Functionalization of alkanes by isolated transition metal boryl complexes

Article abstract of DOI:10.1021/ja002840j

We report the photochemical reaction of transition metal boryl complexes of the type Cp*M(CO)(n)B-(OR)₂ (M = Fe, Ru, W) with alkanes to form alkylboronate esters in yields as high as 85% for reaction of Cp*W(CO)₃[Bcat(Me)₂] {[Bcat(Me)₂] = B-I, 2-O₂C₆H₂-3, 5-Me₂} with pentane. Synthesis of a series of Cp and Cp* catecholboryl complexes showed that sterically blocking or eliminating sp² positions on the metal boryl complex was important for alkane functionalization to occur. The metal boryl complexes reacted exclusively at the terminal C-H position of alkanes. Functionalization of 2-methylbutane occurred preferentially at the least sterically hindered terminal position with a selectivity of 10:1. This selectivity data, in addition to kinetic isotope effects measured for reaction of metal boryls with a mixture of pentane and pentane-d₁₂, argues against radical chemistry. Several experiments were conducted to probe for CO dissociation. An experiment employing added ¹³CO, one conducted under 2 atm pressure of CO, and one conducted in the presence of PMe₃ indicated that the mechanism involves CO dissociation to form a 16-electron intermediate that reacts faster with alkane solvent than it recoordinates CO. The effect of boryl electronics on the functionalization of alkanes was studied by examining the reactions of ruthenium dialkylboryl, dithioboryl, and dialkoxyboryl complexes with pentane. The dialkoxyboryl complexes gave the highest yields of functionalized product. A comparison between reactions of the different boryl complexes in arene and alkane solvents showed that the electronic properties of the boryl group had a greater effect on the reaction of the unsaturated intermediate with alkane than they did on the generation of the intermediate.

Full text of DOI:10.1021/ja002840j

11358
J. Am. Chem. Soc. 2000, 122, 11358-11369
Functionalization of Alkanes by Isolated Transition Metal Boryl
Complexes
Karen M. Waltz and John F. Hartwig*
Contribution from the Department of Chemistry, Yale UniVersity, P.O. Box 208107,
New HaVen, Connecticut 06520-8107
ReceiVed August 1, 2000
Abstract: We report the photochemical reaction of transition metal boryl complexes of the type Cp*M(CO)nB-
(OR)2 (M ) Fe, Ru, W) with alkanes to form alkylboronate esters in yields as high as 85% for reaction of
Cp*W(CO)3[Bcat(Me)2] {[Bcat(Me)2] ) B-1, 2-O2C6H2-3, 5-Me2} with pentane. Synthesis of a series of Cp
and Cp* catecholboryl complexes showed that sterically blocking or eliminating sp2 positions on the metal
boryl complex was important for alkane functionalization to occur. The metal boryl complexes reacted
exclusively at the terminal C-H position of alkanes. Functionalization of 2-methylbutane occurred preferentially
at the least sterically hindered terminal position with a selectivity of 10:1. This selectivity data, in addition to
kinetic isotope effects measured for reaction of metal boryls with a mixture of pentane and pentane-d12, argues
against radical chemistry. Several experiments were conducted to probe for CO dissociation. An experiment
employing added 13CO, one conducted under 2 atm pressure of CO, and one conducted in the presence of
PMe3 indicated that the mechanism involves CO dissociation to form a 16-electron intermediate that reacts
faster with alkane solvent than it recoordinates CO. The effect of boryl electronics on the functionalization of
alkanes was studied by examining the reactions of ruthenium dialkylboryl, dithioboryl, and dialkoxyboryl
complexes with pentane. The dialkoxyboryl complexes gave the highest yields of functionalized product. A
comparison between reactions of the different boryl complexes in arene and alkane solvents showed that the
electronic properties of the boryl group had a greater effect on the reaction of the unsaturated intermediate
with alkane than they did on the generation of the intermediate.
Introduction
to functionalize alkanes is through dehydrogenation, which
recently has been reported to form terminal olefins regioselec-
tively in the presence of a second olefin as hydrogen acceptor.13
Internal olefins were formed in the absence of acceptor,14 or at
long reaction times with acceptor. Metal-catalyzed carbonylation
of pentane is also selective for the primary position,15,16 but
the chemoselectivity for aldehyde product and the alkane
conversion were low. Shilov showed that chloroplatinum salts
could react with alkanes in aqueous solution to form alcohols
and alkyl halides.17 In this case the reactions occur preferentially
at the primary position, but the selectivity for primary vs sec-
ondary positions in linear alkanes, such as pentane and hexane,
is only on the order of 3:1.18 The oxidation of alkanes by
electrophilic metal systems has been enhanced through the use
The efficient conversion of alkanes to products with func-
tionality at the terminal position is a long-standing synthetic
problem.1-3 Hydrocarbons are typically considered inert due
to their high C-H bond strength and low acidity. Thus, reagents
or catalysts that are reactive enough to cleave C-H bonds often
react unselectively or at the weakest C-H bonds, which are
the internal ones.1 Few reagents and even fewer catalysts
selectiVely cleave primary C-H bonds. Metal-catalyzed oxida-
tion of linear alkanes is typically unselective. With only recent
exceptions4-6 involving shape-selective catalysts, the most
selective reactions have occurred at the weakest tertiary C-H
bonds.7,8 In contrast, homogeneous transition-metal systems
which react stoichiometrically with C-H bonds through oxida-
tive addition or σ-bond metathesis pathways generally show
reactivity at primary, over secondary and tertiary, C-H bonds.9
While many homogeneous metal systems are known that
cleave C-H bonds regiospecifically at primary positions,10-12
few functionalize the alkane directly. One of the simplest ways
(8) Sheldon, R. A.; Kochi, J. K. Metal-Catalyzed Oxidations of Organic
Compounds: Mechanistic Principles and Synthetic Methodology Including
Biochemical Processes; Academic Press: New York, 1981.
(9) SelectiVe Hydrocarbon ActiVation; Davies, J. A., Watson, P. L.,
Liebman, J. F., Greenberg, A., Eds.; Wiley-VCH: New York, 1990.
(10) Jones, W. D.; Feher, F. J. Organometallics 1983, 2, 562-563.
(11) Bergman, R. G.; Seidler, P. F.; Wenzel, T. T. J. Am. Chem. Soc.
1985, 107, 4358-4359.
(12) Bennett, J. L.; Wolczanski, P. T. J. Am. Chem. Soc. 1997, 119,
10696-10719.
(13) Liu, F.; Pak, E. B.; Singh, B.; Jensen, C. M.; Goldman, A. S. J.
Am. Chem. Soc. 1999, 121, 4086-4087.
(14) Xu, W.; Rosini, G. P.; Gupta, M.; Jensen, C. M.; Kaska, W. C.;
Krogh-Jespersen, K.; Goldman, A. S. Chem. Commun. 1997, 2273-2274.
(15) Sakakura, T.; Tanaka, M. J. Chem. Soc., Chem. Commun. 1987,
758-759.
(16) Rosini, G. P.; Zhu, K. M.; Goldman, A. S. J. Organomet. Chem.
1995, 504, 115-121.
(17) Gol’dshleger, N. F.; Es’kova, V. V.; Shilov, A. E.; Shteinman, A.
A. Zh. Fiz. Khim. (Engl. Transl.) 1972, 46, 785-786.
(1) Shilov, A. E.; Shul’pin, G. B. Chem. ReV. 1997, 97, 2879-2932.
(2) Stahl, S. S.; Labinger, J. A.; Bercaw, J. E. Angew. Chem., Int. Ed.
1998, 37, 2180-2192.
(3) Arndtsen, B. A.; Bergman, R. G.; Mobley, T. A.; Peterson, T. H.
Acc. Chem. Res. 1995, 28, 154-162.
(4) Thomas, J. M.; Raja, R.; Sankar, G.; Bell, R. G. Nature 1999, 398,
227-230.
(5) Raja, R.; Sankar, G.; Thomas, J. M. Angew. Chem., Int. Ed. 2000,
39, 2313-2316.
(6) Dugal, M.; Sankar, G.; Raja, R.; Thomas, J. M. Angew. Chem., Int.
Ed. 2000, 39, 2310-2313.
(7) ActiVation and Functionalization of Alkanes; Hill, C. L., Ed.; John
Wiley and Sons: New York, 1989.
10.1021/ja002840j CCC: $19.00 © 2000 American Chemical Society
Published on Web 11/01/2000

Metal Boryl Complexes and Alkane Borylation
J. Am. Chem. Soc., Vol. 122, No. 46, 2000 11359
of strongly acidic media.2,19 Sen used this chemistry to develop
the Pd-mediated oxidation of methane in trifluoroacetic acid.20,21
This process was made catalytic by use of hydrogen peroxide
as the stoichiometric oxidant in trifluoroacetic anhydride.22 The
selective oxidation of methane also was carried out using 5%
Pd/C as catalyst, dioxygen as oxidant, and soluble copper salts
as cocatalyst in a mixture of trifluoroacetic acid and water.23
Periana has developed a Pt system which converts methane to
methyl bisulfate in fuming sulfuric acid in high yield and high
selectivity.24 We recently reported the selective catalytic pho-
tochemical25 and thermal26 borylation of alkanes to produce
alkylboronate esters, and these discoveries stemmed from our
initial report of selective, stoichiometric borylation of alkanes.27
The organoboron products are versatile synthetic intermediates
that provide access to a variety of functionalized organic
compounds, such as alcohols, amines, ketones, aldehydes, and
alkyl halides.28 We report here a systematic synthetic and
mechanistic account of the photochemical functionalization of
alkanes by isolated transition metal boryl complexes.
Table 1. Carbonyl-Stretching Frequencies for Ruthenium Boryl,
Alkyl, and Hydride Complexesa
compound
ν )
CO (cm-1
Cp*Ru(CO)2[Bcat(Me)2] (2)
Cp*Ru(CO)2BS2Tol (9)
Cp*Ru(CO)2Bpin (5)
Cp*Ru(CO)2Hb
2012, 1952
2007, 1948
2002, 1940
2000, 1941
1998, 1935
1994, 1931
1984, 1921
Cp*Ru(CO)2Mec
Cp*Ru(CO)2BBN (8a)
Cp*Ru(CO)2BCy2 (8b)
a All IR spectra were recorded in benzene, except for 2 and 5 which
were recorded in benzene-d6. b From ref 76. c From ref 31.
about the M-B bond on the NMR time scale. In the case of
the tungsten monophosphine complex Cp*W(CO)2PXy3[Bcat-
(Me)2], we observed hindered rotation about the M-P bond on
1
the H and 13C NMR time scale. Instead of a single resonance
that would be expected for the six ortho aromatic hydrogens of
the xylyl group, two resonances with an intensity ratio of 2:1
were observed. In addition, the 18 protons of the xylyl methyl
groups appeared as two resonances, again with a 2:1 ratio of
integrated intensity. The 1H NMR spectrum of the molybdenum
analogue showed a similar pattern at room temperature;
however, the spectrum at 64 °C showed a single set of ligand
aromatic resonances, indicating that rotation around the W-P
bond occurred rapidly on the NMR time scale at this elevated
temperature. A possible structure of these complexes at room
temperature would contain one xylyl group aligned with the
metal Cp* axis and the other two xylyl groups canted sym-
metrically from the axis.
IR spectra of the complexes revealed electronic effects of
the metal boryl substituents. Table 1 provides IR stretching
frequencies for the various complexes, with some benchmarks
for the analogous alkyl and hydrido species. These data indicate
that the dialkylboryl groups are the strongest electron donors
overall, whereas the catecholboryl groups are the weakest
electron donors. The dithio- and pinacolboryl groups were
intermediate donors, with the pinacolboryl the stronger donor
of the two. The dialkylboryl groups appear to be stronger donors
and the dithio- and dialkoxyboryl groups weaker donors than
the methyl or hydride groups in Cp*Ru(CO)2Me and Cp*Ru-
(CO)2H. The pinacolboryl complex showed the closest νCO
values to those of Cp*Ru(CO)2H and Cp*Ru(CO)2Me.
Results
Synthesis and Characterization of Transition Metal Boryl
Complexes. The transition metal boryl complexes used for this
work on alkane functionalization were prepared by reacting the
appropriate metal carbonyl anion with haloborane in toluene
or pentane solvent (eq 1). The metal carbonyl anions were
prepared by Na/Hg reduction of the corresponding dimers, and
the metal phosphine carbonyl anions were prepared by depro-
tonation of the metal hydride using n-butyllithium. The pina-
colboryl complexes were prepared from the metal anion and
B-chloropinacolborane, which was prepared and used in situ.
The reactions were monitored by 11B NMR spectroscopy,29
which showed the disappearance of haloborane and the forma-
tion of the metal boryl complexes. The dialkoxyboryl, dialkyl-
boryl, and dithioboryl complexes resonated between δ 45 and
58, between δ 117 and 126, and at δ 78, respectively. The metal
boryl complexes were isolated from boron and metal impurities
by recrystallization and in some cases by sublimation.
Photochemical Reaction of Transition Metal Boryl Com-
plexes with Alkanes. Reaction of Metal Catecholboryl
Complexes with Pentane. Using our previous results on arene
functionalization as a starting point,30 we sought complexes that
(18) Shilov, A. E.; Shteinman, A. A. Coord. Chem. ReV. 1977, 24, 97-
143.
(19) Sen, A. Acc. Chem. Res. 1998, 31, 550-557.
(20) Sen, A. Acc. Chem. Res. 1988, 21, 421-428.
(21) Sen, A.; Gretz, E.; Oliver, T. F.; Jiang, Z. New J. Chem. 1989, 13,
755-760.
(22) Kao, L.; Hutson, A. C.; Sen, A. J. Am. Chem. Soc. 1991, 113, 700-
701.
(23) Lin, M.; Hogan, T.; Sen, A. J. Am. Chem. Soc. 1997, 119, 6048-
6053.
(24) Periana, R. A.; Taube, D. J.; Gamble, S.; Taube, H.; Satoh, T.; Fujii,
H. Science 1998, 280, 560-564.
(25) Chen, H.; Hartwig, J. F. Angew. Chem., Int. Ed. 1999, 38, 3391-
3393.
(26) Chen, H.; Schlecht, S.; Semple, T. C.; Hartwig, J. F. Science 2000,
287, 1995-1997.
(27) Waltz, K. M.; Hartwig, J. F. Science 1997, 277, 211-213.
(28) Pelter, A.; Smith, K.; Brown, H. C. Borane Reagents; Academic
Press: New York, 1988.
(29) No¨th, H.; Wrackmeyer, B. In NMR Basic Principles and Progress;
Diehl, P., Fluck, E., Kosfeld, R., Eds.; Springer-Verlag: New York, 1978;
Vol. 14; p 271.
In most cases, the 1H and 13C NMR spectra of the compounds
were straightforward. In all cases we observed free rotation

