Regioselective Hydroformylation of Internal and Terminal Alkenes via Remote Supramolecular Control

Article abstract of DOI:10.1002/chem.202000620

Regioselective catalytic transformations using supramolecular directing groups are increasingly popular as it allows for control over challenging reactions that may otherwise be impossible. In most examples the reactive group and the directing group are close to each other and/or the linker between the directing group is very rigid. Achieving control over the regioselectivity using a remote directing group with a flexible linker is significantly more challenging due to the large conformational freedom of such substrates. Herein, we report the redesign of a supramolecular Rh–bisphosphite hydroformylation catalyst containing a neutral carboxylate receptor (DIM pocket) with a larger distance between the phosphite metal binding moieties and the DIM pocket. For the first time regioselective conversion of internal and terminal alkenes containing a remote carboxylate directing group is demonstrated. For carboxylate substrates that possess an internal double bond at the Δ-9 position regioselectivity is observed. As such, the catalyst was used to hydroformylate natural monounsaturated fatty acids (MUFAs) in a regioselective fashion, forming of an excess of the 10-formyl product (10-formyl/9-formyl product ratio of 2.51), which is the first report of a regioselective hydroformylation reaction of such substrates.

Full text of DOI:10.1002/chem.202000620

A Journal of
Accepted Article
Title: Regioselective Hydroformylation of Internal and Terminal Alkenes
via Remote Supramolecular Control
Authors: Pim R. Linnebank, Stephan Falcão Ferreira, Alexander M.
Kluwer, and Joost Reek
This manuscript has been accepted after peer review and appears as an
Accepted Article online prior to editing, proofing, and formal publication
of the final Version of Record (VoR). This work is currently citable by
using the Digital Object Identifier (DOI) given below. The VoR will be
published online in Early View as soon as possible and may be different
to this Accepted Article as a result of editing. Readers should obtain
the VoR from the journal website shown below when it is published
to ensure accuracy of information. The authors are responsible for the
content of this Accepted Article.
To be cited as: Chem. Eur. J. 10.1002/chem.202000620
Link to VoR: http://dx.doi.org/10.1002/chem.202000620
Supported by

10.1002/chem.202000620
Chemistry - A European Journal
COMMUNICATION
Regioselective Hydroformylation of Internal and Terminal Alkenes
via Remote Supramolecular Control
Pim R. Linnebank[a], Stephan Falcão Ferreira[a], Alexander M. Kluwer[b], Joost N. H. Reek*[a][b]
Abstract:
Regioselective
catalytic
transformations
using
Toste et al. reported
a
transition metal catalyst
supramolecular directing groups are increasingly popular as it allows
for control over challenging reactions that may otherwise be
impossible. In most examples the reactive group and the directing
group are close to each other and/or the linker between the directing
group is very rigid. Achieving control over the regioselectivity using a
remote directing group with a flexible linker is significantly more
challenging due to the large conformational freedom of such
substrates. Herein, we report the redesign of a supramolecular Rh-
encapsulated in a self-assembled cage that can be used for site
selective hydrogenation of polyenes.[16] Moreover, the selectivity
in hydroformylation reactions can also be controlled by substrate
preorganization via carboxylate directing groups. The
guanidinium functionalized monodentate phosphine ligands
introduced by Breit et al. convert terminal and internal alkenes to
the outermost aldehyde with high regioselectivity, provided that
the carboxylic acid and alkene are in close distance.[24] Our group
reported the regioselective hydroformylation of unsaturated
carboxylates using bisphosphine and bisphosphite ligands, which
contained a neutral anion receptor based on 7,7’-diamido-2,2’-
diindolylmethane(DIM pocket).[20-23,29] This class of ligands was
coined DIMPhos. The rhodium-DIMPhos catalyst based on L1
(Figure 1a) hydroformylates internal alkenes such as 4-hexenoate
with high regioselectivity (78:1 selectivity) but for longer
substrates the regioselectivity is much lower and application of L1
in the hydroformylation of natural fatty acids with a double bond
on the 9-position gives no regioselectivity (vide infra).[22] Currently
there are no hydroformylation catalysts that convert natural
monounsaturated fatty acids (MUFAs) in a regioselective fashion,
whereas such technologies may allow broader applicability of the
biofeedstock.[30-42]
bisphosphite hydroformylation catalyst containing
a
neutral
carboxylate receptor (DIM pocket) with a larger distance between the
phosphite metal binding moieties and the DIM pocket. For the first
time regioselective conversion of internal and terminal alkenes
containing a remote carboxylate directing group is demonstrated. For
carboxylate substrates that possess an internal double bond at the Δ-
9 position regioselectivity is observed. As such, the catalyst was used
to hydroformylate natural monounsaturated fatty acids (MUFAs) in a
regioselective fashion, forming of an excess of the 10-formyl product
(10-formyl/9-formyl product ratio of 2.51), which is the first report of a
regioselective hydroformylation reaction of such substrates.
