CO-Free Enantioselective Hydroformylation of Functionalised Alkenes: Using a Dual Catalyst System to Give Improved Selectivity and Yield

Article abstract of DOI:10.1002/adsc.201900640

The scope of carbon monoxide-free Asymmetric Transfer HydroFormylation (ATHF) procedures using a highly active single catalyst system derived from 1,2-bis-((2,5)-diphenylphospholano)ethane as chiral ligand has been studied. This reveals some highly successful reactions, but also significant limitations. The development of a new protocol in which a catalyst for formaldehyde decomposition to CO and H₂ is combined with the catalyst of choice for the subsequent asymmetric hydroformylation is described. This enables ATHF reactions that were problematic to be significantly improved. The new method has been used in the synthesis of several key precursors to biologically active molecules. (Figure presented.).

Full text of DOI:10.1002/adsc.201900640

Title: CO-free Enantioselective Hydroformylation of Functionalised
Alkenes: Using a Dual Catalyst System to give Improved
Selectivity and Yield
Authors: Rachael Pittaway, Paul Dingwall, José Fuentes, and Matt
Clarke
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: Adv. Synth. Catal. 10.1002/adsc.201900640
Link to VoR: http://dx.doi.org/10.1002/adsc.201900640

10.1002/adsc.201900640
Advanced Synthesis & Catalysis
FULL PAPER
DOI: 10.1002/adsc.201((will be filled in by the editorial staff))
CO-free Enantioselective Hydroformylation of Functionalised
Alkenes: Using a Dual Catalyst System to give Improved
Selectivity and Yield
Rachael Pittaway, a Paul Dingwall,b José. A. Fuentes, a Matthew. L. Clarke a*
a
School of Chemistry, University of St Andrews, St Andrews, Fife, UK, KY16 9ST: mc28@st-andrews.ac.uk; +44 1334
463850
b
Current Address: School of Chemistry and Chemical Engineering, Queen’s University Belfast, Belfast, BT9 5AG (UK)
Received: ((will be filled in by the editorial staff))
Supporting information for this article is available on the WWW under http://dx.doi.org/10.1002/adsc.201######.((Please
delete if not appropriate))
Abstract. The scope of carbon monoxide-free Asymmetric
Transfer HydroFormylation (ATHF) procedures using a
highly active single catalyst system derived from 1,2-bis-
((2,5)-diphenylphospholano)ethane as chiral ligand has been
studied. This reveals some highly successful reactions, but
also significant limitations. The development of a new
protocol in which a catalyst for formaldehyde decomposition
to CO and H2 is combined with the catalyst of choice for the
subsequent asymmetric hydroformylation is described. This
enables ATHF reactions that were problematic to be
significantly improved. The new method has been used in
the synthesis of several key precursors to biologically active
molecules.
Keywords: Hydroformylation; Asymmetric Synthesis;
Carbon monoxide surrogates; Regioselectivity; Rhodium
system could be used to conduct linear selective
hydroformylations of alkenes using formaldehyde as
CO source.[3] More recently, our group and the
Introduction
There is no argument that the ‘conventional’ rhodium
catalysed hydroformylation reaction is a valuable
reaction; several examples are practised at huge scale
for commodity and fine chemicals.[1] While
recognised as a tool in organic synthesis,[1] it is
Morimoto
group
reported
in
2015
that
enantioselective hydroformylation can be carried out
using formaldehyde as a surrogate for syngas.[4,5]
These Asymmetric Transfer Hydroformylations are
given the acronym ATHF, since AHF is sometimes
perhaps underused in research-scale organic synthesis. used to refer to Asymmetric HydroFormylations.
One reason for this could be the initial nervousness of
the non-expert in using CO gas, and the requirement
to invest in some equipment to operate the reactions.
For this reason, improved protocols for various types
of CO-free hydroformylations are needed urgently.
