The one-electron oxidation of an azazirconacyclobutene in the presence of B(C6F5)3 [7]

Full text of DOI:10.1021/ja984290j

7274
J. Am. Chem. Soc. 1999, 121, 7274-7275
The One-Electron Oxidation of an
Azazirconacyclobutene in the Presence of B(C6F5)3
C. Jeff Harlan,§ Tony Hascall,§ Etsuko Fujita,‡ and
Jack R. Norton*,§
Department of Chemistry
Columbia UniVersity
New York, New York 10027
Chemistry Department
1
Figure 1. The 300 MHz H NMR spectrum (CD2Cl2; 293 K) of 2 (a),
BrookhaVen National Laboratory
and 2 (45 mM) with 16 mM B(C6F5)3 (b). The taller peaks are truncated.
Upton, New York 11973
ReceiVed December 14, 1998
ReVised Manuscript ReceiVed May 16, 1999
Erker has used B(C6F5)3 to generate active olefin polymeriza-
tion catalysts by opening zirconacycles such as 1.1 In an effort to
extend this work to heteroatom-substituted zirconacycles, we
treated 22 with B(C6F5)3. To our surprise, the result is a partial
oxidation to a radical cation, 2+•.
Figure 2. EPR spectra (9.775 GHz) of 2+• in CD3C6D5. (a) 2 (49 mM)
+ B(C6F5)3 (12 mM); (b) 2 (47 mM), 2+• (0.45 mM); (c) 2+• (0.45 mM).
Complex 2+• in (b) and (c) was generated via oxidation of 2 with
[(CpMe)2Fe]+ (see text). Simulations used aN ) 7.0 G; aHa ) 3.2 G; aHb
) 1.0 G; aHc ) 3.7 G; line broadening (G): (d) 0.65; (e) 0.85; (f) 0.40.
conditions. In eq 2, ∆(T2ex-1) is directly proportional to the rate
constant for electron exchange k; in eq 3, ∆(T2ex-1) is inversely
porportional to k. Thus, resonances in the large hyperfine limit
will broaden as the temperature is raised, while those in the small
hyperfine limit will sharpen.
A typical 1H NMR spectrum of 2 after the addition of B(C6F5)3
is shown in Figure 1. One set of phenyl resonances has broadened
t
to the point of disappearance; the Bu signal has broadened
∆(T2ex-1) ) k[Radical]
(2)
(3)
substantially; the C5H5 resonance and the other phenyl resonances
do not broaden significantly; the chemical shifts remain un-
changed. Less broadening is observed when smaller amounts of
B(C6F5)3 are added. There is no immediate change in the 19F NMR
spectrum of the B(C6F5)3.
∆(T2ex-1) ) ([Radical]/[Neutral]2)(1/4)(a2)k-1
In the present case the neutral species is 2, the radical is 2+•,
and k is the rate constant for electron exchange between them
(eq 4). The phenyl resonances (Ha, Hb, and Hc) of solutions of 2
Such selective broadening has been seen in the 1H NMR
spectrum of partially oxidized chlorophyll a and explained by
electron exchange with the corresponding radical cation.3 Indeed,
the EPR spectrum of the radical cation 2+• is readily detectable
at g ) 2.0118 when solutions of 2 are treated with B(C6F5)3
(Figure 2a). The observable hyperfine coupling constants (G) are
aN ) 7.0, aHa ) 3.2, aHb ) 1.0, aHc ) 3.7; those for the other
protons are too small to be resolved.
1
that contain 2+• show8 the increase in H NMR line width with
temperature expected in the large hyperfine limit (although
fNτR a2/4 ≈ 1).9
2
k
2 + 2+• h 2+• + 2
(4)
1
While examining the H NMR spectrum of partially reduced
t
The Bu and C5H5 resonances of solutions of 2 that contain
2+• sharpen with temperature in the manner expected from eq 3,
and are thus characteristic of the small hyperfine limit. The
p-xylene, de Boer and MacLean derived the basic equations for
the effect of electron exchange on NMR spectra.4 Equation 1
describes the line broadening due to exchange of an electron,
∆(T2ex-1), as a function of the mole fractions of the neutral and
radical (fN and fR), the lifetime of the radical τR, the hyperfine
coupling constant a, and the electron spin lattice relaxation time,
T1e.
(2) (a) Walsh, P. J.; Hollander, F. J.; Bergman, R. G. J. Am. Chem. Soc.
