C-H Activation and Functionalization of Unsaturated Hydrocarbons by Transition-Metal Boryl Complexes

Article abstract of DOI:10.1021/om990113v

Transition-metal boryl complexes of the form Cp′Fe(CO)LBcat and (CO)₅MBcat, where Cp′ = C₅H₅, C₅Me₅, M = Mn, Re, L = CO, PMe₃, and cat = 1,2-O₂C₆H₄, were synthesized by reaction of ClBcat with [Cp′Fe(CO)L]^- or [M(CO)₅]^-. X-ray crystal structures of CpFe-(CO)₂Bcat, Cp*Fe(CO)₂Bcat, and (CO)₅MnBcat were obtained. Upon irradiation, these metal boryl complexes reacted with arenes and alkenes to form aryl- and vinylboronate ester products in moderate to high yields. Monosubstituted arenes with methyl, chloro, trifluoromethyl, methoxy, and dimethylamino substituents were used as substrates, and the resulting ratios of ortho- to meta- to para-substituted arene products were measured. No significant electronic effects were observed, indicating that the chemistry is not occurring through a typical electrophilic aromatic substitution pathway. Competition experiments between toluene and other substituted arenes were conducted. Reactivity differences were small, but anisole was found to have the fastest rate of reaction. Kinetic isotope effects were measured for the reaction of CpFe(CO)₂Bcat, (CO)₅MnBcat, or (CO)₅ReBcat with benzene/ benzene-d₆ mixtures and were found to be 3.3 ± 0.4, 2.1 ± 0.1, and 5.4 ± 0.4, respectively. This difference in isotope effect along with differences in selectivities with substituted arsenic reagents rules out a mechanism by which a common free Beat radical attacks free substrate. Several experiments were also conducted to probe for CO loss. A ¹³CO-labeling experiment, CO inhibition experiment, and PMe₃ trapping experiment indicate that the mechanism most likely proceeds through irreversible CO loss to form a 16-electron intermediate. Functionalization of alkenes to form vinylboronate esters was also observed, and mechanistic studies showed the absence of a measurable kinetic isotope effect for reaction of CpFe(CO)₂Bcat or (CO)₅ReBcat with ethylene/ethylene-d₄ mixtures or for reaction with ethylene-d₂.

Full text of DOI:10.1021/om990113v

Organometallics 1999, 18, 3383-3393
3383
C-H Activa tion a n d F u n ction a liza tion of Un sa tu r a ted
Hyd r oca r bon s by Tr a n sition -Meta l Bor yl Com p lexes
Karen M. Waltz, Clare N. Muhoro, and J ohn F. Hartwig*
Department of Chemistry, Yale University, P.O. Box 208107,
New Haven, Connecticut 06520-8107
Received February 19, 1999
Transition-metal boryl complexes of the form Cp′Fe(CO)LBcat and (CO)5MBcat, where
Cp′ ) C5H5, C5Me5, M ) Mn, Re, L ) CO, PMe3, and cat ) 1,2-O2C6H4, were synthesized by
reaction of ClBcat with [Cp′Fe(CO)L]- or [M(CO)5]-. X-ray crystal structures of CpFe-
(CO)2Bcat, Cp*Fe(CO)2Bcat, and (CO)5MnBcat were obtained. Upon irradiation, these metal
boryl complexes reacted with arenes and alkenes to form aryl- and vinylboronate ester
products in moderate to high yields. Monosubstituted arenes with methyl, chloro, trifluo-
romethyl, methoxy, and dimethylamino substituents were used as substrates, and the
resulting ratios of ortho- to meta- to para-substituted arene products were measured. No
significant electronic effects were observed, indicating that the chemistry is not occurring
through a typical electrophilic aromatic substitution pathway. Competition experiments
between toluene and other substituted arenes were conducted. Reactivity differences were
small, but anisole was found to have the fastest rate of reaction. Kinetic isotope effects were
measured for the reaction of CpFe(CO)2Bcat, (CO)5MnBcat, or (CO)5ReBcat with benzene/
benzene-d6 mixtures and were found to be 3.3 ( 0.4, 2.1 ( 0.1, and 5.4 ( 0.4, respectively.
This difference in isotope effect along with differences in selectivities with substituted arsenic
reagents rules out a mechanism by which a common free Bcat radical attacks free substrate.
Several experiments were also conducted to probe for CO loss. A 13CO-labeling experiment,
CO inhibition experiment, and PMe3 trapping experiment indicate that the mechanism most
likely proceeds through irreversible CO loss to form a 16-electron intermediate. Function-
alization of alkenes to form vinylboronate esters was also observed, and mechanistic studies
showed the absence of a measurable kinetic isotope effect for reaction of CpFe(CO)2Bcat or
(CO)5ReBcat with ethylene/ethylene-d4 mixtures or for reaction with ethylene-d2.
In tr od u ction
Thus, systems that can selectively deliver a functional
group to a hydrocarbon are current targets. Some
examples of functionalization processes mediated by
homogeneous transition-metal systems include dehy-
drogenation of alkanes,7-9 carbonylation of benzene,10-12
carbonylation of pentane,13 and aldimine formation from
isonitrile insertion into benzene.14 Recently, oxidation
of hydrocarbons by electrophilic late transition metals15
has shown renewed promise.16 A Pt system developed
recently by Periana et al. selectively catalyzes the
conversion of methane to methyl bisulfate in 70% yield
at temperatures as low as 100 °C.17 This method of
The functionalization of unreactive C-H bonds has
been a major goal of organometallic chemistry.1-3 Many
homogeneous organometallic systems undergo intermo-
lecular C-H activation reactions with unsaturated as
well as saturated organic substrates, but few lead to
unbound functionalized products.3 Of the systems that
cleave hydrocarbon C-H bonds, many contain an aro-
matic supporting ligand such as cyclopentadienide. For
example, unsaturated metal fragments of the form
CpIrLx were shown in the early 1980s to react with
hydrocarbons to form isolable metal alkyl hydride
complexes, providing some of the first evidence for
oxidative addition of alkane C-H bonds.4-6 In contrast,
no metal carbonyl fragments of the form M(CO)xR or
even CpM(CO)2R (M ) Fe, Ru, Os) have been reported
to react with free arene or alkane C-H bonds to date.
Future studies may reveal such reactivity.
(7) Burk, M. J .; Crabtree, R. H. J . Am. Chem. Soc. 1987, 109, 8025-
8032.
(8) Maguire, J . A.; Boese, W. T.; Goldman, A. S. J . Am. Chem. Soc.
1989, 111, 7088-7093.
(9) Maguire, J . A.; Goldman, A. S. J . Am. Chem. Soc. 1991, 113,
6706-6708.
(10) Kunin, A. J .; Eisenberg, R. J . Am. Chem. Soc. 1986, 108, 535-
536.
(1) Arndtsen, B. A.; Bergman, R. G.; Mobley, T. A.; Peterson, T. H.
Acc. Chem. Res. 1995, 28, 154-162.
(11) Kunin, A. J .; Eisenberg, R. Organometallics 1988, 7, 2124-
2129.
(2) Crabtree, R. H. Chem. Rev. 1985, 85, 245-269.
(3) Activation and Functionalization of Alkanes; Hill, C. L., Ed.;
Wiley: New York, 1989.
(12) Fisher, B. J .; Eisenberg, R. Organometallics 1983, 2, 764-767.
(13) Sakakura, T.; Tanaka, M. J . Chem. Soc., Chem. Commun. 1987,
758-759.
(4) Hoyano, J . K.; McMaster, A. D.; Graham, W. A. G. J . Am. Chem.
Soc. 1983, 105, 7190-7191.
(14) J ones, W. D.; Foster, G. P.; Putinas, J . M. J . Am. Chem. Soc.
1987, 109, 5047-5048.
(5) J anowicz, A. H.; Bergman, R. G. J . Am. Chem. Soc. 1983, 105,
3929-3939.
(15) Shilov, A. E.; Shul’pin, G. B. Chem. Rev. 1997, 97, 2879.
(16) Stahl, S. S.; Labinger, J . A.; Bercaw, J . E. Angew. Chem., Int.
Ed. 1998, 37, 2180-2192.
(6) J ones, W. D.; Feher, F. J . Organometallics 1983, 2, 562-563.
