Remote supramolecular control of catalyst selectivity in the hydroformylation of alkenes

Article abstract of DOI:10.1002/anie.201005173

In the pocket: The supramolecular interactions between a Rh phosphine catalyst equipped with an anion-binding pocket and alkenes that contain anionic functionalities (see picture) provide an excellent design concept to achieve remote control of the regioselectivity in hydroformylation reactions. The 4-pentenoate and 3-butenylphosphonate, which fit tightly between the Rh center and the pocket, were hydroformylated with unprecedented selectivity.

Full text of DOI:10.1002/anie.201005173

Communications
DOI: 10.1002/anie.201005173
Supramolecular Catalysis
Remote Supramolecular Control of Catalyst Selectivity in the
Hydroformylation of Alkenes**
Paweł Dydio, Wojciech I. Dzik, Martin Lutz, Bas de Bruin, and Joost N. H. Reek*
Dedicated to Professor Janusz Jurczak on the occasion of his 70th birthday
Immense progress in the field of transition-metal catalysis has
been achieved over the past few decades, and the contribu-
tions of the ligands that are coordinated to the metals are now
well understood.[1] Despite notable insights made into various
reaction mechanisms, the prediction of the selectivity that a
new catalyst will display is still beyond our abilities. This
becomes a particularly difficult issue when the reaction
pathways that lead to the isomeric products are nearly
identical in energy, or worse, if the pathway to the desired
isomer is higher in energy. For these challenging reactions, the
trial-and-error approach is still dominant in the search for
appropriate catalysts, and thus, combinatorial methods and
high-throughput screening of ligands and catalysts have been
developed.[2] Supramolecular ligands that form by self-
assembly of smaller components appear suitable for this
approach, as the modular synthesis efficiently generates wide
libraries of ligands.[3]
Enzyme mimicry represents an alternative route to
selective catalysts. Inspired by the properties and working
principles of enzymes, a great effort in catalyst development
has been applied to the incorporation of cavitands that can
bind guest molecules to an active site and promote reactions
that are typically displayed by enzymes.[4] Remarkable
examples of highly selective oxidation reactions catalyzed
by metalloporphyrins, where hydrophobic interactions allow
for substrate preorganization, have been developed by
Breslow and co-workers.[5] However, simple hydrogen-bond-
ing interactions between a substrate and the functional groups
of a catalyst can also be used to greatly improve the selectivity
of a reaction. This principle was elegantly demonstrated by
hybridized carbon atoms.[6] Recently, our research group
showed that even a single hydrogen bond between a catalyst
and a substrate leads to improved activities in cyclopropana-
tion reactions,[7] and excellent enantioselectivity in the hydro-
genation of the Roche ester precursor.[8] A similar supra-
molecular substrate preorganization strategy has been
reported by Breit and co-workers, who applied guanidi-
nium-functionalized phosphines to the regioselective hydro-
formylation of vinylacetic acid and its analogues.[9]
As supramolecular interactions can be arranged relatively
easily, selective installation of functional groups can provide a
powerful tool for the rational design of selective catalysts that
operate predicatively. The approach requires a set of recep-
tors that can address a wide range of functional groups. In
addition, the impact would be larger if the selectivity could
also be controlled by noncovalent interactions that are more
remote from the catalytic center. Herein we report a
bisphosphine ligand based on an anion receptor backbone.
As predicted, the ligand can be used in a regioselective
rhodium-based hydroformylation catalyst for substrates that
have anionic groups. Remarkably high regioselectivities are
obtained for those substrates that precisely span the distance
between receptor and rhodium center, and substrates with
various anionic groups can be used. DFT calculations of the
essential intermediates show that the hydride migration
pathway to the undesired product is blocked by the anion
binding, whereas that for the desired product is lowered in
energy.
