1,2-Azaborolyls, isoelectronic analogues of the ubiquitous cyclopentadienyl ligand: Synthesis of B-heteroatom-substituted 1,2-azaborolyl complexes and an assessment of their electronic features

Article abstract of DOI:10.1002/1521-3773(20020104)41:1<174::AID-ANIE174>3.0.CO;2-7

The substituent on boron allows the electronic nature of the heterocycle of B-heteroatom-substituted (R = H, O, S, N, P, and F) 1,2-azaborolyl complexes to be modulated. Given that isoelectronic with cyclopentadienyl (see scheme), one of the most widely u

Full text of DOI:10.1002/1521-3773(20020104)41:1<174::AID-ANIE174>3.0.CO;2-7

COMMUNICATIONS
single precursor, we can synthesize azaborolyl complexes that
bear a wide array of substituents (hydrogen, carbon, nitrogen,
oxygen, fluorine, phosphorus, and sulfur). We have structur-
1,2-Azaborolyls, Isoelectronic Analogues of the
Ubiquitous Cyclopentadienyl Ligand: Synthesis
of B-Heteroatom-Substituted 1,2-Azaborolyl
Complexes and an Assessment of Their
Electronic Features**
À
ally characterized a B OR adduct, and through electro-
chemical studies we have established that the group on boron
exerts a significant impact on the electronic nature of the
metal complex.
Shih-Yuan Liu, Michael M.-C. Lo,* and Gregory C. Fu*
In our initial investigation, we chose to focus on the
synthesis of azaborolyl iron complexes, thereby allowing
direct comparison with much-studied, isoelectronic ferroce-
nes.[1b] Transmetalation of the previously reported stannacycle
1[13] with BCl3 affords the B-chloroboracycle 2, which is then
complexed to iron (Scheme 2).[14] Treatment of this h5-(1,2-
azaborolyl) adduct 3 with TlCp[15] furnishes ferrocene ana-
logue 4, the chloride of which is then abstracted by AgOTf
(OTf ˆ OSO2CF3) to provide the more reactive triflate
complex 5.[16]
The cyclopentadienyl group is one of the most widely used
ligands in organometallic chemistry, and metal complexes that
bear cyclopentadienyl (Cp) ligands have been applied across a
broad spectrum of fields.[1] One particularly noteworthy use of
cyclopentadienyl complexes (e.g., zirconocene- and titano-
cene-based systems) is as catalysts for Ziegler Natta poly-
merizations of olefins.[2]
The desire to modulate the reactivity of Ziegler Natta
catalysts has led to growing interest in the development of
variations of Cp-based Group 4 metallocenes.[2, 3] For exam-
ple, a number of recent studies have pursued the use of boron-
based heterocycles as alternatives to cyclopentadienyl.[4]
Particularly noteworthy are the investigations of Bazan and
Ashe, who have shown that boratabenzene zirconium com-
plexes can furnish reactivity distinct from cyclopentadienyl-
zirconium complexes and that the electronic nature of the
t Bu
N
t Bu
B
OTf
N
SnMe2
Fe
1
5
8]
boron substituent dictates the catalyst×s course of action.[5
1,2-Azaborolyls are isoelectronic with cyclopentadienyls
(Scheme 1). As with boratabenzenes,[9] the boron of azabor-
olyls provides a potentially straightforward point of attach-
ment for substituents that can modulate the electronic nature
of the boron heterocycle.[10]
BCl3
AgOTf
t Bu
t Bu
N
1) Fe(CO)5
2) I2
N
t Bu
B Cl
B
Cl
TlCp
N
Cl
Fe
B
Fe
OC
OC
I
4
2
3
Scheme 2. Synthesis of B-heteroatom-substituted h5-(1,2-azaborolyl) com-
plexes.
N
B
B
cyclopentadienyl
1,2-azaborolyl
boratabenzene
Complex 5 reacts with anionic nucleophiles to produce a
wide array of B-substituted adducts 6 in good to excellent
yields (Table 1). Organometallic reagents (Table 1, entries 1
and 2), hydride (Table 1, entry 3),[17] alkoxides and thiolates
Scheme 1. Cyclopentadienyl and related ligands.