1988, 110, 8729. (b) Walsh, P. J.; Hollander, F. J.; Bergman, R. G.
Organometallics 1993, 12, 3705.
(3) Waterton, J. C.; Sanders, J. K. M. J. Am. Chem. Soc. 1978, 100, 4044.
(4) (a) de Boer, E.; MacLean, C. J. Chem. Phys. 1966, 44, 1334. (b) de
Boer, E.; van Willigen, H. Prog. Nucl. Magn. Reson. 1967, 2, 111.
(5) These have been referred to as the slow exchange4 (or strong pulse6)
limit and rapid exchange4 (or weak pulse7) limit cases, respectively. We prefer
the terms large hyperfine and small hyperfine because there is only one k for
exchange.
(6) McConnell, H. N.; Berger, S. B. J. Chem. Phys. 1957, 27, 230.
(7) Solomon, I.; Bloembergen, N. J. Chem. Phys. 1956, 25, 261.
(8) The solutions used for these experiments were prepared by oxidizing 2
with [(C5H4Me)2Fe][B(C6F5)4] (as mentioned later in the text). The amount
of the radical cation 2+• in solutions prepared from the reaction of 2 and
B(C6F5)3 increases with time, along with slow formation of products that are
diamagnetic but have not been identified.
fRτRa2/4
∆(T2ex-1) )
(1)
1 + fNτR a2/4 + 2τRT1e
2
-1
Equation 1 has two limiting cases. In the large hyperfine
limit,4-7 a is large enough that fNτR a2/4 . 1 + 2τRT1e-1; in the
2
2
small hyperfine limit,4-7 a is small enough that fNτR a2/4 +
2τRT1e-1 , 1. Equations 2 and 3 show the consequences of these
(9) If we use the k obtained from eq 5 at a typical [2] of 50 mM, τR is 1.33
× 10-7 s; fN ≈ 1 and a(Hc) ) 3.7 G (1 × 107 Hz) make fNτR2a2/4 equal to
0.48sless than 1, and not in the large hyperfine limit. A quantitative test of
eq 2 is difficult because the Ha, Hb, and Hc resonances in the 1H NMR spectrum
§ Columbia University.
‡ Brookhaven National Laboratory.
(1) Karl, J.; Erker, G.; Fro¨hlich, R. J. Am. Chem. Soc. 1997, 119, 11165
and references therein.
of 2 broaden significantly even at very small concentrations of 2+•
.
10.1021/ja984290j CCC: $18.00 © 1999 American Chemical Society
Published on Web 07/27/1999

Communications to the Editor
J. Am. Chem. Soc., Vol. 121, No. 31, 1999 7275
temperature dependence of the width of the C5H5 resonance of 2
(Table S-1 in Supporting Information) implies an Ea of 4.1(1)
kcal/mol for k.
of B(C6H5)3 rapidly turns blue, reflecting the formation of the
-• 16
radical anion B(C6H5)3
. In the presence of Na or Na/K alloy
a THF solution of B(C6F5)3 remains colorless, and we see no
Solutions of 2+• that are free of 2 can be generated by oxidation.
The CV of 2 shows a reversible one-electron oxidation at E°1/2
) -0.07 V vs SCE in CH2Cl2 ([nBu4N][PF6], 0.1 N; 50 mV/s),
and the 1,1′-dimethylferricinium cation oxidizes 2 quantitatively
to 2+• (which can be reduced cleanly back to 2 by cobaltocene).
The EPR of 2+• generated with [(C5H4Me)2Fe][B(C6F5)4] (Figure
2c) is sharper in the absence of exchange with 2 and broadens
upon addition of 2 (Figure 2b). From the excess line width in
Figure 2b, ∆W, eq 510 gives an exchange rate constant k of 1.5
× 108 M-1 s-1 at 293 K, implying from eq 3 and the width of
evidence of a radical anion via EPR or 19F NMR.17
-•
The radical anion B(C6F5)3 does appear in the negative ion
mass spectrum of B(C6F5)3, but CV experiments on B(C6F5)3 have
failed to provide evidence for the formation of the radical anion
by reversible reduction in solution. (Electrochemistry on B(C6F5)3
-
is complicated by the fact that it interacts with the anionssPF6
,
BF4-, and ClO4-sof common supporting electrolytes, as shown
by 19F NMR.) Reversible reduction is also unobservable for
B(C6H5)3; the radical anion apparently becomes adsorbed on
electrode surfaces.16
t
1
No radical other than 2+• can be detected when the reaction of
2 with B(C6F5)3 is examined by EPR. Because so little of the 2
is converted to 2+•, the possibility that impurities play a role
cannot be excluded. Addition of (H2O)B(C6F5)318 (traces of which
remain even after sublimation of commercial B(C6F5)3) or
[PhNMe2H][B(C6F5)4] to solutions of 2 causes formation of 2+•,
although less cleanly than does B(C6F5)3 alone. We have always
observed the formation of some 2+• from the reaction of 2 and
B(C6F5)3 no matter how carefully the reagents are purified
(recrystallization, sublimation, etc.).
