Steric parameters in the Ir-catalyzed regio- and diastereoselective isomerization of primary allylic alcohols

Article abstract of DOI:10.1021/ol403023x

The iridium-catalyzed diastereo- and regioselective isomerization of primary allylic alcohols using Crabtree's catalyst or sterically modified analogs is reported. The importance of the size of the substituents on either the substrates or the catalysts has been rationalized by linear free energy relationships.

Full text of DOI:10.1021/ol403023x

ORGANIC
LETTERS
2013
Vol. 15, No. 24
6170–6173
Steric Parameters in the Ir-Catalyzed
Regio- and Diastereoselective
Isomerization of Primary Allylic Alcohols
ꢀ
Houhua Li and Clement Mazet*
University of Geneva, Department of Organic Chemistry, 30 quai Ernest Ansermet,
1211 Geneva-4, Switzerland
Clement.mazet@unige.ch
Received October 21, 2013
ABSTRACT
The iridium-catalyzed diastereo- and regioselective isomerization of primary allylic alcohols using Crabtree’s catalyst or sterically modified
analogs is reported. The importance of the size of the substituents on either the substrates or the catalysts has been rationalized by linear free
energy relationships.
The systematic evaluation of the influence of electronic
and steric parameters in the outcome of selective catalytic
transformations is of prime importance, as it permits
rationales and predictive models to be elaborated. Since
the emergence of asymmetric catalysis in the late 1960s,
such exercises have been regularly practiced.1 Interestingly
the main focus has been placed on enantioselective cata-
lysis rather than diastereoselective catalysis.2 Similarly, the
quantification of electronic rather than steric parameters
has often been favored. Recently, Sigman and co-workers
have demonstrated that Charton and Sterimol parameters
could be appropriately used in the context of enantiose-
lective catalysis.3,4 They convincingly established that the
log of the enantiomeric ratio (er) of various asymmetric
transformations can be correlated with descriptors of the
size of the substituents of either the chiral ligands or the
substrates. Noticeably, examples of such analysis for
diastereoselective transformations are less common.5
As part of our program on the iridium-catalyzed en-
antioselective isomerization of primary allylic alcohols,6À8
we have shown that excellent Linear Free Energy
€
(5) (a) Adam, W.; Mitchell, C. M.; Saha-Moller, C. R. Eur. J. Org.
€
€
Chem. 1999, 785. (b) Meynhardt, B.; Luning, U.; Wolff, C.; Nather, C.
Eur. J. Org. Chem. 1999, 2327. (c) You, L.; Berman, J. S.;
Lucknasawichien, A.; Anslyn, E. V. J. Am. Chem. Soc. 2012, 134, 7126.
(6) For recent reviews, see: van der Drift, R. C.; Bouwman, E.; Drent,
ꢀ
ꢀ
E. J. Organomet. Chem. 2002, 650, 1. (b) Uma, R.; Crevisy, C.; Gree, R.
Chem. Rev. 2003, 103, 27. (c) Fu, G. C.; Modern Rhodium-Catalyzed
Organic Reactions; Evans, P. A., Ed.; WileyÀVCH: Weinheim, 2005;
Chapter 4. (d) Cadierno, V.; Crochet, P.; Gimeno, J. Synlett 2008, 1105.
(e) Mantilli, L.; Mazet, C. Chem. Lett. 2011, 40, 341. (f) Ahlsten, N.;
Bartoszewicz, A.; Martin-Matute, B. Dalton Trans. 2012, 41, 1660.
(7) For contributions from our group, see: (a) Mantilli, L.; Mazet, C.
