Regioselectivity of the rhodium-catalyzed hydroboration of vinyl arenes: Electronic twists and mechanistic shifts

Article abstract of DOI:10.1002/anie.200702563

Any substituent does it: The hydroboration of vinyl arenes with pinacol borane (HBPin) and cationic rhodium complexes selectively placed the boron proximal to the aryl rather than phenyl ring, regardless of whether this ring bears electron-donating or electron-withdrawing substituents. In competition experiments between styrene and various vinyl arenes, preferential hydroboration also occured at the substituted arene (see scheme). Hammett plots indicate a break in the mechanism. (Chemical Equation Presented).

Full text of DOI:10.1002/anie.200702563

Angewandte
Chemie
DOI: 10.1002/anie.200702563
Metal-Catalyzed Hydroborations
Regioselectivity of the Rhodium-Catalyzed Hydroboration of Vinyl
Arenes: Electronic Twists and Mechanistic Shifts**
David R. Edwards, Yonek B. Hleba, Christopher J. Lata, Larry A. Calhoun, and
Cathleen M. Crudden*
In memory of Yoshihiko Ito
The hydroboration of alkenes represents a versatile synthetic
method for the functionalization of organic compounds.[1] The
is generated. This unique regioselectivity has been attributed
to the stabilizing effect of a p-benzyl interaction available
only to the branched isomer 3.[2] p-Benzyl complexes of a
À
À
installed carbon–boron bond can be converted into C O, C
[1b]
À
N, and C C bonds with retention of stereochemistry.
Several highly enantioselective catalysts have been developed
by the groups of Hayashi,[2] Brown,[3] Knochel,[4] Togni,[5] and
Takacs[6] among others, allowing the hydroboration to be
employed in the preparation of a variety of organic com-
pounds including NSAIDs (non-steroidal anti-inflammatory
drugs), such as Ibuprofen and Naproxen.[7] Remarkably, not
only does the catalyst affect the introduction of asymmetry, it
can also alter the chemo- and regioselectivity of the reac-
tion.[1,8] For example, the uncatalyzed hydroboration of
styrene with reagents such as 9-BBNoccurs with formally
anti-Markovnikov selectivity yielding linear borane 1. Con-
versely, the rhodium-catalyzed reaction affords predomi-
nantly regioisomer 2 with up to 99% selectivity.[2]
variety of transition metals[10–15] including rhodium[16–21] have
been isolated, however, there is no empirical evidence to
definitively ascribe the regioselectivity in the rhodium-
catalyzed hydroboration to this interaction.
In order to probe the importance of p-benzyl interactions
in the regiodefining event of the hydroboration reaction, we
carried out hydroborations of mono-para-substituted cis-
stilbenes 6a–g (Table 1).[22] Only para-substituted stilbenes
Table 1: Hydroboration of cis-stilbene derivatives 6a–g.[a]
However, the branched selectivity observed in this
catalyzed reaction occurs with only a few types of olefins,
among which vinyl arenes are the most prominent.[1,8] In the
case of simple aliphatic olefins, linear selectivity or even
“chain running”, where internal olefins are isomerized to the
corresponding terminal isomer prior to hydroboration[9] can
be observed. The branched selectivity observed with vinyl
arene substrates is of particular interest since a chiral product
Substrate, X
s
Ratio 7/8 Yield [%][b]
Mass balance [%][b]
6a, CF3
6b, Cl
6c, F
6d, Me
6e, OMe
6 f, OiPr
6g, NMe2
0.54
0.23
0.06
3.00:1
1.63:1
1.12:1
1.10:1
1.20:1
1.25:1
1.89:1
67 (63)
43 (43)
78 (86)
49 (49)
81 (78)
61 (59)
18
93 (87)
87 (97)
99 (94)
91 (96)
90 (95)
98 (91)
50
À0.17
À0.27
À0.29
À0.83
[a]See Experimental Section for details; dppb =1,4-bis(diphenylphos-
phanyl)butane. [b]Determined by 1H NMR spectroscopy versus internal
standard, yields of isolated product in parentheses.
[*]D. R. Edwards, Y. B. Hleba, C. J. Lata, Prof. C. M. Crudden
Queen’s University
Kingston, Ontario, K7L 3N6 (Canada)
Fax: (+1)1-613-533-6669
E-mail: cruddenc@chem.queensu.ca
were employed to prevent complications due to steric effects.