11360 J. Am. Chem. Soc., Vol. 122, No. 46, 2000
Waltz and Hartwig
Table 2. Yields of 1-Pentylboronate Ester and 1-Pentylborane
from Photochemical Reaction of Transition Metal Boryl Complexes
with Pentane
accessible aromatic C-H bonds. As a result, reaction of the
metal complex with alkane is observed. The complex 1b with
the smaller methyl groups to block the catecholate aromatic
C-H bonds gave a similar 20% yield. Only primary alkylbo-
ronate esters were formed from reaction of either 1a or 1b; no
products from functionalization at the secondary positions were
observed by GC. The 2- and 3-pentylboronate ester products
were prepared independently as a mixture of the two isomers
from hydroboration of 2-pentene. To determine if the blocking
groups were necessary on both Cp and catecholate moieties,
the complexes CpFe(CO)2[Bcat(tBu)2] (1c) and Cp*Fe(CO)2Bcat
(1d) were prepared. Neither of these species reacted with
pentane.
compound
% yield
CpFe(CO)2(Bcat)
<1
15
20
<1
<1
40
85
20
7
78
72
20
8
Cp*Fe(CO)2[Bcat(tBu)2] (1a)
Cp*Fe(CO)2[Bcat(Me)2] (1b)
CpFe(CO)2[Bcat(tBu)2] (1c)
Cp*Fe(CO)2(Bcat) (1d)
Cp*Ru(CO)2[Bcat(Me)2] (2)
Cp*W(CO)3[Bcat(Me)2] (3a)
Cp*W(CO)3Bcat (3b)
Cp*Mo(CO)3[Bcat(Me)2] (4)
Cp*Ru(CO)2Bpin (5)
Cp*W(CO)3Bpin (6)
Cp*Fe(CO)2BBN (7)
To improve the yield of alkylboronate ester, the ruthenium
analogues of iron complexes 1a and 1b were prepared. Cp*Ru-
(CO)2[Bcat(tBu)2] was difficult to isolate because of its high
solubility, but Cp*Ru(CO)2[Bcat(Me)2] (2) [Bcat(Me)2 ) B-1,2-
O2C6H2-3,5-Me2] was less soluble and more readily isolated.
Complex 2 reacted photochemically with pentane to give the
1-pentylboronate ester in 40% yield. This yield was measurably
higher than that for reaction of the analogous iron complex 1b.
Because the second-row ruthenium complex gave higher
yields of functionalized alkane than the first-row iron complex,
a third-row analogue was sought. Because Cp*W(CO)3- is more
Cp*Ru(CO)2BBN (8a)
Cp*Ru(CO)2BCy2 (8b)
Cp*Ru(CO)2BMe2 (8c)
Cp*Ru(CO)2BS2Tol (9)
Cp*W(CO)2(PXy3)[Bcat(Me)2] (10a)
Cp*W(CO)2(PMe3)[Bcat(Me)2] (10b)
<1
<1
<1
59
32
would react with alkanes and that would allow for an under-
standing of the factors that controlled their reactivity and
selectivity toward alkane substrates. Our results on the develop-
ment of isolated boryl complexes that functionalize alkanes are
summarized in eq 2 and Table 2. Transition metal boryl
accessible than Cp*Os(CO)2 ,
- 36,37 we studied boryl complexes
derived from the tungsten anion. The tungsten complex Cp*W-
(CO)3[Bcat(Me)2] (3a) showed remarkably high reactivity
toward pentane and formed 1-pentylboronate ester in 85% yield
with exclusive selectivity for reaction at the terminal position
of the alkane.
To determine if blocking groups were also important in this
highly reactive tungsten system, Cp*W(CO)3Bcat (3b) was
prepared. This complex did react with pentane, but the pentyl-
boronate ester was formed in only 20% yield. Again, this yield
was much lower than that obtained from reaction of the
catecholate-substituted complex 3a.
To compare metal boryl complexes of the chromium triad
directly with those of the iron triad, the molybdenum complex
Cp*Mo(CO)3[Bcat(Me)2] (4) was prepared. This complex
reacted with pentane to give pentylboronate ester in only 7%
yield. This yield is much lower than that obtained from reaction
of the analogous second-row ruthenium complex 2. This result
suggests that the lower-coordinate complexes of the iron triad
are more reactive toward alkanes than are the higher-coordinate
complexes in the chromium triad.
Reaction of Pinacol-, Dialkyl-, and Dithioboryl Complexes
with Pentane. To determine which features of the boryl com-
plexes lead to the alkane functionalization chemistry, we pre-
pared boryl complexes containing several different substituents
with varying electronic properties. These studies focused on the
ruthenium system because substantially lower or higher yields
versus those of dimethylcatecholate complex 2 could be
observed. In selected cases the iron and tungsten analogues also
were evaluated. Yields for reactions of these complexes are
included in Table 2.
complexes of the type (Cp′)M(CO)nB(OR)2 (Cp′ ) Cp or Cp*)
were irradiated with a medium pressure mercury arc lamp in
pentane solvent. The alkylboronate ester products from these
reactions were identified by comparison of spectral data of the
functionalized products to those of samples synthesized inde-
pendently by hydroboration of the corresponding alkenes. The
initial metal-containing products were the metal hydrides Cp′M-
(CO)nH. At early reaction times, Cp*Ru(CO)2H31 and Cp*W-
(CO)3H32 were observed by 1H NMR spectroscopy. These
complexes are photochemically unstable and decompose upon
33-35
irradiation into the dimeric species [Cp′M(CO)n]2,
which
were the metal products observed after all of the starting
complex had reacted.
Our studies were initiated with CpFe(CO)2-boryl complexes,
and this chemistry revealed structural features that were
important for observing chemistry with alkanes. In contrast to
its reaction with unsaturated hydrocarbons, irradiation of the
metal boryl complex CpFe(CO)2Bcat in alkane did not produce
functionalized product. The iron complex reacted completely,
but no dominant main group product was observed. In contrast,
the related complex Cp*Fe(CO)2[Bcat(tBu)2] (1a) [Bcat(tBu)2
) B-1,2-O2C6H2-3,5-tBu2] reacted with pentane under photo-
chemical conditions to form 1-pentylboronate ester in 15% yield.
This complex, and therefore the entire system, contains no
The pinacolboryl complex Cp*Ru(CO)2Bpin (pin ) 1,2-O2C2-
Me4) (5), which is analogous to the catecholboryl complex 2,
reacted photochemically with pentane to form the pentylboronate
ester in 78% yield. This value is significantly higher than that
for reaction of 2. To determine if the reactivity of the tungsten
pinacolboryl complex toward alkane was also high, Cp*W-
(30) Waltz, K. M.; Muhoro, C. N.; Hartwig, J. F. Organometallics 1999,
18, 3383-3393.
(31) Stasunik, A.; Wilson, D. R.; Malisch, W. J. Organomet. Chem. 1984,
270, C18-22.
(32) Kazlauskas, R. J.; Wrighton, M. S. J. Am. Chem. Soc. 1982, 104,
6005-15.
(33) Alt, H. G.; Mahmoud, K. A.; Rest, A. J. Angew. Chem., Int. Ed.
Engl. 1983, 22, 544-545.
(34) Mahmoud, K. A.; Rest, A. J.; Alt, H. G. J. Chem. Soc., Dalton
Trans. 1984, 187-197.
(36) Kawano, Y.; Tobita, H.; Ogino, H. Organometallics 1994, 13,
3849-3853.
(35) Shackleton, T. A.; Mackie, S. C.; Fergusson, S. B.; Johnston, L. J.;
Baird, M. C. Organometallics 1990, 9, 2248-2253.
(37) Hoyano, J. K.; May, C. J.; Graham, W. A. G. Inorg. Chem. 1982,
21, 3095-3099.

Metal Boryl Complexes and Alkane Borylation
J. Am. Chem. Soc., Vol. 122, No. 46, 2000 11361
Table 3. Photochemical Reaction of Cp*W(CO)2L[Bcat(Me)2]
Complexes with Hydrocarbons
Scheme 1
(CO)3Bpin (6) was prepared and irradiated in pentane solvent.
The tungsten pinacolboryl complex 6 reacted with pentane in
72% yield, which is similar in magnitude to that of tungsten
catecholboryl 3a.
Dialkylboryl complexes were much less reactive toward
alkanes. The ruthenium complex 8a containing a BBN (BBN
) 9-borabicyclo[3.3.1]nonyl) group reacted with pentane in a
low 8% yield. This product was identified by comparison of
spectral data with that of a sample prepared independently by
hydroboration of 1-pentene. Moreover, the dicyclohexyl and
dimethyl boryl complexes 8b and 8c did not produce any alkane
functionalization product, although the starting boryl complexes
were fully consumed within 1 h of irradiation. No dominant
boron-containing product was observed from these reactions.
We were concerned that the low yield resulted from metal-
catalyzed decomposition of the trialkylborane product. To check
the stability of the product, pure pentylBBN was irradiated in
benzene-d6 in the presence of [Cp*Ru(CO)2]2. No decomposition
was observed after 2 h of irradiation, which was the time
required for reaction of 8a. To determine if the low yields were
specific for the ruthenium system, the iron complex Cp*Fe-
(CO)2BBN (7) was reacted with pentane under photochemical
conditions. This reaction formed pentylBBN in 20% yield,
which is slightly higher than that for the ruthenium BBN com-
plex, but similar to that for the iron catecholboryl complexes.
To compare the reactivity of a boryl complex with a structure
more similar to the catecholboryl complex 2, but with increased
Lewis acidity, the dithioboryl complex Cp*Ru(CO)2BS2Tol (9)
(BS2Tol ) B-1,2-S2C6H3-4-Me) was prepared. Upon irradiation
in pentane, the dithioboryl complex did not form pentylboronate
ester. To distinguish between low reactivity of the intermediate
formed from photolysis and low yield for formation of the
intermediate, complex 9 was irradiated in benzene-d6. In this
case, complex 9 produced phenyl-d5 dithioborane in 57% yield.
This arene functionalization product was identified by com-
parison with an independently prepared sample.38 To test again
the product stability toward the reaction conditions and medium,
the phenylboronate dithioester was irradiated in the presence
of [Cp*Ru(CO)2]2. Again, no significant decomposition occurred
after 1 h (the time scale of the reaction), but extended irradiation
for 12 h did result in decomposition of about half of the
phenylboronate dithioester.
in functionalization of the terminal position of the ethyl group
in 74% yield and showed no functionalization of the cyclohexyl
ring. Reaction of 3a with cyclohexane gave the cyclohexylbo-
ronate ester in only 22% yield. Reaction of 3a with arenes
occurred in high yields. Irradiation in benzene produced 86%
yield of phenylboronate ester.
Reaction of Cp*W(CO)2L[Bcat(Me)2] (L ) phosphine)
with Alkanes and Arenes. Tungsten boryl complexes contain-
ing one phosphine ligand were also capable of functionalizing
a variety of branched or unbranched aliphatic and aromatic
hydrocarbons, as shown in Tables 2 and 3. Cp*W(CO)2(PXy3)-
[Bcat(Me)2] (10a), which contains a bulkier triarylphosphine,
reacted with pentane to give functionalized product in 59% yield,
which is lower than that for reaction of tricarbonyl complex
3a. Cp*W(CO)2(PMe3)[Bcat(Me)2] (10b), which contains the
smaller, more basic PMe3 ligand, produced the 1-pentyl boronate
ester in an even lower 32% yield. Complexes 10a and 10b
showed similar selectivity upon reaction with 2-methylbutane,
forming the two terminally functionalized isomers in a 10((1):1
ratio, with functionalization favoring the less sterically hindered
terminal methyl group.
To determine if the low yield from reaction of 10b with
pentane was due to low reactivity of the intermediate formed
from photolysis or low yield for formation of the intermediate,
complex 10b was irradiated in benzene-d6. Phenyl-d5Bcat(Me)2
was formed in 54% yield. For comparison, the tungsten
tricarbonyl boryl 3a was also photolyzed in benzene-d6, afford-
ing the arylboronate ester in a considerably higher 84% yield.
Reaction of either 10a or 10b with toluene formed the same
1:6:1 ratio of meta and para functionalized products, uncorrected
for the different number of meta and para C-H bonds.
Mechanistic Studies. To obtain mechanistic information,
kinetic isotope effects were measured for several of the metal
boryl complexes, as shown in Table 4. Photochemical reaction
of the metal boryl complexes with an equimolar mixture of
pentane and pentane-d12 resulted in a mixture of pentyl[Bcat-
(Me)2] and pentyl-d11[Bcat(Me)2], as shown in eq 3. The ratio
Reaction of Cp*W(CO)3[Bcat(Me)2] with Nonlinear Al-
kanes and Arenes. The reaction of tungsten complex 3a with
several nonlinear alkanes was studied, as shown in Scheme 1.
Reaction of 3a with 2-methylbutane resulted in functionalization
of the least hindered terminal position of the alkane in 53%
yield. A small amount of functionalization at the more hindered
terminal position was observed. The ratio of these two products
was 11((1):1. Reaction of 3a with ethylcyclohexane resulted
(38) Wieber, M.; Ku¨nzel, W. Z. Anorg. Allg. Chem. 1974, 403, 107-
115.
of these products, which was measured by GC/MS, provides