“There is no definite knowledge of either the nature of the
counterion or the fate of the electrons”19 in many reactions that
generate radical cations. For example, various Lewis acid/solvent
combinations (AlCl3/CH2Cl2, AlCl3/CH3NO2, BF3/SO2, SbCl5/
PhNO2, etc.)19,20 oxidize neutral organic molecules by one
electron, but [as Bard, Ledwith, and Shine have noted for aromatic
substrates] “the final state of the electron acceptor is not too well-
known”, and the “Lewis-acid anion radical has never been
detected”.19 The formation of some 2+• from 2 can also be effected
by another Lewis acid, methyl alumoxane (Aldrich, 10% in
toluene).
t
the Bu H NMR resonance at 293 K a hyperfine aBu of 0.071 G;
the hyperfine of the C5H5 protons is similarly calculated to be
0.018 G. Oxidation of a solution of 2 with small amounts of
ferricinium permits the determination of the 1H NMR line
broadening produced by a given amount of 2+•, and implies that
less than 1% of the 2 is oxidized to 2+• in a B(C6F5)3 experiment.11
∆W
[Neutral]
k ) 1.54 × 107
(5)
In an effort to generate a radical anion from B(C6F5)3 we have
treated it with cobaltocene. No radical anion is observed, but the
products 312 and 4 are formed cleanly in a one-to-one ratio (eq
6); the structure of 4 has been confirmed by X-ray diffraction.13
The oxidation of two cobalts to Co(III) has produced the hydride
ligand in the anion of 3, suggesting an electrophilic attack on a
Cp ring of Cp2Co.14,15
The above results raise the possibility that one-electron transfer
is involved in other reactions of B(C6F5)3.
Acknowledgment. We acknowledge Dr. John Birmingham, Boulder
Scientific Co., and NSF Grant CHE-98-96151 for financial support of
this work. We thank Professor N. Turro, M. Kleinman, and E. Karatekin
for help with the EPR experiments. We are grateful to G. Eaton, G.
Huttner, C. Mirkin, G. Parkin, G. Pez, C. Reed, and A. Rappe´ for valuable
discussions. The work at Brookhaven National Laboratory was performed
under contract DE-AC02-98CH10886 with the U.S. Department of Energy
and supported by its Division of Chemical Sciences, Office of Basic
Energy Sciences.
Further experiments have shown that B(C6F5)3 is remarkably
difficult to reduce. In the presence of alkali metals a THF solution
(10) (a) Ward, R. L.; Weissman, S. I. J. Am. Chem. Soc. 1957, 79, 2086.
(b) Sorensen, S. P.; Bruning, W. H. J. Am. Chem. Soc. 1972, 94, 6352.
1
t
(11) The broadening of the H NMR Bu resonance produced by addition
of B(C6F5)3 solutions (∼12 mM) to solutions of 2 (50 mM) varies between 5
and 25 Hz, whereas the broadening produced by 150 µM of ferricinium
(enough to oxidize only 0.3% of the 2) is 44 Hz.
(12) 1H NMR (CD2Cl2) of 3: δ 5.65 (s, 10 H); 3.65 (q, 1H, JB-H ) 90
Hz). [HB(C6F5)3]- is known: (a) Yang, X.; Stern, C. L.; Marks, T. J. Angew.
Chem., Int. Ed. Engl. 1992, 31, 1375. (b) Yang, X.; Stern, C. L.; Marks, T.
J. J. Am. Chem. Soc. 1994, 116, 10015. (c) Temme, B.; Erker G. J. Organomet.
Chem. 1995, 488, 177. (d) Ro¨ttger, D.; Schmuck, S.; Erker, G. J. Organomet.
