Structures and MO Diagrams of 3aB/3bB
Scheme 1. Relationship between Dihedral Angle θ and Energy
Level of the Lowest Unoccupied MO Concerning to Lewis Acid
) 0°) gives the highest MO energy level because of effective
conjugation between the p-orbitals on O and B. Gradual
change in the MO level can be realized by varying θ, even
with three B-O bonds, keeping the structure in-plane.2 This
result shows our concept is promising for precise control.
Next, the structures of two types of 3B (R ) Me, 3aB; R
An interesting theoretical result in Table 1 and Scheme 2
prompted us to synthesize triphenolic silanes 3H3 (R ) Me,
i
3aH3; R ) Bu, 3bH3) which are precursors of sila-cage-
shaped complexes 3B. Although we obtained (o-MeOC6H4)3-
SiMe by the reaction of o-lithioanisol with MeSiCl3 accord-
ing to our previously reported procedure for 1B,2 its
deprotection by BBr3 failed and gave the undesired product
owing to weak Si-aryl bonds. Among many protecting
groups examined, dimethylcarbamate worked well for 3aH3
as shown in Scheme 3. Protection of 2-bromophenol 4 and
its subsequent lithiation5 followed by treatment with MeSiCl3
gave the triarylmethylsilane 6a. Deprotection of 6a by
LiAlH4 effectively afforded 3aH3 without cleavage of Si-
aryl bonds. Once the general and reliable synthetic procedure
for 3H3 was established, the derivatives were obtained to
i
) Bu, 3bB) were estimated from theoretical calculations
(Table 1 and Scheme 2). The boron had three oxygens in-
plane (sum of the three OBO angles: 3aB, 360.0°; 3bB,
359.9°). The dihedral angles (C-O-B-O) θ were 45.4° and
43.6° for 3aB and 3bB, respectively. These values were
intermediate to those of carbon-tethered cage-shaped borate
1B (48.4°) and the open-shaped configuration 2 (2.0°).2 3aB
had a slightly larger angle than 3bB. As expected from this
(3) Similar cage-shaped compounds were reported as follows. As a Lewis
base: Dinger, M. B.; Scott, M. J. Inorg. Chem. 2001, 40, 856-864. As a
titanium complex: Akagi, F.; Matsuo, T.; Kawaguchi, H. J. Am. Chem.
Soc. 2005, 127, 11936-11937.
(4) The described MOs are the lowest unoccupied orbitals with suitable
lobes on boron as a Lewis acid.
(5) Godard, A.; Robin, Y.; Queguiner, G. J. Organomet. Chem. 1987,
336, 1-12.
930
Org. Lett., Vol. 10, No. 5, 2008

the C-aryl bonds (average 1.519 Å) in 1B‚Py. It is worth
noting that the silicon in 3aB‚Py has an almost tetrahedral
structure (sum of Ar-Si-Ar; 330.3°), while the carbon in
1B‚Py has a distorted structure with the sum of the bond
Scheme 3. Triphenolic Silane 3H3 and Cage-Shaped Borate 3B
i
angles equal to 343.3°. The BuSi-tethered borate 3bB‚Py
was also analyzed by X-ray crystallography (Figure 1b). It
had a framework similar to that of 3aB‚Py. The sums of the
bond angles for 3bB‚Py were as follows: O-B-O, 342.5°;
N-B-O, 312.6°; Ar-Si-Ar, 328.0°. The average Si-aryl
bond length was 1.879 Å.