11362 J. Am. Chem. Soc., Vol. 122, No. 46, 2000
Waltz and Hartwig
Table 4. Kinetic Isotope Effects for Photochemical Reaction of
Metal Boryl Complexes with Pentane/Pentane-d12
at determining if the phosphine or carbonyl group dissociated
from 10b. Reaction in the presence of 10 equiv of added PMe3
led to slightly lower conversions as a function of time than a
side-by-side reaction in the absence of added PMe3. More
striking, the sample with added phosphine gave much lower
yields of functionalized product. After consumption of 10b, no
major species that could be attributed to a bisphosphine complex
Cp*W(CO)(PMe3)2[Bcat(Me)2] were observed.
metal boryl complex
kH/kD
Cp*Fe(CO)2[Bcat(Me)2 (1b)
Cp*Ru(CO)2[Bcat(Me)2] (2)
Cp*W(CO)3[Bcat(Me)2] (3a)
Cp*W(CO)2PXy3[Bcat(Me)2] (10a)
Cp*W(CO)2PMe3[Bcat(Me)2] (10b)
1.9
2.2
5.1
4.9
5.1
Discussion
the kH/kD for reaction of the photochemically generated inter-
mediate with alkane. All of the boryl systems showed primary
kinetic isotope effects. The iron and ruthenium complexes 1b
and 2 showed values close to 2. For tungsten complexes 3a,
10a, and 10b, an isotope effect of 5.0 ( 0.1 was observed,
regardless of whether CO, PMe3, or PXy3 was the third dative
ligand.
Several experiments were performed to probe for reversible
CO dissociation. Tungsten boryl complex 3a was photolyzed
in pentane solvent under 2 atm of CO side by side with an
identical sample without added CO. The reactions showed the
same conversions as a function of irradiation time; thus, no
inhibition was observed in the presence of added CO. Complex
3a was also irradiated in pentane solvent with 2 atm of added
13CO gas. After half of the starting material had been consumed,
labeled CO was not observed in the remaining starting material,
but was detected in the metal carbonyl products Cp*W(CO)3H
and [Cp*W(CO)3]2. This incorporation of labeled CO into the
products may occur through CO dissociation during the reaction
or from CO exchange with the tungsten hydride product.
Photolysis of Cp*W(CO)3H alone in pentane in the presence
of 13CO resulted in formation of labeled Cp*W(CO)3H and
[Cp*W(CO)3]2. Thus, 13CO may be incorporated before or after
the formation of Cp*W(CO)3H.
Effect of Boryl Group on Reactivity. Steric Effects. Studies
on metal boryl complexes that photochemically functionalize
arenes30 were used as a starting point to develop complexes
that would react with alkanes to form alkylboronate esters.
Simple metal catecholboryl complexes with the formula CpM-
(CO)nBcat did not react with alkanes. Attack of one metal boryl
complex on the sp2-hybridized C-H bonds of the catecholate
or cyclopentadienyl groups of another metal boryl complex
would explain the lack of reactivity of CpFe(CO)2Bcat and
related compounds with alkanes.
To prevent this reactivity at sp2-hybridized C-H bonds, Cp*
was used as ligand, and a 3,5-dialkylcatecholate substituent was
placed on the boryl group. The presence of only aromatic C-H
bonds with two ortho substituents should prevent reaction at
the catecholate group. Metal catecholboryl complexes, such as
CpFe(CO)2Bcat, were shown to react with substituted arenes
predominantly in the m- and p- positions.30 Two ortho substit-
uents may be necessary because the position ortho to an oxygen
substituent is unusually reactive toward certain metal-boryl
complexes. For example, CpFe(CO)2Bcat reacted with anisole
at the ortho, meta, and para positions, and Re(CO)5Bcat reacted
preferentially at the ortho position.30 As designed, Cp*Fe(CO)2-
[Bcat(tBu)2], Cp*Fe(CO)2[Bcat(Me)2], Cp*Ru(CO)2[Bcat(Me)2],
and Cp*W(CO)3[Bcat(Me)2] all reacted with alkanes to form
alkylboronate esters.
To test whether blocking of the sp2-hybridized positions was
truly the important factor that allowed for reactivity with alkanes,
a variety of boryl complexes with different substitution patterns
were prepared and reacted photochemically with pentane
solvent. Within the set of complexes CpFe(CO)2[Bcat(tBu)2]
(1c), Cp*Fe(CO)2Bcat (1d), and Cp*Fe(CO)2[Bcat(tBu)2] (1a),
only complex 1a, with both cyclopentadienyl and catecholate
blocking groups, resulted in functionalization of pentane. In
addition, a comparison of the reactions of pentane with tungsten
dimethylcatecholboryl complex 3a and with complex 3b, which
contains unsubstituted catecholate, showed that 3a gave a high
yield of pentylboronate ester, but 3b gave low yields of the
functionalized product. The high reactivity of the tungsten
complex toward alkanes, in conjunction with the vast difference
in concentration of alkane versus catecholate C-H bonds, did
allow for some functionalization of alkane C-H bonds by
compound 3b.
These experiments were more revealing than those directed
toward isolating or detecting borylated metal complex in reaction
mixtures. Although the experiments below argue strongly that
accessible sp2 sites on the metal boryl complex interfere with
alkane functionalization, the products formed by the proposed
borylation of the catecholate ring of CpFe(CO)2Bcat, 1d, or 3b
were not observed directly. Of course, this borylated metal boryl
species would most likely display similar photochemistry to that
of the starting complex, and complex reaction mixtures would
result. In a related system, however, borylation of the cyclo-
pentadienyl ligands of CpM(CO)3 catalysts was observed. The
irradiation of CpRe(CO)3 with B2pin2 in pentane25 formed the
Irradiation of 3a was performed with added phosphine to
provide more definitive evidence for or against CO dissociation.
Photolysis of tungsten boryl complex 3a in pentane solvent in
the presence of PMe3 resulted in formation of phosphine
complex 10b and pentyl[Bcat(Me)2]. A side-by-side photolysis
of 3a in pentane with 2 equiv and 4 equiv of PMe3 resulted in
roughly twice the amount of 10b when the concentration of
PMe3 was doubled, as shown in eq 4. These ratios were
measured when half of the starting material had reacted, and
these data provide strong evidence for reaction that is initiated
by dissociation of CO, as detailed in the Discussion section.
The identity of 10b was confirmed by independent synthesis
and complete characterization. A single CO resonance in its
13C NMR spectrum, along with a higher frequency symmetric
CO band of medium intensity and a lower frequency asymmetric
band of stronger intensity,39 suggested the trans geometry. This
complex could exist as a mixture of rapidly interconverting cis
and trans isomers; however, this process is slow for similar
CpW(CO)2PR3X complexes, and we observed only one set of
infrared CO bands.
In addition to studies on selectivities of 3a, 10a, and 10b,
we performed reactions with added phosphine that were aimed
(39) Cotton, F. A.; Wilkinson, G. AdVanced Inorganic Chemistry, 5th
ed.; Wiley & Sons: New York, 1988; pp 1035-1037.