Surprisingly, 1,2-azaborolyls have not been widely inves-
tigated. Nearly all of the work to date is due to Schmid, whose
pioneering studies of azaborolyls were initiated two decades
ago.[11, 12] Only two substituents on boron, both of which are
carbon-based (methyl and phenyl), have been described.
In 1998, we initiated a program directed at expanding the
diversity of accessible 1,2-azaborolyls, with a particular focus
on the boron substituent. Herein, we demonstrate that, from a
Table 1. Synthesis of
complexes by nucleophilic substitution.
a diverse array of B-substituted 1,2-azaborolyl
t Bu
t Bu
N
N
B
OTf
M
Nu
B
Nu
Fe
Fe
5
6
[*] Prof. Dr. G. C. Fu, S.-Y. Liu, M. M.-C. Lo[‡]
Department of Chemistry
Entry
Yield [%][a]
À
M Nu
Massachusetts Institute of Technology
Cambridge, Massachusetts 02139 (USA)
Fax : (‡1)617-258-7500
1
2
Li-nBu
MgBr
84
88
3
4
5
6
LiAlH4
91
83[b]
89
85
75
E-mail: gcf@mit.edu
‡
Na-OMe
Na-SBn
Li-NMe2
[ ] X-ray crystal structure analysis.
[**] We thank Prof. Timothy M. Swager, Dr. Takeshi Shioya, and J. D.
Tovar for assistance with the electrochemistry. Support has been
provided by Bristol-Myers Squibb, Merck, Novartis, and Pfizer.
7K-PPh
8
2
K-F
87
Supporting information for this article is available on the WWW under
http://www.angewandte.com or from the author.
[a] Yield of isolated product (average of two runs). [b] 95% purity.
174
¹ WILEY-VCH Verlag GmbH, 69451 Weinheim, Germany, 2002
1433-7851/02/4101-0174 $ 17.50+.50/0
Angew. Chem. Int. Ed. 2002, 41, No. 1

COMMUNICATIONS
(Table 1, entries 4 and 5), amides and phosphides (Table 1,
entries 6 and 7), as well as fluoride (Table 1, entry 8) all
cleanly displace the triflate, presumably via an addition eli-
mination pathway. The syntheses illustrated in Table 1 and
Scheme 2 describe the first examples of heteroatom-substi-
tuted 1,2-azaborolyls.
To gain insight into how the boron substituent affects the
electronic nature of the iron, we have measured the oxidation
potential of these new azaborolyl complexes (Table 2).[18] As
expected, the NMe2 and OMe groups are the best donors
among those that we have examined (Table 2, entries 1 and 2).
The nBu, SBn, F, and allyl substituents appear to be modestly
electron-donating (Table 2, entries 3 6) relative to H (Ta-
ble 2, entry 7), whereas PPh2 is electron-withdrawing (Table 2,
entry 8).
geometry consistent with p bond-
ing between oxygen and boron
(Figure 2).[21]
Finally, we have determined the
impact that replacing two carbons
of a cyclopentadienyl ligand with
the corresponding isoelectronic
À
B N unit has on a metal. Thus,
electrochemistry indicates that 1,2-
azaborolyl is somewhat more elec-
tron-donating than cyclopentadien-
yl (Scheme 3).[22]
Figure 2. Molecular struc-
ture of azaborolyl complex
6 (Nu ˆ OMe) (ORTEP il-
lustration, with thermal el-
lipsoids drawn at the 35%
probability level).
In summary, we have developed
a synthetic route that provides
access to a diverse array of the first
B-heteroatom-substituted (H, N,
O, F, P, S, and Cl) 1,2-azaborolyl complexes. In addition, we
have established that the substituent on boron can modulate
the reactivity of the complexes, specifically, their susceptibility
to one-electron oxidation. Furthermore, we have determined
Table 2. Oxidation potential of 1,2-azaborolyliron complexes as a function
of the substituent on boron (0.0026m; 0.10m Bu4NPF6; CH2Cl2; 20 mVsÀ1
;
‡
‡
potentials relative to Ag/Ag with E1/2 ˆ 0.23 V for Fc/Fc ).