(1) (a) Comprehensive Asymmetric Catalysis; Jacobsen, E. N., Pfaltz,
A., Yamamoto, I., Eds.; Springer: Berlin, 1999. (b) Fundamentals of
Asymmetric Catalysis; Kozlowski, M. C., Walsh, P. J., Eds.; University
Science Books: Sausalito, CA, 2009. (c) Modern Physical Organic Chem-
istry; Anslyn, E. V., Dougherty, D. A., Eds.; University Science Books: Mill
Valley, CA, 2006. (c) Hansch, C.; Leo, A.; Taft, R. W. Chem. Rev. 1991, 91,
165. (d) Bunten, K. A.; Chen, L.; Fernandez, A. L.; Poe, A. J. Coord.
Chem. Rev. 2002, 233À234, 41.
ꢀ
Tetrahedron Lett. 2009, 50, 4141. (b) Mantilli, L.; Gerard, D.; Torche, S.;
Besnard, C.; Mazet, C. Angew. Chem., Int. Ed. 2009, 48, 5143. (c)
Mantilli, L.; Mazet, C. Chem. Commun. 2010, 46, 445. (d) Mantilli, L.;
ꢀ
Gerard, D.; Torche, S.; Besnard, C.; Mazet, C. Chem.;Eur. J. 2010, 16,
12736. (e) Quintard, A.; Alexakis, A.; Mazet, C. Angew. Chem., Int. Ed.
(2) Mahatthananchai, J.; Dumas, A. M.; Bode, J. W. Angew. Chem.,
Int. Ed. 2012, 51, 10954.
ꢀ
2011, 50, 2354. (f) Mantilli, L.; Gerard, D.; Besnard, C.; Mazet, C. Eur.
(3) (a) Miller, J. J.; Sigman, M. S. Angew. Chem., Int. Ed. 2008, 47,
771. (b) Sigman, M. S.; Miller, J. J. J. Org. Chem. 2009, 74, 7633. (c)
Harper, K. C.; Sigman, M. S. Proc. Natl. Acad. Sci. U.S.A. 2011, 108,
2179. (d) Harper, K. C.; Sigman, M. S. Science 2011, 333, 1875. (e)
Harper, K. C.; Bess, E. N.; Sigman, M. S. Nat. Chem. 2012, 4, 366. (f)
Harper, K. C.; Vilardi, S. C.; Sigman, M. S. J. Am. Chem. Soc. 2013, 135,
2482.
J. Inorg. Chem. 2012, 20, 3320.
(8) For other examples of enantioselective isomerization of primary
allylic alcohols, see: (a) Botteghi, C.; Giacomelli, G. Gazz. Chim. Ital.
1976, 106, 1131. (b) Tanaka, K.; Qiao, S.; Tobisu, M.; Lo, M. M.-C.; Fu,
G. C. J. Am. Chem. Soc. 2000, 122, 9870. (c) Tanaka, K.; Fu, G. C.
J. Org. Chem. 2001, 66, 8177. (d) Chapuis, C.; Barthe, M.; de Saint
Laumer, J.-Y. Helv. Chim. Acta 2001, 84, 230. (e) Li, J.-Q.; Peters, B.;
Andersson, P. G. Chem.;Eur. J. 2011, 17, 11143. (f) Arai, N.; Sato, K.;
Azuma, K.; Ohkuma, T. Angew. Chem., Int. Ed. 2013, 52, 7500.
(4) (a) Wu, J. H.; Zhang, G.; Porter, N. A. Tetrahedron Lett. 1997, 38,
2067. (b) Quintard, A.; Alexakis, A. Org. Biomol. Chem. 2011, 9, 1407.
r
10.1021/ol403023x
Published on Web 11/12/2013
2013 American Chemical Society

Relationships (LFER) could be obtained by correlating
log(er) to the Charton values of the substrate substituent.
This eventually led tothe designofan improved generation
of catalysts.7d As a direct continuation of this work we
questioned whether LFER could also be elaborated for the
diastereoselective isomerization of primary allylic alcohols
with a vicinal stereocenter using achiral iridium catalysts.
We anticipated that, if successful, such a correlation might
certainly serve as a predictive tool in the design of complex
molecules synthesis.