10.1021/om990113v CCC: $18.00 © 1999 American Chemical Society
Publication on Web 07/17/1999

3384 Organometallics, Vol. 18, No. 17, 1999
Waltz et al.
generating an ester of methanol eliminates the common
problem of higher reactivity of the product and over-
oxidation. A recent iridium complex with a PCP-type
ligand leads to acceptorless dehydrogenation of cyclic
alkanes or transfer dehydrogenation of linear alkanes
to form R-olefins as the predominant product at early
reaction times.18 More traditional methods to function-
alize hydrocarbons are based on radical or superacid
reagents. However, these reagents react with poor
selectivity for different types of C-H bonds. Even
enzymatic systems or models of them usually show poor
selectivity upon oxidation of linear alkanes.19 Homoge-
neous transition-metal systems should be able to over-
come this selectivity problem because they react pref-
erentially with primary C-H bonds.
Resu lts
A. Syn th esis a n d Ch a r a cter iza tion of Tr a n si-
tion -Meta l Bor yl Com p lexes. Transition-metal boryl
complexes of the general formula Cp′Fe(CO)LBcat (Cp′
) C5H5, C5Me5; L ) CO, PMe3; cat ) 1,2-O2C6H4) were
prepared by addition of ClBcat to a suspension of
M[Cp′Fe(CO)L] (M ) Na, Li) in toluene solvent as
shown in eq 1. With the exception of Li[CpFe(CO)-
We have found that the presence of a boryl ligand in
the M(CO)x and CpM(CO)2 (M ) Fe, Ru) metal systems
that previously showed no intermolecular C-H activa-
tion creates a system that is reactive not only toward
the cleavage of hydrocarbon C-H bonds but also toward
elimination of functionalized organoboronate ester prod-
ucts. We have previously communicated the function-
alization of arenes and alkenes by CpFe(CO)2Bcat and
(CO)5MBcat (M ) Mn, Re)20 to form aryl- and vinylbo-
ronate esters. This functionalization chemistry is un-
usual for organometallic systems, and the initial C-H
bond cleavage must result from the presence of the boryl
group.
(PMe3)], all metal anions used in the syntheses of the
metal boryl complexes were prepared by Na/Hg reduc-
tion of the corresponding metal dimers in THF. The
reactions were monitored by 11B NMR spectroscopy,
which showed the disappearance of ClBcat at δ 29, and
the formation of a new peak between δ 44 and 57
corresponding to the Cp′Fe(CO)LBcat complex. In the
case of the parent CpFe(CO)2Bcat complex (1), a satu-
rated pentane solution of the crude product was crystal-
lized at -30 °C for 20-30 min to give a 73% yield of
yellow product. Yellow Cp*Fe(CO)2Bcat (2) was isolated
in 71% yield from Na[Cp*Fe(CO)2], and CpFe(CO)-
(PMe3)Bcat (3) was isolated in 33% yield from Li[CpFe-
(CO)(PMe3)] (4) prepared by deprotonation of the known
CpFe(CO)(PMe3)H27 with n-BuLi. The bis(phosphine)
complex CpFe(PMe3)2Bcat (5) was synthesized by ir-
radiation of 1 in excess PMe3 in pentane. Recrystalli-
zation from toluene removed traces of CpFe(CO)(PMe3)-
Bcat impurity and gave 5 in 52% yield.
The boryl group’s effect on the reactivity and regio-
chemistry for HX bonds was initially demonstrated by
the reactions of metal boryl complexes with protic
reagents.21 In these transformations, a metal hydride
was formed and an aminoborane, alkoxoborane, or
haloborane was formed from reaction with an amine,
alcohol, or mineral acid. Reaction with hydrocarbons
occurs with similar regiochemistry to produce a metal
hydride and a hydrocarbon with a boryl functional
group. The boryl ligand’s dramatic effect on the reactiv-
ity of the metal center may be due to its electrophilic
property caused by the empty p-orbital or the presumed
strong σ-donating ability of the anionic boryl ligand.
Alternatively, functionalization of hydrocarbons with
boryl complexes may result from the thermodynamic
driving force provided by formation of strong boron-
carbon bonds.22,23 We report here a full synthetic and
mechanistic account of the functionalization of arenes
and alkenes by transition-metal boryl complexes to form
aryl- and vinylboronate esters. These organoboronate
ester compounds are useful organic intermediates that
can be converted into a variety of functionalized organic
molecules, such as alcohols, amines, aldehydes, and
halides.24-26
The metal boryl complexes (CO)5MBcat (M ) Mn (6),
Re (7)) were prepared in a similar fashion by addition
of ClBcat to a suspension of Na[M(CO)5] in toluene (eq
2). These reactions were also monitored by 11B NMR
spectroscopy, which showed new peaks for 6 and 7 at δ
49 and 44. Crystallization from toluene layered with
pentane afforded 6 and 7 in 72% and 30% yields,
respectively.
(17) Periana, R. A.; Taube, D. J .; Gamble, S.; Taube, H.; Satoh, T.;
Fujii, H. Science 1998, 280, 560-564.
(18) Liu, F.; Pak, E. B.; Singh, B.; J ensen, C. M.; Goldman, A. S. J .
Am. Chem. Soc. 1999, 121, 4086-4087.
(19) Cook, B. R.; Reinert, T. J .; Suslick, K. S. J . Am. Chem. Soc.
1986, 108, 7281-7286.
(20) Waltz, K. M.; He, X.; Muhoro, C.; Hartwig, J . F. J . Am. Chem.
Soc. 1995, 117, 11357-11358.
(24) Brown, H. C. Boranes in Organic Chemistry; Cornell University
Press: Ithaca, NY, 1972.
(25) Brown, H. C. Organic Synthesis via Boranes; Wiley: New York,
1975.
(21) Hartwig, J . F.; Huber, S. J . Am. Chem. Soc. 1993, 115, 4908-
4909.
(22) Rablen, P. R.; Hartwig, J . F.; Nolan, S. P. J . Am. Chem. Soc.
1994, 116, 4121.
(23) Rablen, P. R.; Hartwig, J . F. J . Am. Chem. Soc. 1996, 118,
4648-4653.
(26) Pelter, A.; Smith, K.; Brown, H. C. Borane Reagents; Academic
Press: New York, 1988.
(27) Treichel, P. M.; Komar, D. A. J . Organomet. Chem. 1981, 206,
77-88.

C-H Activation by Transition-Metal Boryl Complexes
Organometallics, Vol. 18, No. 17, 1999 3385
Ta ble 1. Cr ysta llogr a p h ic Da ta for Cp F e(CO)2Bca t (1), Cp *F e(CO)2Bca t (2), a n d (CO)5Mn Bca t (6)
CpFe(CO)2Bcat
13H9O4BFe
Cp*Fe(CO)2Bcat
(CO)5MnBcat
C11H4O7BMn
empirical formula
fw
C
C18H19O4BFe
366.00
295.87
313.90
cryst color, habit
cryst dimens (mm)
cryst syst
space group
temp (K)
lattice params
a (Å)
pale yellow parallelogram
0.20 × 0.20 × 0.45
orthorhombic
P212121 (No. 19)
296
pale yellow prism
0.19 × 0.29 × 0.46
monoclinic
P21/a (No. 14)
183
golden yellow cut block
0.35 × 0.35 × 0.42
monoclinic
P21/a (No. 14)
296
9.5178(6)
9.608(1)
13.713(2)
8.016(1)
24.49(1)
8.737(1)
96.06(2)
1706(1)
4
12.865(9)
7.019(6)
15.296(4)
113.17(2)
1270(2)
4
b (Å)
c (Å)
â (deg)
V (Å3)
1254.0(4)
4
1.567
Z
D
calcd (g/cm3)
1.425
1.642
residuals
R
Rw
0.030
0.032
1.27
0.034
0.043
2.55
0.038
0.040
2.24
goodness of fit indicator
Ta ble 2. Selected Bon d Dista n ces for
Cp F e(CO)2Bca t (1) a n d Cp *F e(CO)2Bca t (2)a
CpFe(CO)2Bcat
Cp*Fe(CO)2Bcat
Fe-B
1.959(6)
1.734(5)
1.728(6)
1.7155(7)
1.401(6)
1.420(7)
1.980(2)
1.749(3)
1.743(2)
1.72(1)
1.404(3)
1.410(3)
Fe-C7
Fe-C8
Fe-Cp′ centroid
B-O1
B-O2
a
Distances are in angstroms. Estimated standard deviations
in the least significant figure are given in parentheses.