Supramolecular approaches in transition metal catalysis offer
unique tools to achieve selectivity in transformations that are
otherwise difficult to control.[1–8] Unrivalled selectivity by
a)
supramolecular strategies has been demonstrated for
wide )array of organic and organometallic transformations.[9-28]
a
A
frequently applied strategy involves the use a functional group on
a substrate that serves as a directing group to control the
substrate coordination at the metal center, allowing for
differentiation of reactive sites that are otherwise indistinguishable
for transition metal catalysts. This strategy, coined substrate
preorganization, has been broadly demonstrated for substrates in
which the directing group is relatively close to the reactive
group.[13,14,22-26] It remains an open question if such a strategy can
be extended to substrates in which the functional group is remote
from the directing group, which may be especially challenging for
long flexible alkyl chain type substrates due to the
[Rh(9-decenoate)ÌL1)(CO) H]
[Rh(9-decenoate)ÌL2)(CO) H]
b)
large conformational freedom of such compounds.
Recently, Costas et al. reported a system in which
protonated aliphatic amines were oxidized by a manganese
catalyst functionalized with crown ether recognition sites, leading
to selective oxidation of the C-H carbons to yield a mixture of
position
8
and
9
oxidation products using substrate
preorganization.28
[a]
[b]
P. R. Linnebank, S.Falcão Ferreira, Prof. Dr. J. N. H. Reek
Homogeneous, Supramolecular and Bio-Inspired Catalysis
Van’t Hoff Institute for Molecular Sciences University of Amsterdam
Science Park 904, 1098 XH Amsterdam (The Netherlands)
E-mail: j.n.h.reek@uva.nl
A. M. Kluwer, J. N. H. Reek
InCatT B.V.
Science Park 904, 1098 XH Amsterdam (The Netherlands)
Figure 1. a) Redesign of a rhodium-monophenyl (L1) to a rhodium biphenyl (L2)
DIMPhos complex to match the distance between the acid directing group and
alkene functionality in typical fatty acids b) Modeling (DFT) of 9-decenoate as a
fatty acid model, bound ditopically to [Rh(L1)(H)(CO)] (left) and [Rh(L2)(H)(CO)]
(right). For clarity, the 9-decanoate substrate was coloured green.
1
This article is protected by copyright. All rights reserved.

10.1002/chem.202000620
Chemistry - A European Journal
COMMUNICATION
In this paper we report the redesign of DIMPhos ligand L1 to L2
in which the distance between the active metal and the binding
site matches that of typical natural fatty acids (Figure 1a), and
demonstrate that the concept of substrate orientation to control
the regioselectivity in hydroformylation also works when the
directing group is remote from the double bond.
average hydrodynamic radius (8.4 Å) (see Figure S11) is close to
that of the [Rh(L2)(acac)] species, indicating that carboxylate
binding in the DIM pocket preorganizes the two phosphorous
moieties for the formation of mononuclear species.[29]
A series of (deprotonated) ω-unsaturated carboxylic acids
with varying length between the alkene reactive group and the
carboxylate directing group was hydroformylated using [Rh(L2] as
a catalyst (Figure 2). As substrates we reacted 4-pentenoic acid
(n=2) up to10-undecenoic acid (n=8) both in the presence and the
absence of base.
