Anionic multisubstituted 1,2-azaborolyl ligands: Syntheses, characterization, and coordination chemistry

Article abstract of DOI:10.1021/om8001075

Two synthetic methods for the preparation of tri- or tetrasubstituted 1,2-azaborolyls (Ab) are described. Both triand tetrasubstituted Ab rings can be conveniently attained via either dilithiation-directed or Cp₂Zr ^II-mediated cycliz

Full text of DOI:10.1021/om8001075

2
408
Organometallics 2008, 27, 2408–2410
Anionic Multisubstituted 1,2-Azaborolyl Ligands: Syntheses,
Characterization, and Coordination Chemistry
†
Xiangdong Fang* and Jalil Assoud
Department of Chemistry, UniVersity of Waterloo, Waterloo, Ontario, Canada N2L 3G1
ReceiVed February 7, 2008
Summary: Two synthetic methods for the preparation of tri- or
tetrasubstituted 1,2-azaborolyls (Ab) are described. Both tri-
and tetrasubstituted Ab rings can be conVeniently attained Via
converts the Cp ring into an Ab ligand. The B-N unit in
aminoboranes has substantial double-bond character, as evi-
denced by the fact that the rotational barriers about the B-N
bonds in various aminoborane molecules has been determined
by variable-temperature NMR spectroscopy in the range of
II
either dilithiation-directed or Cp2Zr -mediated cyclization fol-
lowed by transmetalation in good yields. In particular, anionic
-1 6
1
,2,4-trimethyl-1,2-azaborolyl and 1,2,3,4-tetramethyl-1,2-
10-24 kcal mol . Taking the much higher rotational barrier
-1 7
azaborolyl, readily prepared through our methods, haVe been
demonstrated as good supporting ancillary ligands in group
IV metal complexes.
of CdC double bonds (∼63 kcal mol ) into consideration,
one can expect that π electrons are more localized in the B-N
π bond than in the CdC double bond, which consequently
contributes to an Ab ring with less π-electron delocalization,
relative to Cp systems. Therefore, the HOMO in the Ab system
has more character of a filled nitrogen atomic orbital, while the
LUMO of the Ab system resembles more the empty 2p orbital
Cyclopentadienyl (Cp) and its derivatives constitute one of
the most important groups of ancillary ligands utilized in
transition-metal complexes, as demonstrated in olefin polym-
1
erization and selective organic synthesis. It has been noted that
8
at boron. In this regard, the Ab ligands distinguish themselves
in many Cp-based half-sandwich and metallocene complexes,
simple modification of Cp-ring substituents may induce sig-
from their Cp analogues with more electron-donating and less
electron-accepting characters.
2
nificant changes in reactivity and selectivity of the metal center.
In fact, Ab ligands have been investigated as the potential
For example, Chirik et al. have demonstrated that in dinitrogen
activation the end-on versus side-on N2 coordination is subtly
influenced by the substitution pattern of the Cp ligand in low-
8–10
replacement ligand for Cp group in various metal complexes,
and it has been observed that Ab ligand is more electron-
9
donating than its Cp rival in metal complexes. However, those
3
valent group IV metallocene chemistry, while the recent
previous studies were rather synthetically limited. For instance,
the scope with respect to the ring substituents afforded by current
5
work by Hou et al. has established that cationic [(η -C5Me4-
+
SiMe3)Sc(CH2SiMe3)(THF)] half-sandwich species exhibit
synthetic methods is almost entirely restricted to B- and
olefin polymerization properties that are distinctly different
8–10
N-substituted Ab ligands.
To the best of our knowledge,
5
from those of a comparable {[η -C5H3(SiMe3)2]Sc(CH2SiMe3)-
the report for C-substituted Ab ligand was extremely sporadic,
as only one such compound has been isolated and characterized
+
4
(
THF)} complex. On the other hand, the desire to modulate
and control the reactivity and selectivity of the metal catalysts
has stimulated the design and syntheses of numerous Cp
11
to date. To address this synthetic challenge, we have estab-
lished that both unprecedented tri- and tetrasubstituted Ab rings
can be conveniently obtained through either dilithiation-directed
5
analogues. In this respect, we have initiated a program
directed toward the development of multisubstituted monoan-
ionic 1,2-azaborolyl (Ab) ligands that closely relate to the
omnipresent Cp counterparts. The application of these Ab
ligands in metal complexes may ultimately offer new
directions in designing new metal catalysts with novel
synthetic properties.
II
or Cp2Zr -mediated ring cyclization followed by B/Sn or B/Zr
transmetalation, respectively.