Ta ble 3. Selected Bon d An gles for Cp F e(CO)2Bca t
(1) a n d Cp *F e(CO)2Bca t (2)a
CpFe(CO)2Bcat
Cp*Fe(CO)2Bcat
C7-Fe-C8
C7-Fe-B
C8-Fe-B
Fe-B-O1
Fe-B-O2
O1-B-O2
92.9(3)
83.5(3)
87.1(2)
127.0(4)
125.0(4)
108.0(5)
96.3(1)
87.5(1)
82.82(9)
126.8(2)
124.4(2)
108.8(2)
F igu r e 1. Ortep diagram of CpFe(CO)2Bcat (1).
bond length was found to be 1.959(6) Å. The difference
in dihedral angle between the Cp centroid-Fe-B and
O1-B-O2 planes is 7.9°. This structure suggests that
a π-interaction may exist between the iron and boron,
since the catecholboryl ligand is oriented to allow
overlap of the empty boron p-orbital with the CpFe(CO)2
HOMO.28,29
The crystal structure for the Cp* complex 2 is shown
in Figure 2. The Fe-B bond length of 1.980(2) Å is
similar to that of the Cp complex 1. The difference in
dihedral angle between the Cp centroid-Fe-B and O1-
B-O2 planes is 26.7°, which is significantly larger than
that in complex 1. This larger angle in complex 2 is most
likely due to the steric bulk of the pentamethylcyclo-
pentadienyl group that forces the catecholate substitu-
ent further out of the Cp centroid-Fe-B plane. This
larger angle, coupled with the slight lengthening of the
Fe-B bond compared with complex 1, may indicate a
decrease in the π-interaction between the iron and boryl
ligand.
a
Angles are in degrees. Estimated standard deviations in the
least significant figure are given in parentheses.
Ta ble 4. Selected Bon d Dista n ces for
(CO)5Mn Bca t (6)a
Mn-B
2.108(6)
1.870(7)
1.841(6)
1.837(7)
Mn-C10
Mn-C11
B-O1
1.847(7)
1.832(8)
1.373(6)
1.398(6)
Mn-C7
Mn-C8
Mn-C9
B-O2
a
Distances are in angstroms. Estimated standard deviations
in the least significant figure are given in parentheses.
Ta ble 5. Selected Bon d An gles for
(CO)5Mn Bca t (6)a
C7-Mn-B
C8-Mn-B
C9-Mn-B
C10-Mn-B
85.5(2)
178.8(3)
86.3(3)
82.1(3)
C11-Mn-B
Mn-B-O1
Mn-B-O2
O1-B-O2
83.0(3)
124.3(4)
125.0(4)
110.7(4)
a
Angles are in degrees. Estimated standard deviations in the
least significant figure are given in parentheses.
X-r a y Cr ysta l Str u ctu r es for Cp F e(CO)2Bca t (1),
Cp *F e(CO)2Bca t (2), a n d (CO)5Mn Bca t (6). Tables
1-5 provide data collection and refinement parameters
and selected bond lengths and angles for compounds 1,
2, and 6. The crystal structure for iron boryl complex 1
has been previously reported in communication form,21
and an ORTEP diagram is shown in Figure 1. The Fe-B
The crystal structure for manganese boryl complex 6
has been previously reported20 and is shown in Figure
3. The Mn-B bond length of 2.108(6) Å is longer than
(28) Schilling, B. E. R.; Hoffmann, R.; Faller, J . W. J . Am. Chem.
Soc. 1979, 101, 592-598.
(29) Schilling, B. E. R.; Hoffmann, R.; Lichtenberger, D. L. J . Am.
Chem. Soc. 1979, 101, 585-591.

3386 Organometallics, Vol. 18, No. 17, 1999
Waltz et al.
scopy using dodecahydrotriphenylene as an internal
standard. The reaction was monitored by 11B NMR
spectroscopy, which showed the formation of a new peak
at δ 32 for the phenylboronate ester product. This
product was further characterized by 1H NMR spec-
troscopy and GC/MS. Comparison of these spectral data
with those of an independently prepared sample of
PhBcat36 confirmed the identity of the product. The
transition-metal product was [CpFe(CO)2]2, as deter-
1
mined by H NMR and IR spectroscopy. Presumably,
the reaction initially produces CpFe(CO)2H, which is
known to decompose at room temperature to [CpFe-
(CO)2]2.37 Similarly, irradiation of a benzene solution of
Cp* iron boryl complex 2 resulted in formation of
[Cp*Fe(CO)2]2 and PhBcat in 99% yield.
To determine if the Cp* moiety had an effect on the
rate of reaction, equimolar solutions of iron complexes
1 and 2 in benzene were irradiated side by side, and
the conversion to product was determined qualitatively
by 11B NMR spectroscopy at various times. Reactions
of the Cp* complex 2 and of the Cp version 1 displayed
equal conversions at various reaction times.
F igu r e 2. Ortep diagram of Cp*Fe(CO)2Bcat (2).
Iron phosphine boryl complexes 3 and 5 also con-
verted solvent benzene to arylboronate esters in high
yields. Irradiation of CpFe(PMe3)(CO)Bcat (3) in benzene-
d6 afforded Ph-d5Bcat in 94% yield. One of the several
metal products from photolysis in benzene was deter-
mined to be the known hydride CpFe(PMe3)(CO)H.27
The bis(phosphine) complex CpFe(PMe3)2Bcat (5) re-
acted cleanly in benzene-d6 to give Ph-d5Bcat in 90%
yield. The sole metal-containing product from photolysis
in benzene was the known hydride CpFe(PMe3)2H.38
Photolysis of (CO)5MnBcat (6) and (CO)5ReBcat (7)
resulted in formation of PhBcat in 45% and 50% yields
respectively as shown in eq 4. In most cases the
F igu r e 3. Ortep diagram of (CO)5MnBcat (6).
the metal-boron distances in other metal carbonyl boryl
complexes 1, 2, and (CO)4Fe(Bcat-t-Bu)2 (2.028(7) Å).30
The Mn-CO bond length trans to the boryl ligand
(1.841(6) Å) falls in the range of the other four M-CO
distances (1.832(8)-1.870(7) Å). This similarity in M-CO
bond distances contrasts the distinct M-CO distances
in metal carbonyl complexes of the type (CO)5MR (M )
Mn, Re; R ) alkyl, H),31-35 where the M-CO bond
length trans to R is shorter than those cis to the R
group.
B. Rea ction of Tr a n sition -Meta l Bor yl Com -
p lexes w ith Ar en es. F u n ction a liza tion of Ben zen e.
Photolysis of iron boryl complex 1 in neat benzene
solvent (eq 3) resulted in formation of PhBcat in
remainder of the boron consisted of multiple products
that we have not been able to identify. The transition-
metal products of these reactions were Mn2(CO)10 or Re2-
(CO)10, with small amounts of accompanying Re3-
(CO)12H3.39 These transition-metal products were identi-
1
fied by IR and H NMR spectroscopy.
F u n ction a liza tion of Su bstitu ted Ar en es. Reac-
tion of the metal boryl complexes with arene C-H bonds
was observed to occur in the presence of accompanying
(35) Masters, A. P.; Richardson, J . F.; Sorensen, T. S. Can. J . Chem.
1990, 68, 2221-2227.
1
quantitative yield, as determined by H NMR spectro-
(36) PhBcat was synthesized by refluxing PhB(OH)2 (prepared
according to: Bean, F. R.; J ohnson, J . R. J . Am. Chem. Soc. 1932, 54,
4415) and catechol in benzene while azeotropically removing the water
with a Dean-Stark apparatus.
(37) Shackleton, T. A.; Mackie, S. C.; Fergusson, S. B.; J ohnston,
L. J .; Baird, M. C. Organometallics 1990, 9, 2248-2253.
(38) Green, M. L. H.; Wong, L.-L. J . Chem. Soc., Dalton Trans. 1987,
411-416.
(39) Andrews, M. A.; Kirtley, S. W.; Kaesz, H. D. Inorg. Synth. 1977,
17, 66.
(30) He, X.; Hartwig, J . F. Organometallics 1996, 15, 400-407.
(31) Davies, J . A.; El-Ghanam, M.; Pinkerton, A. A. J . Organomet.
Chem. 1991, 409, 367-376.
(32) LaPlaca, S. J .; Hamilton, W. C.; Ibers, J . A.; Davison, A. Inorg.
Chem. 1969, 8, 1928-1935.
(33) Lee, G.-H.; Peng, S.-M.; Yang, G.-M.; Lush, S.-F.; Liu, R.-S.
Organometallics 1989, 8, 1106-1111.
(34) Lindner, E.; Pabel, M.; Fawzi, R.; Mayer, H. A.; Wurst, K. J .
Organomet. Chem. 1992, 435, 109-121.