The distance between rhodium and the 7,7’-diamido-2,2’-
diindolylmethane anion receptor for L1 (the DIM pocket, 6.8 Å,
Figure 1a) is significantly shorter than the carboxylate-alkene
distance of fatty acids (12.4 Å) and this mismatch was proposed
to be the reason for the low selectivity observed for long
infra).[22]
Indeed,
DFT
calculations
Figure 2. Hydroformylation of ω-unsaturated carboxylic acids using rhodium
complexes based on L2[a]
substrates
(vide
(BLYP,DZP,D3BJ) show that 9-decenoate, used as a model for
natural fatty acids, needs to fold significantly to bind ditopically to
[Rh(L1)(H)(CO)] (see Figure 1b).[43-45] It was hypothesized that an
extended ligand binds substrates with large carboxylate-alkene
distances in less folded manners and as a result leads to a higher
control over the regioselectivity. To achieve this goal, we
designed a ligand that has a biphenyl linker ((Figure 1a, L2)
between the DIM pocket and the phosphite donor atoms, instead
of the phenyl linker that is present in the original DIMphos ligand
(L1). DFT calculations of 9-decenoate ditopically bound to
[Rh(L2)(H)(CO)] (Figure 1b) indeed shows less folding compared
to binding to [Rh(L1)(H)(CO)]. This is also reflected in the lower
folding energy of 9-decenoate bound to a [Rh(L2)(H)(CO)] (9.4
kcal/mol for [Rh(9-decenoate) Ì L2)(CO)H] vs. 15.4 kcal/mol for
[Rh(9-decenoate) Ì L1)(CO)H] (see Table S6 in the supporting
info(SI)). Also binding enthalpies of aliphatic deprotonated ω-
unsaturated carboxylic acids with various lengths (3-butenoate to
10-undecenoate) to [Rh(L2)(H)(CO)] were calculated using DFT
(see Figure S74). These studies show that 8-nonenoate fits best
and has the highest binding enthalpy. In addition, 9-decenoate,
which is a better model for natural fatty acids, also binds well to
[Rh(L2)(H)(CO)]. Encouraged by these results, we synthesized
the L2 ligand using a similar synthetic strategy as previously
reported for L1 (see SI).[22]
30
25
20
15
10
5
27
23
14
14
5.7
5.4
4.8
3.0
3.7
3.5
3.4
2.1
2.2
1.5
0
n =
2
3
4
5
6
7
8
substrate 5.9 Å
length
7,2 Å
8,5 Å
9,7 Å
11,1 Å
12,4 Å
13,6 Å
[a] Reagents and conditions: [substrate] = 0.2 M, DIPEA (15 equiv. (blue bars)),
[Rh(CO)2(acac)] (1 mol %), L2 (1.1 mol %), 20 bar CO/H2 (1:1), 40°C, 24 h.
Conversion and regioselectivity determined by 1H NMR analysis of the crude
reaction mixture. For full experimental details, see the SI. The blue bars are
experiments in presence of base, and the red bars are in absence of base as
control experiment.
To investigate if ligand L2 formed a stable metal complex
with rhodium it was mixed with [Rh(acac)(CO)2], which is the
precursor complex of the hydroformylation catalyst, in a 1:1 ratio
In the presence of base, the carboxylate functional group of
the substrate binds in the DIM pocket and thus substrates long
enough to span the DIM pocket-rhodium distance are expected to
react with improved regioselectivity. In absence of base, the
protonated carboxylic acids do not bind in the DIM pocket and as
a result the directing group, the carboxylic acid, cannot be used
for substrate preorganization with the [Rh(L2)] catalyst and
should react with lower regioselectivity.[20,21] Because of this, the
protonated substrates were used as control experiments. All
substrates tested gave full conversion to the aldehyde and the
linear/branched (l/b) ratios of the aldehyde products were
determined by 1H-NMR spectroscopy (Figure 2). The catalytic
results show that the long anionic substrates (with n > 4, 7-
octenoate and longer) display much higher l/b ratios than the
protonated analogues. The distance between the carboxylate and
alkene function in these substrates is at least 9.7 Å, and this
shows this catalyst is able to control the regioselectivity via
substrate preorganization on remote distance. The highest l/b
ratio of 27 is obtained for the 8-nonenoate (n = 6), which is the
substrate that binds strongest to the catalyst as it fits perfectly
according to our modelling studies (vide supra). The longer
substrates that easily span the distance between the DIM pocket
and the rhodium center (n = 7 and 8) are also converted with
improved regioselectivity when preorganized, albeit with a lower
1
in CD2Cl2. H NMR studies show a well-defined complex formed
after mixing (see Figure S2) and DOSY spectroscopy reveals the
formation of a single species with a hydrodynamic radius of 7.9 Å
in line with the size of a mononuclear complex (see Figure S3).[46]
Upon addition of 1.5 equivalents of tetrabutylammonium acetate,
the N-H protons are downfield shifted, which shows the
carboxylate group hydrogen binds to the DIM pocket of the
[Rh(L2)(acac)] in a similar fashion as reported for complex
[Rh(L1)(acac)] (Figure S4).[21] Pressurization 5 bar of syngas
(H2:CO (1:1)) to the solution of the [Rh(acac)(L2)] complex
provides the corresponding pentacoordinate [Rh(L2)(CO)2H]
species as evidenced by the rhodium hydrido signal (δ = -10.5
ppm).[47] Next to the well-defined signal, also a broad rhodium-
hydrido signal (δ = -10.7 ppm) appears in the NMR spectrum,
which becomes larger over time (see Figure S4). DOSY
spectroscopy of the [Rh(L2)(CO)2H] complex under 5 bar CO/H2
(1:1) in CD2Cl2 gave a larger average hydrodynamic radius (12.1
Å) than for the [Rh(acac)(L2)] complex suggesting that the broad
signal is due to formation of dinuclear and/or oligomeric species
under these conditions (see Figure S10). However, under
identical conditions but in the presence of 4 equivalents guest that
binds in the DIMpocket (tetrabutylammonium acetate) the
2
This article is protected by copyright. All rights reserved.