Table 1. Diastereoselective Isomerization of Primary Allylic
Alcohols 1aÀg
R2
yieldb
(%)
drc,d
At the outset, because alkyl/alkyl and aryl/alkyl primary
allylic alcohols behave differently in the enantioselective
isomerization,6 racemic substrates belonging to these two
classes were synthesized. Starting from the appropriate
R-substituted ketones, the geometrically pure (E) primary
allylic alcohols 1aÀc (R1 = Ph, R2 = alkyl) and 1dÀg
(R1 = Me, R2 = alkyl) were prepared in practical yields
following a HornerÀWadsworthÀEmmons olefination/
selective reduction sequence (see Supporting Information
(SI) for details). The seven substrates were then subjected
to the standard isomerization protocol developed in our
group using 5 mol % of the Crabtree catalyst 2a (BArF as
anion, activation by H2 (1 min) followed by degassing).
The products were systematically isolated in excellent
entrya
1
R1
(Charton value, ν)
(syn/anti)
1
2
3
4
5
6
7
1a
1b
1c
1d
1e
1f
Ph
Ph
Ph
Me
Me
Me
Me
Me (0.52)
Et (0.56)
nPr (0.68)
Me (0.52)
Et (0.56)
iPr (0.76)
Cy (0.87)
78
89
81
94
96
92
94
1/2.6
1/2.3
1/2.1
1/2.4
1./1.1
22/1
1g
30/1
a Average of two runs (0.1 mmol of 1aÀg). b Isolated yield of the
corresponding alcohol after reduction. c Determined by 1H NMR.
d Relative configuration assigned by chemical correlation or by analogy.
1
yield, and the dr’s were measured by H NMR spectro-
scopy. For the aryl/alkyl substrates 1aÀc, the LFER
between log(dr) and the Charton steric parameter was
satisfactory, though the syn/anti ratios were moderate
across the series, with the anti isomer being always slightly
favored (Table 1, entries 1À3 and Figure 1 (top)). For
1dÀg, from a modest 1/2.4 syn/anti ratio for 1d, the
selectivity increased to 30/1 in favor of the syn isomer
for 1g (Table 1, entries 4À7). Remarkably, an excellent
linear correlation between log(dr) and the corresponding
Charton values ν was obtained (Figure 1 (bottom)).9,10
To evaluate the potential influence of the ligand struc-
ture on the diastereoselectivity of the isomerization reac-
tion, we prepared a series of Crabtree catalyst analogs
using ortho substituted pyridyl derivatives (R = Me, iPr,
and CHEt2) according to the one-pot protocol recently
developed in our laboratory.7f,11 Complexes 2bÀd were
isolated as air-stable orange solids in excellent yields
(72À97%). When 8-methyl-quinoline was employed,
an unexpected monohydride complex 2e was isolated in
89% yield after chromatography (Scheme 1). Its molecular
Figure 1. Charton plot for the diastereoselective isomerization
of 1aÀc (top) and 1dÀg (bottom) with catalyst 2a.
structure was unambiguously determined by 1H, 31P, 13C,
and 2D NMR spectroscopy (see SI). It is noteworthy that
cyclometalation occurred by a C(sp3)ÀH bond activation
at room temperature.12
(9) (a) Charton, M. J. Am. Chem. Soc. 1969, 91, 615. (b) Charton, M.
J. Am. Chem. Soc. 1975, 97, 1552. (c) Charton, M. J. Am. Chem. Soc.
1975, 97, 3691. (d) Charton, M. J. Am. Chem. Soc. 1975, 97, 3694.
(10) Excellent correlations were also obtained using the Sterimol
parameters. See Supporting Information.