Entry
Boron substituent
Epa [V][a]
Nu on 6
t Bu
C
t Bu
N
1
2
3
4
NMe2
OMe
nBu
SBn
F
À 0.26
À 0.08
0.07
0.08
0.08
C
H
B H
Fe
Fe
5
6
allyl
0.13
E1/2 = 0.19 V
E1/2 = 0.14 V
7H
8
0.20
PPh2
0.29
Scheme 3. Direct electrochemical comparison between a 1,2-azaborolyl
and a cyclopentadienyl complex (0.0026m; 0.10m Bu4NPF6; CH2Cl2;
[a] Epa ˆ anodic peak potential.
‡
‡
20 mVsÀ1; potentials relative to Ag/Ag with E1/2 ˆ 0.23 V for Fc/Fc ).
Using the data in Table 2 and a two-parameter Hammett
analysis (sI ˆ inductive component; sR ˆ resonance compo-
nent),[19, 20] we have determined an excellent correlation
between observed and calculated oxidation potentials for
that a 1,2-azaborolyl is more electron-rich than an isostruc-
tural cyclopentadienyl ligand. In view of the ubiquity and the
utility of cyclopentadienyl metal complexes, we anticipate
that the observations described in this study will stimulate the
development of applications of h5-(1,2-azaborolyl) ligands in
metal-catalyzed processes.
these substituted 1,2-azaborolyl complexes (Figure 1; Epa
0.42sI ‡ 0.94sR ‡ 0.20).
ˆ
We have confirmed our structural assignment for azabor-
olyl complexes 6 through an X-ray crystallographic study of
the B-OMe adduct (Figure 2). As expected on the basis of our
electrochemical investigations, the OMe group adopts a
Received: September 7, 2001 [Z17871]
[1] a) Metallocenes, Vol. 1 2 (Eds.: A. Togni, R. L. Halterman), Wiley,
New York, 1998; b) Ferrocenes (Eds.: A. Togni, T. Hayashi), VCH,
New York, 1995.
[2] For recent reviews, see: a) Metallocene-Based Polyolefins (Eds.: J.
Scheirs, W. Kaminsky), Wiley, New York, 2000 b) W. Kaminsky, M.
Arndt in Applied Homogeneous Catalysis with Organometallic Com-
pounds (Eds: B. Cornils, W. A. Hermann), VCH, New York, 1996;
c) Ziegler Catalysis (Eds.: G. Fink, R. M¸lhaupt, H. H. Brintzinger),
Springer, Berlin, 1995; d) H. H. Brintzinger, D. Fischer, R. M¸lhaupt,
B. Rieger, R. M. Waymouth, Angew. Chem. 1995, 107, 1255 1283;
Angew. Chem. Int. Ed. Engl. 1995, 34, 1143 1170; Erratum: Angew.
Chem. 1995, 107, 1652; Angew. Chem. Int. Ed. Engl. 1995, 34, 1368.
[3] For a review of the influence of cyclopentadienyl-ring substituents on
Ziegler Natta catalysts based on Group 4 metallocenes, see: P. C.
Mˆhring, N. J. Coville, J. Organomet. Chem. 1994, 479, 1 29.
[4] For example, see: a) A. J. Ashe III, X. Fang, J. W. Kampf, Organo-
metallics 2000, 19, 4935 4937; b) A. J. Ashe III, X. Fang, J. W. Kampf,
Organometallics 1999, 18, 1821 1823.
Figure 1. Observed and calculated (by Hammett analysis) oxidation
potentials for substituted 1,2-azaborolyl complexes (Epa(calcd) ˆ 0.42 sI
‡ 0.94 sR ‡ 0.20).
[5] a) G. C. Bazan, G. Rodriguez, A. J. Ashe III, S. Al-Ahmad, C. M¸ller,
J. Am. Chem. Soc. 1996, 118, 2291 2292; b) G. C. Bazan, G.
Rodriguez, A. J. Ashe III, S. Al-Ahmad, J. W. Kampf, Organometal-
Angew. Chem. Int. Ed. 2002, 41, No. 1
¹ WILEY-VCH Verlag GmbH, 69451 Weinheim, Germany, 2002
1433-7851/02/4101-0175 $ 17.50+.50/0
175

COMMUNICATIONS
lics 1997, 16, 2492 2494; c) J. S. Rogers, G. C. Bazan, C. K. Sperry, J.