10.1002/chem.202000620
Chemistry - A European Journal
COMMUNICATION
linear/branched ratio of 23 and 14, respectively, in line with the
binding energy calculated for these substrates. The smaller
substrates (n = 2 and 3) are not able to bind in a ditopic fashion
to [Rh(L2)] and thus the difference in l/b ratios between the
anionic and the protonated substrates is very small. Consistent
with our design model, the L2 system is indeed more selective
than the L1 system for long substrates (e.g. 9-decenoate: l/b of
7/1 for L1 and l/b 23/1 for L2, see Figure S18 for full comparison
of l/b ratios of L1 and L2).[21]
the catalyst that can also preorganize but is optimized for smaller
substrates, only slightly better selectivities are obtained than with
the PPh3 based catalysts. The same trend was observed in the
hydroformylation of 9-undecenoic acid; only the [Rh(L2)] catalyst
provides the product with high regioselectivity, yielding a 10-
formyl/9-formyl ratio of 6.9. Importantly, the redesigned [Rh(L2)]-
catalyst clearly outcompetes [Rh(L1)] with respect to
regioselectivity and conversion for the longer substrates, as a
result of more favourable ditopic binding.
Having established our redesigned [Rh(L2)]-catalyst is able
to control the regioselectivity of remote internal alkenes on
position D8 and D9 through preorganization, we extended our
system to naturally occurring monounsaturated fatty acids
(palmitoleic acid and myristoleic acid, Table 2). When myristoleic
acid is hydroformylated using the PPh3 based rhodium catalyst,
equal amounts of the 10-formyl and the 9-formyl products are
formed, in line with previous reports.[39-41] Furthermore, when the
same reaction was carried out with the [Rh(L1)]-catalyst, equal
amounts of the two regioisomers are obtained. In contrast, the
[Rh(L2)]-catalyst provides a 10-formyl/9-formyl ratio of 1.61 and
shows this catalyst can control the regioselectivity of this
substrate via substrate preorganization. For the fatty acids also
minor amounts of isomerization/hydroformylation products were
observed with the [Rh(L1)] and [Rh(L2)] catalysts, lowering the
overall selectivity.
Table 1. Selective hydroformylation of 8-decenoic acid and 9-undecenoic acida
Substrate
Ligand
Conversion
(%)
9-formyl/ 8-
formyl
9-formyl/all
other
isomers
Table 2. Hydroformylation of natural fatty acids[a]
8-decenoic
acid
8-decenoic
acid
8-decenoic
acid
L1
74
2.6
8.8
1.6
1.9
5.8
1.6
L2
97
PPh3
100
Substrate
Ligand
Conversion
(%)
10-
formyl/9-
formyl
10-formyl/all
other
isomers
L1
70
2.2
6.9
1.3
1.7
5.0
1.3
9-undecenoic
acid
9-undecenoic
acid
L2
96
Ligand
Conversion
(%)
10-
10-formyl /all
other isomers
formyl/9-
formyl
1.03
PPh3
100
9-undecenoic
acid
Myristoleic acid
Myristoleic acid
Myristoleic acid
Palmitoleic acidb
Palmitoleic acidb
Palmitoleic acidb
Oleic acidc
L1
27
0.79
1.23
1.0
[a] Reagents and conditions: [substrate]
= 0.2 M, DIPEA (1.5 equiv.),
L2
69
1.61
1.0
[Rh(CO)2(acac)] (2 mol %), L1 and L2 (2.2 mol %), PPh3 (6.6 mol%), 20 bar
CO/H2 (1:1), 60°C, 96 h. Conversion and regioselectivity determined by 1H NMR
analysis of the reaction mixture. For full experimental details, see the SI.
PPh3
L1
100
23
~1.0
~1.5
~1.0
n.d.
n.d.
n.d.
~0.8
~1.2
~1.0
n.d.
n.d.
n.d.