Metal Boryl Complexes and Alkane Borylation
J. Am. Chem. Soc., Vol. 122, No. 46, 2000 11363
product from borylation of the cyclopentadienyl ring of CpRe-
(CO)3, in addition to forming alkylboronate ester. Similar
borylation of a cyclopentadienyl group was observed in ana-
logous reactions mediated by (MeCp)Mn(CO)3. In these cases
the borylated metal complexes were stable enough to be
identified in the reaction mixture by GC/MS.25
Although we cannot provide detailed arguments for the
differences in reactivity between complexes of all the boryl
groups, several general conclusions can be drawn from the
studies of electronic effects using the ruthenium complexes.
First, the electronic properties of the boryl group did not
dramatically alter the photochemical generation of the pro-
posed 16-electron intermediate. Instead, the electronic prop-
erties of the boryl ligands influenced the reactivity of the
intermediate toward alkane substrate. Second, a highly Lewis
acidic boryl group is not necessary for alkane functionalization;
among the ruthenium complexes, the least Lewis acidic boryl
complex, pinacolboryl 5, gave the highest yields of function-
alized alkane.
Effect of Dative Ligands on Reactivity. In addition to
examining the effect of the boryl group on reactivity, we
examined the effect of altering the dative ligand. The reactions
of tungsten tricarbonyl and phosphine carbonyl complexes 3a,
10a, and 10b with several substrates were examined. The
reactions could occur through dissociation of one dative ligand,
followed by C-H bond cleavage and B-C bond formation. If
phosphine complexes 10a and 10b undergo dissociation of
phosphine, the same 16-electron intermediate Cp*W(CO)2[Bcat-
(Me)2] would be formed as in reactions of complex 3a. In this
case, all three complexes would show similar selectivities,
particularly if these selectivities were determined by steric
effects. Moreover, added phosphine should inhibit the formation
of pentylboronate ester. If the reaction proceeds through CO
dissociation, three different 16-electron intermediates of the
general formula Cp*W(CO)(L)[Bcat(Me)2] would be formed,
where L ) CO, PMe3, and PXy3. In this case, the three different
intermediates would give different selectivities because of the
large difference in steric properties between CO, PMe3, and
PXy3. Moreover, added phosphine should convert 10b into the
bis-phosphine complex Cp*W(CO)(PMe3)2[Bcat(Me)2] in com-
petition with formation of pentylboronate ester.
Previous studies of the photochemistry of mixed carbonyl
phosphine systems suggest that the lability of the CO or
phosphine ligand upon irradiation depends on the identity of
the metal and covalent ligand. Photolysis of Cp*Re(CO)2(PMe3)
led to CO dissociation, but irradiation of Cp*Re(CO)(PMe3)2
led to phosphine dissociation.11 Photolysis of CpFe(CO)(PPh3)-
(COMe) resulted in CO dissociation,40 but irradiation of the
analogous tungsten acetyl complex CpW(CO)2(PPh3)(COMe)
resulted in phosphine dissociation.41 Moreover, the Cp*, instead
of Cp analogue of the iron acetyl complex, and the methyl,
instead of acetyl, analogue underwent phosphine dissociation
upon photolysis.40 Thus, it is difficult to predict if CO or
phosphine will dissociate from complexes 10a and 10b.
Our experiments suggest that the functionalization chemistry
of 3a, 10a, and 10b occurred predominantly through the same
Cp*W(CO)2[Bcat(Me)2] intermediate, which would be formed
by CO, PMe3, or PXy3 dissociation. The reaction of PXy3 and
PMe3 complexes 10a and 10b with isopentane gave similar
ratios of isomeric products as those obtained from reactions of
3a with isopentane. Moreover, the m-:p- ratios from reaction
of 10a and 10b with toluene were identical. Complexes 3a, 10a,
and 10b also gave isotope effects that were indistinguishable
from each other when these three complexes were photolyzed
with a mixture of pentane and pentane-d12. Finally, the presence
of added PMe3 in photochemical reactions of 10b in pentane
did not lead to the formation of the bis-phosphine complex
Electronic Effects. In addition to probing the ability of steric
effects to inhibit reaction at aromatic C-H bonds, we studied
the influence of the boryl groups’ electronic properties on alkane
functionalization. One might propose that the boryl complexes
are highly reactive toward alkane C-H bonds because the boryl
groups are strong σ-donors; alternatively, one might propose
that they are reactive because of Lewis acidic character. To
address these issues, the reactivities of metal boryl complexes
containing the groups B(OR)2, B(SR)2, and BR2 (R ) alkyl)
were examined. The electron-donating ability of these boryl
groups can be gauged by the CO stretching frequencies of their
complexes with Cp*Ru(CO)2, shown in Table 1. As mentioned
in the Results section, the dialkylboryl groups are apparently
the strongest electron donors, while the catecholboryl group is
the weakest electron donor. The electron-donating ability of the
dialkylboryl complexes presumably results from strong σ-dona-
tion. The electronics of the boryl groups can also be ordered
according to Lewis acidity, based on conventional properties
of the analogous boranes. The dialkylboryl groups would be
the most Lewis acidic, while the pinacolboryl groups would be
the least Lewis acidic.
To evaluate the electronic effects of the boryl group, we
focused on the ruthenium boryl complexes. The ruthenium
pinacolboryl complex 5 reacted with pentane in twice the yield
as the catecholboryl complex 2. This large increase in yield
may result from either the increased electron donation or
decreased Lewis acidity of the pinacolboryl moiety relative to
the catecholboryl group. A similar comparison between tungsten
pinacolboryl 6 and catecholboryl 3a showed that both complexes
provided clean reactions with alkanes.
In contrast to catecholboryl 2, tungsten dithiocatecholboryl
9 did not form any pentylboronate ester upon irradiation in
pentane. Yet, complex 9 did react photochemically with ben-
zene to form PhBS2Tol in moderate yield. The functionaliza-
tion of benzene by 9 shows that the reactive intermediate was
successfully formed upon irradiation; thus, the inability of
9 to functionalize alkanes most likely results from low reac-
tivity of the metal boryl intermediate toward alkane C-H bonds,
and not from a low yield for formation of the reactive
intermediate.
In addition to pinacol- and dithioboryl ligands, we studied
dialkylboryl groups that would be more electron-donating, and
also more Lewis acidic, than catecholboryl ligands. The lack
of C-H bonds of sp2-hybridized sites on these ligands should
make them suitable components of complexes that functionalize
alkanes. However, the BBN complex 8a was the only dialkyl-
boryl complex among the ruthenium dialkylboryl series that
formed any pentylboronate ester, and even this complex formed
functionalized product with yields that were much lower than
those obtained with the catecholboryl complex 2. The analogous
iron BBN complex 7 also gave low yields upon photochemical
reaction with pentane. Yet, complex 7 reacted with arenes to
form arylboranes,30 again demonstrating that the photochemistry
generates a 16-electron intermediate, but that the intermediate
is unreactive toward alkanes. The low yields for reactions of
all iron boryl complexes with alkanes precluded any firm
conclusions concerning the effect of electronic variations on
the reactions of the iron complexes.
(40) Aplin, R. T.; Booth, J.; Compton, R. G.; Davies, S. G.; Jones, S.;
McNally, J. P.; Metzler, M. R.; Watkins, W. C. J. Chem. Soc., Perkin Trans.
2 1999, 5, 913-922.
(41) Tyler, D. R. Inorg. Chem. 1981, 20, 2257-2261.

11364 J. Am. Chem. Soc., Vol. 122, No. 46, 2000
Waltz and Hartwig
Cp*W(CO)(PMe3)2[Bcat(Me)2] and led to lower yields of
functionalized product.
Scheme 2
However, the variation in reaction yield suggests that a
parallel path involving CO dissociation from 10a and 10b leads
to a reduction in yield. Reactions of 10a, 10b, and 3a showed
significant differences in yields, but similar regioselectivities
and isotopic effects. Although there are other explanations for
the effect of dative ligands, these results may be rationalized
by competitive formation of Cp*W(CO)2[Bcat(Me)2] and
Cp*W(CO)(L)[Bcat(Me)2] as intermediates and by a low
reactivity of Cp*W(CO)(PR3)[Bcat(Me)2] toward alkanes. If
Cp*W(CO)(PR3)[Bcat(Me)2] is unreactive toward alkanes then
the regioselectivities and isotope effects for formation of func-
tionalized product by 3a, 10a, and 10b would not be perturbed
by generation of the unsaturated, phosphine-ligated intermediate,
but the reaction yields would.
Scheme 3
Photochemical reactions of 3a and 10b with benzene-d6 are
consistent with this conclusion. Tricarbonyl boryl 3a reacted
with benzene-d6 in higher yield than did the monophosphine
boryl 10b. In addition, the yield for reaction of 10b with arene
is higher than that for its reaction with alkane. The yields for
reaction of 3a with alkanes and arenes are similar. These data
are consistent with a similar high yield for reaction of the Cp*W-
(CO)2[Bcat(Me)2] intermediate with alkanes and arenes, but
substantially higher yields for reaction of the Cp*W(CO)(L)-
[Bcat(Me)2] intermediate with arenes than with alkanes. The
greater yields for reaction of Cp*W(CO)(L)[Bcat(Me)2] with
alkanes when L)CO instead of a phosphine run counter to the
general observation that increased electron density at the metal
favors oxidative addition. However, these reactions may be
highly sensitive to steric effects and cyclometalation of the phos-
phine may compete with the intermolecular C-H activation.
Mechanistic Considerations. Probe for Radical Chemistry.
Additional mechanistic information was obtained through
measurement and comparison of kinetic isotope effects for
reaction of the iron, ruthenium, and tungsten [Bcat(Me)2]
complexes 1b, 2, and 3a with a mixture of pentane and pentane-
d12. If the functionalization were to occur through a free [Bcat-
(Me)2] radical, then the values for the isotope effect would be
the same for all three systems, because the reactivity of the boryl
radical should be independent of its origin. The different values
obtained for reactions of complexes 1b, 2, and 3a indicate that
the alkane functionalization does not occur by reaction of a free
boryl radical with free alkane.
irradiation of 3a in pentane in the presence of 13CO showed no
formation of labeled starting material Cp*W(CO)2(13CO)[Bcat-
(Me)2]. One could argue that the lack of CO inhibition and
incorporation could be the result of poor CO solubility in alkane
solvent. The solubility of 1 atm of CO in heptane at room
temperature has been reported as 0.012 M.42 Our experiments
used 2 atm of CO, and the concentration of tungsten boryl was
about 0.018 M. We estimate that only 10% incorporation of
labeled CO would be necessary to detect 13CO incorporation
by 13C NMR spectroscopy. We observed labeled CO in two
different metal products, and their concentrations were less than
the concentration of the remaining tungsten boryl after 50%
conversion. Thus, we would detect reversible CO dissociation
with this experiment, and our experiments indicate that CO does
not dissociate from 3a reversibly.
However, the formation of pentylboronate ester and phosphine
complex 10b upon photolysis of tungsten complex 3a in the
presence of PMe3 in pentane indicated that irreversible dis-
sociation of CO did occur. The ratio of 10a to alkylboronate
ester depended directly on the concentration of added phosphine.
Because the amounts of these products depended on the amount
of added phosphine, the pentylboronate ester and trapped
intermediate 10b most likely originate from the same intermedi-
ate, as summarized in Scheme 2. The likely intermediate is the
16-electron species Cp*W(CO)2[Bcat(Me)2]. The trapped com-
plex 10b reacted photochemically with pentane, but this process
was inhibited by the presence of added phosphine. Thus, the
pentylboronate ester, formed in competition with 10b in the
presence of PMe3, originated predominately from reaction of
the starting material 3a with alkane.
The exceptional selectivity of metal boryl complexes toward
C-H bonds of terminal, unhindered methyl groups also provides
mechanistic information. C-H activation processes that occur
through radical pathways typically show the highest reactivity
at tertiary C-H bonds at tertiary sites, followed by secondary,
and finally primary. All of the reactions of metal boryl com-
plexes with alkanes show functionalization exclusively at the
C-H bonds of primary carbons. In the absence of primary C-H
bonds, complex 3a does react with secondary C-H bonds but
in low yields. This observed selectivity argues against a pathway
involving radical chemistry and supports the conclusion of a
nonradical mechanism drawn from the kinetic isotope effects.
The most plausible mechanistic pathway suggested by these
data is described in Scheme 3 using 3a as an example.
Photochemical CO dissociation occurs to form the reactive 16-
electron intermediate Cp*W(CO)2[Bcat(Me)2]. This 16-electron
intermediate reacts with substrate much faster than it reassociates
CO. The 16-electron species most likely undergoes addition of
an alkane C-H bond to form an unusual cyclopentadienyl
tungsten (IV) carbonyl complex, which would reductively
eliminate alkylboronate ester. The remaining metal hydride
fragment Cp*W(CO)2H would then recombine with CO to form
the observed metal product Cp*W(CO)3H.
Probe for CO Dissociation. The reaction mechanism can
be envisioned to occur through CO dissociation to form a 16-
electron intermediate or through a direct sigma bond metathesis.
A series of experiments were conducted to determine if CO
dissociation occurred on the reaction pathway. Reaction of
Cp*W(CO)3[Bcat(Me)2] (3a) with pentane in the presence of
added CO showed no difference in the conversion to pentyl-
boronate ester as a function of irradiation time. Moreover,
(42) Gerrard, W. Solubility of Gases and Liquids; Plenum Press: New
York, 1976; p 73.