A simple and effective preparatory method, which provides
multisubstituted Ab ligands on a large synthetic scale, would
greatly facilitate the exploration of the coordination chemistry
of these ligands. We have now extended Schmid’s dilithiation
In principle, the formal replacement of a CdC bond unit in
a Cp ligand with an isoelectronic and isolobal B-N unit
1
2
method to the preparation of trisubstituted Ab ligands 6-8,
as illustrated in Scheme 1. Dilithiation of N-tert-butyl-N-
*
To whom correspondence should be addressed. E-mail: xdfang@
uwaterloo.ca. Tel: 01-519-888-4567, ext. 36229. Fax: 01-519-746-0435.
(6) (a) Imbery, D.; Jaeschke, A.; Friebolin, H. Org. Magn. Reson. 1970,
2, 271. (b) Friebolin, H.; Rensch, R.; Wendel, H. Org. Magn. Reson. 1976,
8, 287.
†
To whom correspondence concerning X-ray crystallographic data should
be addressed.
(
1) (a) Brintzinger, H. H.; Fischer, D.; M u¨ lhaupt, R.; Rieger, B.;
(7) Lange’s Handbook of Chemistry, 13th ed.; Dean, J. A., Ed.; McGraw-
Hill: New York, 1985.
Waymouth, R. M. Angew. Chem., Int. Ed. Engl. 1995, 34, 1143. (b) Titanium
and Zirconium in Organic Synthesis; Marek, I., Ed.; Wiley: New York,
(8) Schmid, G. In ComprehensiVe Heterocyclic Chemistry II; Shinkai,
I., Ed.; Elsevier: Oxford, U.K., 1996; Vol. 3, Chapter 3.17.
(9) (a) Liu, S. Y.; Lo, M. M. C.; Fu, G. C. Angew. Chem., Int. Ed.
2002, 41, 174. (b) Liu, S. Y.; Hills, I. D.; Fu, G. C. Organometallics 2002,
21, 4323.
2
002.
2) (a) Arndt, S.; Okuda, J. Chem. ReV. 2002, 102, 1953. (b) Lin, S.;
Waymouth, R. M. Acc. Chem. Res. 2002, 35, 765.
3) (a) Chirik, P. J. Dalton Trans. 2007, 16. (b) Bernskoetter, W. H.;
(
(
Lobkovsky, E.; Chirik, P. J. Angew. Chem., Int. Ed. 2007, 46, 2858. (c)
Pool, J. A.; Lobkovsky, E.; Chirik, P. J. Nature 2004, 427, 527.
(10) (a) Ashe, A. J., III; Fang, X. D. Org. Lett. 2000, 2, 2089. (b) Ashe,
A. J., III; Hong, Y.; Fang, X. D.; Kampf, J. W. Organometallics 2002, 21,
4578.
(4) (a) Li, X. F.; Baldamus, J.; Hou, Z. M. Angew. Chem., Int. Ed. 2005,
4
1
4, 962. (b) Luo, Y. L.; Baldamus, J.; Hou, Z. M. J. Am. Chem. Soc. 2004,
26, 13910.
(11) Schmid, G.; Zaika, D.; Boese, R. Angew. Chem., Int. Ed. Engl.
1985, 24, 602.
(
5) (a) Gibson, V. C.; Spitzmesser, S. K. Chem. ReV. 2003, 103, 283.
(12) Schulze, J.; Schmid, G. Angew. Chem., Int. Ed. Engl. 1980, 19,
54.
(
b) Bolton, P. D.; Mountford, P. AdV. Synth. Catal. 2005, 347, 355.
1
0.1021/om8001075 CCC: $40.75  2008 American Chemical Society
Publication on Web 04/29/2008

Communications
Organometallics, Vol. 27, No. 11, 2008 2409
Scheme 3
Figure 1. Comparison of B-N and CdC bonds.
Scheme 1. Synthesis via Dilithiation
Interestingly, in the sterically more hindered system 12, the
analogous [1,3]H shift was not observed and only directly
transmetalated product 13 was isolated in 70% yield. Subsequent
deprotonation with KN(SiMe3)2 or LDA gives the expected
ligands 11 and 14, respectively.
II
Scheme 2. Synthesis via Cp
2
Zr Cyclization
(
1)
The versatility of the synthetic methods established above
has been tested in the syntheses of the unprecedented anions
1,2,4-trimethyl-1,2-azaborolyl (20) and 1,2,3,4-tetramethyl-1,2-
1
3
methallylamine with 2 equiv of BuLi followed by addition of
Bu2SnCl2 affords the corresponding stannacycle 1 in 68% yield.