We anticipated that neutral anion receptors are excellent
candidates for the design of substrate-directing motifs in
supramolecular catalysts. The relatively strong and highly
directional interactions of these receptors with anionic
substrates[10] allow the predictable orientation of a reactive
functionality close to a catalytic metal center. In this study, we
fused a 7,7’-diamido-2,2’-diindolylmethane (DIM) scaffold—
a tailor-made receptor[11] for carboxylate and phosphate
anions—with two triphenylphosphine moieties to generate a
new bidentate ligand (DIMPhos, 1, Figure 1). DFT calcula-
tions show that ligand 1 forms a rigid mononuclear Rh
complex, which can orient functionalized alkenes for selective
hydride migration as a part of the hydroformylation cycle.
The ligand DIMPhos (1) is prepared easily by hydro-
genation of 7,7’-dinitro-2,2’-diindolomethane 2[11] using H2
and Pd/C and subsequent condensation with 4-(diphenyl-
phosphino)benzoic acid by following a standard protocol,[12]
which provided bisphospine 1 in an overall yield of 62%
(Scheme 1).
Crabtree, Brudvig, and co-workers in a dimanganese catalyst
3
À
for the highly selective functionalization of C H bonds at sp -
[*] P. Dydio, W. I. Dzik, Dr. B. de Bruin, Prof. Dr. J. N. H. Reek
Van’t Hoff Institute for Molecular Sciences
University of Amsterdam, Science Park 904
1098 XH, Amsterdam (The Netherlands)
Fax: (+31)20-525-56-04
E-mail: j.n.h.reek@uva.nl
Homepage: http://www.science.uva.nl/research/imc/HomKat/
Dr. M. Lutz
Bijvoet Center for Biomolecular Research
Utrecht University (The Netherlands)
[**] We kindly acknowledge the NRSC-C for financial support, Dr. T.
Gadzikwa for fruitful discussions, and R. Bellini and Dr. M. Pilar del
Rio Varea for assistance with the high-pressure and low-temperature
NMR experiments.
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/anie.201005173.
396
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2011, 50, 396 –400

group of the anionic guest points towards the metal center,
as is consistent with our design.
High-pressure (HP) NMR experiments show that a 1:1
mixture of ligand 1 and [Rh(acac)(CO)2] (acac = acetylacet-
onate) under hydroformylation conditions, 5 bar CO/H2 (1:1),
forms exclusively a mononuclear trigonal bipyramidal rho-
dium complex [Rh(1)(CO)2H].[14] Low-temperature NMR
studies reveal that bisphosphine ligand 1 is coordinated
predominantly in an equatorial–equatorial (ee) fashion. At
room temperature, this complex is in fast equilibrium on the
NMR timescale with the minor equatorial–apical (ea)
isomer.[15] Indeed, high-pressure infrared (IR) spectra using
either H2/CO or D2/CO (both 1:1) revealed four bands in the
carbonyl region that correspond to the ee and ea isomeric
complexes.[14,16] Moreover, HP IR studies indicate a fast
catalyst activation process, the conversion of [Rh(1)(acac)]
into [Rh(1)(CO)2H] takes less than 2 hours, even at room
temperature. HP NMR experiments show that the coordina-
tion geometry around the Rh center does not change in the
presence of carboxylate (acetate) or phosphate ions
Figure 1. Structure of ligand 1 (left) and general concept of anionic
substrate preorganization by a Rh catalyst that bears a ligand
furnished with an anion-binding pocket (right).
À
(H2PO4 ). Under these conditions, the signals of the NH
groups are shifted towards lower fields in the 1H NMR spectra
(Dd = 2.5–3.1 ppm), thus signifying the formation of strong
hydrogen bonds between the anion and the pocket.[11] The
carbonyl bands in the HP IR spectra also show a small shift to
lower wavenumbers (Dn up to 5 cmÀ1) upon anion binding,
indicating an increased electron density at the phosphorus
atoms.
Scheme 1. Synthesis of ligand 1: a) H2, Pd/C(10%), MeOH, RT;
b) 4-(Ph2P)PhCOOH, diisopropylcarbodiimide (DIC), 4-dimethylamino-
pyridine (DMAP), 4-pyrrolidinopyridine, CH2Cl2, RT.