Remarkably, we found that regioisomer 7, in which boron was
placed proximal to the substituted aryl ring, was preferred in
all cases, regardless of whether the substituent was electron-
donating or electron-withdrawing.
The hydroboration reaction was carried out using pinacol
borane (HBPin) as the hydroborating reagent on mono-para-
substituted geometrically pure Z-stilbenes 6a–g (Table 1).
Following oxidation with hydrogen peroxide, the resulting
alcohols were isolated by column chromatography as mix-
Dr. L. A. Calhoun
Department of Chemistry, University of New Brunswick
Fredericton, New Brunswick, E3B 5E2 (Canada)
[**]The Natural Sciences and Engineering Research Council of Canada
(NSERC) is acknowledged for support of this research in terms of
operating grants to C.M.C. Queen’s University, the Ontario
government, and the Sumner foundation are acknowldeged for
grants to D.R.E., Queen’s University is acknowledged for support to
Y.B.H., and the NSERC is acknowledged for support of C.J.L. in
terms of an undergraduate research scholarship.
Angew. Chem. Int. Ed. 2007, 46, 7799 –7802
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
7799

Communications
tures of isomers.[23] The identities of the individual regioisom-
styrene and para-substituted styrene derivatives were
ers were verified by comparison to authentic samples
prepared by Grignard addition to the requisite aldehydes.
With the one exception of p-dimethylamino stilbene, the
reactions were quite clean with mass balances generally
exceeding 90%.[24] Recovered starting material was always
devoid of the cis isomer, presumably due to a concurrent
isomerization to the more stable trans isomer.
exposed to a limiting amount of HBPin, Equation (3). This
As previously noted, the major isomer always corre-
sponded to borylation proximal to the substituted aromatic
ring, irrespective of the electronic influence of the para-
substituent. The greatest selectivity was observed for the most
electron-donating, NMe2, and most electron-withdrawing,
CF3, para-substituents. A graph of log(isomer ratio) versus s
revealed a linear free energy relationship with a minimum
centered at the origin (Figure 1). Hammett plots such as this
study provides information about the relative reactivities of
the two substrates by comparing the relative conversions of
each substrate. From these data, the kX/kH ratio was
calculated and a graph of log(kX/kH) versus s composed
(Figure 2). Remarkably, a minimum centered at the origin is
Figure 1. Hammett plot for unsymmetrically substituted stilbenes.
Figure 2. Hammett plot for styrene derivatives.
featuring a minimum are usually interpreted as resulting from
a competition between two mechanisms, one favoured by
electron-donating, and the other by electron-withdrawing
substituents.[25,26] Interestingly, in this case, the two different
mechanisms lead to the preferential formation of the same
regioisomer.
As shown in Figure 1, the correlation coefficients for the
Hammett plots were very high (0.998 and 0.999) indicating
that the fit to the linear regression is good in both cases. The
data were also analyzed using s+ and sÀ parameters, but the
best correlation was observed using the original Hammett s
parameter.
The hydroboration of similar unsymmetrical stilbenes
with catechol borane (HBCat) in place of pinacol borane
using quinap (1-(2-diphenylphosphanyl-1-naphthyl)isoquino-
line) as the ligand has been recently reported by Brown et al.
to give hydroboration proximal to the less electron-rich arene
in all cases.[22,27,28] We believe this stems from significant
mechanistic differences between hydroborations with HBCat
and HBPin.[29]
In addition to stilbene derivatives, we examined the
reaction of styrenes, which are more common hydroboration
substrates. In this case, intermolecular competition experi-
ments were carried out in which equimolar amounts of
also observed, indicating a break in the mechanism for these
substrates as well. A simple interpretation of these results
would be that both the rate and the regioselectivity of the
reaction are determined at the same point in the multi-step
reaction and that both are influenced by the ability of aryl
substituents to dissipate charge.