Some particularly desirable carbon monoxide
surrogates are easily transportable and are available
from waste biomass (e.g. formaldehyde, methanol,
oxalates, formates etc.). It is therefore also
conceivable that some larger scale hydroformylation
could one day be operated using a CO surrogate
should such methods be competitive with the use of
syngas. The use of formaldehyde as a surrogate for
both CO and hydrogen is highly appealing and has
been the topic of research for some time.[2-5] Probably
the seminal work in this arena was by Morimoto and
co-workers who discovered that a dual catalyst
Labelling studies by both groups came to the same
conclusion; that the ATHF is transfer
hydroformylation from a practical point of view, but
mechanistically a tandem process in which the
catalyst is adept at both decomposition of
formaldehyde to syngas, and a highly reactive
catalyst for low-pressure hydroformylation.
a
Both research studies relied on the use of the same
ligand, Ph-BPE (1) (Fig. 1),[6] combined with
different Rh sources: [Rh(acac)(CO)2][4] or
[{RhCl(cod)}2].[5] In order to fully appreciate the
rarity of catalysts that can conduct the two steps of
formaldehyde decomposition and hydroformylation
in ATHF, Fig 1 shows the best conversion to
aldehydes found for some well established
hydroformylation ligands in this first phase of work.
1
This article is protected by copyright. All rights reserved.

10.1002/adsc.201900640
Advanced Synthesis & Catalysis
chosen was quite a poorly reactive alkene. It is
difficult since one of the rate-determining parts of a
conventional hydroformylation is the exchange of a
CO ligand for an alkene. Anything other than a 1-
alkene tends to compete less successfully with CO. In
fact the results obtained with [Rh(acac)(CO)2] / Ph-
BPE on this less reactive alkene are quite remarkable
since conventional hydroformylation using this
catalyst after several attempts using normally optimal
conditions did not give higher than 74% e.e. The
Ph
Ph
% aldehyde in transfer
hydroformylation of cis-stilbene
P
P
Ph
Ph
(R,R)-Ph-BPE, 1
tBu
80% (95% e.e.)
ATHF
therefore
shows
slightly
different
Ph
O
O
characteristics from conventional hydroformylation,
probably as a result of the gradually produced low
concentrations of CO being quite hard to achieve
using a normal hydroformylation protocol in a sealed
vessel.
P
O
P
Ph
tBu
(S ,S,S)-bobphos, 2
ax
8%
CHO
1mol% [Rh(acac)(CO)2], 3 mol% Ligand
tBu
tBu
Ph
Ph
Ph
Ph
O
O
6 equiv. [CH2O]n, toluene
120 oC, 45 mins.
O
P
P
O
7
8
O
O
[Rh(acac)(CO)2]
80% product, (72% isolated), 95 % ee, (R)
tBu
tBu
Ligand = (R,R)-1
[Rh(m-Cl)(COD)]2
Ligand = (R,R)-1
2% product.
0%
(R,R)-Kelliphite, 3
O
Scheme 1. ATHF of cis-stilbene using different Rh
sources.
R
R
R =
N
Ph
O
O
H
P
N
P
Using the same Ph-BPE ligand and [RhCl(cod)2]
instead of [Rh(acac)(CO)2] leads to almost no product
(Scheme 1). The combination of [RhCl(cod)2] and
Ph-BPE is an excellent catalyst for decomposition of
formaldehyde: based on simple experiments, using
the pressure sensor in a chemical microwave to
measure pressure build up without alkene present, it
is superior to [Rh(acac)(CO)2] / Ph-BPE. However, it
seems likely that [RhCl(cod)2] and Ph-BPE only
produce low levels of the most active
hydroformylation catalyst, [RhH(CO)2(Ph-BPE)], or
that this mixture converts to a less reactive
hydroformylation catalyst of a different form. With a
highly reactive terminal alkene substrate, like styrene,
used in reference 5, then this catalyst works
especially well (and we have been able to reproduce a
typical protocol from reference 5 well). However, we
had found that our most successful combination
[Rh(acac)(CO)2] / Ph-BPE does not promote ATHF
of styrene very well due to low regioselectivity. From
this we infer that that our preferred catalyst produced
CO and hydrogen at a slow rate relative to styrene
hydroformylation; this can lead to linear selectivity in
the hydroformylation of styrene.[7] However, for other
alkenes, the use of [Rh(acac)(CO)2], which is more
commonly used in conventional hydroformylation
catalysis, is more reactive and can deliver high
selectivity.
N
R
R
N
N
O
O
Bis[(R,R,S)-DiazaPhos-SPE], 4
10%
O
PPh2
PPh2
0%
0%
Xantphos, 5
rac-BINAP, 6
BINAP= 2,2'-bis(diphenylphosphinyl)-1,1'-binaphthalene
Figure 1. Ligands used in this study, and conversion to
aldehyde obtained in the transfer hydroformylation of cis-
stilbene using 1 mol% Rh catalysts, 3 mol% ligand, 6
equivalents of paraformaldehyde, at 120 oC for 45 minutes,
as communicated in ref. 4. Toluene was used as solvent,
and in those unsuccessful reactions significant amount of
trans-stilbene were detected (see ref 4).