The diastereoselective isomerization of 1a using these
five precatalysts was investigated next. Whereas complex
2e was found to be completely inactive using our activation
protocol, 2bÀd all delivered quantitatively the desired
aldehyde after 4 h at room temperature. Interestingly,
the syn/anti ratio decreased gradually from 1/2.6 (2a) to
1/1.7 (2d) as the size of the ortho substituent of the pyridine
ligand increased (Table 2).13 This trend is particularly well
visualized on the plot of log(dr) as a function of the
Charton values as displayed in Figure 2 for which, again,
an excellent correlation was obtained.14
(11) (a) Crabtree, R. H. Acc. Chem. Res. 1979, 12, 331. (b) Crabtree,
R. H.; Gautier, A.; Giordano, G.; Kahn, T. J. Organomet. Chem. 1977,
€
141, 113. (c) Wustenberg, B.; Pfaltz, A. Adv. Synth. Catal. 2008, 350, 174.
(12) Cyclometallated cationic iridium hydride complexes are not
common and are usually the result of a C(sp2)ÀH bond activation.
For relevant examples, see: (a) Crabtree, R. H.; Lavin, M.; Bonneviot, L.
J. Am. Chem. Soc. 1986, 108, 4032. (b) Patel, B. P.; Crabtree, R. H.
J. Am. Chem. Soc. 1996, 118, 13105. (c) Li, X.; Incarvito, C. D.;
Crabtree, R. H. J. Am. Chem. Soc. 2003, 125, 3698. (d) Esteruelas,
M. A.; Fernandez-Alvarez, F. J.; Olivan, M.; Onate, E. Organometallics
2009, 28, 2276. (e) Song, G.; Su, Y.; Periana, R. A.; Crabtree, R. H.; Han,
K.; Zhang, H.; Li, X. Angew. Chem., Int. Ed. 2010, 49, 912.
Org. Lett., Vol. 15, No. 24, 2013
6171

a necessary requirement for productive isomerization of
the aldehyde (Figure 3).6,7 In line with this hypothesis,
when isomerization of 1e by Crabtree catalyst 2a was
performed in the presence of controlled amounts of water
(5À100 mol %, Table 3, entries 5À6), a ca. 80:20 mixture of
aldehyde 3e and homoallylic alcohol 4e was measured.15
Similar results were obtained when CH3CN was used as
additive(5 mol %, Table 3, entry7), further suggesting that
2-point binding of the allylic alcohol might also be per-
turbed by competitive coordination of polar solvents.16,17
Furthermore, the electronically disfavored insertion of
iridium hydride in R styrenyl positions might account for
the distinct regioselectivity observed for aryl/alkyl and
alkyl/alkyl allylic alcohols.18
Scheme 1. Synthesis of Crabtree Catalyst Analogs
Table 2. Diastereoselective Isomerization of 1a with Catalysts
2aÀe
loading
R
yieldb
(%)
drb
entrya
2
(mol %) (Charton value, ν)
(syn/anti)
Figure 2. Charton plot for the diastereoselective isomerization
of 1a with catalysts 2aÀd.
1
2
3
4
5
2a
2b
2c
2d
2e
5.0
7.5
7.5
10.0
7.5
H (0.52)
Me (0.56)
iPr (0.76)
CHEt2 (1.51)
À
>99
>99
>99
>99
nrc
1/2.6
1/2.3
1/1.9
1/1.7
ndd
Table 3. Regioselective Isomerization of 1e into Aldehyde 3e
and Homoallylic Alcohol 4e
a Average of two runs (0.1 mmol of 1a). b Determined by 1H NMR.
c No reaction. d Not determined.