Unlike uncatalyzed hydroborations, the presence of a
metal catalyst induces the sequential addition of boron and
hydrogen across the olefin.[1,28] The precise mechanism by
which this occurs was the subject of considerable debate in the
1990s.[30–33] Potential mechanisms include initial hydride
transfer followed by reductive carbon–boron bond formation,
boryl transfer followed by reductive elimination of hydrogen
and sigma-bond metathesis.[34]
Theoretical work by Ziegler et al. provides support for
both hydride and boryl insertion[35] mechanisms and the
feasibility of boryl insertion under mild conditions is demon-
strated by the elegant diboration studies of Morken,[36]
Baker,[37] Marder,[37c] Westcott,[38] and Fernandez and
Peris.[39] These two potential catalytic pathways are depicted
in Scheme 1.[40]
7800 www.angewandte.org
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2007, 46, 7799 –7802

Angewandte
Chemie
A completely different picture emerges from the reaction
of HBPin with deuterated vinyl arenes or DBPin with protic
substrates [Equation (5), the numbers in parentheses indicate
the total observable label at each site based on initially added
HBPin]. The “H” from HBPin is transferred mainly to the
terminal position of the product but some label is observed at
the benzylic position as well. In addition, recovered starting
material is observed to have incorporated 29% of the label.
Since 5 equivalents of the vinyl arene are employed, this
implies that the hydride addition is reversible, although the
forward reaction occurs on a similar time scale. These results
demonstrate the dramatic mechanistic differences that can be
observed when changing from catechol to pinacol borane. In
this case, the reversibility and selectivity of the metal hydride
addition is significantly different when the two ostensibly
similar boranes are employed. This may signify a completely
different mechanism, or at least a change in the rate-
(turnover)-limiting step with a change in borane.
In conclusion, the reactions of vinyl arenes with pinacol
borane catalyzed by cationic Rh species proceed through
different mechanisms depending on the electronic properties
of the arene. This is based on the observation of breaks in the
respective Hammett plots for both intramolecular systems
(unsymmetrical stilbenes) and intermolecular competitions
between styrene and vinyl arenes. In concert with these
results, deuterium labeling studies indicate that the mecha-
nism of the hydroboration of olefins with HBPin is signifi-
cantly different from that observed with HBCat. The reasons
for these differences and the elucidation of the two mecha-
nisms at play in the hydroboration of olefins with HBPin are
the subject of current investigations.
Scheme 1. Potential reaction pathways.[40]
In order to obtain more information on the nature of the
product-forming step in our system, a 13C KIE study was
carried out using the natural abundance method as described
by Singelton.[41] This technique provides information about
the first irreversible step of the catalytic cycle.[41] The KIE
study was performed for p-methoxystyrene and p-chlorostyr-
ene in order to provide mechanistic information for the two
different pathways. Since the hydroboration reaction is
accompanied by a trace of hydrogenation (ca. 2.5%) which
would complicate recovery of the starting material, we chose
to carry out the KIE studies by examination of the product at
low conversion.
Approximately equal KIEs are observed at the benzylic
and methyl carbons of p-methoxy phenethanol. In the case of
the p-chloro derivative, the KIE observed at the methyl
position is approximately half that observed at the methyl in
the p-methoxy product (Scheme 2), but unfortunately the
Scheme 2. Results of KIE studies.
Experimental Section
Hydroboration of cis-stilbenes 6a–g: The reaction was set up inside a
glove box. To a sealed tube were added [Rh(cod)2]BF4 (10.15 mg, 5
mol%) and dppb (10.7 mg, 5 mol%). This mixture was dissolved in
THF (4 mL) and stirred for 5 min. To the stirring suspension was
succesively added cis-4-fluorostilbene (99.1 mg, 0.50 mmol) and
pinacolborane (87 mL, 0.60 mmol). The reaction tube was sealed
within the glove box and removed to a fume hood. Reaction mixtures
were then heated in an oil bath set at 808C for 3 days. The reaction
flask was cooled to ambient temperature and 30% H2O2 (2 mL) and
2m NaOH (2 mL) were added. The reaction was stirred vigorously for
2 h. The mixture was then subjected to an aqueous work-up
procedure and extracted with ether. The crude product obtained
from the ether phase was loaded onto a small biotage loading
cartridge and chromatographed using a gradient elution 1–20%
EtOAc in hexanes. 1H NMR spectra were used to determine the
isomer ratio by comparison with authentic samples of 7a–g prepared
independently.
small magnitude of the KIEs obtained and the errors that
accompany the method complicates the comparison of these
numbers. In the absence of additional data,[42] a more detailed
mechanistic analysis is not possible.