There are several reasons for the very low yields of
aldehyde product for all of the catalysts except those
derived from Ph-BPE: some catalysts might not
survive the relatively high temperatures, or could lead
to extensive isomerisation before hydroformylation
takes place. Some catalysts might not operative well
at the relatively low pressures of syngas generated,
and/or some catalysts might not be able to produce
sufficient syngas to operate properly. The substrate
In this paper we report on a project aimed at
exploring the scope of the Rh/ Ph-BPE system, and
the development of conditions that allow the use of
other chiral hydroformylation ligands in a syngas-free
ATHF procedure. We hope that, in addition to
2
This article is protected by copyright. All rights reserved.

10.1002/adsc.201900640
Advanced Synthesis & Catalysis
providing useful catalytic methods for various targets,
the findings should be enabling to other researchers
who wish to develop a syngas-free hydroformylation
reaction.
1,3,5,7-tetramethyl-2,4,8-trioxa-6-
phosphaadamantane[12] was carried out. This gave a
preliminary indication of the regioselectivity that
these substrates might favour. These results are
shown and discussed in more detail in the ESI;
conversion and chemselectivity was good, and
selectivities were around 2:1. The two regioisomers
were found to be inseparable. The aldehydes were
reduced with NaBH4 to an inseparable mix of the
regioisomeric alcohols, which could be resolved by
analytical chiral HPLC. The identity of the major and
minor aldehyde (or alcohol) could be assigned
through the use of 1H-13C HMBC NMR.
Results and Discussion
The [Rh(acac)(CO)2] / Ph-BPE catalyst was
reported in our original communication to give high
yields in the transfer hydroformylation of a range of
alkenes, but probably the most striking results were
the ATHF of cis-stilbene and a few derivatives. The
scope of this catalyst was first studied on the ATHF
of some asymmetrically substituted cis-stilbene
derivatives. Despite some achievements,[8] there is a
lack of predictability, and further research needed on
the hydroformylation of non-symmetric alkenes in
general.
Conventional asymmetric hydroformylation using
syngas and chiral Rh catalysts was studied and is
described in the ESI; despite many variations in
conditions, AHF did not lead to promising results
with these challenging substrates. The long reaction
times necessary for AHF often led to full
consumption of the starting material but also
isomerisation to the (E)-alkene and low
enantioselectivity. ATHF was studied next, which
gave higher conversion, as described below.
H
R2
[PdCl2(PPh3)2] (4 mol %)
+
piperidine, 80 °C, 1 h
R2
R1
9-12
X
The initial ATHF reactions were explored with
alkene 13. The standardised conditions (120 °C, 1 h,
[CH2O]n (6 eq.) observed 55% conversion to the
aldehydes, with a 64:36 regioselectivity favouring
aldehyde 17 with 41% e.e but 90% e.e for the minor
component, 18 (Table 1, entry 1). The lower e.e. for
17 is presumed to be a result of the more electron-
withdrawing aryl ring adjacent to the formyl group
stabilising the enol tautomer, leading to racemisation
under the reaction conditions. This decrease in
selectivity (and conversion) is emphasised for longer
reaction periods (Table 1, entry 2 & 3), despite using
a lower temperature (Table 1, entry 2). This is
consistent with the observation that slower AHF
reactions were unsuccessful with these substrates.
Larger amounts of the trans-alkene are also
noticeable at long reaction times, but with no increase
(70-78% yield)
R1
X = I, Br
cis-selective reduction
(see ESI)
R2
R1
O
OMe
MeO
OMe
MeO
14
13
O
OMe
OMe
MeO
OMe
MeO
in
aldehyde
yield.
The
possibility
of
16
15
O
dehydroformylation of the aldehydes to the more
stable (E)-isomer as the cause was disproved (see
ESI). Therefore, an alternative explanation may relate
to decomposition of the catalyst over time, meaning
only the simpler process of alkene isomerisation is
possible. As a longer reaction time had a negative
impact on the selectivity, we shortened the reaction
time to 30 minutes, which led to increase in
conversion to the aldehydes (61%), regioselectivity
(68:32) and enantioselectivity for 17 to 84% e.e.