A strikingly different outcome was observed when the
same study was performed using an alkyl/alkyl substrate
such as 1e. Whereas Crabtree catalyst 2a delivered 3e
quantitatively after 4 h at room temperature (syn/anti =
1/1.1) (Table 3, entry 1), a mixture of aldehyde 3e and of
homoallylic alcohols 4e (E/Z = 1:1) was obtained with the
sterically more demanding catalysts 2bÀd (Table 3, entries
2À4). The homoallylic alcohols were always measured as
the major products, the best regioselectivity being obtained
with catalyst 2c (3e/4e = 15/85; Table 3, entry 3).
loading
(mol %)
yield
(%)b
entrya
2
additive
3e/4eb
100/0
1
2
3
4
5
6
7
2a
2b
2c
2d
2a
2a
2a
5.0
7.5
7.5
7.5
5.0
5.0
5.0
À
À
À
À
>99
>99
>99
>99
>99
>99
97
38/62
15/85(58)c
21/79
H2O (5 mol %)
81/19
To account for this observation, we propose that the
ortho substituent of the pyridine ligand in 2bÀd might
project toward the first coordination sphere of the metal
center and prevent 2-point binding of the allylic alcohol,
H2O (100 mol %)
CH3CN (5 mol %)
78/22
69/31
a Average of two runs (0.1 mmol of 1e). b Determined by 1H NMR.
c Isolated yield of 4e.
(13) The catalyst loading was adjusted to achieve complete conver-
sion of the starting material. The reduced catalytic activity for sterically
demanding iridium complexes is in line with our previous results using
chiral catalysts. See ref 7 for details.
(14) The CHEt2 substituent has two Charton values: 1.28 and 1.51.
The latter was used in this study, but correlation with the second value
was equally good. See Supporting Information.
(16) The effect of added water clearly indicates that the aged catalyst
may result not only in lower performances but also in varying selectivity
(if any). We indeed found that the iridium catalysts with the BArF anion
are quite hygroscopic. Their performances are not altered with time if
kept in a desiccator or a glovebox.
(15) Similar results were obtained if H2O or CH3CN were added
prior or after activation of the iridium catalyst by molecular hydrogen.
(17) Preliminary investigations showed that 1d and 1fÀg behave
similarly.
6172
Org. Lett., Vol. 15, No. 24, 2013

19% yield (syn/anti = 1/1.1), clearly indicating that 4e is
not an intermediate in the isomerization of 1e to 3e but is
rather produced by an independent catalytic manifold.
Despite the low yield, this result is remarkable because
the double bond in 4e is tetrasubstituted and previous
attempts to isomerize homoallylic alcohols with Crabtree
catalyst were met with failure.6À8
In conclusion, we have shown that the diastereoselective
isomerization of primary allylic alcohols can be quantified
using steric descriptors for both the substrate substituents
and the catalyst substituents. We also found that by using
sterically demanding iridium catalysts with an alkyl/alkyl
allylic alcohol the regioselectivity of the reaction can be
switched toward the formation of homoallylic alcohols
preferentially over the formation of aldehydes. General-
ization of these observations and extension of this study
to enantiopure substrates are currently underway in our
laboratory.
Figure 3. Plausible explanation for the switch in regioselectivity
using bulkier analogs of Crabtree catalyst with alkyl/alkyl
primary allylic alcohols.
Scheme 2. Isomerization of Homoallylic Alcohols 4ea
Acknowledgment. This work was supported by the
University of Geneva, the Swiss National Foundation
(Project PP00P2_133482), Roche, and the Schmidheiny
Foundation. We thank Johnson-Matthey for a generous
gift of the iridium precursor.
a Average of two runs (0.1 mmol of 4e). b Determined by 1H NMR.
Finally, homoallylic alcohol 4e was isolated as a 1/1 E/Z
mixture and resubmitted to the standard isomerization
conditions using Crabtree catalyst 2a (Scheme 2). After 4 h
at room temperature, aldehyde 3e was obtained in only
Supporting Information Available. Experimental pro-
cedures and spectroscopic and analytical data. This ma-
terial is available free of charge via the Internet at http://
pubs.acs.org.
ꢀ
(18) Mazet, C.; Gerard, D. Chem. Commun. 2011, 47, 298.
The authors declare no competing financial interest.
Org. Lett., Vol. 15, No. 24, 2013
6173