C-H Activation by Transition-Metal Boryl Complexes
Organometallics, Vol. 18, No. 17, 1999 3387
Ta ble 6. Ra tio of P r od u ct Isom er s fr om Rea ction
of Cp F e(CO)2Bca t (1) w ith C6H5X
Ta ble 7. P r od u ct Ra tios fr om Com p etition
Rea ction s of Cp F e(CO)2Bca t (1) w ith Mixtu r es of
Su bstitu ted Ar en es
product isomer ratio
X
o
m
p
Me
OMe
Cl
CF3
NMe2
1.1
1.6
1.5
1.5
1.0
1.0
1.1
1.0
1.0
8.0
1.0
alkyl, trifluoromethyl, chloro, methoxy, and dimethyl-
amino groups on the arene, as shown in eq 5. This
ratios as those products produced by reaction with 1.
Like iron complex 1, 7 also formed ortho-substituted
products in addition to meta- and para-substituted
products upon reaction with anisole. In this case, the
major product in the reaction of 7 with anisole was the
ortho-substituted arylboronate ester. Overall yields of
functionalized product were not measured for 7.
chemistry allowed for the evaluation of the electronic
interactions involved in the functionalization by using
arene substituent effects. Irradiation of CpFe(CO)2Bcat
(1) in a variety of monosubstituted arene solvents led
to formation of arylboronate ester products with the
ortho, meta, and para ratio of products and yields shown
in Table 6. All reactions were monitored by 11B NMR
spectroscopy, which showed a new peak at δ 32-33,
indicative of ArBcat products. The different isomers of
the various arylboronate ester products were identified
Com p etition Exp er im en ts. To determine the rela-
tive rates for reaction of the different substituted arenes
with a given metal boryl complex, competition experi-
ments were carried out in which a metal boryl was
irradiated in equimolar amounts of two substituted
arenes. Ratios of arylboronate ester products shown in
Table 7 were determined by GC/MS or 1H NMR spec-
troscopy. The product ratios were constant throughout
the reaction, suggesting that the products were stable
to the reaction conditions and that product ratios were
not altered by selective decomposition of products.
Irradiation of 1 with equimolar amounts of toluene
and chlorobenzene resulted in roughly 1:1 ratios of tolyl
to chlorophenyl boronate esters. A similar product ratio
was determined for reactions conducted in the presence
of both toluene and R,R,R-trifluorotoluene. N,N-Di-
methylaniline and anisole were the only two substrates
which showed an enhanced rate of reaction over that of
toluene. N,N-Dimethylaniline reacted 2.4 times faster
than toluene, and anisole reacted 3 times faster. Com-
plex 1 was also irradiated with equimolar amounts of
tert-butylbenzene and N,N-dimethylaniline to determine
if steric effects of the substituents affected the rate of
reaction. N,N-Dimethylaniline was found to react about
3.4 times faster than tert-butylbenzene. The similarity
of this product ratio to that of the reactions involving
N,N-dimethylaniline and toluene rules out substantial
steric effects on the product ratios.
1
by H NMR spectroscopy, GC/MS, and comparison of
thesespectraldata with independentlyprepared samples.40
Product ratios and yields were determined by integra-
tion of 1H NMR resonances versus an internal standard
of dodecahydrotriphenylene.
Reaction of 1 with substituted arenes resulted in
formation of only meta- and para-substituted arylbor-
onate esters for all substrates except anisole, which
showed substantial amounts of ortho-substituted prod-
uct. Irradiation of 1 in p-xylene solvent showed very low
yields of arylboronate ester product, which is consistent
with the lack of ortho products observed in the reactions
of most of the monosubstituted arenes. The meta:para
ratio of products for all substrates except N,N-dimethyl-
aniline and anisole were found to be roughly the same
regardless of the substituent. N,N-Dimethylaniline
showed a preference for reaction at the para position,
while anisole formed all three product isomers. The
yields of arylboronate esters of the various substituted
1
arenes were determined by either H NMR spectroscopy
or GC. The highest yields were obtained for toluene
(70%), chlorobenzene (55%), and anisole (52%). The
substrates R,R,R-trifluorotoluene and N,N-dimethyl-
aniline gave product yields of 33% and 30%, respec-
tively.
Irradiation of rhenium boryl 7 with toluene and R,R,R-
trifluorotoluene resulted in only meta- and para-
substituted arylboronate esters in roughly the same
Mea su r em en t of Kin etic Isotop e Effects. Isotope
effects of the benzene functionalization reaction were
measured in three different ways. First, a competition
experiment between benzene and benzene-d6 was con-
ducted by irradiation of the metal boryl complexes 1, 6,
and 7 in equimolar mixtures of benzene and benzene-
d6. Relative amounts of deuterated and protiated phe-
nylboronate ester products were determined by the ratio
of the abundance of the molecular ions in the mass
(40) Morgan, J .; Pinhey, J . T. J . Chem. Soc., Perkin Trans. 1 1990,
715-720.

3388 Organometallics, Vol. 18, No. 17, 1999
Waltz et al.
Ta ble 8. Kin etic Isotop e Effects for P h otolysis in
spectroscopy. This result demonstrates that CO is either
lost irreversibly or not at all.
Ben zen e/Ben zen e-d 6
To determine if CO dissociation occurs, the photolyses
were conducted in the presence of the more soluble and
more favorable ligand PMe3. A solution of 1 in benzene-
d6 was irradiated with 9 equiv of PMe3 and monitored
by 11B NMR spectroscopy (eq 8). In addition to small
complex
kH/kD
CpFe(CO)2Bcat (1)
(CO)5MnBcat (6)
(CO)5ReBcat (7)
3.3 ( 0.4
2.1 ( 0.1
5.4 ( 0.4
spectrum. Kinetic isotope effects kH/kD were found to
be 3.3 ( 0.4, 2.1 ( 0.1, and 5.4 ( 0.4, respectively (Table
8).
A second experiment involved a comparison of con-
version of complex 1 in benzene and benzene-d6. Com-
plex 1 in benzene was irradiated side by side with an
equally concentrated sample of complex 1 in benzene-
d6. Conversions of the boryl complexes to phenylbor-
onate ester product were qualitatively evaluated at
various time points in deuterated and protiated solvents
by 11B NMR spectroscopy. The conversions were found
to be indistinguishable at each time.
amounts of phenylboronate ester product, a new boryl
species was observed by 11B, 31P, and 1H NMR spec-
troscopy. The identity of this complex was determined
to be CpFe(CO)(PMe3)Bcat (3) by matching the spectral
data to those of material prepared independently by the
route described in section A on the synthesis of metal
boryl complexes. Continued irradiation converted this
material into a new complex, which was identified as
CpFe(PMe3)2Bcat (5), this time by matching its spectral
features to those of a sample prepared by a large-scale
photolysis of 1 in the presence of PMe3, as described
above in section A.
An intramolecular isotope effect was also measured
by irradiation of 1 with 10 equiv of 1,3,5-trideuterioben-
zene in pentane solvent (eq 6). The ratio of the two
The rate of disappearance of 1 upon photolysis was
found to be independent of added phosphine. Two
equimolar solutions of 1 in benzene-d6 were irradiated
side by side, one with 10 equiv of PMe3 and one without
added PMe3. The rate of disappearance of 1 was
monitored qualitatively by 11B NMR spectroscopy and
found to be indistinguishable for the two samples over
the course of the reaction.
phenylboronate ester products formed from either C-H
or C-D cleavage was determined by the relative abun-
dances of the molecular ions in the mass spectrum. This
intramolecular isotope effect kH/kD was found to be 3.4,
which matches the intermolecular isotope effect ob-
served for reaction of 1 with the mixture of benzene and
benzene-d6.
P r obe for CO Dissocia tion . Several experiments
were performed to determine whether CO loss is occur-
ring on the reaction pathway, and if so, whether the
process is reversible or irreversible. Two samples of 1
in a 25% benzene-d6/pentane solution were irradiated
side by side, one with 5 atm of added CO, and the other
without added CO. The relative conversions of the boryl
complexes to arylboronate ester products in the presence
and absence of CO were evaluated qualitatively by 11B
NMR spectroscopy at various time points. The sample
without added CO appeared to react slightly faster than
the sample with added CO; however, the difference in
conversions was almost immeasurable. This lack of CO
inhibition on product formation suggests that CO is
either lost irreversibly or not at all.