Am. Chem. Soc. 1997, 119, 9305 9306; d) R. A. Lee, R. J. Lachicotte,
G. C. Bazan, J. Am. Chem. Soc. 1998, 120, 6037 6046; e) J. S. Rogers,
R. J. Lachicotte, G. C. Bazan, J. Am. Chem. Soc. 1999, 121, 1288
1298; f) G. C. Bazan, W. D. Cotter, Z. J. A. Komon, R. A. Lee, R. J.
Lachicotte, J. Am. Chem. Soc. 2000, 122, 1371 1380.
Total Synthesis of Ambruticin**
Eun Lee,* Seung Jib Choi, Hahn Kim, Hee Oon Han,
Young Keun Kim, Sun Joon Min, Sung Hee Son,
Sang Min Lim, and Won Suk Jang
[6] For studies of cyclotrimerization reactions catalyzed by cobalt bor-
atabenzene complexes, see: a) H. Bˆnnemann, W. Brijoux, R.
Brinkmann, W. Meurers, Helv. Chim. Acta 1984, 67, 1616 1624;
b) H. Bˆnnemann, Angew. Chem. 1985, 97, 264 279; Angew. Chem.
Int. Ed. Engl. 1985, 24, 248 262.
[7] For pioneering studies of boratabenzene chemistry, see: a) G. E.
Herberich, G. Greiss, H. F. Heil, Angew. Chem. 1970, 82, 838 839;
Angew. Chem. Int. Ed. Engl. 1970, 9, 805 806; b) A. J. Ashe III, P.
Shu, J. Am. Chem. Soc. 1971, 93, 1084 1085.
Ambruticin (1) was isolated from fermentation extracts of
the Myxobacteria species Polyangium cellulosum var. fulvum.
It is an orally active antifungal agent showing in vitro and in
vivo activity against a variety of pathogenic fungi, including
Histoplasma capsulatum, Coccidioides immitis, and Blasto-
myces dermatitides, as well as the dermatophytic filamentous
fungi.[1] Ambruticin features unique cis-2,6-disubstituted
tetrahydropyran and dihydropyran ring systems together with
a methylcyclopropane moiety. In spite of considerable interest
[8] For reviews on boratabenzene chemistry, see: a) G. E. Herberich, H.
Ohst, Adv. Organomet. Chem. 1986, 25, 199 236. b) G. C. Fu, Adv.
Organomet. Chem. 2001, 47, 101 119.
[9] For a versatile synthesis of B-substituted boratabenzenes, see: S. Qiao,
D. A. Hoic, G. C. Fu, J. Am. Chem. Soc. 1996, 118, 6329 6330.
[10] For an application of 1,2-azaborolyl complexes in olefin polymer-
ization, see: S. Nagy, R. Krishnamurti, B. P. Etherton, PCT Int. Appl.
WO 9634021, 1996; [Chem. Abstr. 1997, 126, 19432j].
[11] For overviews of 1,2-azaborolyl chemistry, see: a) G. Schmid in
Comprehensive Heterocyclic Chemistry II, Vol. 3 (Ed.: I. Shinkai),
Elsevier, Oxford, 1996, chap. 3.17; b) G. Schmid, Comments Inorg.
Chem. 1985, 4, 17 32.
O
O
HO2C
OH
OH
1 (+)-Ambruticin
[12] For a recent contribution to 1,2-azaborolyl chemistry, see: A. J.
Ashe III; X. Fang, Org. Lett. 2000, 2, 2089 2091.
[13] Stannacycle 1 can be synthesized in two steps from commercially
available materials: D. H‰nssgen, E. Odenhausen, Chem. Ber. 1979,
112, 2389 2393.
[14] For precedent with B-alkyl- or B-arylboracycles, see: a) J. Schulze, R.
Boese, G. Schmid, Chem. Ber. 1980, 113, 2348 2357; b) J. Schulze, G.
Schmid, J. Organomet. Chem. 1980, 193, 83 91.
[15] For a review, see: A. G. Lee, Organomet. React. 1975, 5, 1 99.
[16] For certain nucleophiles (e.g., LiNMe2 and LiAlH4), not only the B
OTf complex (5; vide infra), but also the B Cl complex (4), serves as
a suitable substrate for displacement reactions at boron.