(Table 1, entry 4). The major regioisomer of the
aldehyde, (17) forms adjacent to the electron-
withdrawing methyl-ester; this is possibly due to the
stabilisation of the negative charge built up during C-
H bond formation by the electron poor arene.
Scheme 2. The preparation of cis-stilbene derivatives
used in this study.
For this reaction, we studied this aspect using four
carefully chosen alkenes (Scheme 2). A two-step
methodology was used to make the substrates: a
Sonogashira cross-coupling reaction[9] to give alkynes
9-12, followed by a cis-selective reduction. The latter
step required different methods depending on the
alkynes used, with the planned Lindlar reduction
being replaced by a Ti mediated[10] or a Pt
catalysed[11] reduction adapted from the literature for
15 and 16 respectively. The synthesis of these alkenes
is discussed in more detail in the ESI, but alkenes 13-
18 were prepared in sufficient quantity and purity to
carry out the studies discussed.
Conventional hydroformylation of alkenes 13 -16
with a catalyst derived from [Rh(acac)(CO)2] and
3
This article is protected by copyright. All rights reserved.

10.1002/adsc.201900640
Advanced Synthesis & Catalysis
Table 1. Asymmetric Transfer HydroFormylation of
asymmetrically substituted Z-alkenes.
regioselectivity again places the aldehyde away from
an ortho substituent and also on the more electron
deficient of the alkene carbons. Enantioselectivity
was excellent: 95% e.e for the major isomer and 98%
e.e for the minor isomer, reflecting good
configuration stability (Table 1, entry 6).
[Rh(acac)(CO)2] (2 mol%)
Ph-BPE (3 mol%)
Ar1
CHO
Ar2
CHO
Ar1
+
[CH2O]n (6 eq.), 120 °C
toluene
Ar1
Ar2
Ar2
OHC
CHO
Table 2. Asymmetric Transfer Hydroformylation of alkene
15.
MeO
OMe
MeO
[Rh(acac)(CO)2] (2 mol%)
Ph-BPE (3 mol%)
MeO
MeO
OMe
18
MeO
MeO
OMe
17
O
Ar2
O
Ar2
+
[CH2O]n (6 eq.), 120 °C
toluene
O OHC
O
CHO
Ar2
*
CHO
*
15
OHC
OMe
23
Ar2 =
24
OMe
OMe
20
19
21
OHC
OMe
CHO
MeO
MeO
e.e.
minor
e.e.
major
%
(CH2O)n
(equiv.)
Entry
Aldehyde
[yield][a]
23 : 24
t(h)
1
OMe
OMe
22
1
37[13]
38:62
98
95
6
O
O
2
2
47[27]
63[25]
39:61
42:58
94
91
88
88
2
6
4
4
4
e.e.
minor
e.e.
major
%
3
Aldehyde
[yield][a] selectivity
Regio-
Entry
t(h)
Alkene
4
4
85[30]
21[6]
45:55
76:24
96
88
93
92
4
5[b]
ratio: 17: 18
1
55[37]
64:36
41
90
1
13
[a] Remaining mass balance is mainly starting material, except entry 2
(15%) and entry 5 (57%) E isomer of alkene
% aldehyde determined by NMR spectroscopy against cyclooctane as
internal standard. Isolated yield in square brackets are for the
corresponding alcohols obtained after both hydroformylation and then
NaBH4 reduction and chromatography (e.e. determined at this stage).
3[b] 8[6]
60:40
65:35
31
38
88
89
2
3
13
13
2
48[38]
61[43]
0.5
68:32
84
90[c]
4
13
[b] Ligand 4 (Fig 1) used in place of (R,R)-Ph-BPE
ratio: 19: 20
5
14
1
1
31[27]
68:32
90
n.d.
Hydroformylation of alkene 15 is a possible new
route to (S)-Equol {(3S)-3-(4-hydroxyphenyl)-7-
chromanol}.[13] ATHF of alkene 15, gave relatively
low conversion to the aldehydes (37%) using the
standard conditions (120 °C, [CH2O]n (6 eq.), 1 h), as
well as regioselectivity to the undesired regioisomer,
24 as the major product (38:62). This was also
observed with AHF using [Rh]/Ph-BPE (see ESI).