Thus we turned to deuterium labeling as a method to
assess at least the first step of the reaction. It has been
previously shown by Evans and Fu that with deuterated
catechol borane, the combination of metal hydride(deuter-
ide) addition and olefin decomplexation is irreversible.[32,33]
Deuterium is incorporated cleanly into the methyl group of
the product, and when the reaction is run with an excess of
alkene, no deuterium is incorporated into the recovered
alkene [Eq. (4)].[43]
Styrene competition experiments: In a glove box, a round-
bottomed flask was charged with [Rh(cod)2]BF4 (10.15 mg, 5 mol%)
and dppb (10.7 mg, 5 mol%) in THF (2 mL). After stirring for 5 min,
styrene (0.50 mmol) and para-substituted styrene (0.50 mmol) were
Angew. Chem. Int. Ed. 2007, 46, 7799 –7802
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
7801

Communications
added simultaneously in THF (2 mL). Lastly pinacolborane (15 mL,
[25] a) Y. Apeloig, M. Nakash, J. Am. Chem. Soc. 1996, 118, 9798;
b) S. E. Denmark, T. Bai, J. Org. Chem. 2005, 70, 10393.
[26] T. H. Lowry, K. S. Richardson, Mechanism and Theory in
Organic Chemistry, 2nd ed., Harper & Row, New York, 1981.
[27] This is consistent with the palladium-catalyzed hydrostannyla-
tion of unsymmetrical para-substituted tolanes: M. Rubin, A.
Trofimov, V. Gevorgyan, J. Am. Chem. Soc. 2005, 127, 10243.
[28] J. D. Hewes, C. W. Kreimendahl, T. B. Marder, M. F. Hawthorne,
J. Am. Chem. Soc. 1984, 106, 5757.
0.10 mmol) was added. The reaction mixture was stirred in the glove
box for 24 h. Upon removal of the flask from the glove box, the
reaction was immediately quenched with aqueous ether. 80 mL decane
was added as GC internal standard. The relative conversion of each
substrate was determined by GC. Results described in Table 1 are the
averages of two runs.
Received: July 12, 2007
[29] a) C. M. Crudden, Y. Hleba, A. C. Chen, J. Am. Chem. Soc. 2004,
126, 9200; b) A. M. Segarra, E. Daura-Oller, C. Claver, J. M.
Poblet, C. Bo, E. Fernꢀndez, Chem. Eur. J. 2004, 10, 6456.
[30] K. Burgess, W. A. Van der Donk, A. M. Kook, J. Org. Chem.
1991, 56, 2949.
[31] K. Burgess, W. A. van der Donk, S. A. Westcott, T. B. Marder,
R. T. Baker, J. C. Calabrese, J. Am. Chem. Soc. 1992, 114, 9350.
[32] D. A. Evans, G. C. Fu, J. Org. Chem. 1990, 55, 2280.
[33] D. A. Evans, G. C. Fu, B. A. Anderson, J. Am. Chem. Soc. 1992,
114, 6679.
[34] Sigma-bond metathesis is another possibility, generally observed
with lanthanide catalysts: K. N. Harrison, T. J. Marks, J. Am.
Chem. Soc. 1992, 114, 9220.
[35] C. Widauer, H. Grützmacher, T. Ziegler, Organometallics 2000,
19, 2097.
[36] a) J. B. Morgan, S. P. Miller, J. P. Morken, J. Am. Chem. Soc.
2003, 125, 8702; b) S. Trudeau, J. B. Morgan, M. Shrestha, J. P.
Morken, J. Org. Chem. 2005, 70, 9538; c) N. F. Pelz, A. R.
Woodward, H. E. Burks, J. D. Sieber, J. P. Morken, J. Am. Chem.
Soc. 2004, 126, 16328.
[37] a) R. T. Baker, J. C. Calabrese, S. A. Westcott, P. Nguyen, T. B.
Marder, J. Am. Chem. Soc. 1993, 115, 4367; b) G. Mann, K. D.
John, R. T. Baker, Org. Lett. 2000, 2, 2105; c) R. T. Baker, P.
Nguyen, T. B. Marder, S. A. Westcott, Angew. Chem. 1995, 107,
1451; Angew. Chem. Int. Ed. Engl. 1995, 34, 1336.
[38] C. A. G. Carter, C. M. Vogels, D. J. Harrison, M. K. J. Gagnon,
D. W. Norman, R. F. Langler, R. T. Baker, S. A. Westcott,
Organometallics 2001, 20, 2130.
[39] V. Lillo, J. A. Mata, A. M. Segarra, E. Peris, E. Fernandez,
Chem. Commun. 2007, 2184.
[40] It should be noted that under various conditions, the linear
isomer 2 is formed, making pathway b’ a reasonable option.