To quantify the effects of added phosphine on PhBcat
formation, two equimolar solutions of 1 in benzene-d6
were irradiated side by side in the presence of 5 and 10
equiv of PMe3. After conversion of approximately half
of the starting material, the reactions were analyzed by
1H NMR spectroscopy to determine the ratio of phos-
phine-ligated boryl complex to PhBcat product. The
reaction with 10 equiv of PMe3 gave a 5.7:1 ratio of iron
phosphine boryl species to PhBcat, whereas the reaction
with 5 equiv of PMe3 gave a ratio of 2.9:1. Although the
metal phosphine species CpFe(CO)(PMe3)Bcat (3) and
CpFe(PMe3)2Bcat (5) react photochemically with ben-
zene to form PhBcat in the absence of phosphine as
described above, the presence of free phosphine inhibits
this reactivity. Independent experiments showed that
complex 3 reacted with phosphine to generate 5 rather
than reacting with solvent to generate PhBcat. More-
over, the reaction of 5 toward arene was strongly
inhibited by the presence of added phosphine, presum-
ably because of reversible, photochemical ligand dis-
sociation. Thus, the PhBcat measured in the above
competition reactions arises only from reactions of the
CpFe(CO)2Bcat (1) starting material.
To verify the result of the CO inhibition experiment,
a solution of 1 in benzene-d6 was irradiated in the
presence of added 13CO until half of the starting
material remained (eq 7). 13CO was not incorporated
C. Rea ction of Tr a n sition Meta l Bor yl Com -
p lexes w ith Alk en es. Ter m in a l Alk en es. The transi-
tion-metal boryl complexes described above also react
photochemically with alkenes to form vinylboronate
esters. Irradiation of 1 with ethylene in alkane solvent
produced ethenylBcat as the major product in 88% yield
into the starting material, as determined by 13C NMR

C-H Activation by Transition-Metal Boryl Complexes
Organometallics, Vol. 18, No. 17, 1999 3389
and ethylBcat in 7% yield. As shown in eq 9, irradiation
The intramolecular isotope effect was also determined
for iron complex 1 and rhenium boryl complex 7 by
irradiation of the metal boryl complexes in pentane with
trans-1,2-dideuterioethylene (eq 11). The value of kH/
of 1 dissolved in 1-hexene formed terminal hexenylbo-
ronate ester as the major product in 90% yield, in
addition to small amounts of hexylboronate ester. These
reactions were monitored by 11B NMR spectroscopy,
which showed peaks at δ 32 and 36, indicative of
vinylboronate ester and alkylboronate ester products,
respectively. The identity of the products was further
kD in this experiment is equal to the ratio of ethenyl-
Bcat-d2 to ethenylBcat-d, which was determined by the
ratio of the molecular ions in the mass spectrum. In
these two systems, there was again no detectable isotope
effect.
1
confirmed by GC/MS, H NMR spectroscopy, and com-
parison of these spectral data with those of indepen-
dently prepared samples of ethenylBcat, ethylBcat,
hexenylBcat, and hexylBcat.41 The rhenium boryl com-
plex 7 also reacted photochemically with alkenes. Ir-
radiation in neat 1-hexene formed the terminal hexen-
ylboronate ester in 55% yield, as well as hexylboronate
ester in 20-25% yield.
Discu ssion
Mech a n ism of Ar en e F u n ction a liza tion . Several
potential pathways for arene functionalization are
shown in Scheme 1. Path A involves formation of a Bcat
radical that reacts with arene to give the functionalized
product. The Bcat radical would be expected to be
electrophilic and may, therefore, react initially with the
arene π-system. If the chemistry did occur by attack of
a free Bcat radical on free solvent, then one must
observe identical isotope effects for all metal boryl
systems studied. This free radical would, of course,
cleave a C-H or C-D bond with the same selectivity
regardless of the origin of the radical. The isotope effects
measured for reaction of metal boryl complexes 1, 6, and
7 in benzene/benzene-d6 are clearly not equal; thus, a
mechanism involving reaction of free Bcat radical with
free solvent can be ruled out.
An alternative mechanism would be reaction of a Bcat
radical with arene solvent that is coordinated to the
metal center. This mechanism seems unreasonable,
because coordination of arene to the metal radical would
generate a 19-electron species and additional loss of CO
would require a two-photon process involving highly
reactive intermediates.
In ter n a l Olefin s. The reaction with internal olefins
was not as selective as the reaction with terminal
olefins. Irradiation of 7 in neat trans-4-octene resulted
in vinylboronate ester, at least three isomeric products,
and octylboronate ester, as determined by GC/MS.
Reaction of cyclohexene led to the formation of cyclo-
hexylboronate ester and one cyclohexenylboronate ester.
The 11B NMR spectrum showed only one peak at δ 36,
indicating that no vinylboronate ester had been formed.
Instead, an isomer of the vinylboronate ester product
had formed that contains an sp3-carbon-boron bond.
This observation was confirmed by the absence of a peak
in the GC/MS corresponding to the vinylboronate ester.
Irradiation of the rhenium complex with the bicyclic
olefin norbornene in pentane provided a single vinyl-
boronate ester product, as well as two norbornylbor-
onate ester products. These products were identified by
GC/MS and comparison to a sample of norbornylbor-
onate ester prepared by hydroboration of norbornene.
Mea su r em en t of Kin etic Isotop e Effects. Complex
1 in pentane was irradiated with equal volumes of
ethylene and ethylene-d4 (eq 10). The reaction was
Our isotope effect experiments also probed for ben-
zene coordination prior to C-H bond cleavage. If ir-
reversible precoordination occurred, we would expect a
small intermolecular isotope effect for reaction with
benzene/benzene-d6, since the η2-coordination should be
similar for benzene and benzene-d6. In contrast, the
intramolecular isotope effect for reaction with 1,3,5-
trideuteriobenzene would be large, since the selectivity
for reaction with a C-H or C-D bond occurs after the
precoordination of arene. Our experiments showed that
the inter- and intramolecular isotope effects for CpFe-
monitored by 11B NMR spectroscopy, which showed the
formation of vinylboronate ester, with small amounts
of alkylboronate ester products as for higher olefins. The
ratio of ethenylBcat and ethenylBcat-d3 was determined
by the ratio of the abundance of the molecular ions in
the mass spectrum. This value of kH/kD was 1.0 ( 0.2,
indicating that there is no intermolecular kinetic isotope
effect for this system.
Sch em e 1
(41) EthylBcat was prepared by Rh(PPh3)3Cl-catalyzed hydrobora-
tion of ethylene. EthenylBcat was formed as a minor product of this
hydroboration. HexylBcat was prepared by hydroboration of 1-hexene
with HBcat. HexenylBcat was prepared by reaction of hexenyllithium
with ClBcat.

3390 Organometallics, Vol. 18, No. 17, 1999
Waltz et al.
Sch em e 2
strates are virtually identical. Methyl and chloro groups
are both ortho,para-directors, whereas the trifluoro-
methyl substituent is a meta-director. If the reaction
were occurring through electrophilic aromatic substitu-
tion, we would expect to see arylboronate ester products
with m-trifluoromethyl, p-methyl, and p-chloro substit-
uents as the major products. The similarity of the
product ratios for varying substituents is strong evi-
dence against an electrophilic aromatic substitution
pathway.
(CO)2Bcat were almost identical (3.3 and 3.4, respec-
tively). Thus, coordination of the arene substrate to the
metal center prior to C-H bond cleavage either is fully
reversible or does not occur.
The formation of ortho-substituted products with only
anisole may be due to an interaction of the oxygen atom
with the boryl ligand or transition-metal center, which
would lead to C-H bond cleavage at the sterically
hindered ortho-position. The ortho-directing ability of
the methoxy group is well-established in directed meta-
lation chemistry with alkyllithium reagents.42 Consis-
tent with coordination of the methoxy group, the less
electron rich Re complex provides the ortho-product as
the major isomer, whereas the more electron rich Fe
complex gives the ortho-isomer as the minor product.
Two additional mechanisms involve reaction with
either a photochemically generated excited state (path
B) or a 16-electron intermediate generated by CO
dissociation (path C). Our studies under CO and 13CO
atmospheres showed that added CO does not inhibit the
formation of product and that 13CO is not incorporated
into the starting material during the reaction. Neither
of these experiments distinguish between the absence
of CO dissociation and irreversible dissociation. The
reactions conducted in the presence of PMe3, however,
provide strong evidence for CO dissociation as the initial
step of the arene functionalization. The formation of
phosphine-ligated boryl complex in competition with
arylboronate ester indicates either a common interme-
diate that reacts with solvent or phosphine (Scheme 2)
or separate competitive reactions with PMe3 and arene
solvent. A mechanism involving initial dissociation of
CO in competition with direct reaction with phosphine
would predict higher conversions of the starting boryl
complex as a function of the time of irradiation. Experi-
ments performed with and without added phosphine
showed no difference in conversion of 1 at different
times, ruling out a separate pathway initiated by
reaction with phosphine. In addition, the ratio of metal
phosphine boryl complex and arylboronate ester species
produced by the photochemical reaction depended on the
concentration of phosphine. These data strongly favor
a pathway involving photochemical CO dissociation to
generate a 16-electron boryl complex that is unusually
reactive toward hydrocarbon C-H bonds.