We continued our catalytic studies using internal alkenes
with a remote carboxylate group as substrates, which served as
models for natural monounsaturated fatty acids that posses an
internal double bond at the D9-position. Initial investigations were
conducted with 8-decenoate, which is the internal alkene
analogue of the most selective terminal alkene substrate(vide
supra), and 9-undecenoate, which has the exact alkene-
carboxylic acid distance as natural fatty acids (Table 1).[9,10,21]
When 8-decenoate was hydroformylated using the PPh3 based
catalyst the two aldehyde products were formed with a small
excess for the 9-formyl product.[9,10,21,39] Performing the same
reaction with the rhodium catalyst based on L2 that preorganizes
the substrate leads to high conversion with a high regioselectivity
to produce the 9-formyl product in excess (9-formyl/8-formyl ratio
is 8.8).
L2
66
PPh3
L1
100
21
Oleic acidc
L2
76
Oleic acidc
PPh3
100
[a] Reagents and conditions: [substrate]
= 0.2 M, DIPEA (1.5 equiv.),
[Rh(CO)2(acac)] (2 mol %), L1 and L2 (2.2 mol %), PPh3 (6.6 mol%), 20 bar
CO/H2 (1:1), 60°C, 96 h. Conversion determined by 1H NMR analysis of the
reaction mixture and the regioselectivity was determined by GC analysis after
methylation of the reaction mixture.[b] Methyl 9- and 10-formylpalmitate could
not be baseline separated on GC, therefore a larger error in the determined
regioselectivity is expected.[c] 9- and 10-formyl stearic acid not separable after
methylation on GC and therefore the regioselectivity was not determined (n.d.).
For full experimental details, see the SI.
The linear aldehyde product is also observed under these
conditions, which arises from an isomerization/hydroformylation
sequence, which is not uncommon for bisphosphite-based
catalysts.47 When we applied the rhodium catalyst based on L1,
Palmitoleic acid was converted with similar levels of
regioselectivity as observed for myristoleic acid, with the [Rh(L2)]-
catalyst being the only catalyst capable of controlling the
regioselectivity (10-formyl/9-formyl ratio is ~1.5 for [Rh(L2)] and
3
This article is protected by copyright. All rights reserved.

10.1002/chem.202000620
Chemistry - A European Journal
COMMUNICATION
~1.0 for [Rh(L1)] and [Rh(PPh3)](Table 2)). For the fatty acids also
minor amounts of isomerization/hydroformylation products were
observed with the [Rh(L1)] and [Rh(L2)] catalysts, lowering the
overall selectivity. Palmitoleic acid was converted with similar
levels of regioselectivity as observed for myristoleic acid, with the
[Rh(L2)]-catalyst being the only catalyst capable of controlling the
regioselectivity (10-formyl/9-formyl ratio is ~1.5 for [Rh(L2)] and
~1.0 for [Rh(L1)] and [Rh(PPh3)](Table 2)). A major platform
chemical, oleic acid, was also hydroformylated using these
catalysts. Similar conversion was observed with oleic acid as with
myristoleic acid and palmitoleic acid using the respective
catalysts. However, the regioisomers could not be separated on
GC and therefore the regioselectivity could not be determined.
longer substrates.21,48,49 In the experiment where the catalyst
concentration was reduced by a factor 10 (entry 4) the
regioselectivity further increased (10-formyl/9-formyl ratio to 2.31).
Somewhat counterintuitively, the conversion was also higher in
the experiment, reflecting the complicated kinetics of the system.