Upon addition of [RhCl(CO)2]2 to a CH2Cl2 solution of 1,
which had an acetate ion bound in the DIM pocket, an Rh
complex was formed with two P donors coordinated to the Rh
center in a mutual trans orientation. This coordination
geometry was further supported by the X-ray structure of
TBA[Rh(1·AcO)(CO)Cl] (Figure 2; TBA = tetrabutylam-
monium).[13] Importantly, the acetate ion remains bound in
the DIM pocket of the Rh complex of ligand 1. As
anticipated, all four NH groups are engaged in acetate
binding and the relative short distances indicate that the
hydrogen bonds formed are relatively strong (the N–O
distances are 2.737(3) and 3.006(3) ꢀ, for the amide and
indole N atoms, respectively). Importantly, the aliphatic
The binding constants of carboxylate and phosphate
anions in the DIM pocket of [Rh(1)(CO)2H] were determined
from titration experiments carried out at 5 bar CO/H2 (1:1) in
CH2Cl2 by monitoring the chemical shift of various protons by
using HP NMR spectroscopy. These titration studies reveal
that only one anion is bound in the DIM receptor of
[Rh(1)(CO)2H], and that the association constants for
CH3COOÀ and H2PO4 are higher than 105 and around
À
103.7, respectively. In contrast to these anionic species, non-
deprotonated CH3COOH and H3PO4, as well as their methyl
and ethyl esters, are not bound in the pocket of ligand 1.
Next, we applied ligand 1 in the rhodium-catalyzed
hydroformylation of a series of (deprotonated) w-unsaturated
carboxylic acids with a range of aliphatic chain lengths
between the carboxylic functionality and the double bond,
that is, from 3-butenoic to 8-nonenoic acid. Calculations show
À
that 3-butenoate anion (C3-CO2 ) is too small to bind in the
pocket, while its double bond is coordinated to the metal
À
center, but the 4-pentenoate anion (C4-CO2 ) fits precisely,
and the other substrates easily span the distance between
metal and binding site. The neutral acids and methyl esters,
that is, substrates that do not bind in the pocket, were used as
control experiments. The linear/branched selectivity (l/b)
determined by 1H NMR spectroscopy is shown in Figure 3.[14]
For the small substrates based on 3-butenoic acid, we hardly
observe any difference between the anion and the acid and
ester analogues, and the l/b ratio is in the expected range
between 1.6 and 2.6. In contrast, the 4-pentenoate ion, which
is the substrate that precisely spans the distance between the
metal and the pocket, is hydroformylated with unprecedented
selectivity for the linear aldehyde (l/b ratio of 40). If the
Figure 2. Crystal structure of the supramolecular complex
[Rh(1·AcO)(CO)Cl]À; TBA+ counterion and disordered solvent mole-
cules are omitted for clarity.
Angew. Chem. Int. Ed. 2011, 50, 396 –400
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
397

Communications
This result suggests that the reaction barrier for formation of
the linear aldehyde is lowered by the substrate-preorganizing
binding event, while the barrier to the branched aldehyde is
increased. To further verify that the anion binding unit and
the catalytic center must be present as an integrated system,
we performed a control experiment using a mixture of the
anion receptor (4) and triphenylphosphine (3). In this case, as
expected, the 4-pentenoate ion (5) is hydroformylated with
the typical selectivity displayed by catalysts based on triphe-
nylphospine (3; Table 1, entries 2 and 3). In the absence of a
phosphorus ligand, substrate 5 is hydroformylated with poor
selectivity (l/b = 1.8) and low convrsion (Table 1, entry 4).
Importantly, for non-anionic substrates, the addition of excess
acetate ion that binds in the DIM pocket of [Rh(1)(CO)2H]
has no effect on the (low) regioselectivity.[17]
The substrates that are longer than the one that precisely
fits between the metal and the binding site—5-hexenoate
through 8-nonenoate anion—also experience an effect of
binding in the DIM pocket. These substrates are hydro-
formylated with higher selectivity for linear products than
their acid or ester analogues. The l/b factors are all greater
than 20 for the anionic substrates, and between 5 and 10 for
the substrates that do not bind to the DIM. Importantly, the
highest regioselectivity is obtained for the 4-pentenoate ion,
which is the substrate that fits best in the catalytic pocket.