One possible reaction mode for the unsaturated
intermediate is an electrophilic aromatic substitution.
The boron center in the Bcat moiety, of course, displays
some degree of electrophilic character, making an elec-
trophilic aromatic substitution reaction a reasonable
pathway for the arene functionalization. Classic elec-
trophilic aromatic substitution reactions show large
substituent effects on selectivity for reaction at the
different positions of the aromatic ring. If our arene
functionalization reactions are occurring through this
type of mechanism, then we would expect to observe
electronic effects that parallel the directing effects in
simple halogenations, nitrations, or Friedel-Crafts
alkylations of aromatic systems.
The rate of electrophilic aromatic substitution reac-
tions is also influenced by the electron-donating or
-withdrawing ability of the substituent on the aromatic
substrate. The rate of substitution should be faster for
substrates with activating groups and slower for sub-
strates with deactivating groups. Of our substituents,
activating groups include OMe, NMe2, Me, and CMe3;
deactivating groups include Cl and CF3. Our competition
experiments showed an enhancement in rate of reaction
of the reactive intermediate with anisole and N,N-
dimethylaniline over toluene. Chlorobenzene and tri-
fluorotoluene, however, reacted with this intermediate
at nearly the same rate as toluene. An electrophilic
aromatic substitution pathway is consistent with the
enhanced rate of reaction for substrates with the more
donating groups OMe and NMe2; however, the magni-
tude of our effect is much smaller than typically
observed for halogenations and nitrations,43 and the
similar rates for reaction of the boryl complexes with
Cl- and CF3-substituted benzenes are inconsistent with
this pathway. These data lead us to rule out the
electrophilic aromatic substitution pathway, although
the reactive boryl intermediate may have some weak
electrophilic character. The small electronic effect on the
functionalization process presented here is similar to
some previous studies concerning electronic effects on
the rate of σ-bond metathesis reactions with substituted
arenes44 and regioselectivity for oxidative addition of
toluene by rhodium complexes.45
Another possible reaction pathway for the unsatur-
ated intermediate involves an unusual C-H activation
process that leads to coupling of the metal-bound
covalent ligand with the aryl portion of the substrate.
The C-H activation can occur by either (1) oxidative
addition of a C-H bond to the unsaturated metal center
Although the ratios of products in classic electrophilic
aromatic substitution chemistry depend on reaction
conditions, our results are clearly distinct from standard
substituent effects on electrophilic aromatic substitu-
tion. For example, the meta:para ratio of products
resulting from reaction of the boryl complexes with
toluene, chlorobenzene, and R,R,R-trifluorotoluene sub-
(42) Mallan, J . M.; Bebb, R. L. Chem. Rev. 1969, 69, 693-755.
(43) Carey, F. A.; Sundberg, R. J . In Advanced Organic Chemistry;
Plenum Press: New York, 1990; p 539.
(44) Thompson, M. E.; Baxter, S. M.; Bulls, A. R.; Burger, B. J .;
Nolan, M. C.; Santarsiero, B. D.; Schaefer, W. P.; Bercaw, J . E. J . Am.
Chem. Soc. 1987, 109, 203-219.
(45) J ones, W. D.; Feher, F. J . J . Am. Chem. Soc. 1984, 106, 1650-
1663.

C-H Activation by Transition-Metal Boryl Complexes
Organometallics, Vol. 18, No. 17, 1999 3391
Sch em e 3
with ethylene/ethylene-d4 and ethylene-d2 if the C-H
bond cleavage is irreversible.
Reaction of CpFe(CO)2Bcat with either ethylene/
ethylene-d4 or ethylene-d2 showed no measurable iso-
tope effects. This result was surprising because it is
inconsistent with both reaction pathways. One of several
explanations could account for the immeasurably small
isotope effect. First, the reaction could occur through
rate-determining alkene insertion into the metal-boron
bond46-50 and subsequent â-hydrogen elimination of
vinylboronate ester. Isotope effects for â-hydrogen elimi-
nation are small in some cases.51-58 Alternatively, a
σ-bond metathesis mechanism with an early transition
state in which a minimal amount of C-H cleavage has
occurred could provide a small isotope effect. Finally a
pathway involving the reversible oxidative addition of
alkene and subsequent irreversible B-C bond-forming
reductive elimination of vinylboronate ester could ac-
count for the small kH/kD. In this case, the isotope effect
would be generated by only equilibrium and secondary
isotope effects.
Con clu sion s. The formation of arylboronate esters
from reactions of transition-metal boryl complexes with
arenes is unusual for two reasons. First, the process
provides functionalized products rather than simply
transition-metal aryl complexes. Second, the boryl
ligand must alter the properties of the iron, manganese,
and rhenium complexes in a fashion that makes them
more reactive toward hydrocarbon C-H bonds. One
explanation for this enhanced reactivity is that the
metal center is more electron rich than it is in more
conventional organometallic compounds because of the
strong σ-donation by a formally anionic boryl ligand. A
second possibility is that the electrophilicity of the boron
p-orbital, albeit weak in the case of Bcat ligands, helps
to activate the hydrocarbon substrates. Finally, the
favorable thermodynamics for formation of B-C bonds22
may provide a driving force for product formation that
is stronger than the driving force for C-C bond forma-
tion in conventional systems. These thermodynamic
factors may serve to trap a low concentration of C-H
activation product that may constantly equilibrate with
the starting carbonyl complexes. This hypothesis would
predict that C-H activation could be observed with
these metal fragments in the form of exchanges between
and subsequent reductive elimination of organoborane
or (2) σ-bond metathesis (Scheme 3). Oxidative addition
is the generally accepted pathway for C-H activation
by late-metal systems;4,5 however, oxidative addition/
reductive elimination and σ-bond metathesis mecha-
nisms are difficult to distinguish in the absence of
observed aryl hydride intermediates.
Mech a n ism of Alk en e F u n ction a liza tion . The two
pathways in Scheme 4 can be envisioned for the func-
tionalization of alkenes. The first pathway involves
direct C-H activation of the olefin by the metal center,
either by an oxidative addition/reductive elimination
sequence or by a concerted σ-bond metathesis mecha-
nism. The second pathway occurs by insertion of the
alkene into the metal-boron bond, followed by â-hy-
drogen elimination to give vinylboronate ester and metal
hydride. Both of these mechanisms initially proceed
through photochemical loss of CO to form an unsatur-
ated intermediate, which could then coordinate the
alkene substrate in an η2-fashion prior to C-H bond
cleavage.
Many mechanisms are possible for formation of the
minor alkylboronate ester products. One simple mech-
anism is metal-catalyzed hydrogenation of the vinylbo-
ronate ester. A second involves formation of catechol-
borane by known hydrogenolysis of the metal boryl
complex20 followed by metal-catalyzed hydroboration of
the alkenes.
Our mechanistic information for formation of the
major vinylboronate ester products was not clear-cut.
Isotope effects for reaction of the metal boryl complex
with a mixture of ethylene/ethylene-d4 or with ethylene-
d2 could distinguish these paths. If the mechanism
occurs through the insertion pathway, we would not
expect to see an intermolecular isotope effect for reaction
of the metal boryl complex with the ethylene/ethylene-
d4 mixture, since protiated and deuterated substrates
would have similar rates for insertion into the metal-
boron bond. Reaction with ethylene-d2, however, would
be likely to show an intramolecular primary kinetic
isotope effect because the metal center could perform
either a â-hydrogen or â-deuterium elimination. If the
direct C-H activation pathway is followed, we would
expect to see both an inter- and intramolecular kinetic
isotope effect for reaction of the metal boryl complex
(47) Westcott, S. A.; Blom, H. P.; Marder, T. B.; Baker, R. T. J . Am.
Chem. Soc. 1992, 114, 8863-8869.
(48) Westcott, S. A.; Marder, T. B.; Baker, R. T. Organometallics
1993, 12, 975-979.
(49) Burgess, K.; van der Donk, W. A.; Westcott, S. A.; Marder, T.
B.; Baker, R. T.; Calabrese, G. C. J . Am. Chem. Soc. 1992, 114, 9350-
9359.
(50) Brown, J . M.; Lloyd-J ones, G. C. J . Chem. Soc., Chem. Commun.
1992, 710-712.
(51) Bullock, R. M. In Transition Metal Hydrides; Dedieu, A., Ed.;
VCH: New York, 1992; pp 263-307.