Metal Boryl Complexes and Alkane Borylation
J. Am. Chem. Soc., Vol. 122, No. 46, 2000 11365
Alternatively, the 16-electron intermediate could react with
an alkane C-H bond through σ-bond metathesis.12 This type
of mechanism is typical of early transition metal, lanthanide,
and actinide systems which cannot undergo oxidative addition.3
Although a σ-bond metathesis mechanism is unconventional for
mid- to late-metal systems, the electrophilicity of boron may
allow this main group site to play the role of the early transition
metal center, resulting in a σ-bond metathesis with the opposite
regiochemistry of conventional σ-bond metatheses.
Nevertheless, we favor a mechanism involving oxidative
addition. Bergman has reported the low-temperature activation
of alkanes by [Cp*Ir(PMe3)(CH3)]+.43 Theoretical calculations
on a model of this complex showed that oxidative addition is a
low-energy pathway for reaction of Ir(III) to form Ir(V), and
that σ-bond metathesis is unlikely to occur.44 Bergman’s direct
observation of C-H oxidative addition at an Ir(III) center to
form an isolable Ir(V) species has provided further experimental
evidence to support these calculations.45
Finally, the thermodynamics for alkane borylation are unusu-
ally favorable. Because all transformations appear to occur
thermally after photochemical ligand dissociation, an examina-
tion of the bond dissociation energies (BDEs) involved, albeit
for the saturated 18-electron analogue of the 16-electron
intermediate, can reveal why alkane borylation is favorable
thermodynamically. The BDEs for W-H, C-B, and C-H
bonds have been measured or calculated previously, and a value
for the W-B bond may be estimated based on the reported 5
kcal/mol difference in Ir-B and Ir-H bond energies.50 For this
analysis, only relative BDEs are important; thus, the roughly 5
kcal/mol difference estimated between the W-B and W-H
BDEs is the important value, not the absolute BDE. The total
energy of the bonds that are formed [68 kcal/mol (W-H)59
+
112 kcal/mol (C-B)50] is then 9 kcal/mol greater than the energy
of the bonds that are broken [∼73 kcal/mol (W-B) + 98 kcal/
mol (C-H)60]. Thus, the strength of the B-C bond dominates
the thermodynamics and provides the driving force for the
reactivity of the boryl complexes with alkanes. Even if the
oxidative addition of a C-H bond to the Fe(II), Ru(II), or W(II)
intermediate is endothermic, the highly exothermic combination
of reductive elimination to form the B-C bond and reassociation
of CO would make the overall reaction exothermic.
Conclusions
We conclude by reflecting on the features of the boryl group
that allow transition metal boryl complexes to functionalize
alkanes, while the analogous metal alkyl complexes do not.
Upon irradiation of metal alkyl complexes containing â-hydro-
gens, â-H elimination of the alkyl group is known to occur after
CO dissociation.32,46,47 The pinacol and catecholboryl complexes,
which provided the best yields of functionalized alkane, cannot
undergo â-hydrogen elimination. Irradiation of metal alkyl
complexes is also known to generate alkyl radicals.46,48,49 We
showed previously that metal-boron bonds are unusually
strong50 and in the present work that metal boryl complexes do
not react through radical chemistry. It is likely that the strong
metal-boron bond prevents the generation of boryl radicals.
Thus, the major pathways for decomposition of metal-carbonyl
alkyl complexes are inaccessible for metal boryl complexes.
We propose that the modest electrophilic properties of the
boron in the boryl group assist in providing favorable kinetics
for B-C bond formation. The coupling of electrophilic and
nucleophilic groups from a metal center are favorable. For
example, acyl-alkyl couplings are faster than alkyl-alkyl
couplings.51 Favorable matching of nucleophile and electrophile
have also been demonstrated in C-N, C-S, and C-O coupling
processes.52-54 Thus, the electrophilicity of the boron may make
coupling of the alkyl and boryl groups facile. Rapid B-C bond
formation has been implied by a number of experimental ob-
servations in other systems.55,56 Alkylmetal boryl complexes are
unknown, and aryl or vinylmetal boryl complexes are rare.57,58
On the basis of this logic, the thermal dissociation of a dative
ligand would provide a route to alkane functionalization that
occurs in the absence of photochemical irradiation. This
prediction was recently fulfilled by the development of a
rhodium complex which catalyzes the borylation of alkanes
under purely thermal conditions.26
Experimental Section
General Methods. Unless otherwise noted, all manipulations were
conducted in an inert atmosphere glovebox or by using standard Schlenk
techniques. 1H NMR spectra were recorded on either a General Electric
QE-300 or Bruker AM-500 NMR spectrometer, and 11B- and 31P{H}
NMR spectra were recorded on an Omega 300 NMR spectrometer.
11B- and 31P{H} 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 assigned positive chemical
shifts. Proton chemical shifts are reported in ppm relative to residual
protiated solvent as internal standard. Gas chromatography was
performed on an HP5890A GC fitted with a 30 m × 0.25 mm ID
capillary column coated with a 0.25 µm film of EC-5 and connected
to a 2 m deactivated fused silica guard column. The GC was connected
to an HP5971A mass spectrometer. Runs were ramped from 50 to 210
°C at 10 °C/min. UV/vis spectra were acquired on a Cary 3E UV/vis
spectrophotometer. Pentane, THF, benzene, and toluene solvents were
distilled from sodium/benzophenone ketyl prior to use. Alkanes for
use as substrates were treated with concentrated sulfuric acid to remove
olefin impurities and then washed with a saturated NaHCO3 solution
and water. The olefin-free alkanes were dried first with MgSO4 and
then with sodium/benzophenone ketyl, and were finally transferred by
vacuum techniques and degassed. Benzene-d6 was dried over sodium/
benzophenone ketyl and degassed before use. The complexes CpFe-
(CO)2Bcat, Cp*Fe(CO)2Bcat (1d),30 [Cp*Fe(CO)2]2,61 [Cp*Ru(CO)2]2,62,63
(43) Arndtsen, B. A.; Bergman, R. G. Science 1995, 270, 1970-1973.
(44) Strout, D. L.; Zarie´, S.; Niu, S.; Hall, M. B. J. Am. Chem. Soc.
1996, 118, 6068-6069.
(45) Klei, S. R.; Tilley, T. D.; Bergman, R. G. J. Am. Chem. Soc. 2000,
2000, 1816-1817.
(46) Alt, H. G. Angew. Chem., Int. Ed. Engl. 1984, 23, 766-782.
(47) Mahmoud, K. A.; Rest, A. J.; Alt, H. G.; Eichner, M. E.; Jansen,
B. M. J. Chem. Soc., Dalton Trans. 1984, 175-186.
(48) Giese, G.; Thoma, G. HelV. Chim. Acta 1991, 74, 1143-1155.
(49) Gismondi, T. E.; Rausch, M. D. J. Organomet. Chem. 1985, 284,
59-71.
(50) Rablen, P. R.; Hartwig, J. F.; Nolan, S. P. J. Am. Chem. Soc. 1994,
116, 4121-4122.
(51) Collman, J. P.; Hegedus, L. S.; Norton, J. R.; Finke, R. G. Principles
and Applications of Organotransition Metal Chemistry, 2nd ed.; University
Science Books: Mill Valley, 1987; pp 332-333.
(52) Driver, M. S.; Hartwig, J. F. J. Am. Chem. Soc. 1997, 119, 8232-
8245.
(55) Baker, R. T.; Calabrese, J. C.; Westcott, S. A.; Marder, T. B. J.
Am. Chem. Soc. 1995, 117, 8777-8784.
(56) Baker, R. T.; Calabrese, J. C. J. Am. Chem. Soc. 1993, 115, 4367-
4368.
(57) Rickard, C. E. F.; Roper, W. R.; Williamson, A.; Wright, L. J.
Angew. Chem., Int. Ed. 1999, 38, 1110-1113.
(58) Knorr, J. R.; Merola, J. S. Organometallics 1990, 9, 3008-3010.
(59) Barbini, D. C.; Tanner, P. S.; Francone, T. D.; Furst, K. B.; Jones,
J., Wayne E. Inorg. Chem. 1996, 35, 4017-4022.
(53) Baran˜ano, D.; Hartwig, J. F. J. Am. Chem. Soc. 1995, 117, 2937-
2938.
(54) Widenhoefer, R. A.; Buchwald, S. L. J. Am. Chem. Soc. 1998, 120,
6504-6511.
(60) McMillen, D. F.; Golden, D. M. Annu. ReV. Phys. Chem. 1982, 33,
493-532.
(61) Catheline, D.; Astruc, D. Organometallics 1984, 3, 1094-1100.
(62) Nelson, G. O.; Sumner, C. E. Organometallics 1986, 5, 1983-1990.

11366 J. Am. Chem. Soc., Vol. 122, No. 46, 2000
Waltz and Hartwig
Cp*Ru(CO)2H,31 Cp*W(CO)3H,32 Cp*Mo(CO)3H,64 Li[Cp*W(CO)2-
PMe3],65 Cl[Bcat(tBu)2],66 ClBS2Tol,67 PhBS2Tol,38,68 3,5-dimethylcat-
1-[Bcat(tBu)2]pentane: 1H NMR (C6D6) δ 7.22 (d, J ) 1.6 Hz, 1H),
7.19 (d, J ) 1.6 Hz, 1H), 4.27 (s, 5H), 1.52 (s, 9H), 1.35-1.13 (m,
15H), 1.60 (t, J ) 7.3 Hz, 2H); 13C{1H} NMR (C6D6) δ 149.48, 146.12,
144.93, 135.22, 116.66, 108.43, 35.41 (4 °C of tBu), 35.08 (CH2), 34.95
(4 °C of tBu), 32.29 (3 Me of tBu), 30.33 (3 Me of tBu), 24.32 (CH2),
23.18 (CH2), 14.58 (CH3), 11.33 (CH2B); 11B NMR (C6D6) δ 36.
Direct Hydroboration of Alkene. The following compounds were
prepared by heating borane in neat alkene according to the literature
procedure.75
echol,69 and PXy3 were prepared according to literature procedures.
70
All other chemicals were used as received from commercial suppliers.
We report the 1H NMR spectrum of PhBcatS2Tol in benzene-d6,
complete with coupling constants, which have not appeared previously
in the literature. Photolyses were carried out with a medium pressure
450-W Hanovia mercury arc lamp in a quartz immersion well using
Pyrex reaction vessels placed flush with the immersion well, ap-
proximately 0.75 in. from the lamp.
1-[Bcat(Me)2]pentane: 1H NMR (C6D6) δ 6.83 (s, 1H), 6.51 (s, 1H),
2.25 (s, 3H), 2.10 (s, 3H), 1.56 (m, 2H), 1.35-1.08 (m, 6H), 0.83 (t,
J ) 6 Hz, 3H); 11B NMR (C6D6) δ 37; GC/MS (16.07 min) m/z ) 218
((M+), 173, 160, 148, 133, 91. HRMS (EI) calcd for C13H19BO2:
218.1478. Found: 218.1470.
Preparation of Anionic Metal Complexes in THF. Na[CpFe-
(CO)2], Na[Cp*Fe(CO)2], and Na[Cp*W(CO)3]. 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. The THF was removed as described
below, resulting in quantitative yield of the orange/brown sodium salts.
Mixture of 2-[Bcat(Me)2]pentane and 3-[Bcat(Me)2]pentane: 11
B
NMR (C6D6) δ 37; GC/MS Two distinct peaks were observed. (15.15
min) m/z ) 218 (M+), 176, 161, 148, 133. (15.39 min) m/z ) 218
(M+), 173, 162, 148, 133, 91.
KCp*Ru(CO)2. [Cp*Ru(CO)2]2 (1.60 g, 2.74 mmol) was suspended
in THF and refluxed with K metal (535.0 mg, 13.68 mmol) under
nitrogen for 12 h with stirring, as reported in the literature.71 The K
metal was removed by filtration, and the THF was removed as described
below. The resulting solid was washed with toluene to remove dark
brown impurities, affording the yellow potassium salt in 86% yield
(1.5673 g).
1
Cyclohexyl[Bcat(Me)2]: H NMR (C6D6) δ 6.83 (s, 1H), 6.50 (s,
1H), 2.25 (s, 3H), 2.10 (s, 3H), 1.90 (m, 2H), 1.60 (m, 5H), 1.45-1.15
(m, 4H); 11B NMR (C6D6) δ 37; GC/MS (18.02 min) m/z ) 230 (M+),
173, 159, 148, 133, 91. HRMS (EI) calcd for C14H19BO2: 230.1478.
Found: 230.1474.
1
1-[Bcat(Me)2]-2-methylbutane: H NMR (C6D6) δ 6.82 (s, 1H),
6.50 (s, 1H), 2.25 (s, 3H), 2.11 (s, 3H), 1.45-1.03 (m, 5H), 0.97 (d, J
) 6 Hz, 3H), 0.83 (t, J ) 6 Hz, 3H); 11B NMR (C6D6) δ 37; GC/MS
(15.38 min) m/z ) 218 (M+), 203, 189, 173, 161, 148, 133, 91.
1-BBNpentane: 1H NMR (C6D6) δ 2.00-1.10 (m, 22H), 0.93 (t, J
) 6 Hz, 3H); 11B NMR (C6D6) δ 89.
LiCp*W(CO)3. LiCp*, which was generated in situ from Cp*H (1
mL, 6.39 mmol) and n-BuLi (2.6 mL of a 2.46 M solution in n-hexane,
6.40 mmol) at -78 °C in 30 mL of THF, was refluxed with W(CO)6
(2.4786 g, 7.04 mmol) in 30 mL of THF for 2 d as reported in the
literature.72 The THF was removed as described below, resulting in
quantitative yield of the yellow/orange lithium salt (3.0831 g).
H[Bcat(Me)2]. A solution of 3,5-dimethylcatechol (4.0882 g, 0.0296
mol) in 15 mL of ether was added slowly to a stirred solution of BH3‚
Me2S (2.81 mL, 0.0296 mol) in 10 mL of ether at 0 °C. After complete
addition of the catechol, the reaction was stirred at room temperature
for 3 h. Volatile materials were removed under vacuum, resulting in a
sticky white solid in 96% yield (4.22 g) that was suitably pure for the
Removal of THF from Metal Salts. 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 of the solvent had been removed.
This procedure was repeated three times. The resulting solid was washed
with toluene to remove dimeric impurities, affording the metal salt.
1
hydroboration reaction. H NMR (C6D6) δ 6.75 (s, 1H), 6.47 (s, 1H),
2.17 (s, 3H), 2.06 (s, 3H); 11B NMR (C6D6) δ 28.4 (d, J ) 195.4 Hz).
Cl[Bcat(Me)2]. A three-necked round-bottom flask equipped with
a stirbar and nitrogen inlet was charged with 3,5-dimethylcatechol (3.82
g, 0.0277 mol) in 200 mL of pentane and submerged in an ice bath.
Neat BCl3 (5 mL) was added via cannula with vigorous stirring,
resulting in rapid gas evolution. After complete addition of BCl3, the
reaction was stirred at room temperature for 3 h. Volatile materials
were removed under vacuum, resulting in a white solid in 96% yield
(4.82 g) which was spectroscopically pure. 1H NMR (C6D6) δ 6.56 (s,
1H), 6.38 (s, 1H), 2.04 (s, 3H), 1.99 (s, 3H); 11B NMR (C6D6) δ 29.2.
Cp*Fe(CO)2[Bcat(tBu)2] (1a). A solution of ClBcat(tBu)2 (209 mg,
0.785 mmol) in 5 mL of toluene was added to a stirred suspension of
Na[Cp*Fe(CO)2] (213.6 mg, 0.791 mmol) in 10 mL of toluene. After
30 min of stirring, all of the ClBcat(tBu)2 had been consumed. The
solvent was removed under reduced pressure, and the residue was
extracted with pentane. The extracts were filtered through glass wool,
condensed under vacuum, and crystallized at -30 °C. Two additional
recrystallizations were necessary to remove colored impurities, affording
a pale yellow crystalline solid in 30% yield (108 mg). 1H NMR (C6D6)
δ 7.26 (d, J ) 1.9 Hz, 1H), 7.15 (d, J ) 1.9 Hz, 1H), 1.610 (s, 9H),
1.605 (s, 15H), 1.30 (s, 9H); 13C{1H} NMR (C6D6) δ 216.51, 151.66,
147.18, 144.74, 133.75, 115.43, 107.52, 95.89, 35.35, 34.97, 32.41,
30.43, 10.35; 11B NMR (C6D6) δ 53.7; IR (pentane) νCO 2003 (s), 1949
(s) cm-1. Anal. calcd for C26H35BO4Fe: C, 65.30; H, 7.38. Found: C,
64.97; H, 7.47.
Independent Synthesis of Organoboronate Esters and Orga-
noboranes. Catalytic Hydroboration of Alkene. The following
compounds were prepared by rhodium-catalyzed hydroboration of
alkene in THF according to the literature procedure using Wilkinson’s
catalyst.73,74
1
1-[Bcat(Me)2]-3-methylbutane: H NMR (C6D6) δ 6.83 (s, 1H),
6.51 (s, 1H), 2.25 (s, 3H), 2.11 (s, 3H), 1.46 (m, 3H), 1.13 (m, 2H),
1.84 (m, 6H); 11B NMR (C6D6) δ 36; GC/MS: (15.51 min) m/z ) 218
(M+), 203, 175, 160, 148, 133, 91.
1
1-[Bcat(Me)2]ethylcyclohexane: H NMR (C6D6) δ 6.84 (s, 1H),
6.51 (s, 1H), 2.25 (s, 3H), 2.10 (s, 3H), 1.80-1.45 (m, 7H), 1.30-
1.00 (m, 6H), 0.82 (t, J ) 12 Hz, 2H); 11B NMR (C6D6) δ 36; GC/
MS: (20.69 min) m/z ) 258 (M+), 175, 160, 148, 133, 91.
(63) King, R. B.; Iqbal, M. Z.; King, J., A. D. J. Organomet. Chem.
1979, 171, 53-63.
(64) Kubas, G. J.; Kiss, G.; Hoff, C. D. Organometallics 1991, 10, 2870-
2876.
(65) Schmitzer, S.; Weis, U.; Ka¨b, H.; Buchner, W.; Malisch, W.; Polzer,
T.; Posset, U.; Kiefer, W. Inorg. Chem. 1993, 32, 303-309.
(66) He, X.; Hartwig, J. F. Organometallics 1996, 15, 400-407.
(67) Srivastava, G. J. Chem. Soc., Perkin Trans. 1 1974, 8, 916-917.
(68) Goetze, R.; No¨th, H.; Pommerening, H.; Sedlak, D.; Wrackmeyer,
B. Chem. Ber. 1981, 114, 1884-1893.
Cp*Fe(CO)2[Bcat(Me)2] (1b). A solution of ClBcat(Me)2 (215.0
mg, 1.178 mmol) in 5 mL of pentane was added to a stirred suspension
of Na[Cp*Fe(CO)2] (327.4 mg, 1.213 mmol) in 10 mL of pentane.
After 15 min of stirring, all of the ClBcat(Me)2 had been consumed.
The reaction mixture was filtered through glass wool, condensed under
vacuum, and crystallized at -30 °C for 15 min, resulting in yellow
crystalline product. Crystallization for longer periods of time resulted
(69) Weller, D. D.; Stirchak, E. P. J. Org. Chem. 1983, 48, 4873-4879.
(70) Culcasi, M.; Berchadsky, Y.; Gronchi, G.; Tordo, P. J. Org. Chem.
1991, 56, 3537-3542.
(71) Guerchais, V.; Lapinte, C.; The´pot, J.-Y.; Toupet, L. Organome-
tallics 1988, 7, 604-612.
(72) Lindsell, W. E.; McCullough, K. J.; Plancq, S. J. Organomet. Chem.
1995, 491, 275-287.
(73) Beletskaya, I.; Pelter, A. Tetrahedron 1997, 53, 4957-5026.
(74) Mannig, D.; No¨th, H. Angew. Chem., Int. Ed. Engl. 1985, 24, 878-
879.
(75) Brown, H. C.; Gupta, S. K. J. Am. Chem. Soc. 1975, 97, 5249-
5255.
(76) Nelson, G. O. Organometallics 1983, 2, 1474-1475.