Chem. 1996, 508, 263.
Supporting Information Available: Experimental details, 1H NMR
line widths, and X-ray structural information on 4 (PDF). An X-ray
crystallographic file on 4, in CIF format, is also available. This material
is available free of charge via the Internet at http://pubs.acs.org.
(13) Crystal data for 4: monoclinic space group P21/n (No. 14), a )
15.998(2) Å, b ) 10.116(1) Å, c ) 16.401 (2) Å, â ) 110.918(2)°, V )
2479.4(4) Å3, Z ) 4, T ) 203 K, R1 (I > 2σ(I)) ) 6.66%, GOF ) 1.023. 1H
NMR (CD2Cl2) for 4: δ 5.44 (br, 2H), 5.41 (br, 2H) 5.17 (s, 5H); 19F NMR:
δ -128.1 (d, JF-F ) 22.6 Hz), -159.9 (t, JF-F ) 21.2 Hz), -164.8 (t, JF-F
) 18.4).
JA984290J
(16) (a) Leffler, J. E.; Watts, G. B.; Tanigaki, T.; Dolan, E.; Miller, D. S.
J. Am. Chem. Soc. 1970, 92, 6825-6830. (b) Eisch, J. J.; Dluzniewski, T.;
Behrooz, M. Heteroatom Chem. 1993, 4, 235-241. (c) DuPont, T. J.; Mills,
J. L. J. Am. Chem. Soc. 1975, 97, 6375-6382. (d) Schulz, A.; Kaim, W.
Chem. Ber. 1989, 122, 1863-1868.
(17) The reaction of B(C6F5)3 with Na or Na/K alloy in a variety of solvents
(diglyme, toluene, THF, Et2O, and toluene/THF mixtures) resulted in slow
decomposition of the B(C6F5)3 (19F NMR). These reactions were not noticeably
exothermic, but extreme caution should be exercised in the reaction of
perfluorinated organics with alkali metals (see the Caution in Marsella, J. A.;
Gilicinski, A. G.; Coughlin, A. M.; Pez, G. P. J. Org. Chem. 1992, 57, 2856-
2860).
(14) The reaction of cobaltocene with Lewis acids has been examined in
an effort to explain the mechanism of eq 7.15
The initial step in reaction 7 is generally thought15 to be an electron transfer,
forming the RX-• radical anion. The enhanced reactivity of BBr3 (relative to
CHBr3 and Me3N f BBr3) toward Cp2Co has been interpreted15 in terms of
formation of BBr3-•. However, it is possible that the Cp2Co/BBr3 reaction
begins with electrophilic attack by BBr3 on cobaltocene (before formation of
the charged intermediate), and it is possible that reaction 6 gives rise to a
radical anion at some stage.
(15) (a) Herberich, G. E.; Schwarzer, J. J. Organomet. Chem. 1972, 34,
C43-C47. (b) Herberich, G. E.; Carstensen, T.; Klein, W.; Schmidt, M. U.
Organometallics 1993, 12, 1439-1441. (c) Sheats, J. E. J. Organomet. Chem.
Library 1979, 7, 461-521. (d) Ruwwe, J.; Erker, G.; Fro¨hlich, R. Angew.
Chem., Int. Ed. Engl. 1996, 35, 80-82.
(18) Water adducts of B(C6F5)3 are mentioned by: (a) Siedle, A. R.;
Newmark, R. A.; Lamanna, W. M.; Huffman, J. C. Organometallics 1993,
12, 1491-1492 (ref 5). (b) Bradley, D. C.; Harding, I. S.; Keefe, A. D.;
Motevalli, M.; Zheng, D. H. J. Chem. Soc., Dalton Trans. 1996, 3931-3936.
(c) Piers, W. E.; Chivers, T. Chem. Soc. ReV. 1997, 26, 345-354. (d)
Danopoulos, A. A.; Galsworthy, J. R.; Green, M. L. H.; Cafferkey, S.; Doerrer,
L. H.; Hursthouse, M. B. J. Chem. Soc., Chem. Comm. 1998, 2529-2530.
(19) Bard, A. J.; Ledwith, A.; Shine, H. J. AdV. Phys. Org. Chem. 1976,
13, 156-278.
(20) Bock, H.; Kaim, W. Acc. Chem. Res. 1982, 15, 9-17 and references
therein.