This is hypothesised to be due to the lack of
electronic or steric bias. However, enantioselectivity
was excellent, achieving 98% e.e for the major
(undesired) isomer and 95% for the minor (desired)
(Table 2, entry 1). Using 4 equivalents of
paraformaldehyde and increasing the reaction time,
optimally to 4 hours, led to a higher conversion
(85%) (Table 2, entry 4). The conventional AHF of
alkene 15 using the [Rh]/Bisdiazaphos, 4 catalyst14
led to good regioselectivity towards the desired
regioisomer, 2 (see ESI), but conversion to the
aldehydes never exceeded 40% (up to 14% isolated
yield of alcohol). The good selectivity made transfer
hydroformylation using this ligand of great interest,
since we hoped the lower pressures of CO might
improve reactivity, whilst retaining selectivity. The
ATHF reaction was run at 120 °C for 4 hours and
delivered 74:26 regioselectivity towards the desired
ratio: 21: 22
65[35]
65:35
95
98
6
16
[a] Remaining mass balance is mainly E isomer of alkene, except final 2
entries which are starting material.
% aldehyde determined by NMR spectroscopy against cyclooctane as
internal standard. Isolated yield in square brackets are for the
corresponding alcohols obtained after both hydroformylation and then
NaBH4 reduction and chromatography (e.e. determined at this stage).
[b] 110 oC; [c] result reported in communication (ref. 4).
ATHF of methyl (Z)-2-(4-methoxystyryl)benzoate,
14, gave lower product yield, and favours aldehyde
19 31%, 19:20 = 68:32). Both of these observations
can be rationalised as being due to the sterically
bulky ortho substituent hindering the coordination of
the alkene to the metal centre. Enantioselectivity was
excellent (Table 1, entry 5). ATHF of alkene 16
obtained 65% conversion to the aldehydes and a
65:35 regioselectivity favouring aldehyde 21. This
4
This article is protected by copyright. All rights reserved.

10.1002/adsc.201900640
Advanced Synthesis & Catalysis
regioisomer, but gave low conversion and was
accompanied by significant isomerisation to the (E)-
alkene (Table 2, entry 5). The low conversion was
ascribed to Bisdiazaphos being a poor ligand for the
decarbonylation of paraformaldehyde into syngas.
most catalysts give no selectivity or favour the linear
aldehyde in hydroformylation of unbiased alkenes of
this type, with the main exception being catalysts
derived from ligand 2 that can give around 75-80%
selectivity.[15]
ATHF of allyl benzene was studied, with a dual
catalyst system. Simply combining a Rh source with
This result highlights an issue with using a single
ATHF catalyst: so few ligands are able to promote
both the formaldehyde decomposition and
hydroformylation reactions. As another example that
illustrates this, branched selective hydroformylation
of allyl benzene derivatives is of interest, but the Rh /
Ph-BPE system has been reported to give low
regioselectivity and low e.e. in AHF of this
substrate.[15a]
It would therefore be a step forward if the asymmetric
transfer hydroformylation reaction could be
developed to utilise a dual catalyst system, where the
decarbonylating ability of Rh/ Ph-BPE could be
combined with the selectivity of a different catalyst,
in order to achieve optimum conversion and
selectivity.
ligand
2
along with
a
better formaldehyde
decomposition catalyst A, [RhCl(cod)2] and (S,S)-Ph-
BPE (Table 3, entry 1) was not effective. The key to
getting improved results is the physical separation of
the catalysts from the alkene inside a single vessel,
allowing the decarbonylation reaction to generate
syngas at high temperature, prior to mixing of the
alkene with the hydroformylation catalyst after
cooling to a lower temperature optimal for high
selectivity. While such procedures are well-known
functions that can be added to a commercial pressure
reactor, a simple solution for laboratory scale
procedures is described in the ESI and is what has
been used here. This leads to an improvement from
40% aldehydes and low selectivity (Table 3, entry 1)
to high yield, b:l of 2.1:1 and 79% e.e. (Table 3 entry
2). If catalyst B, derived from (Sax,S,S)-2 is used
without an addition formaldehyde decomposition
catalyst, then a low yield is observed (Table 3, entry
3). As a control experiment, if both formaldehyde
decomposition catalyst, A and hydroformylation
catalyst B are derived from (S,S)-Ph-BPE, then the
selectivity changes to favour the linear aldehyde
slightly, along with producing some prop-1-en-yl
benzene (Table 3, entry 4). The enantioselectivity
was low, as was observed in previous work using
syngas as reagent.[15a] It stands to reason that racemic
Ph-BPE would be equally effective as the single
enantiomer in the formaldehyde decomposition, but
since this work was aimed at organic synthesis
laboratories, we have only used the commercially
available single enantiomer. Unfortunately, changing
catalyst A to utilise the achiral ligand dppe (dppe =
1,2-bis diphenylphosphinoethane), reduces yield
(Table 3, entry 5, see ESI for some other examples).