Metal Boryl Complexes and Alkane Borylation
J. Am. Chem. Soc., Vol. 122, No. 46, 2000 11367
in concurrent crystallization of yellow Cp*Fe(CO)2Bcat with red
[Cp*Fe(CO)2]2 impurity. An additional recrystallization was necessary
to remove colored impurities, affording a yellow crystalline solid (239
mg). Additional pure material was obtained from the mother liquor
after two recrystallizations. The total yield of recrystallized product
was 81% (374 mg). 1H NMR (C6D6) δ 6.89 (s, 1H), 6.51 (s, 1H), 2.34
(s, 3H), 2.15 (s, 3H), 1.58 (s, 15H); 13C{1H} NMR (C6D6) δ 216.52,
151.11, 147.99, 131.05, 124.00, 120.82, 110.04, 96.11, 21.69, 15.26,
(C6D6) δ 223.93, 219.59, 151.80, 122.40, 111.98, 104.56, 11.14; 11B
NMR (C6D6) δ 52.4; IR (pentane) νCO 2010 (s), 1931 (s), 1914 (s)
cm-1. Anal. calcd for C19H19BO5W: C, 43.72; H, 3.67. Found: C,
43.91; H, 3.72.
LiCp*Mo(CO)3. A 2.5 M solution of n-BuLi in n-hexane (1.3 mL,
3.25 mmol) was added dropwise to a stirred solution of Cp*Mo(CO)3H
(1.0047 g, 3.18 mmol) in 15 mL of pentane. Upon addition, the clear
yellow solution turned bright cloudy yellow. After 1 h of stirring, the
reaction mixture was filtered, and the solid residue was washed several
times with pentane. The bright yellow lithium salt was obtained in
quantitative yield (1.0923 g).
Cp*Mo(CO)3[Bcat(Me)2] (4). A solution of ClBcat(Me)2 (179.5 mg,
0.984 mmol) in 5 mL of pentane was added to a suspension of
LiCp*Mo(CO)3 (318.7 mg, 0.990 mmol) in 5 mL of pentane. After
1.5 h of stirring, the solution was filtered through glass wool, and the
solid residue was extracted with at least 30 mL of pentane. The filtrate
and extracts were combined, condensed, and recrystallized at -30 °C.
Two additional recrystallizations gave pure, pale yellow crystalline
product in 32% yield (144.7 mg). 1H NMR (C6D6) δ 6.82 (s, 1H), 6.48
(s, 1H), 2.29 (s, 3H), 2.12 (s, 3H), 1.75 (s, 15H); 13C{1H} NMR (C6D6)
δ 233.48, 227.98, 151.29, 148.37, 131.68, 124.43, 121.20, 110.30,
106.06, 21.62, 15.14, 11.21; 11B NMR (C6D6) δ 55.4; IR (C6D6) νCO
2009 (s), 1920 (s), cm-1. Anal. calcd for C21H23BO5Mo: C, 54.58; H,
5.02. Found: C, 54.72; H, 5.06.
10.31; 11B NMR (C6D6) δ 54; IR (C6D6) νCO 1996 (s), 1939 (s) cm-1
;
UV/vis (pentane) λmax ) 277 nm; ꢀ277 ) 6000 cm-1 M-1. Anal. calcd
for C20H23BO4Fe: C, 60.96; H, 5.88. Found: C, 61.06; H, 5.71.
CpFe(CO)2[Bcat(tBu)2] (1c). A solution of ClBcat(tBu)2 (274 mg,
1.03 mmol) in toluene was added to a suspension of NaCpFe(CO)2
(205 mg, 1.03 mmol) in toluene and stirred for 1 h. The toluene solvent
was removed under vacuum, and the resulting residue was extracted
with pentane. These pentane extracts were condensed under vacuum,
filtered through glass wool, and crystallized at -30 °C to give a pale
yellow solid. A second crop was obtained from the mother liquor. An
additional recrystallization of both crops gave pure pale yellow product
in 69% yield (289 mg). 1H NMR (C6D6) δ 7.25 (d, J ) 1.91 Hz, 1H),
7.16 (d, J ) 1.94 Hz, 1H), 4.27 (s, 5H), 1.58 (s, 9H), 1.31 (s, 9H);
13C{1H} NMR (C6D6) δ 214.55, 151.31, 146.87, 145.16, 134.00, 115.71,
107.53, 84.22, 35.39, 35.03, 32.41, 30.36; 11B NMR (C6D6) δ 51; IR
(pentane) νCO 2022 (s), 1969 (s) cm-1. Anal. calcd for C21H25BO4Fe:
C, 61.81; H, 6.17. Found: C, 61.85; H, 6.12.
Cp*Ru(CO)2Bpin (5). A 1.0 M solution of BCl3 in heptane (1.8
mL, 1.8 mmol) was added via syringe to a vigorously stirred solution
of pinacol (142.0 mg, 1.20 mmol) in pentane (40 mL) at 0 °C. After
all of the BCl3 was added, the reaction was stirred at room temperature
for 1 h. The solution was condensed under vacuum until a few mL’s
of liquid remained in the flask. It was important not to condense to
dryness because ClBpin is volatile. 11B NMR spectroscopy of this crude
reaction mixture confirmed the clean formation of ClBpin, which
resonates at δ 28. The mixture was filtered through glass wool to
remove white insoluble impurities and was used immediately to avoid
thermal decomposition of ClBpin. The crude pentane solution of ClBpin
was added to a stirred suspension of KCp*Ru(CO)2 (400.0 mg, 1.21
mmol) in pentane. After 45 min of stirring, the ClBpin was completely
consumed as determined by 11B NMR spectroscopy. The reaction
mixture was filtered through glass wool to remove all solids, and the
pentane solution was condensed under vacuum. Crystallization at -30
°C produced yellow crystals, which were then recrystallized to remove
colored metal impurities and boron impurities which resonate at δ 23
in the 11B NMR spectrum. Pure Cp*Ru(CO)2Bpin was isolated as an
off-white crystalline solid (138.4 mg). Additional material was ob-
tained from the mother liquor by sublimation (25 °C, 0.01 Torr) for
at least 1 d, followed by recrystallization from pentane at -30 °C
to remove yellow impurities. The total yield of pure material was
36% (183.4 mg). 1H NMR (C6D6) δ 1.79 (s, 15H), 1.16 (s, 12H);
13C{1H} NMR (C6D6) δ 204.24, 99.96, 82.46, 25.61, 10.91; 11B
Cp*Ru(CO)2[Bcat(Me)2] (2). A solution of ClBcat(Me)2 (145.0 mg,
0.795 mmol) in 5 mL of pentane was added to a stirred suspension of
K[Cp*Ru(CO)2] (264.3 mg, 0.798 mmol) in 10 mL of pentane. After
an hour of stirring, all of the ClBcat(Me)2 had been consumed. The
reaction mixture was filtered through glass wool, and the solid residue
was extracted with 50 mL of pentane. The filtrate and pentane extracts
were combined, condensed under vacuum, and crystallized at -30 °C.
An additional recrystallization failed to remove all boron impurities
which appear at δ 23 in the 11B NMR spectrum. These impurities were
sublimed overnight (45-55 °C, 0.01 Torr) from the recrystallized
material, leaving a slightly impure ruthenium boryl compound which
was recrystallized from pentane to give pure, pale yellow product in
1
40% yield (140.0 mg). H NMR (C6D6) δ 6.90 (s, 1H), 6.52 (s, 1H),
2.35 (s, 3H), 2.15 (s, 3H), 1.71 (s, 15H); 13C{1H} NMR (C6D6) δ
203.29, 150.76, 147.64, 131.11, 124.05, 120.89, 110.16, 100.61, 21.70,
15.25, 10.79; 11B NMR (C6D6) δ 48; IR (C6D6) νCO 2012 (s), 1952 (s)
cm-1. Anal. calcd for C20H23BO4Ru: C, 54.69; H, 5.28. Found: C,
54.77; H, 5.15.
Cp*W(CO)3[Bcat(Me)2] (3a). A pentane solution of ClBcat(Me)2
(206.6 mg, 1.13 mmol) was added to a suspension of NaCp*W(CO)3
(482.3 mg, 1.13 mmol) in pentane. After an hour of stirring, the reaction
mixture was filtered through glass wool, and the solid residue was
extracted with 50 mL of pentane. The filtrate and pentane extracts were
combined, condensed under vacuum, and crystallized at -30 °C. A
second recrystallization afforded pure, amber crystalline material in
67% yield. Additional material was obtained from the mother liquor.
1H NMR (C6D6) δ 6.83 (s, 1H), 6.49 (s, 1H), 2.30 (s, 3H), 2.13 (s,
3H), 1.82 (s, 15H); 13C{1H} NMR (C6D6) δ 224.07, 219.67, 151.45,
148.48, 131.57, 124.19, 121.16, 110.29, 104.65, 21.63, 15.17, 11.17;
11B NMR (pentane) δ 53; IR (benzene) νCO 2002 (m), 1920 (s), 1900
(vs) cm-1; UV/vis (pentane) λmax ) 246, 286, 295 nm; ꢀ246 ) 13 000
cm-1 M-1. Anal. calcd for C21H23BO5W: C, 45.81; H, 4.21. Found:
C, 45.69; H, 4.17.
NMR (C6D6) δ 44.7; IR (Nujol) νCO 2009(s), 1950(s), 1920(w) cm-1
.
Anal. calcd for C18H27BO4Ru: C, 51.56; H, 6.49. Found: C, 51.61; H,
6.47.
Cp*W(CO)3Bpin (6). The same procedure was followed as for the
preparation of Cp*Ru(CO)2Bpin, except that LiCp*W(CO)3 (516.0 mg,
1.26 mmol) was reacted with the ClBpin generated in situ from 1.0 M
BCl3 (1.8 mL, 1.8 mmol) and pinacol (144.0 mg, 1.22 mmol). Four
recrystallizations were necessary to remove boron impurities and orange
metal impurities, affording pure, pale yellow boryl complex. Additional
material was obtained from the mother liquor by sublimation (45 °C,
0.01 Torr) overnight, followed by recrystallization from pentane at -30
°C. The total yield of pure product was 10% (64.1 mg). 1H NMR (C6D6)
δ 1.85 (s, 15H), 1.14 (s, 12H); 13C{1H} NMR (C6D6) δ 224.66, 220.52,
104.36, 83.92, 24.87, 11.18. 11B NMR (C6D6) δ 50; IR (pentane) νCO
1999 (s), 1916 (m), 1900 (s) cm-1. Anal. calcd for C19H27BO5W: C,
43.05; H, 5.13. Found: C, 42.94; H, 5.00.
Cp*Fe(CO)2BBN (7). A solution of BrBBN (185.6 mg, 0.925 mmol)
in 5 mL of pentane was added to a stirred suspension of Na[Cp*Fe-
(CO)2] (258.7 mg, 0.959 mmol) in 10 mL of pentane. After an hour of
stirring, all of the BrBBN had been consumed. The reaction mixture
was filtered through glass wool, condensed under vacuum, filtered again
to remove dark red [Cp*Fe(CO)2]2, and crystallized at -30 °C. Two
Cp*W(CO)3Bcat (3b). A solution of ClBcat (373.1 mg, 2.416
mmol) in toluene was added to a stirred suspension of LiCp*W(CO)3
(1.001 g, 2.443 mmol) in 15 mL of toluene. The reaction was monitored
by 11B NMR spectroscopy, and after 1 h all of the ClBcat was
consumed. The reaction mixture was filtered through glass wool to
remove all solids, and the toluene solvent was removed under vacuum.
The residue was extracted with pentane, and the resulting solution was
condensed under vacuum. Crystallization at -30 °C afforded beige
crystals. Several more recrystallizations from pentane were necessary
to remove all traces of colored metal impurity and boron impurities
which resonate at δ 23 in the 11B NMR spectrum. Cp*W(CO)3Bcat
1
was isolated as a light beige solid in 47% yield (591.3 mg). H NMR
(C6D6) δ 7.10 (m, 2H), 6.76 (m, 2H), 1.77 (s, 15H); 13C{1H} NMR