We extended the scope of substrates to alkenes function-
alized with a phosphate group (Table 2). Again, we expected
that 3-butenylphosphonate ion (C4-PO32À) would be hydro-
formylated with the highest selectivity, as this substrate
Figure 3. Hydroformylation of w-unsaturated carboxylic acids and
methyl esters. [Rh(acac)(CO)2]/1/substrate=1:3:100;
c0(substrate)=0.2m, 20 bar CO/H2 (1:1), CH2Cl2, 408C, 24 h. N,N-
diisopropylethylamine (DIPEA) was used as a base for anionic
substrate generation (cDIPEA =0.3m). Regioselectivity and conversion
1
(%) were determined by analysis of the reaction mixture by H NMR
spectroscopy. No side reactions (isomerisation, hydrogenation, etc.)
were observed.
reaction is carried out at room temperature (instead of 408C)
the l/b ratio exceeds 50 (Table 1, entry 1). For the protonated
or the methyl ester analogues, substrates of the same size that
Table 2: Hydroformylation of w-unsaturated phosphonic acids and ethyl
esters.[a]
Table 1: Hydroformylation of 5.[a]
Entry Substrate [n] Form (OR) Conversion [%] Regioselectivity
[l/b ratio]
1
2
3
4
5
6
1
1
1
2
2
2
ester (OEt)
acid (OH)
3
–
0.6
–
[b]
anion (OÀ) 10 (69)[c]
– (1.6)[c]
4.0
ester (OEt) 12
[b]
acid (OH)
–
–
>40
Entry
Ligand
Conversion [%]
Regioselectivity
[l/b ratio]
anion (OÀ) 100
[a] [Rh(acac)(CO)2]/1/substrate=1:3:100; c0(substrate)=0.2m, 20 bar
CO/H2 (1:1), DCM, RT, 24 h, DIPEA used as a base for anionic substrate
generation (cDIPEA =0.6m). Regioselectivity and conversion (%) were
determined by 1H and 13C NMR analysis of the reaction mixture.
[b] Additional experiments with 1-octene showed that the catalyst is
inactive under these strongly acidic conditions. [c] The conversion is too
low at RT to determine the selectivity; values between parentheses are for
the reaction at 408C.
1
2
1
95 (80)[b]
100
100
40 (>50)[b]
2.9
3.1
1.8
4/3 (1:2)
3
–
3[c]
4
13
[a] [Rh(acac)(CO)2]/1(4 and/or 3)/5/DIPEA=1:3(3 and/or 6):100:150;
c0(5)=0.2m, 20 bar CO/H2 (1:1), DCM, 408C, 24 h. Regioselectivity and
conversion (%) were determined by 1H NMR analysis of the reaction
mixture. [b] Reaction at RT, 72 h. [c] The results for other studied acids
(Figure 3 are similar to those (see the Supporting Information for
details).
exactly spans the distance between the binding site and the
metal center. Indeed, this substrate is hydroformylated by
[Rh(1)(CO)2H] to form the linear aldehyde with excellent
regioselectivity (l/b > 40). Again, the high selectivity and
higher conversion are observed only when the anionic
substrate is used (Table 2, entries 4–6). The shorter allyl-
do not bind in the DIM pocket, typical low l/b values of
around 3 are obtained, confirming the importance of the
anion binding. Besides the higher selectivity, the conversion is
also much higher when the substrate is bound in the pocket.