Such complicated kinetics was previously reported for [Rh(L1)],
in which the catalytically active species is in equilibrium with
dormant state complexes in which carboxylate groups of the
substrate and product are directly coordinated to rhodium.23,50
Increasing the rhodium:ligand ratio from 1:1.1 to 1:2 (entries 5 and
6) further improved the regioselectivity to yield a 10-formyl/9-
formyl ratio of 2.43 under dilute conditions (entry 6). Changing the
solvent from DCM to THF or DMF (entries 7 and 8) led to lower
activity and selectivity. Experiments performed at syngas
pressures of 50 bar instead of 20 bar (entries 9 and 10), but
otherwise identical conditions, led to an improved regioselectivity
of 2.10 and 2.51 for entries 9 and 10 respectively. Notably, also
the overall selectivity improved to 1.91 and 2.33 respectively,
which is explained by lower levels of isomerization of the alkene,
commonly observed for hydroformylation reactions carried out at
higher CO concentration.[51]
In conclusion, supramolecular substrate orientation is a
powerful tool to control selectivity in transition metal catalysis,
which has been mainly demonstrated for substrates in which the
supramolecular functional group is close to the reactive group. In
this paper we demonstrate that supramolecular substrate can
also work when this group is remote from the reactive group,
thereby increasing the scope of the approach. In order to show
this a previously reported hydroformylation catalyst with an
integrated anion receptor, DIMPhos [Rh(L1)], was redesigned to
accommodate larger substrates. This hydroformylation catalyst
([Rh(L2)]) converts substrates with high regioselectivity when the
carboxylate directing group is remote from the alkene group,
including monounsaturated fatty acids (MUFAs) and their model
substrates. The [Rh(L2)]-catalyst provides the hydroformylation
product with a 10-formyl/9-formyl ratio of 2.51 for myristoleic acid,
which represents the first selective catalyst for this biobased
compound. These results show that catalysts that operate via
supramolecular substrate preorganization can be redesigned to
provide selective catalysts for substrates of different sizes, and as
such we are able to make a catalyst that can convert fatty acids
in a regioselective fashion. This paves the way for the design of
other challenging conversions for which no catalysts exist yet.
With
a
regioselective hydroformylation catalyst for
monounsaturated fatty acids in hand, we optimized the reaction
conditions using myristoleic acid as a substrate to further improve
the selectivity (Table 3). We commenced our optimization studies
by lowering the reaction temperature to 40°C. This resulted in an
improvement of 10-formyl/9-formyl ratio from 1.61 to 1.87,
although at a lower conversion (69% vs. 33%) (see entry 1 of
Table 3). Under the same conditions (40°C) but at lower substrate
concentrations (compare entries 1-3) the regioselectivity was
further enhanced yielding 10-formyl/9-formyl ratios of 1.99 and
2.20 at a substrate concentration of 0.1 M and 0.02 M,
respectively.
Table 3. Optimization of regioselectivity of myristoleic acid[a]
En
try
[substr
ate] (M)
[catalyst]
(M)
Conver
sion
(%)
10-/9-
formyl
tetradeca
noic acid
10-formyl
tetradecanoic
acid /all other
isomers
1
0.2M
4mM
33
66
85
76
35
77
20
>1
32
63
1.87
1.99
2.20
2.31
1.82
2.43
1.42
n.d.
1.58
1.61
1.72
1.90
1.63
1.95
1.42
n.d.
2
0.1M
4mM
3
0.02M
0.02M
0.2M
4mM
4
0.4mM
4mM
5b
6b
7c
8d
9e
10e
0.02M
0.02M
0.02M
0.2M
0.4mM
0.4mM
0.4mM
4mM
Acknowledgements
2.10
2.51
1.91
2.33
NWO, the Dutch science foundation, is acknowledged for
financial support. We also would like to thank InCatT for financial
support and useful discussions.
0.02M
0.4mM
[a] Reagents and conditions: DCM, DIPEA (1.5 equiv. with respect to acid),
catalyst = [Rh(CO)2(acac)] L2 in a 1:1.1 ratio, substrate = myristoleic acid, 20
bar CO/H2 (1:1), 40°C, 96 h. Conversion determined by 1H NMR analysis of the
reaction mixture and the regioselectivity was determined via GC analysis
following methylation of the reaction mixture. [b] Rhodium:ligand ratio 1:2. [c]
THF used as solvent instead of DCM [d] DMF was used instead of DCM. For
full experimental details, see the SI.[e] 50 bar syngas (H2:CO)(1:1) was used
instead of 20 bar of syngas.
Keywords: Hydroformylation, Supramolecular chemistry,
Regioselectivity, Substrate preorganization, Fatty acid
functionalization
[1]
[2]
[3]
[4]
H. J. Davis, R. J. Phipps, Chem. Sci. 2017, 8, 864–877.
J. Meeuwissen, J. N. H. Reek, Nat. Chem. 2010, 2, 615–621.
P. Dydio, J. N. H. Reek, Chem. Sci. 2014, 5, 2135–2145.
S. S. Nurttila, P. R. Linnebank, T. Krachko, J. N. H. Reek, ACS Catal.
2018, 8, 3469–3488.
Most likely, the lower selectivity at higher substrate
concentration results from unselective hydroformylation reactions
in which the substrate is not ditopically bound, which is more
dominant at higher substrate concentrations, especially for these
4
This article is protected by copyright. All rights reserved.

10.1002/chem.202000620
Chemistry - A European Journal
COMMUNICATION
[5]
[6]
[7]
M. Raynal, P. Ballester, A. Vidal-Ferran, P. W. N. M. van Leeuwen, Chem.
Soc. Rev. 2014, 43, 1660–1733.