(52) McCarthy, T. J .; Nuzzo, R. G.; Whitesides, G. M. J . Am. Chem.
Soc. 1981, 103, 3396-3403.
(53) Ozawa, F.; Ito, T.; Yamamoto, A. J . Am. Chem. Soc. 1980, 102,
6457-6463.
(54) Komiya, S.; Morimoto, Y.; Yamamoto, A.; Yamamoto, T. Orga-
nometallics 1982, 1, 1528-1536.
(55) Brainard, R. L.; Whitesides, G. M. Organometallics 1985, 4,
1550-1557.
(56) Evans, J .; Schwartz, J .; Urquhart, P. W. J . Organomet. Chem.
1974, 81, C37-C39.
(57) Ikariya, T.; Yamamoto, A. J . Organomet. Chem. 1976, 120,
257-284.
(46) Baker, R. T.; Calabrese, J . C.; Westcott, S. A.; Nguyen, P.;
Marder, T. B. J . Am. Chem. Soc. 1993, 115, 4367-4368.
(58) Burger, B. J .; Thompson, M. E.; Cotter, W. D.; Bercaw, J . E. J .
Am. Chem. Soc. 1990, 112, 1566-1577.

3392 Organometallics, Vol. 18, No. 17, 1999
Waltz et al.
Sch em e 4
scopy and stirred for about 1 h until all of the ClBcat was
consumed. The reaction mixture was filtered through glass
wool to remove all solids and the toluene solvent removed
under vacuum. The crude product was extracted with pentane
and filtered to remove the less soluble dark red impurity,
[CpFe(CO)2]2. Cooling to -30 °C for a very short period of time
(less than 20 min) resulted in yellow crystalline product.
Crystallization for longer than 20 min resulted in concurrent
crystallization of yellow CpFe(CO)2Bcat with red [CpFe(CO)2]2
impurity. Recrystallization from pentane at -30 °C was
necessary to remove traces of boron impurities which resonate
at δ 23 in the 11B NMR spectrum and are believed to be
catBOBcat.60 CpFe(CO)2Bcat was isolated in 73% yield (419.8
mg). 1H NMR (C6D6): δ 7.13 (m, 2H), 6.81 (m, 2H), 4.20 (s,
5H). 13C{1H} NMR (C6D6): δ 213.96, 150.71, 121.92, 111.48,
83.80. 11B NMR (C6D6): δ 51.8. IR (pentane): νCO 2024, 1971
cm-1. Anal. Calcd for C13H9BO4Fe: C, 52.77; H, 3.07. Found:
C, 52.81; H, 3.14.
arenes and Mn, Re, and Fe aryl complexes. Experiments
to distinguish between these potential contributions of
the boryl ligands to C-H activation are in progress.
Exp er im en ta l Section
Gen er a l Meth od s. Unless otherwise noted, all manipula-
tions were conducted in an inert-atmosphere glovebox or by
using standard Schlenk line techniques. 1H NMR spectra were
recorded on either a General Electric QE-300 or Bruker AM-
500 NMR spectrometer, and 11B and 31P{1H} NMR spectra
were recorded on an Omega 300 NMR spectrometer. 11B and
31P{1H} chemical shifts are reported in parts per million
relative to external standards of BF3‚Et2O and 85% H3PO4,
respectively. Resonances downfield of the standard are as-
signed positive chemical shifts. Proton chemical shifts are
reported in ppm relative to residual protiated solvent as
internal standard. Pentane, THF, benzene, and toluene sol-
vents were distilled from sodium/benzophenone ketyl prior to
use. Benzene-d6 and anisole were dried over sodium/benzophe-
none ketyl and degassed before use. Chlorobenzene and R,R,R-
trifluorotoluene were dried over calcium hydride and degassed
prior to use. N,N-Dimethylaniline was dried over phosphorus
pentoxide and distilled prior to use. The metal complexes
Cp *F e(CO)2Bca t (2). ClBcat (369.3 mg, 2.39 mmol) was
reacted with Na[Cp*Fe(CO)2] (677.4 mg, 2.51 mmol) using the
same procedure for the synthesis of CpFe(CO)2Bcat (1). Two
recrystallizations from pentane at -30 °C were necessary to
remove all traces of red impurity. Additional pure material
was obtained from the mother liquor, resulting in a total yield
of 71% (620.1 mg) of yellow crystalline product. 1H NMR
(C6D6): δ 7.15 (m, 2H), 6.80 (m, 2H), 1.55 (s, 15H). 13C{1H}
NMR (C6D6): δ 216.37, 151.43, 122.00, 111.70, 96.07, 10.26.
CpFe(CO)(PMe3)H27 and [Cp*Fe(CO)2]2 were prepared ac-
59
cording to literature procedures. All other chemicals were used
as received from commercial suppliers. Photolyses were carried
out with a 450 W Hanovia mercury arc lamp in a quartz
immersion well using Pyrex reaction vessels placed flush with
the immersion well, approximately 0.75 in. from the lamp.
Yields were determined by 1H NMR through use of internal
standards of either trimethoxybenzene or dodecahydrotri-
phenylene.
P r ep a r a tion of Na [Re(CO)5], Na [Mn (CO)5], Na [Cp F e-
(CO)2], a n d Na [Cp *F e(CO)2] by Na /Hg Red u ction . The
appropriate metal dimer was dissolved in THF and reacted
with 1% sodium-mercury amalgam (2.5 equiv of Na) for 12 h
with vigorous stirring. The solution was removed from the
amalgam and filtered through Celite to remove a gray powdery
solid. Because the presence of THF led to the decomposition
of the metal boryl complexes, it was necessary to remove as
much THF from the metal anions as possible. Thorough
removal of THF was achieved by first removing the solvent
under vacuum to yield a sticky solid, which was ground into a
fine powder in the presence of toluene. A suspension of this
powder in toluene was heated to 80 °C under vacuum until
all the solvent had been removed. This was repeated three
times. The resulting solid was washed with toluene to remove
dimeric impurities, affording the sodium salt in quantitative
yield.
11B NMR (C6D6): δ 54.3. IR (pentane): νCO 1996, 1940 cm-1
.
Anal. Calcd for C18H19BO4Fe: C, 59.07; H, 5.23. Found: C,
59.08; H, 5.23.
Li[Cp F e(CO)(P Me3)] (4). To a stirred solution of CpFe-
(CO)(PMe3)H (277.0 mg, 1.23 mmol) in pentane was added 0.47
mL of 2.6 M n-BuLi (1.22 mmol) in hexane. The solution
turned from clear dark yellow to cloudy orange. After the
mixture was stirred for 20 min, the orange solid was collected
on a fritted funnel and washed several times with pentane,
1
giving the product in 92% yield (261.2 mg). H NMR (C6D6):
δ 4.51 (s, 5H), 1.47 (d, J HP ) 7.4 Hz, 9H). 31P NMR (C6D6): δ
29.3. IR (C6D6): νCO 1915 cm-1
.
Cp F e(CO)(P Me3)Bca t (3). ClBcat (286.8 mg, 1.86 mmol)
was reacted with Li[CpFe(CO)(PMe3)] (430.6 mg, 1.86 mmol)
using the same procedure for the synthesis of CpFe(CO)2Bcat
(1). Crystallization from pentane at -30 °C afforded a powder
containing both black and yellow solids. Only one compound
was observed by NMR spectroscopy. The yellow material was
separated by sublimation (65 °C, 0.01 mmHg) and subsequent
recrystallization from pentane to give pale yellow crystalline
1
material in 33% yield (210.2 mg). H NMR (C6D6): δ 7.18 (m,
2H), 6.84 (m, 2H), 4.31 (d, J HP ) 1.5 Hz, 5H), 0.92 (d, J HP
)
9.4 Hz, 9H). 13C{1H} NMR (C6D6): δ 219.17 (d, J CP ) 28.6 Hz),
151.47, 121.54, 111.24, 81.77, 22.18 (d, J CP ) 30.5 Hz). 11B
Cp F e(CO)2Bca t (1). A solution of ClBcat (301.8 mg, 1.95
mmol) in 5 mL of toluene was added slowly to a stirred
suspension of Na[CpFe(CO)2] (397.7 mg, 1.99 mmol) in 10 mL
of toluene. The reaction was monitored by 11B NMR spectro-
(60) Clegg, W.; Lawlor, F. J .; Marder, T. B.; Nguyen, P.; Norman,
N. C.; Orpen, A. G.; Quayle, M. J .; Rice, C. R.; Robins, E. G.; Scott, A.
J .; Souza, F. E. S.; Stringer, G.; Whittell, G. R. J . Chem. Soc., Dalton
Trans. 1998, 301-309.