11368 J. Am. Chem. Soc., Vol. 122, No. 46, 2000
Waltz and Hartwig
additional recrystallizations were necessary to remove colored impuri-
ties, affording a yellow/orange crystalline solid in 21% yield (72.8 mg).
1H NMR (C6D6) δ 2.68 (s, 2H), 1.95-2.20 (m, 10H), 1.56 (s, 15H),
1.47 (m, 2H); 13C{1H} NMR (C6D6) δ 218.26, 96.28, 43.77, 34.05,
23.86, 10.40; 11B NMR (C6D6) δ 125.6; IR (pentane) νCO 1981, 1927
(d, J ) 8.1 Hz, 1H), 7.26 (s, 1H), 7.13 (m, 3H), 6.72 (d, J ) 8.1 Hz,
1H), 1.99 (s, 3H). 11B NMR (C6D6) δ 59.6.
Cp*W(CO)2PXy3H. A glass reaction vessel fused to a Kontes
vacuum adaptor was equipped with a stirbar and charged with Cp*W-
(CO)3H (563.6 mg, 1.395 mmol) and PXy3 (663.0 mg, 1.92 mmol)
dissolved in 20 mL of benzene. The solution was heated at 140 °C for
cm-1; UV/vis (pentane) λmax ) 253, 332 nm; ꢀ253 ) 7000 cm-1 M-1
.
1
3 d and monitored by H and 31P{1H} NMR. After all of the tungsten
Anal. calcd for C20H29BO2Fe: C, 65.26; H, 7.94. Found: C, 64.99; H,
8.00.
hydride was consumed, the volatile materials were removed under
vacuum, and the oily residue was washed with several 2 mL portions
of pentane to solidify the product and remove unreacted PXy3.
Additional free phosphine was removed by sublimation for 2 d (50
°C, 0.01 Torr). The remaining yellow solid was recrystallized from
pentane at -30 °C, affording pure, yellow crystalline product in 60%
Cp*Ru(CO)2BBN (8a). A solution of BrBBN (301.0 mg, 1.50
mmol) in 5 mL of pentane was added to a stirred suspension of
K[Cp*Ru(CO)2] (501.4 mg, 1.51 mmol) in 15 mL of pentane. After 5
min of stirring, all of the BrBBN was consumed. The reaction mixture
was filtered through glass wool, condensed under vacuum, and
crystallized at -30 °C. Two additional recrystallizations were necessary
to remove colored impurities, affording a yellow crystalline solid in
1
yield (604.0 mg). H NMR (C6D6) δ 7.36 (d, J ) 11.0 Hz, 6H), 6.70
(s, 3H), 2.05 (s, 18H), 1.88 (s, 15H), -5.91 (d, J ) 67.8 Hz, 1H);
13C{1H} NMR (C6D6) δ 139.46 (d, JCP ) 48.3 Hz, ipso C), 137.79 (d,
JCP ) 10.3 Hz, m-), 132.22 (d, JCP ) 12.0 Hz, o-), 131.71(p-), 101.32,
21.76, 11.79, the CO resonance was not observed at room tempera-
ture; 31P{1H} NMR (C6D6) δ 43.7 (s, JPW ) 268.6 Hz); IR (C6D6) νCO
1920 (s), 1834 (m) cm-1. Anal. calcd for C36H43O2PW: C, 59.84; H,
6.00. Found: C, 59.78; H, 5.86.
1
28% yield (171.5 mg). H NMR (C6D6): δ 2.62 (s, 2H), 1.90-2.20
(m, 10H), 1.67 (s, 15H), 1.46 (m, 2H); 13C{1H} NMR (C6D6): δ 205.41,
100.81, 43.49, 33.41, 23.94, 10.74; 11B NMR (C6D6): δ 117; IR
(pentane): νCO 2002 (s), 1944 (s) cm-1; UV/VIS (pentane): λmax
)
270 nm; ꢀ270 ) 7000 cm-1M-1; Anal. Calc’d for C20H29BO2Ru: C,
58.12; H, 7.07. Found: C, 58.09; H, 7.13.
Cp*Ru(CO)2BCy2 (8b). A 1.0 M solution of ClBCy2 (0.97 mL, 0.97
mmol) in hexanes was added to a stirred suspension of KCp*Ru(CO)2
(337.3 mg, 1.02 mmol) in 10 mL of pentane. The reaction was
monitored by 11B NMR spectroscopy and stirred for 2 h until all of the
ClBCy2 was consumed. The reaction mixture was filtered through glass
wool to remove all solids, and the resulting pentane solution was
condensed under vacuum. Crystallization at -30 °C afforded 314 mg
of pale yellow crystals. Two more recrystallizations from pentane were
necessary to remove all traces of colored metal impurity. Cp*Ru-
(CO)2BCy2 was isolated as a pale yellow crystalline solid in 55% yield
LiCp*W(CO)2PXy3. A 2.46 M solution of n-BuLi in hexanes (0.075
mL, 0.185 mmol) was added dropwise to a stirred solution of Cp*W-
(CO)2PXy3H (133.5 mg, 0.185 mmol) in pentane. After 1 min, the clear
yellow solution turned cloudy. After 20 min of stirring, the reaction
mixture was filtered, and the solid residue was washed with pentane.
The bright yellow lithium salt was obtained in quantitative yield (134.2
mg). Immediate refrigeration at -30 °C was necessary to prevent
decomposition to a brown solid. 1H NMR (THF-d8) δ 7.23 (d, J ) 9.8
Hz, 6H), 6.72 (s, 3H), 2.18 (s, 18H), 1.81 (s, 15H); 31P{1H} NMR
(THF-d8) δ 58.8 (s, JPW ) 468.8 Hz); IR (THF) νCO 1775 (m), 1655
1
(s), cm-1
.
(250.3 mg). H NMR (C6D6) δ 2.35 (tt, J ) 2.8 Hz, 12.0 Hz, 2H),
1.68 (s, 15H), 1.4-2.0 (m, 20H); 13C{1H} NMR (C6D6) δ 206.50,
100.73, 48.31, 28.39, 28.05, 28.01, 10.79; 11B NMR (C6D6) δ 120; IR
(Nujol) νCO 1989 (s), 1929 (s) cm-1. Anal. calcd for C24H37BO2Ru: C,
61.41; H, 7.94. Found: C, 61.59; H, 7.67.
trans-Cp*W(CO)2PXy3[Bcat(Me)2] (10a). A solution of ClBcat-
(Me)2 (86.0 mg, 0.471 mmol) in 5 mL of pentane was added to a stirred
suspension of Li[Cp*W(CO)2PXy3] (343.7 mg, 0.472 mmol) in 10 mL
of pentane. After 30 min of stirring, the reaction mixture was filtered,
and the solid residue was extracted with at least 50 mL of pentane.
The filtrate and pentane extracts were combined, and the pentane solvent
removed under vacuum. The residue was recrystallized twice from
toluene layered with pentane at -30 °C. Another recrystallization from
toluene at -30 °C over the course of several days resulted in yellow
crystalline material that contained traces of toluene solvent. Removal
of toluene was achieved by lyophilizing the product in frozen benzene,
resulting in a yellow powder in 20% yield (51.5 mg). 1H NMR (C6D6)
δ 7.88 (d, J ) 11.6 Hz, 2H), 7.37 (d, J ) 11.1 Hz, 4H), 6.97 (s, 1H),
6.75 (s, 3H), 6.56 (s, 1H), 2.43 (s, 3H), 2.22 (s, 3H), 2.20 (s, 6H), 2.06
(s, 12H), 1.96 (s, 15H); 13C{1H} NMR (C6D6) δ 230.68 (d,2JCP ) 19.4
Hz, CO), 152.10 (C-O in catechol), 149.18 (C-O in catechol), 138.05
(d, JCP ) 10.8 Hz, m- C), 136.67 (d, JCP ) 43.0 Hz, ipso C), 133.09
(d, JCP ) 10.1 Hz, o-), 132.41 (d, J ) 12.0 Hz, o-), 132.20 (p-), 130.58
(C-Me in cat), 123.52 (C-H in cat), 120.36 (C-Me in cat), 109.92 (C-H
in cat), 102.66 (Cp*), 21.95 (Me on cat), 21.72 (Me in PXy3), 15.57
(Me on cat), 11.58 (Cp*); 11B NMR (C6D6) δ 58; 31P{1H} NMR (C6D6)
Cp*Ru(CO)2BMe2 (8c). Neat BrBMe2 (0.092 mL, 0.943 mmol) was
added to a stirred suspension of KCp*Ru(CO)2 (313.7 mg, 0.947 mmol)
in pentane. After 45 min of stirring, the reaction mixture was filtered
through glass wool, condensed, and crystallized at -30 °C, producing
a dark red solid which contained no 11B NMR signals. The mother
liquor was condensed and crystallized again, this time affording a
lighter-colored solid containing boron, as determined by 11B NMR
spectroscopy. Sublimation (70 °C, 0.01 Torr) of this solid separated a
yellow solid from red impurities. Two recrystallizations of this yel-
low solid from pentane removed most of the Cp*Ru(CO)2H and
colored impurities. The product was obtained as a very pale yellow,
1
slightly impure solid. H NMR (C6D6) δ 1.60 (s, 15H), 1.58 (s, 6H);
13C{1H} NMR (C6D6) δ 205.14, 99.51, 28.19, 10.52; 11B NMR (C6D6)
δ 118.7.
Cp*Ru(CO)2BS2Tol (9). A solution of ClBS2Tol (198.5 mg, 0.991
mmol) in pentane was added to a stirred suspension of KCp*Ru(CO)2
(330.3 mg, 0.997 mmol) in 15 mL of pentane. The reaction was
monitored by 11B NMR spectroscopy and stirred for 30 min until all
of the ClBS2Tol was consumed. The reaction mixture was filtered
through glass wool to remove all solids, and the pentane solvent was
condensed under vacuum. Crystallization at -30 °C afforded pale
yellow crystals. Two more recrystallizations from pentane were
necessary to remove all traces of colored metal impurity. Cp*Ru-
(CO)2BS2Tol was isolated as an off-white solid in 34% yield (155 mg).
1H NMR (C6D6) δ 7.58 (d, J ) 8.0 Hz, 1H), 7.48 (s, 1H), 6.74 (d, J )
8.0 Hz, 1H), 2.04 (s, 3H), 1.63 (s, 15H); 13C{1H} NMR (C6D6) δ 203.32,
147.13, 143.88, 134.54, 126.17, 125.65, 124.71, 100.30, 21.22, 10.61;
11B NMR (C6D6) δ 78; IR (pentane) νCO 2014(s), 1962(s) cm-1; UV/
vis (pentane) λmax ) 247, 286, 295 nm; ꢀ247 ) 16 000 cm-1 M-1. Anal.
calcd for C19H21BO2S2Ru: C, 49.90; H, 4.63. Found: C, 49.88; H,
4.68.
δ 42.4 (s, JPW ) 289.3 Hz); IR (C6D6) νCO 1914 (m), 1832 (s) cm-1
.
Anal. calcd for C44H50BO4PW: C, 60.85; H, 5.80. Found: C, 61.10;
H, 6.05.
trans-Cp*W(CO)2(PMe3)[Bcat(Me)2] (10b). A solution of ClBcat-
(Me)2 (120.0 mg, 0.658 mmol) in 5 mL of pentane was added to a
stirred suspension of Li[Cp*W(CO)2PMe3] (304.0 mg, 0.664 mmol)
in 10 mL of pentane. After 2 h of stirring, the reaction mixture was
filtered, and the solid residue was extracted with at least 50 mL of
pentane. The filtrate and pentane extracts were combined, condensed,
and crystallized at -30 °C. One additional recrystallization from pentane
afforded a pure, bright yellow crystalline solid in 38% yield (148 mg).
1H NMR (C6D6) δ 6.90 (s, 1H), 6.52 (s, 1H), 2.40 (s, 3H), 2.15 (s,
3H), 1.92 (s, 15H), 1.29 (d, JHP ) 8.9 Hz, 9H); 13C{1H} NMR (C6D6)
δ 229.63 (d,2JCP ) 19.7 Hz, 1JCW ) 143.0 Hz), 152.03, 149.03, 130.43,
123.37, 120.23, 109.96, 102.65, 21.74, 20.35 (d, JCP ) 33.0 Hz), 15.42,
11.79; 11B NMR (C6D6) δ 58; 31P{1H} NMR (C6D6) δ -10.61 (s, JPW
) 273.4 Hz); IR (C6D6) νCO 1906 (m), 1823 (s) cm-1. Anal. calcd for
C23H32BO4PW: C, 46.19; H, 5.39. Found: C, 46.30; H, 5.39.
PhBS2Tol. This complex was prepared according to the literature,38
except that benzene solvent was used instead of methylene chloride.
1H NMR data is reported in the literature,68 but the aromatic resonances
1
were not resolved. H NMR (C6D6) δ 7.88 (d, J ) 6.7 Hz, 2H), 7.39