The reaction of 1 with BCl3 in CH2Cl2 gives the B/Sn
transmetalated product 2 in 73% yield, which serves as a general
precursor for various trisubstituted Ab rings 3-5. Deprotonation
of compounds 3-5 with KN(SiMe3)2 or LDA furnishes the
desired trisubstituted Ab ligands 6-8.
azaborolyl (23) (Scheme 3), which are both isoelectronic and
isostructural with 1,2,4-trimethylcyclopentadienyl and tetram-
ethylcyclopentadienyl ligands, respectively. In this respect,
compound 15 is obtained in 60% yield by the sequential reaction
of N-methyl-N-methallylamine with 2 equiv of BuLi and
Bu2SnCl2. After BCl3 transmetalation with 15, 16 can be isolated
by careful vacuum distillation, which quickly crystallizes at -78
°C. However, the characterization of 16 proves to be extremely
difficult, due to its fast decomposition at room temperature.
Thus, colorless crystalline 16 is dissolved in ether at -78 °C
and subsequently treated with various nucleophiles to afford
17-19, which can be deprotonated by LDA to give trisubstituted
Ab ligands 20-22. In particular, 20 can be further methylated
and deprotonated to form the tetramethylated Ab ligand 23 in
50% yield.
Unfortunately, attempts to apply the above dilithiation
protocol in terminally methylated allylamine systems proved
14
to be unsuccessful (eq 1). This prompted us to further explore
other synthetic routes for tri- and tetrasubstituted Ab systems
such as 11 and 14. Gratifyingly, we found that B/Zr transmeta-
lation of the 1,2-azazircona-4-cyclopentenes 9 and 12, readily
II
available from Cp2Zr -mediated cyclization of R,ꢀ-unsaturated
1
5
imine molecules, generally produced the conjugate acids 10
and 13 in good yields, as shown in Scheme 2. Complex 9
transmetalates with PhBCl2 to yield a mixture of 10a and 10b
1
13
11
It is of interest to compare the H, C, and B NMR
chemical shift values for lithium tri- and tetramethylated Ab
1
1
(
10a:10b ) 1:3) after vacuum distillation at 70 °C. The
ligands 20 and 23 (Figure 2). The B NMR chemical shift
values of 20 and 23 are almost identical (20, δ 26.8; 23, δ 26.7),
which indicates that there is significant π donation of negative
isomerization between 10a and 10b is likely to be the result of
a [1,3]H shift, because the controlled NMR experiment indicates
that the direct B/Zr transmetalation unambiguously gives 10a
as the sole product. Thus, the formation of 10b must occur
during the step of vacuum distillation at elevated temperature.
9
13
charge to boron in both cases. In C NMR spectra, the signals
(20, δ 86.7; 23, δ 96.5) for the intra-ring C groups which are
R to boron are shifted much more upfield, relative to those of
other intra-ring carbons (δ 104.6-122.5), suggesting a major
negative charge resting at the ring C(3). Interestingly, the
introduction of a methyl group at C(3) in 23 seems to further
delocalize the negative charge between C(3) and C(5).
(
(
13) Denmark, S. E.; Amburgey, J. J. Am. Chem. Soc. 1993, 115, 10386.
14) (a) Jacobson, M.; Keresztes, I.; Williard, P. G. J. Am. Chem. Soc.
2
005, 127, 4965. (b) Burns, S. A.; Corriu, R. J. P.; Huynh, V.; Moreau,
J. J. E. J. Organomet. Chem. 1987, 333, 281.
15) Davis, J. M.; Whitby, R. J.; Jaxa-Chamiec, A. J. Chem. Soc., Chem.
Commun. 1991, 1743.
The functionalization of activated group IV transition-metal
dinitrogen complexes with nonpolar reagents such as dihydrogen
(

2
410 Organometallics, Vol. 27, No. 11, 2008
Communications
1
13
11
Figure 2. Comparison of H, C (in parentheses), and B NMR
in circles) chemical shift values (in ppm) of lithium salts 20 and
3 in THF-d at 20 °C.