(59) Catheline, D.; Astruc, D. Organometallics 1984, 3, 1094-1100.

C-H Activation by Transition-Metal Boryl Complexes
Organometallics, Vol. 18, No. 17, 1999 3393
NMR (C6D6): δ 57.5. 31P{1H} NMR (C6D6): δ 39.2. IR (C6D6):
νCO 1927 cm-1. Anal. Calcd for C15H18BO3PFe: C, 52.38; H,
5.28. Found: C, 52.57; H, 5.37.
phenylboronate esters by the abundance of their molecular
ions, which is a direct measurement of the kinetic isotope
effect, kH/kD. A similar procedure was used to determine the
intramolecular isotope effect for reaction with 1,3,5-trideute-
riobenzene.
(b) Alk en e Su bstr a te. The metal boryl complex (5 mg) was
dissolved in pentane (0.5 mL) and transferred to an NMR tube.
trans-1,2-Dideuterioethylene (20 equiv) was condensed into the
tube using vacuum techniques, and the tube was flame-sealed.
The solution was irradiated and monitored by 11B NMR
spectroscopy until the starting material was completely con-
sumed. The products were analyzed by GC/MS, and the value
for the kinetic isotope effect was determined by the ratio of
the abundance of the molecular ions of ethenyl-d2-Bcat and
ethenyl-d-Bcat. A similar procedure was used for reaction of
the metal boryl with equal amounts of ethylene and ethylene-
d4.
Com p etition Exp er im en ts. The metal boryl complex was
added to an NMR tube as a solid. Equimolar amounts of
substituted arene were added into the NMR tube by syringe,
and the tube was capped with either a septum or screw cap.
The solution was irradiated until all the starting material was
consumed, as determined by 11B NMR spectroscopy. The
solution was then analyzed by GC/MS to determine the
relative ratio of arylboronate ester products derived from the
two substrates.
Cp F e(P Me3)2Bca t (5). PMe3 (5 equiv) was transferred by
vacuum techniques into a solution of CpFe(CO)2Bcat (1; 416.8
mg, 1.41 mmol) in 30 mL of pentane in a glass bomb equipped
with a stir bar. This yellow solution was irradiated with
stirring for at least 6 h. The reaction was monitored by 11B
NMR spectroscopy until all of the monosubstituted product,
CpFe(CO)(PMe3)Bcat (3) at δ 57, was converted to the disub-
stituted product at δ 60. When the reaction was complete, all
volatile materials were removed under vacuum, and the
residue was extracted with pentane. The extracts were filtered
through glass wool, condensed, and crystallized at -30 °C to
yield dark yellow-amber crystals. Further crystallization from
toluene was necessary to remove all traces of monosubstituted
product impurity, affording the bis(phosphine) complex 5 in
52% yield (286.7 mg). An additional 24% (132.3 mg) of product
in roughly 95% purity was obtained from the toluene mother
1
liquor. H NMR (C6D6): δ 7.19 (m, 2H), 6.86 (m, 2H), 4.14 (t,
J HP ) 1.7 Hz, 5H), 1.05 (vt, 18H). 13C{1H} NMR (C6D6): δ
151.91, 120.80, 110.54, 78.09, 24.72 (vt). 11B NMR (C6D6): δ
60. 31P{1H} NMR (C6D6): δ 38.0. Anal. Calcd for C17H27BO2P2-
Fe: C, 52.09; H, 6.94. Found: C, 52.40; H, 6.75.
(CO)5Mn Bca t (6). ClBcat (208.8 mg, 1.35 mmol) was
reacted with Na[Mn(CO)5] (294.5 mg, 1.35 mmol) using the
same procedure for the synthesis of CpFe(CO)2Bcat (1). After
initial filtration of solid, the resulting yellow toluene solution
was condensed, layered with pentane, and cooled to -30 °C,
affording square colorless crystals. Additional material was
obtained from the mother liquor to give an overall yield of 77%
(327.0 mg). 1H NMR (C7D8): δ 6.97 (m, 2H), 6.74 (m, 2H). 13C-
{1H} NMR (C6D6): δ 210.68, 209.62, 150.34, 122.86, 112.29.
11B NMR (C6H6): δ 49. IR (Nujol): νCO 2112 (w), 2009 (v. broad,
vs) cm-1. Anal. Calcd for C11H4BO7Mn: C, 42.09; H, 1.28.
Found: C, 41.80; H, 1.38.
(CO)5ReBca t (7). ClBcat (325.9 mg, 2.11 mmol) was reacted
with Na[Re(CO)5] (737.6 mg, 2.11 mmol) using the same
procedure for synthesis of CpFe(CO)2Bcat (1). After initial
filtration of solid, the yellow toluene solution was condensed,
layered with pentane, and cooled to -30 °C. The resulting
white crystals were recrystallized two more times from pen-
tane to give analytically pure material in 30% yield (277.4 mg).
1H NMR (C6D6): δ 7.09 (m, 2H), 6.77 (m, 2H). 13C{1H} NMR
(C6D6): δ 183.58, 182.21, 150.48, 122.76, 112.34. 11B NMR
(C6D6): δ 44. IR (C6D6): νCO 2129, 2051, 2016 (vs) cm-1. Anal.
Calcd for C11H4BO7Re: C, 29.68; H, 0.91. Found: C, 29.75; H,
0.92.
CO In h ibition Exp er im en t. The metal boryl complex (10
mg) was dissolved in pentane (0.6 mL). Benzene was added to
this solution (0.2 mL) by syringe, and the solution was then
added to two NMR tubes (0.4 mL in each). Carbon monoxide
(5 atm) was introduced into one of the tubes. 11B NMR spectra
of the two samples were recorded before irradiation side by
side. During the course of irradiation, 11B NMR spectra were
periodically recorded of both samples, and the amount of
product was qualitatively evaluated at each time point.
13CO Exp er im en t. Iron boryl complex 1 (6 mg) was
dissolved in C6D6 and transferred to a Young NMR tube. 13CO
1
(1.8 atm) was introduced into the tube. 11B, H, and 13C NMR
spectra were recorded for the starting material. The solution
was irradiated until roughly half the starting material re-
mained by 11B NMR spectroscopy. 13C NMR spectroscopy of
this sample showed no incorporation of the labeled CO into
the starting material.
P Me3 Tr a p p in g Exp er im en t. Iron boryl complex 1 (5 mg)
was dissolved in benzene-d6 (0.5 mL) and transferred to an
NMR sample tube. PMe3 (9 equiv) was condensed into the tube
using vacuum techniques, and the tube was flame-sealed. 11B,
31P, and 1H NMR spectra were recorded of the starting
material. The solution was irradiated and periodically moni-
tored by 11B, 31P, and 1H NMR spectroscopy.
Gen er a l P r oced u r e for P h otoch em ica l Rea ction of
Meta l Bor yls w ith Ar en e/Alk en e Su bstr a tes. The metal
boryl complex (5 mg) was dissolved in 0.5 mL of neat substrate
and the solution transferred to an NMR tube. For the ethylene
substrate, the metal boryl complex (5 mg) was dissolved in
0.5 mL of pentane and transferred to a thick-walled NMR tube,
and 100 equiv of ethylene was condensed into the tube, which
was then flame-sealed. 11B NMR spectra were recorded for the
starting material. The sample was irradiated with periodic
monitoring of the reaction by 11B NMR spectroscopy.
Isotop e Effect Mea su r em en ts. (a ) Ar en e Su bstr a te.
The metal boryl complex (5 mg) was added to an NMR sample
tube as a solid. Equimolar amounts of benzene (0.30 mL) and
benzene-d6 (0.30 mL) were added to the tube by syringe, and
the tube was capped with a septum. The sample was irradiated
Ack n ow led gm en t. We gratefully acknowledge sup-
port from the National Science Foundation (CHE-
9617171), a Camille Dreyfus Teacher/Scholar Award,
and a National Science Foundation Young Investigator
Award. J .F.H. is a fellow of the Alfred P. Sloan Founda-
tion.
Su p p or tin g In for m a tion Ava ila ble: Tables of crystal
data, structure solution and refinement, positional parameters,
bond lengths and angles, torsion angles, Cartesian coordinates,
and anisotropic thermal parameters for Cp*Fe(CO)2Bcat (2).
This material is available free of charge via the Internet at
http://pubs.acs.org.
until all starting material was consumed as determined by 11
B
NMR spectroscopy. The solution inside the tube was analyzed
by GC/MS to determine the ratio of protiated and deuterated
OM990113V