It is clear that catalyst B alone carries out the
hydroformylation step, despite the presence of
another Rh catalyst in the same solution, since the
mix of catalyst B and achiral dppe-derived catalyst A
in entry 5 gives similar enantioselectivity to that
shown in entry 2. To further support this, if the
enantiomer of ligand 2 in catalyst B is switched,
while keeping the (S,S) enantiomer of Ph-BPE in
catalyst A, similar enantioselectivity towards the
opposite enantiomer of product is observed (Table 3,
entry 6). In summary, while the normal conditions are
not at all useful for an isomerisation prone substrate
like allyl benzene, this dual catalyst system with the
alkene mixed only after syngas has been generated
leads to significantly improved results, if not
matching optimised conditions using syngas; this was
then examined on other substrates to see if it has
general use.
Table 3 Transfer hydroformylation of allyl benzene
0.8% Catalyst A
Ph
Me
2.1 % Catalyst B
Ph(CH2)3CHO
*
CHO
[CH2O]n (5 eq.), toluene
Ph
b
l
(i) 120 °C, 30 min; (ii) 40 oC, 16h
e.e.
(%)
Catalyst B
b:l
Catalyst A
Product
(%)[a]
0.8% [RhCl(COD)]2 2.1 % [Rh(acac)(CO)2]
40[b]
96[c]
55:45
69:31
(1)
27
79 (S)
0.9% (S,S)-Ph-BPE 2.6 % (S ,S,S)-2
ax
0.8% [RhCl(COD)]2 2.1 % [Rh(acac)(CO)2]
(2) 0.9% (S,S)-Ph-BPE 2.6 % (S ,S,S)-2
ax
2.1 % [Rh(acac)(CO)2]
2.6 % (S ,S,S)-2
35[c]
63:27 73 (S)
(3)
(4)
N/A
ax
0.8% [RhCl(COD)]2 2.1 % [Rh(acac)(CO)2]
0.9% (S,S)-Ph-BPE 2.6 % (S,S)-Ph-BPE
69[c,d] 40:60 2 (S)
0.8% [RhCl(COD)]2 2.1 % [Rh(acac)(CO)2]
28[c]
67[c]
71:29 79 (S)
74:26 81(R)
(5) 0.9% dppe
2.6 % (S,S)-Ph-BPE
0.8% [RhCl(COD)]2
2.1 % [Rh(acac)(CO)2]
(6) 0.9% (S,S)-Ph-BPE
2.6 % (Rax,R,R)-2
[a] % aldehyde determined by NMR spectroscopy against methyl
naphthalene as internal standard. e.e. determined on the corresponding
alcohol obtained after both hydroformylation and NaBH4 reduction.
[b] alkene, formaldehyde and pre-mixed catalyst solution all added prior
to step (i).
[c] alkene only mixed with catalysts after step (i)
[d] 10% isomerised internal alkene detected
Allyl benzene was chosen as a challenging model
substrate for our optimisation of this process, since
isomerisation to the internal conjugated alkene, prop-
1-en-yl benzene is a downhill process. In addition,
5
This article is protected by copyright. All rights reserved.

10.1002/adsc.201900640
Advanced Synthesis & Catalysis
In 2017, we published work describing a tandem
hydroformylation-cyclisation reaction of allylglycine
derivatives[16] to either hemi-aminal or hemi-acetals;
the products could be converted to intermediates for
the synthesis of an intermediate to the antibiotic,
Nemonoxacin.[17] The AHF of (S)-allylglycinol, 25,
using a Rh / (R,R,R)-2 catalyst at 5 bar syngas
pressure achieves high conversion and selectivity
(>99% conversion to products, b:l = 4.9:1, d.r = 93:7)
(Table 4, entry 1). The original ATHF conditions
using the [Rh]/Ph-BPE catalyst were tested for
comparison. Standard ATHF conditions using
[Rh]/Ph-BPE gave a b:l of just 60:40 (Table 4, entry
2). The dual catalyst procedure, as described in Table
3, entry 2, with (R,R,R)-2 as the hydroformylation
component of the catalyst gave 91% consumption of
the starting material to obtain 75% hemiacetal, 26 and
16% linear aldehyde, 27 (i.e. a B:L of 82:18, Table 4,
entry 3). With an effective b:l of 4.7:1, this is
comparable to the results achieved in AHF (4.9:1).