[41] J. Boulanger, A. Ponchel, H. Bricout, F. Hapiot, E. Monflier, Eur. J. Lipid
Sci. Technol. 2012, 114, 1439–1446.
M. Raynal, P. Ballester, A. Vidal-Ferran, P. W. N. M. van Leeuwen, Chem.
Soc. Rev. 2014, 43, 1734–1787.
[42] J. Bianga, K. Schlipköter, T. Gaide, A. J. Vorholt, A. Behr, ACS Catal.
2017, 7, 4163–4171.
D. M. Vriezema, M. C. Aragonès, J. A. A. W. Elemans, J. J. L. M.
Cornelissen, A. E. Rowan, R. J. M. Nolte, Chem. Rev. 2005, 105, 1445–
1489.
[43] G. te Velde, F. M. Bickelhaupt, E. J. Baerends, C. Fonseca Guerra, S. J.
A. van Gisbergen, J. G. Snijders, T. Ziegler, J. Comput. Chem. 2001, 22,
931–967.
[8]
[9]
D. Vidal, G. Olivo, M. Costas, Chem. Eur. J. 2018, 24, 5042–5054.
V. F. Slagt, J. N. H. Reek, P. C. J. Kamer, P. W. N. M. Van Leeuwen,
Angew. Chem. Int. Ed. 2001, 40, 4271–4274; Angew. Chem. 2001, 113,
4401–4404.
[44] Amsterdam Modeling Suite, Software for Chemistry and Materials,
http://www.scm.com
[45] E. van Lenthe, A. Ehlers, E.-J. Baerends, J. Chem. Phys. 1999, 110,
8943–8953.
[10] M. Kuil, T. Soltner, P. W. N. M. Van Leeuwen, J. N. H. Reek, J. Am.
Chem. Soc. 2006, 128, 11344–11345.
[46] R. Gramage-Doria, J. Hessels, S. H. A. M. Leenders, O. Tröppner, M.
Dürr, I. Ivanović-Burmazović, J. N. H. Reek, Angew. Chem. Int. Ed. 2014,
53, 13380–13384; Angew.Chem. 2014,126, 13598–13602.
[47] A. Van Rooy, P. C. J. Kamer, P. W. N. M. Van Leeuwen, K. Goubitz, J.
Fraanje, N. Veldman, A. L. Spek, Organometallics 1996, 15, 835–847.
[48] A. Mulder, J. Huskens, D. N. Reinhoudt, Org. Biomol. Chem. 2004, 2,
3409–3424.
[11] T. Gadzikwa, R. Bellini, H. L. Dekker, J. N. H. Reek, J. Am. Chem. Soc.
2012, 134, 2860–2863.
[12] M. Yoshizawa, M. Tamura, M. Fujita, Science 2006, 312, 251–254.
[13] Y. Kuninobu, H. Ida, M. Nishi, M. Kanai, Nat. Chem. 2015, 7, 712–717.
[14] H. J. Davis, M. T. Mihai, R. J. Phipps, J. Am. Chem. Soc. 2016, 138,
12759–12762.
[49] The effective concentration for substrate x bound in the DIM pocket was
roughly estimated: The maximum radius between the rhodium center and
the double bond of natural fatty acids is approximately 11 Å which is 1,1
x 10-8dm3. Which translates to 4/3π(1,1 x 10-8dm)3 = 5,5 x 10-24 dm3. Of
this spherical volume it was estimated the alkene could occupy 50%. This
translates to an effective concentration of ≈ 0.6 M. Since the substrate
concentration is 0.2 M for most experiments, non-bound substrates will
likely compete and allow for a non-selective background reaction. This
nicely explains why the regioselectivity increases upon lowering of the
concentration as this background reaction is repressed.
[50] E. Mieczyńska, A. M. Trzeciak, J. J. Ziółkowski, J. Mol. Catal. 1993, 80,
189–200.
[15] R. Breslow, X. Zhang, R. Xu, M. Maletic, R. Merger, J. Am. Chem. Soc.
1996, 118, 11678–11679.
[16] T. A. Bender, R. G. Bergman, K. N. Raymond, F. D. Toste, J. Am. Chem.
Soc. 2019, 141, 11806–11810.
[17] A. Bauer, F. Westkämper, S. Grimme, T. Bach, Nature 2005, 436, 1139–
1140.
[18] P. A. Lichtor, S. J. Miller, Nat. Chem. 2012, 4, 990–995.