398
www.angewandte.org
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2011, 50, 396 –400

positions are inverted (Fig-
ure 4d), and which favors
the rotation of the alkene
toward the transition state
leading to the branched
alkyl intermediate. How-
ever, this conformation has
much
higher
energy
(15.8 kJmolÀ1), which is
even higher than the tran-
sition state that leads to the
linear product (DDG8 =
4.6 kJmolÀ1). Thus, the cal-
culations corroborate our
assertion that the high
regioselectivity
obtained
by the Rh(1) catalyst
stems from substrate orien-
tation by the hydrogen
bonds between the anionic
functionality and the DIM
Figure 4. Calculated reaction pathway (DFT, BP86, SV(P)) of the regioselectivity-determining hydride-migration
step in the hydroformylation of 4-pentenoate by Rh(1) catalyst; catalyst–substrate complex (a), transition state
toward linear product (b) and linear alkyl Rh complex (c), and alternative catalyst–substrate complex that
favors formation of the branched product (d). G298: Gibbs free energy at 298 K (relative to the catalyst–
substrate complex (a)) in kJmolÀ1
.
pocket of the ligand. Bind-
ing of the substrate in the
phosponate ion (C3-PO32À) is hydroformylated with similar
regioselectivity as the analogue that does not bind to the
pocket (Table 2, entries 1 and 3).
To gain a deeper understanding of the origin of the high
selectivity in the hydroformylation of anionic substrates, we
studied the Rh(1) catalytic system with DFT calculations
(BP86, SV(P)). Since, in the absence of isomerization, the
regioselectivity of the rhodium/phosphine-catalyzed hydro-
formylation is defined by insertion of the olefin into the
pocket favors one reaction pathway and hinders the compet-
ing pathway that would lead to the isomeric product by
restricting the movements of the reactive functionality during
the key selectivity-determining step.
In summary, we have introduced DIMPhos (1), which is a
new bidentate phosphorus ligand with an integral anion-
recognition site. We have demonstrated that under hydro-
formylation conditions, well-defined Rh complexes are
formed that strongly bind anionic substrate species. DFT
calculations show that the 4-pentenoate ion exactly spans the
distance between the rhodium center and the receptor site,
and this substrate orientation should lead to selective
reactions. Indeed, high selectivity for the linear aldehyde is
observed (l/b = 40) for this substrate, in contrast to those that
do not bind or are
too small. The substrate scope can be extended to larger
w-unsaturated carboxylic acids, that is, up to 8-nonenoic acid,
and phosphonic acids, and thus provides the first example of
wide-ranging remote control of catalyst selectivity by secon-
dary substrate–ligand interactions. The use of functionalized
bidentate ligands such as DIMPhos (1) has an advantage over
functionalized mondentate ligands that the number of
possible complexes that can be formed is significantly fewer,
which may be important for the design of selective catalysts
based on substrate orientation.[9] We are currently investigat-
ing optimization of the activity of this system, and are
applying supramolecular ligands equipped with the anion
binding pocket to other challenging transformations.
À
Rh H bond to generate the Rh alkyl complex (alkene
hydrometalation),[1c] this step was studied in detail. We first
calculated several possible structures of the complex
[RhH(CO)(1)]–(4-pentenoate), and found that the ee coor-
dination geometry was favored for this complex. The energy-
minimized structure shows that the carboxylate group of the
substrate is strongly bound in the pocket by four hydrogen
bonds (dNÀO = 2.7–2.9 ꢀ), and the coordinated alkene moiety
is already tilted out of the P-Rh-P plane of the trigonal
bipyramidal complex (Figure 4a). This perturbation arises as
the carboxylate ion is anchored in the pocket, in addition
severely restricting the movement of the coordinated double
bond. However, the alkene easily rotates towards the
transition state, leading to the linear alkyl Rh complex. In
fact, the geometry of the complex in the calculated early
transition state (DG° = 11.2 kJmolÀ1) is almost unperturbed
À
(the Rh H bond elongates by only 0.036 ꢀ), with the alkene
rotated only a little further out of the equatorial plane
(Figure 4b). There is no low-barrier reaction pathway
towards the branched alkyl Rh intermediate from this
catalyst–substrate complex conformer, since the double
bond cannot rotate in the necessary direction due to the
carboxylate moiety being anchored within the pocket.[18] This
result indicates that the branched aldehyde that is formed
during the reaction follows a pathway in which the anion is
not bound in the receptor. We also modeled the conformer of
the initial complex in which the carbonyl and hydride
Received: August 18, 2010
Revised: September 30, 2010
Keywords: hydroformylation · hydrogen bonds ·
molecular recognition · P ligands · supramolecular catalysis
.