[17] As Nˆth has noted, boron-containing heteroaromatic compounds that
bear a hydrogen substituent on boron are relatively uncommon: H.
Nˆth, M. Schmidt, Angew. Chem. 1996, 108, 311 312; Angew. Chem.
Int. Ed. Engl. 1996, 35, 292 293.
[18] a) For leading references to electrochemical studies of ferrocene
complexes, see: P. Zanello in Ferrocenes (Eds.: A. Togni, T. Hayashi),
VCH, New York, 1995, chap. 7. b) All of the azaborolyl complexes
depicted in Table 2, except for the F- and PPh2-substituted com-
pounds, display reversible redox behavior.
[19] a) C. Hansch, A. Leo, Substituent Constants for Corrleation Analysis
in Chemistry and Biology, Wiley, New York, 1979; b) C. G. Swain,
E. C. Lupton, Jr., J. Am. Chem. Soc. 1968, 90, 4328 4337; C. G.
Swain, S. H. Unger, N. R. Rosenquist, M. S. Swain, J. Am. Chem. Soc.
1983, 105, 492 502.
in the preparation of 1,[2] we found in the literature only one
total synthesis, reported by Kende in 1990,[3] and the difficulty
in designing a stereoselective total synthesis is manifested in
recent reports dealing with partial syntheses of the molecule.[4]
In our continuing search for new applications of stereo-
selective radical cyclization reactions of b-alkoxyacrylates,[5]
we examined the efficacy of these reactions in a stereo-
controlled synthesis of 1.
In our retrosynthetic analysis, the tetrahydropyran alde-
hyde B was to be prepared from a b-alkoxyacrylate precursor
C, which may be obtained from l-arabinose (2). The
dihydropyran derivative E was envisaged to arise from the
diene F by olefin metathesis.[6] Connection of the parts A and
D by Julia-type olefination would then complete the con-
struction of the carbon framework (Scheme 1).
Selective acetonide protection of the dithioacetal derivative
of l-arabinose (2) and benzylation of the remaining hydroxy
groups gave the acetonide 4 (Scheme 2).[2a] The b-alkoxy-
acrylate 5 was obtained from 4 by acetonide deprotection,
regioselective TBS protection of the primary hydroxy group,
and reaction with methyl propiolate. The aldehyde group
generated from the dithioacetal moiety in 5 was reduced with
NaBH4, and bromide substitution led to the primary bromide
6, which was then stereoselectively transformed into the
[20] G. C. Bazan, W. D. Cotter, Z. J. A. Komon, R. A. Lee, R. J. Lachi-
cotte, J. Am. Chem. Soc. 2000, 122, 1371 1380.
À
[21] The short B O bond length (1.384 ä) is consistent with a significant p
interaction (sum of covalent radii: 1.47ä).
[22] In the case of h6-benzene versus h6-borazine complexes, substitution of
À
À
B N for C C does not appear to significantly change the electronic
character of the metal. For example, see: a) H. Werner, R. Prinz, E.
Deckelmann, Chem. Ber. 1969, 102, 95 103; b) G. Huttner, B. Krieg,
Angew. Chem. 1971, 83, 541 542; Angew. Chem. Int. Ed. Engl. 1971,
10, 512 513.
[*] Prof. Dr. E. Lee, S. J. Choi, H. Kim, H. O. Han, Y. K. Kim, S. J. Min,
S. H. Son, S. M. Lim, W. S. Jang
School of Chemistry and Molecular Engineering
Seoul National University
Seoul 151-747 (Korea)
Fax : (‡82)2-889-1568
E-mail: eunlee@snu.ac.kr
[**] The authors thank the Ministry of Science and Technology, Republic
of Korea, and Korea Institute of Science and Technology Evaluation
and Planning for a National Research Laboratory grant (1999). The
authors also thank Professor A. S. Kende (University of Rochester)
for NMR spectra of natural and synthetic ambruticin, and Professor G.
Hˆfle (GBF) for a sample of natural ambruticin and NMR spectra.
176
¹ WILEY-VCH Verlag GmbH, 69451 Weinheim, Germany, 2002
1433-7851/02/4101-0176 $ 17.50+.50/0
Angew. Chem. Int. Ed. 2002, 41, No. 1