Metal Boryl Complexes and Alkane Borylation
J. Am. Chem. Soc., Vol. 122, No. 46, 2000 11369
Cp*Mo(CO)2(PXy3)H. A glass reaction vessel fused to a Kontes
vacuum adaptor was equipped with a stirbar and charged with Cp*Mo-
(CO)3H (442.0 mg, 1.40 mmol) and PXy3 (502.4 mg, 1.45 mmol)
dissolved in 20 mL of benzene. The solution was heated at 100 °C for
4 d and monitored by 1H and 31P{1H} NMR, after which time
approximately 75% of the starting material had been consumed. The
reaction was driven to completion by heating for an additional day at
140 °C. The volatile materials were removed under vacuum, and the
residue was extracted with several portions of pentane. The pentane
extracts were combined and filtered through glass wool, and the solvent
was removed under vacuum. Unreacted hydride and PXy3 were removed
by sublimation for 2 d (50 °C, 0.01 Torr). The remaining yellow solid
was recrystallized twice from pentane at -30 °C, affording pure, yellow
crystalline product. Additional material was obtained from the mother
liquor, for a total yield of 33% (290.0 mg). 1H NMR (C6D6) δ 7.37 (d,
J ) 11.0 Hz, 6H), 6.72 (s, 3H), 2.06 (s, 18H), 1.82 (s, 15H), -4.56 (d,
J ) 64.4 Hz, 1H); 13C{1H} NMR (C6D6) δ 139.25 (d, JCP ) 38.8 Hz,
ipso C), 137.86 (d, JCP ) 9.5 Hz, m-), 132.02 (d, JCP ) 12.2 Hz, o-),
131.64 (p-), 102.80, 21.74, 11.76; 31P{1H} NMR (C6D6) δ 72.6; IR
(C6D6) νCO 1928 (s), 1846 (ms) cm-1. Anal. calcd for C36H43O2PMo:
C, 68.13; H, 6.83. Found: C, 68.01; H, 6.72.
Li[Cp*Mo(CO)2PXy3]. A 2.46 M solution of n-BuLi in hexanes
(0.13 mL, 0.32 mmol) was added dropwise to a stirred solution of
Cp*Mo(CO)2(PXy3)H (193.8 mg, 0.305 mmol) in 20 mL of pentane.
After a few seconds, the clear yellow solution turned bright cloudy
yellow. After 30 min of stirring, the reaction mixture was filtered, and
the solid residue was washed several times with pentane. The bright
yellow lithium salt was obtained in 94% yield (184.5 mg). Immediate
refrigeration at -30 °C was necessary to prevent decomposition to a
brown solid.
General Procedure for Photochemical Reaction of Metal Boryl
Complexes with Alkane Substrates. The metal boryl complex (5 mg)
was dissolved in 0.5 mL of neat substrate and transferred to an NMR
tube. 11B NMR spectra were recorded for the starting material. The
sample was irradiated with periodic monitoring of the reaction by 11
NMR spectroscopy.
B
Measurement of Yields of Organoboron Compounds. The metal
boryl complex (5 mg) and dodecahydrotriphenylene (1-2 mg) as an
internal standard were dissolved in 0.8 mL of neat substrate. Half of
this solution was transferred to an NMR tube. 11B NMR spectra were
recorded for the starting material. The sample was irradiated with
periodic monitoring of the reaction by 11B NMR spectroscopy. After
the reaction was judged to be complete, the ratio of organoborane to
standard was determined by either GC or 1H NMR spectroscopy. The
other half of the original solution was reduced to dryness under vacuum,
1
and the residue was dissolved in C6D6. A H NMR spectrum of this
sample was recorded to obtain the ratio of the standard to the metal
boryl starting material.
Isotope Effect Measurements. The metal boryl complex (1 mg)
was added as a solid to an NMR sample tube, which was capped with
a rubber septum. Equimolar amounts of pentane (0.10 mL) and pentane-
d12 (0.10 mL) were added to the tube by syringe. The sample was
irradiated until all starting material was consumed as determined by
11B NMR spectroscopy. The solution inside the tube was analyzed by
GC/MS to determine the ratio of protiated and deuterated alkylboronate
esters by the abundance of their molecular ions.
Irradiation of 3a in the Presence of CO. Tungsten boryl complex
3a (6 mg) was dissolved in pentane (1 mL) and the solution added to
two NMR tubes (0.4 mL in each). Carbon monoxide (2 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 of both samples were periodically
recorded, and the amount of product was qualitatively evaluated at each
time point.
Irradiation of 3a in the Presence of 13CO. Tungsten boryl com-
plex 3a (5 mg) was dissolved in pentane and transferred to a Young
NMR tube. 13CO (2 atm) was introduced into the tube. 11B and 13C
NMR spectra of the starting material were recorded. The solution was
irradiated until roughly half the starting material remained as monitored
by 11B NMR spectroscopy. 13C NMR spectroscopy of this sample
showed no incorporation of the labeled CO into the starting material.
trans-Cp*Mo(CO)2(PXy3)[Bcat(Me)2]. A solution of Cl[Bcat(Me)2]
(51.2 mg, 0.281 mmol) in 5 mL of pentane was added to a stirred
suspension of Li[Cp*Mo(CO)2PXy3] (180 mg, 0.281 mmol) in 5 mL
of pentane. The mixture changed color from bright yellow to a darker
yellow. After 30 min of stirring, the reaction was complete as
determined by 11B NMR spectroscopy. The mixture was filtered, and
the solid residue was extracted with at least 50 mL of pentane. The
filtrate and pentane extracts were combined, and the pentane solvent
was removed under vacuum. The residue was crystallized from toluene
layered with pentane at -30 °C. The product was recrystallized from
toluene at -30 °C over the course of several days, resulting in yellow
crystalline material which contained traces of toluene solvent. The
toluene was removed by lyophilizing the product in frozen benzene,
resulting in a yellow powder in 25% yield (55.0 mg). This material
appeared clean by spectroscopic analysis, but attempts to obtain
satisfactory analytical data after repeated recrystallizations were unsuc-
Irradiation of 3a in the Presence of PMe3. Tungsten boryl complex
3a (5 mg) was dissolved in pentane (0.5 mL), and the solution was
added to a Young NMR tube. Using vacuum techniques, 2 equiv of
PMe3 were condensed into the tube. A 11B NMR spectrum of the starting
material was recorded. The solution was irradiated and periodically
monitored by 11B NMR spectroscopy until half of the starting material
remained. The volatile materials were removed using vacuum tech-
niques, and the residue was dissolved in C6D6. The ratio of Cp*W(CO)2-
PMe3Bcat(Me)2 and pentylBcat(Me)2 was determined by 1H NMR
spectroscopy. This procedure was repeated with 4 equiv of PMe3.
1
cessful. H NMR (C6D6) δ 7.85 (d, J ) 11.5 Hz, 2H), 7.39 (d, J )
10.9 Hz, 4H), 6.96 (s, 1H), 6.76 (s, 3H), 6.55 (s, 1H), 2.42 (s, 3H),
1
2.20 (s, 9H), 2.07 (s, 12H), 1.91 (s, 15H); H NMR (64 °C, C6D6) δ
7.25-7.80 (broad, 6H), 6.91 (s, 1H), 6.80 (s, 3H), 6.54 (s, 1H), 2.41
(s, 3H), 2.21 (s, 3H), 2.14 (broad s, 18H), 1.92 (s, 15H); 13C{1H} NMR
(C6D6) δ 237.54 (d,2JCP ) 26.5 Hz, 2 CO), 151.81 (C-O in catechol),
148.99 (C-O in catechol), 138.10 (d, JCP ) 8.4 Hz, m-), 136.75 (d, JCP
) 36.7 Hz, ipso), 132.65 (d, JCP ) 11.7 Hz, o-), 132.35 (d, J ) 12.1
Hz, o-), 132.16 (p-), 130.70 (C-Me in cat), 123.64 (C-H in cat), 120.46
(C-Me in cat), 109.90 (C-H in cat), 104.00 (Cp*), 21.89 (Me on cat),
21.74 (Me in PXy3), 21.71 (Me in PXy3), 15.56 (Me on cat), 11.58
(Me in Cp*); 11B NMR (C6D6) δ 58; 31P{1H} NMR (C6D6) δ 71.6; IR
Acknowledgment. We gratefully acknowledge support from
the National Science Foundation (CHE-9986236) and Shell
Chemicals.
(C6D6) νCO 1916 (ms), 1834 (s) cm-1
.
JA002840J