Angew. Chem. Int. Ed. 2011, 50, 396 –400
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
399

Communications
[7] a) W. I. Dzik, X. Xu, X. P. Zhang, J. N. H. Reek, B. de Briun,
J. Am. Chem. Soc. 2010, 132, 10891 – 10902; b) S. Zhu, J. V.
Ruppel, H. Lu, L. Wojtas, X. P. Zhang, J. Am. Chem. Soc. 2008,
130, 5042 – 5043.
[1] a) J. F. Hartwig, Organotransition Metal Chemistry: From Bond-
ing to Catalysis, University Science Books, 2009; b) Transition
Metals for Organic Synthesis (Eds.: M. Beller, C. Bolm), Wiley-
VCH, Weinheim, 2004; c) P. W. N. M. van Leeuwen, Homoge-
neous Catalysis: Understanding the Art, Kluwer Academic
Publishers, Dordrecht, 2004.
[8] P.-A. R. Breuil, F. W. Patureau, J. N. H. Reek, Angew. Chem.
2009, 121, 2196 – 2199; Angew. Chem. Int. Ed. 2009, 48, 2162 –
2165.
ˇ
[2] a) E. P. Goudriaan, P. W. N. M. van Leeuwen, M.-N. Birkholz,
J. N. H. Reek, Eur. J. Inorg. Chem. 2008, 2939 – 2958; b) B. Breit,
Pure Appl. Chem. 2008, 80, 855 – 860; c) M. T. Reetz, Angew.
Chem. 2008, 120, 2592 – 2626; Angew. Chem. Int. Ed. 2008, 47,
2556 – 2588; d) J. G. de Vries, A. H. M. de Vries, Eur. J. Org.
Chem. 2003, 799 – 811; e) C. Gennari, U. Piarulli, Chem. Rev.
2003, 103, 3071 – 3100; f) W. Tang, X. Zhang, Chem. Rev. 2003,
103, 3029 – 3070; g) C. Jꢁkel, R. Paciello, Chem. Rev. 2006, 106,
2912 – 2942.
[9] a) T. Smejkal, D. Gribkov, J. Geier, M. Keller, B. Breit, Chem.
ˇ
Eur. J. 2010, 16, 2470 – 2478; b) T. Smejkal, B. Breit, Angew.
Chem. 2008, 120, 317 – 321; Angew. Chem. Int. Ed. 2008, 47, 311 –
315.
[10] J. L. Sessler, P. A. Gale, W.-S. Cho, Anion Receptor Chemistry,
Royal Society of Chemistry, Cambridge, 2006.
´
[11] P. Dydio, T. Zielinski, J. Jurczak, Chem. Commun. 2009, 4560 –
´
4562; P. Dydio, T. Zielinski, J. Jurczak, Org. Lett. 2010, 12, 1076 –
1078.
[3] For reviews see: a) M. J. Wilkinson, P. W. N. M. van Leeuwen,
J. N. H. Reek, Org. Biomol. Chem. 2005, 3, 2371 – 2383; b) B.
Breit, Angew. Chem. 2005, 117, 6976 – 6986; Angew. Chem. Int.
Ed. 2005, 44, 6816 – 6825; c) A. J. Sandee, J. N. H. Reek, Dalton
Trans. 2006, 3385 – 3391; d) J. Meeuwissen, J. N. H. Reek, Nat.
Chem. 2010, 2, 615 – 621.
[12] J. Izdebski, D. Kunce, J. Pept. Sci. 1997, 3, 141 – 144.