We extended the dual catalyst methodology to two
other catalysts. For example, we felt a catalyst system
of [Rh(acac)(CO)2]/Bisdiazaphos, 4-[Rh(cod)Cl]2/Ph-
BPE would be particularly advantageous for the
hydroformylation of alkene 15, as a catalyst derived
from ligand 4 produced higher regioselectivity to the
desired isomer for the synthesis of (S)-equol. A
similar dual catalyst protocol as described in Table 3,
entry 2 was used, although the reaction was carried
out at 70 °C for 16 hours; this achieved high
conversion to the aldehydes (81%), excellent
enantioselectivity (96% for the major, 23, 92% for
the minor, 24) and retained the better regioselectivity
to the desired regioisomer (75:25) (Scheme 3, right).
This methodology also benefitted another of the cis-
stilbene derivatives, 14, that only obtained low
conversion (31%) and adequate regioselectivity
(68:32)
in
the
standard
ATHF
using
[Rh(acac)(CO)2]/Ph-BPE. A significant increase in
conversion to the aldehydes (70%), regioselectivity
(90:10) and enantioselectivity (96% for the major
aldehyde, 19) (Scheme 3, left) was achieved after 16
hours at 70 °C. These results suggest that the
[Rh]/Bisdiazaphos, 4 catalyst can be used in a
syngas-free methodology.
However, there was
a
slight decrease in
diastereoselectivity, (d.r = 86:14, determined from 1H
NMR after oxidation to the lactone using
Ag2CO3/Celite). The results from table 3 and 4
illustrate that a dual catalyst system derived from
ligands 1 and 2 is clearly potentially useful. None-
the-less, conversion and selectivity was slightly
below that observed for AHF using Rh / (R,R,R)-2.
This may be due to the partial decomposition of
ligand 2 under the harsher conditions of the activation
(120 °C for 30 minutes).
[Rh(acac)(CO)2] (2.1 mol%)
OMe
MeO
Ar2
(R,S,S)-4 (2.6 mol%)
MeO
[Rh(cod)Cl]2 (0.8 mol%)
(S,S)-Ph-BPE (0.9 mol%)
Ar2
Ar2
[CH2O]n (5 eq.), toluene
120 °C (0.5h), 70 °C, (16 h)
+
*
CHO
*
OHC
14
or 15
19
or 23
20
or 24
Table 4. AHF and ATHF of (S)-allyl glycinol.
OH
Ar2 =
MeO
O
Ar2 =
70% aldehydes,[a, b]
19: 20 = 90:10;
e.e. of 19 = 96%
e.e. of 20 = n.d.
OMe
[Rh(acac)(CO)2]
Ligand
81% aldehydes,[a,c]
23: 24 = 75:25;
e.e. of 23 = 96%
e.e. of 24 = 92%
O
Bn
Bn
Bn
Bn
(CH2)3CHO
OH
N
N
+
CO/H2 or [CH2O]n
toluene
OH
14
N
OMe
Bn
Bn
15
25
27
26
[a] Remaining mass balance is predominantly E-alkene and some alkane.
[b] A yield of 50% for alcohol formed after hydroformylation and NaBH4
reduction.
Procedure
d.r.[a]
93:7
Conversion
>99
b: l
(26: 27)
[c] A yield of 52% for alcohol formed after hydroformylation and NaBH4
reduction.
[b]
Entry 1:
0.4 mol% [Rh(acac)(CO)2],
2 mol% (R,R,R)-2
40 oC, 5 bar syngas, 16h
83:17
Scheme 3. ATHF of cis-stilbenes 15 and 16 using a dual
catalyst system.
Entry 2:
2 mol% [Rh(acac)(CO)2],
3 mol% (S,S)-Ph-BPE
120 oC, 6 equiv. (CH2O)n, 1 h
>99
91
60:40
82:18
85:15
86:14
We also investigated another catalyst in this dual
catalyst methodology: [Rh]/Kelliphite,[18] which has
previously obtained very selective results in the AHF
of Boc-lactam 28 (>99% conv, 29:30 = 7.5:1 (50 °C,
1 h), or 9.4:1 (15 °C, 16 h).[19] This substrate was also
investigated in 2015 using the standard ATHF
reaction using [Rh]/Ph-BPE, and although a full
conversion was observed in only 5 minutes, it is at
the expense of a lower selectivity (>99% conversion,
b:l =2.5:1 on purification).