Complexation of the borate esters with 2,6-dimethyl
γ-pyron 8 was examined by NMR to estimate their Lewis
acidity.9 A chemical shift δ(13C) of C3 in 8 clearly shows
the degree of Lewis acidity (Table 2). For comparison with
finely tune the metal complexes. In fact, synthesis of 3bH3
was successful in the similar procedure. The reaction of 3aH3
with BH3‚THF gave the desired cage-shaped silicon-tethered
borate 3aB as a THF complex, which was confirmed by
NMR spectra.6,7 The pyridine complex 3aB‚Py was obtained
by ligand-exchange. NMR of 3aB‚Py showed a characteristic
shift in the Me group on Si. A downfield-shift of δ(1H) of
the Me group (0.95 f 1.08 ppm) and an upfield-shift of
δ(13C) (-3.0 f -6.4 ppm) were confirmed relative to the
spectra of the original organic ligand 3aH3. A 29Si NMR
shift was observed at the upper field (-18.2 f -21.1 ppm).
The pyridine signals, which appeared in lower fields relative
to those of free pyridine, indicated strong coordination of
the pyridine to the boron in 3aB. Analogous results for
synthesis and spectral analyses were obtained for 3bB.
The pyridine-complex 3aB‚Py was analyzed by X-ray
crystallography (Figure 1a).8 Boron has a distorted tetrahedral
Table 2. Complexation of Boron Compounds with Carbonyl
Compound 8
entry
B compound
∆δ(13C) of C3/ppm
1
2
3
4
5
1B‚THF
3aB‚THF
3bB‚THF
2
6.8
5.8
4.5
2.5
8.7
BF3‚OEt2
other Lewis acids, the planar borate B(OPh)3 2 and a typical
strong Lewis acid BF3‚OEt2 were employed. In fact, the
complexation of 8 with BF3‚OEt2 showed the largest down-
field shift (∆δ(13C) ) 8.7 ppm) in C3. Interestingly, sila-
cage compounds 3aB and 3bB were shifted downfield by
5.8 and 4.5 ppm, respectively, an effect intermediate to those
of 1B and 2. This result is consistent with the Lewis acidity
expected based on the theoretical results in Table 1.
The carbonyl complexes of 1B and 3aB with 8 were
isolated and analyzed by X-ray crystallography (Figure 2).8
Figure 1. ORTEP drawings of (a) 3aB‚Py and (b) 3bB‚Py (some
hydrogens are omitted for clarity).
coordination sphere with the sum of the bond angles as
follows: O-B-O, 343.1° and N-B-O, 311.8° (1B‚Py has
the corresponding angles2 as follows: O-B-O, 342.9° and
N-B-O, 312.2°). The geometries around B in 3aB‚Py and
1B‚Py were nearly identical. The larger silicon tether results
in longer Si-aryl bonds (average 1.871 Å) in 3aB‚Py than
(6) Recent review of ligands bearing phenoxy moieties: Matsuo, T.;
Kawaguchi, H. Chem. Lett. 2004, 33, 640-645.
(7) (a) Livant, P. D.; Northcott, J. D.; Shen, Y.; Webb, T. R. J. Org.
Chem. 2004, 69, 6564-6571. (b) Verkerk, U.; Fujita, M.; Dzwiniel, T. L.;
McDonald, R.; Stryker, J. M. J. Am. Chem. Soc. 2002, 124, 9988-9989.
(c) Fujita, M.; Qi, G.; Verkerk, U. H.; Dzwiniel, T. L.; McDonald, R.;
Stryker, J. M. Org. Lett. 2004, 6, 2653-2656.
Figure 2. ORTEP drawings of (a) 1B‚8 and (b) 3aB‚8 (some
hydrogens and solvents are omitted for clarity). Selected bond
lengths (Å): (a) O(4)-C(20) 1.294(8), (b) O(4)-C(20) 1.280(5).
(8) The X-ray crystallographic data of 3aB‚Py, 3bB‚Py, 1B‚8, and 3aB‚
8 are given in the Supporting Information.
Org. Lett., Vol. 10, No. 5, 2008
931

The solid-state structure of 1B‚8 showed a longer carbonyl
bond than that of 3aB‚8, reflecting the expected Lewis acidity
in the order 1B > 3aB, although the difference was not large.