However under the conditions of this study, none of this product
is observed.
Published online: September 6, 2007
Keywords: Hammett substituent parameter · hydroboration ·
.
kinetic isotope effect · rhodium · vinyl arenes
[1] a) D. Männig, H. Nöth, Angew. Chem. 1985, 97, 854; Angew.
Chem. Int. Ed. Engl. 1985, 24, 878; b) C. M. Crudden, D.
Edwards, Eur. J. Org. Chem. 2003, 4695.
[2] T. Hayashi, Y. Matsumoto, Y. Ito, J. Am. Chem. Soc. 1989, 111,
3426.
[3] J. M. Brown, D. I. Hulmes, T. P. Layzell, J. Chem. Soc. Chem.
Commun. 1993, 1673.
[4] S. Demay, F. Volant, P. Knochel, Angew. Chem. 2001, 113, 1272;
Angew. Chem. Int. Ed. 2001, 40, 1235.
[5] A. Schnyder, L. Hintermann, A. Togni, Angew. Chem. 1995, 107,
996; Angew. Chem. Int. Ed. Engl. 1995, 34, 931.
[6] S. A. Moteki, D. Wu, K. L. Chandra, D. S. Reddy, J. M. Takacs,
Org. Lett. 2006, 8, 3097.
[7] a) A. C. Chen, L. Ren, C. M. Crudden, Chem. Commun. 1999,
611; b) A. C. Chen, L. Ren, C. M. Crudden, J. Org. Chem. 1999,
64, 9704.
[8] I. Beletskaya, A. Pelter, Tetrahedron 1997, 53, 4957.
[9] D. R. Edwards, C. M. Crudden, K. Yam, Adv. Synth. Catal. 2005,
347, 50.
[10] F. A. Cotton, T. J. Marks, J. Am. Chem. Soc. 1969, 91, 1339.
[11] F. A. Cotton, M. D. Laprade, J. Am. Chem. Soc. 1968, 90, 5418.
[12] Y. Becker, J. K. Stille, J. Am. Chem. Soc. 1978, 100, 845.
[13] U. Nettekoven, J. F. Hartwig, J. Am. Chem. Soc. 2002, 124, 1166.
[14] T. Braun, G. Münch, B. Windmüller, O. Gevert, M. Laubender,
H. Werner, Chem. Eur. J. 2003, 9, 2516.
[15] L. E. Crascall, S. A. Litster, A. D. Redhouse, J. L. Spencer, J.
Organomet. Chem. 1990, 394, c35.
[16] a) M. D. Fryzuk, D. H. McConville, S. J. Rettig, Organometallics
1990, 9, 1359; b) M. D. Fryzuk, D. H. McConville, S. J. Rettig, J.
Organomet. Chem. 1993, 445, 245.
[17] H. O. Stuhler, J. Pickardt, Z. Naturforsch. B 1981, 36, 315.
[18] H. Werner, Organometallics 2005, 24, 1036.
[19] P. Krüger, H. Werner, Eur. J. Inorg. Chem. 2004, 481.
[20] E. Bleuel, P. Schwab, M. Laubender, H. Werner, J. Chem. Soc.
Dalton Trans. 2001, 266.
[41] D. A. Singleton, A. A. Thomas, J. Am. Chem. Soc. 1995, 117,
9357.
[42] For a detailed discussion of the challenges of 13C KIE measure-
ments, see: K. P. Gable, F. A. Zhuravlev, J. Am. Chem. Soc. 2002,
124, 3970.
[43] This outcome is also consistent with an irreversible metal boryl
insertion that is not accompanied by any spurious metal hydride
additions.
[21] H. Werner, R. Feser, J. Organomet. Chem. 1982, 232, 351.
[22] A. Black, J. M. Brown, C. Pichon, Chem. Commun. 2005, 5284.
[23] The ratios of boronate esters were also examined in the crude
reaction mixtures and found to be identical to those obtained
after oxidation.
[44] Although the example given employs a neutral Rh catalyst
(Wilkinsonꢁs catalyst), it should be noted that the original papers
by Evans and Fu (references [32] and [33]) comment that the
chemistry observed with Wilkinsonꢁs catalyst is qualitatively
similar to that observed with cationic rhodium and dppb.
[24] The low mass balance in this case is likely due to interactions of
the basic amine with the hydroborating reagent.
7802 www.angewandte.org
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2007, 46, 7799 –7802