[13] CCDC 787323 contains the supplementary crystallographic data
for this paper. These data can be obtained free of charge from
The Cambridge Crystallographic Data Centre via www.ccdc.
cam.ac.uk/data_request/cif.
[14] For details see the Supporting Information.
[4] For reviews, see: a) T. S. Koblenz, J. Wassenaar, J. N. H. Reek,
Chem. Soc. Rev. 2008, 37, 247 – 262; b) A. W. Kleij, J. N. H.
Reek, Chem. Eur. J. 2006, 12, 4218 – 4227; c) M. D. Pluth, R. G.
Bergman, K. N. Raymond, Acc. Chem. Res. 2009, 42, 1650 –
1659; d) M. Yoshizawa, J. Klosterman, M. Fujita, Angew.
Chem. 2009, 121, 3470 – 3490; Angew. Chem. Int. Ed. 2009, 48,
3418 – 3438; e) Artificial Enzymes (Ed.: R. Breslow), Wiley-
VCH, Weinheim, 2006; f) D. M. Vriezema, M. C. Aragonꢂs,
J. A. A. W. Elemans, J. L. M. Cornelissen, A. E. Rowan, R. J. M.
Nolte, Chem. Rev. 2005, 105, 1445 – 1490; for selected contribu-
tions, see: g) H. K. A. C. Coolen, P. W. N. M. van Leeuwen,
R. J. M. Nolte, Angew. Chem. 1992, 104, 906 – 909; Angew.
Chem. Int. Ed. Engl. 1992, 31, 905 – 907; h) P. Thordarson,
E. J. A. Bijsterveld, A. E. Rowan, R. J. M. Nolte, Nature 2003,
424, 915 – 918.
[5] a) J. Yang, B. Gabriele, S. Belvedere, Y. Huang, R. Breslow,
J. Org. Chem. 2002, 67, 5057 – 5067; b) J. Yang, R. Breslow,
Angew. Chem. 2000, 112, 2804 – 2806; Angew. Chem. Int. Ed.
2000, 39, 2692 – 2694; c) R. Breslow, X. Zhang, Y. Huang, J. Am.
Chem. Soc. 1997, 119, 4535 – 4536; d) R. Breslow, X. Zhang, R.
Xu, M. Maletic, R. Merger, J. Am. Chem. Soc. 1996, 118, 11678 –
11679.
[15] a) The relatively small phosphorous–hydride coupling constant
(2JPH = 4.0 Hz) measured at room temperature suggests predom-
ination of the ee isomer over the ea form, see: L. A.
Van der Veen, P. H. Keeven, G. C. Schoemaker, J. N. H. Reek,
P. C. J. Kamer, P. W. N. M. Van Leeuwen, M. Lutz, A. L. Spek,
Organometallics 2000, 19, 872 – 883; b) For low-temperature
NMR studies of a mixture of ee and ea complexes, see: A. Gual,
C. Godard, S. Castillon, C. Claver, Adv. Synth. Catal. 2010, 352,
463 – 477.
[16] L. A. Van der Veen, M. D. K. Boele, F. R. Bregman, P. C. J.
Kamer, P. W. N. M. Van Leeuwen, K. Goubitz, J. Fraanje, H.
Schenk, C. Bo, J. Am. Chem. Soc. 1998, 120, 11616 – 11626.
[17] In control experiments, we investigated the influence of acetate
ion binding within the DIM pocket of the catalyst, as well as the
effect of base (DIPEA) and carboxylic acid (acetic acid), on the
hydroformylation of a nonbinding substrate, methyl 4-pent-
enoate. In all cases, the influence on the selectivity, as compared
with the reaction without additives, was negligible (see the
Supporting Information).
[18] Despite many attempts, we were not able to find a transition
state for the formation of the branched alkyl complex from this
catalyst–substrate complex conformer.
[6] S. Das, C. D. Incarvito, R. H. Crabtree, G. W. Brudvig, Science
2006, 312, 1941 – 1943.
400
www.angewandte.org
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2011, 50, 396 –400