The precise tuning of Lewis acidity was realized by size-
control of ligand.
The ligand-exchange rate, which is based on the com-
plexation stability and the energy level of the unoccupied
MO discussed in Table 1, illustrates another aspect of cage-
shaped complexes. The mixture of the complex 1B‚pyridine-
d5 (or 3aB‚pyridine-d5) with unlabeled pyridine was observed
by NMR during ligand-scrambling. Kinetic study of the
mixtures revealed a faster rate of ligand-exchange in the case
of 3aB compared with 1B (see the Supporting Information).
Activation parameters for ligand dissociation were shown
in Table 3. Almost the same entropies ∆Sq for 3aB and 1B
rate constant k (∼106 times larger at 20 °C) was shown for
3aB compared with 1B. The size of the cage is found to be
quite sensitive to ligand-exchange rate. This result suggests
that the precise tuning is very important for design of metal
complex. For Lewis acid catalysts, balance between ther-
modynamic factor (Lewis acidity) and kinetic factor (rate
of ligand-exchange) is important.10 This controlling meth-
odology by cage-size is promising for precise design of the
catalyst.11
In summary, the cage-shaped borates with different tethers
were synthesized and investigated as Lewis acids based on
thermodynamic and kinetic points of view in the present
study. The general synthetic procedure of the cage-ligands
was established. Cage-size/geometry precisely controlled
Lewis acidity. The rate of ligand-exchange unexpectedly
showed relatively large difference. Other elements to be used
as tethers or to functionalize organic cage moieties to control
the properties of metal complexes will be investigated in
future studies.
Table 3. Kinetic Parameters for Ligand (Pyridine) Dissociation
of Cage Borates 1B and 3aB
Acknowledgment. This work was supported by a Grant-
in-Aid for Scientific Research (Nos. 18065015, 19027038,
19550038) from MEXT. M.Y. acknowledges financial sup-
port from the Tokuyama Science Foundation, the Mitsubishi
Chemical Corporation Fund, and the Kansai Research
Foundation for Technology Promotion (KRF).
∆Hq
∆Sq
∆Gq,a
entry Z-R
(kcal/mol) (cal/K‚mol) (kcal/mol)
ka (s-1
)
Supporting Information Available: Experimental pro-
cedures, listing of absolute energies and geometries for
calculated species, and X-ray data for 3aB‚Py, 3bB‚Py, 1B‚
8, and 3aB‚8 (CIF). This material is available free of charge
via the Internet at http://pubs.acs.org.
1
2
C-H 1B
Si-Me 3aB
26.6
20.9
12.8
12.7
22.8
17.1
5.84 × 10-5
1.01 × 101
a ∆Gq and k are calculated at t ) 20 °C.
OL7030944
indicated similar transition states in both cases. The free
energy of activation ∆Gq mainly depended on ∆Hq. The
smaller ∆Hq of 3aB than 1B can be expected from the result
of the ∆E values in Table 1. Surprisingly, a much higher
(10) (a) Yasuda, M.; Yamasaki, S.; Onishi, Y.; Baba, A. J. Am. Chem.
Soc. 2004, 126, 7186-7187. (b) Yasuda, M.; Saito, T.; Ueba, M.; Baba,
A. Angew. Chem., Int. Ed. 2004, 43, 1414-1416. (c) Onishi, Y.; Ogawa,
D.; Yasuda, M.; Baba, A. J. Am. Chem. Soc. 2002, 124, 13690-13691.
(11) We preliminary examined a catalytic hetero-Diels-Alder reaction.
The use of the silicon-tethered borate 3aB‚THF gave higher yield (85%)
than carbon-tethered borate 1B‚THF (77% ref 2). We are now investigating
catalytic activities of them based on reaction rates.
(9) Although pyridine shifts were observed, the anisotropic effect due
to phenyl rings would confuse the estimates by NMR shifts.
932
Org. Lett., Vol. 10, No. 5, 2008