[19] G. W. B. Siddhartha Das, Christopher D. Incarvito, Robert H.
Crabtree,Science 2004, 16750, 2004–2007.
[20] P. Dydio, W. I. Dzik, M. Lutz, B. de Bruin, J. N. H. Reek, Angew. Chem.
Int. Ed. 2011, 50, 396–400; Angew. Chem. 2011, 123, 416–420.
[21] P. Dydio, J. N. H. Reek, Angew. Chem. Int. Ed. 2013, 52, 3878–3882;
Angew. Chem. 2013, 125, 3970–3974
[51] R. Franke, D. Selent, A. Börner, Chem. Rev. 2012, 112, 5675–5732.
[22] P. Dydio, R. J. Detz, J. N. H. Reek, J. Am. Chem. Soc. 2013, 135, 10817–
10828.
[23] P. Dydio, R. J. Detz, B. de Bruin, J. N. H. Reek, J. Am. Chem. Soc. 2014,
136, 8418–8429.
[24] T. Šmejkal, B. Breit, Angew. Chem. Int. Ed. 2008, 47, 311–315; Angew.
Chem. 2008, 120, 317–321.
[25] W. Fang, B. Breit, Angew. Chem. Int. Ed. 2018, 57, 14817–14821; W.
Fang, B. Breit, Angew. Chem. 2018,130, 15033–15037.
[26] S. T. Bai, V. Sinha, A. M. Kluwer, P. R. Linnebank, Z. Abiri, B. de Bruin,
J. N. H. Reek, ChemCatChem 2019, 11, 1–9.
[27] S. Perdriau, S. Harder, H. J. Heeres, J. G. de Vries, ChemSusChem
2012, 5, 2427–2434.
[28] G. Olivo, G. Farinelli, A. Barbieri, O. Lanzalunga, S. Di Stefano, M.
Costas, Angew. Chem. Int. Ed. 2017, 56, 16347–16351; Angew. Chem.
2017, 129, 16565–16569.
[29] P. Dydio, T. Zieliński, J. Jurczak, Chem. Commun. 2009, 4560–4562.
[30] D. R. Dodds, R. A. Gross, Science. 2007, 318, 1250–1251.
[31] P. J. Deuss, K. Barta, J. G. De Vries, Catal. Sci. Technol. 2014, 4, 1174–
1196.
[32] T. Vanbésien, E. Monflier, F. Hapiot, Eur. J. Lipid Sci. Technol. 2016,
118, 26–35.
[33] A. Behr, D. Obst, A. Westfechtel, Eur. J. Lipid Sci. Technol. 2005, 107,
213–219.
[34] P. N. R. Vennestrøm, C. M. Osmundsen, C. H. Christensen, E. Taarning,
Angew. Chem. Int. Ed. 2011, 50, 10502–10509; Angew. Chem. 2011,
123, 10686–10694.
[35] S. Pandey, D. R. Shinde, S. H. Chikkali, ChemCatChem 2017, 9, 3997–
4004.
[36] T. Seidensticker, A. J. Vorholt, A. Behr, Eur. J. Lipid Sci. Technol. 2016,
118, 3–25.
[37] S. Pandey, S. H. Chikkali, ChemCatChem 2015, 7, 3468–3471.
[38] K. F. Muilwijk, P. C. J. Kamer, P. W. N. M. van Leeuwen, J. Am. Oil Chem.
Soc. 1997, 74, 223–228.
[39] E. N. Frankel, J. Am. Oil Chem. Soc. 1971, 48, 248–253.
[40] E. Benetskiy, S. Lühr, M. Vilches-Herrera, D. Selent, H. Jiao, L. Domke,
K. Dyballa, R. Franke, A. Börner, ACS Catal. 2014, 4, 2130–2136.
5
This article is protected by copyright. All rights reserved.

10.1002/chem.202000620
Chemistry - A European Journal
COMMUNICATION
Entry for the Table of Contents
N
H
N
H
O
OH
OH
N
N
O
O
O
O
H
H
O
H2:CO
H
O
P
O
P
Rh
O
O
O
O
OC
O
Natural
2.51:1 selectivity
fatty acid
Redesigned catalyst for selective fatty
acid hydroformylation via carboxylate
binding on remote distances
Insert text for Table of Contents here.
((The Table of Contents text should give readers a short preview of the main theme of the research and results included in the paper
to attract their attention into reading the paper in full. The Table of Contents text should be different from the abstract and should
be no more than 450 characters including spaces.))
Institute and/or researcher Twitter usernames:
6
This article is protected by copyright. All rights reserved.