(
2
8
is of great interest, because it provides an attractive approach
to convert atmospheric N2 into more value-added N-containing
1
6
organic molecules under ambient conditions. In particular,
Chirik’s hafnocene-N2 complex based on a tetramethylcyclo-
pentadienyl ligand has been observed to have exclusively side-
on coordination with a more elongated N-N bond, which can
3b
promote the direct coupling with CO2. Therefore, our synthetic
studies have been extended to prepare the hafnocene congener
with Ab ligand. Reaction of (C5Me4H)HfCl3 with 1 equiv of
the lithium salt 20 in refluxing toluene for 4 days and subsequent
recrystallization in hexane furnished the desired Ab complex
Figure 3. ORTEP view of the hafnium complex 24. Hydrogen
atoms are omitted for clarity; thermal ellipsoids are drawn at the
30% probability level. Selected bond lengths (Å) and angles (deg):
Hf1-C1 ) 2.480(2), Hf1-C2 ) 2.506(2), Hf1-C3 ) 2.464(2),
Hf1-B1 ) 2.654(3), Hf1-N1 ) 2.465(2), Hf1-C7 ) 2.445(2),
Hf1-C8 ) 2.501(2), Hf1-C9 ) 2.545(2), B1-N1 ) 1.473(3),
C1-C2 ) 1.443(3), C2-C3 ) 1.380(3), B1-C1 ) 1.503(3),
N1-C3 ) 1.402(3), B1-C4 ) 1.574(4), C7-C8 ) 1.417(3),
C8-C9 ) 1.424(3), C9-C10 ) 1.419(3); ∠ Cl1-Hf1-Cl2 )
1
6c
2
4 as colorless block crystals in 65% yield (Scheme 3).
The molecular structure of 24 is illustrated along with selected
bond lengths and bond angles in Figure 3. The Hf center in
complex 24 adopts a pseudotetrahedral coordination geometry,
in which the Hf atom becomes more tightly bound to N and C
atoms (2.46-2.51 Å) but is loosely coordinated to B (2.65 Å)
in the Ab ligand. Thus, the Ab coordination is more shifted
9
3.0(1), ∠ C1-B1-C4 ) 133.5(2), ∠ N1-B1-C1 ) 101.8(2),
N1-B1-C4)124.7(2),∠ B1-N1-C3)110.3(2),∠ C3-N1-C5
∠
)
120.9(2), ∠ B1-N1-C5 ) 127.8(2).
4
toward η . Perhaps the most striking structural feature of 24 is
that the trimethyl Ab ligand possesses two possible different
cavities for the wedge of hafnocene complex 24, as demon-
strated in Figure 4. This type of ligand defect in Chirik’s
bis(tetramethylcyclopentadienyl) metallocene-N2 systems is of
importance, because it promotes side-on coordination of the
dinitrogen molecule by providing suitable space close to the
3
metal center. Interestingly, the defects involving the B and N
atoms in the Ab ligand 20, as measured by the intersecting
angles (in degrees) of two flanking exocyclic σ bonds, are quite
different (150.0(5) and 135.0(5)°) from that observed in the
tetramethylcyclopentadienyl system (142.5(5)°). Thus, it is
tempting to expect that cavity-size tuning may be feasible in
the dinitrogen functionalization with the Ab metallocene
analogues.
Figure 4. Ab ligand defects in complex 24 as measured by the
intersecting angles (in degrees) of two flanking exocyclic σ bonds.
systems. Further studies on the polymerization and copolym-
erization properties of this new family of metal catalysts are in
progress.
In summary, we have developed two synthetic methods for
the preparation of tri- and tetrasubstituted Ab rings, which
provide, for the first time, easy access to a wide array of B-,
N-, and C-substituted Ab ligands. Ligand 20 readily converts
into its hafnium metallocene complex 24. Our study supplies
new strategies for tuning the reactivity and selectivity of metal
catalysts by implementing both electronic and steric controls
in Ab systems. In this regard, we have synthesized a variety of
half-sandwich metal alkyl complexes based on our ligand
Acknowledgment. We thank the National Science and
Engineering Research Council (NSERC) and the University
of Waterloo for financial support. Instrumentation funding
by the Canadian Foundation for Innovation (CFI) and the
Ontario Innovation Trust (OIT) is also acknowledged. We
are also grateful to Dr. Yuanfu Deng for technical support.
Supporting Information Available: Text giving experimental
details and characterization data for all compounds and a CIF file
giving X-ray crystallographic data for complex 24. This material
is available free of charge via the Internet at http://pubs.acs.org.
(
16) (a) Fryzuk, M. D.; Love, J. B.; Rettig, S. J.; Young, V. G. Science
1
997, 275, 1445. (b) Bernskoetter, W. H.; Lobkovsky, E.; Chirik, P. J. J. Am.
Chem. Soc. 2005, 127, 14051. (c) Hirotsu, M.; Fontaine, P. P.; Zavalij,
P. Y.; Sita, L. R. J. Am. Chem. Soc. 2007, 129, 12690.
OM8001075