[c]
Entry 3:
2.1 mol% [Rh(acac)(CO)2],
2.6 mol% (R,R,R)-2
0.8 mol% [RhCl(cod)]2
0.9 mol% (S,S)-Ph-BPE
6 equiv. (CH2O)n,
120 oC (0.5 h), 40 oC 16h
[a] d.r. determined by 1h NMR spectrsocopy on the lactone obtained after
oxidation with Ag2CO3. [b] result reproduced from ref. 16 [c]dual catalyst
process with delayed mixing of alkene.
Conducting the reaction using the dual catalyst
protocol with delayed mixing of alkene introduced
6
This article is protected by copyright. All rights reserved.

10.1002/adsc.201900640
Advanced Synthesis & Catalysis
earlier, using [Rh(acac)(CO)2]/(R,R)-Kelliphite and
[Rh(cod)Cl]2/(S,S)-Ph-BPE led to full conversion to
the aldehydes and high selectivity towards 29 (7.3:1),
isolating the nearly isomerically pure 29 in 64% yield
(Scheme 4). This outperforms the results from the
standard ATHF reaction and is comparable to those
achieved using AHF.
volume of the vessel and the amounts of
formaldehyde used, so any deviations from this
protocol should be carried out in a vessel that can
withstand the pressures encountered. A sealable tube
(8 bar max pressure, 25 mL) was charged with a 8
mm cross stirrer bar, paraformaldehyde (126 mg, 5
eq.) and another small sample vial containing a flea
stirrer bar. The vial was capped with a rubber septum
and placed under three vacuum/Ar cycles. A solution
of the alkene (0.84 mmol) in toluene (1 mL) and 1-
methylnapthalene (0.05 mL) was made up in a
Schlenk vial under an inert atmosphere. This was
transferred into the small sample vial inside the
sealed tube via syringe. A small sample was and
analysed by 1H NMR spectroscopy (C6D6) The
catalyst stock solutions ([Rh(acac)(CO)2] (4.5 mg,
0.018 mmol, 2.1 mol%)/ligand (e.g. for ligand 2: 14.1
mg, 0.022 mmol, 2.6 mol%) in toluene (1 mL) and
[[Rh(cod)Cl]2 (1.7 mg, 0.007 mmol, 0.8 mol%)/Ph-
BPE (3.8 mg, 0.008 mmol, 0.9 mol%) in toluene (1
mL)) were made up separately in two Schlenk tubes
under an inert atmosphere. These were transferred to
the sealed tube (outside of the sample vial) via
syringe and the rubber septum was placed with a
fitted cap immediately. The pressure vial was placed
in an oil bath, and a blast shield was placed in front of
the reaction. It was then heated to 120 °C for 30 min
(stirring at speed 1000 rpm), before cooling for 30
min. (step (ii) in Table 3): The vial was then placed
back into the oil bath for the desired time at a lower
temperature before inverting the vial several times (to
ensure the contents of the smaller vial are fully mixed
in). This was left to stir at the desired lower
temperature for the desired time. The vial was cooled
to r.t and inverted several times before opening the
[Rh(acac)(CO)2] (2.1 mol%)
(R,R)-Kelliphite (2.6 mol%)
[RhCl(cod)]2 (0.8 mol%)
(S,S)-Ph-BPE (0.9 mol%)
OHC
NBoc
O
NBoc
O
29
+
[CH2O]n (5 eq.), toluene
(i) 120 °C (0.5h),
(ii) 40 °C (16 h)
28
NBoc
OHC
O
30
>99% conversion, 29:30 = 7.3:1
64% yield (13.3:1) on purification)
Scheme 4. ATHF of a bicyclic lactam using a dual catalyst
system.
Conclusion
In this paper, we have studied the performance of the
Rh / Ph-BPE catalyst in formal asymmetric transfer
hydroformylations
in
greater
detail.
High
enantioselectivity can be realised in the ATHF of cis-
stilbene derivatives, but sometimes yields or
regioselectivity are not optimal. In order to allow a
broader range of ligands to be used in ATHF, a dual
catalyst system was developed. In this dual catalyst
protocol we have used enantiopure Ph-BPE ligands in
combination with [RhCl(COD)]2 to decompose
paraformaldehyde into CO and hydrogen. The use of
a single enantiomer here is purely since such ligands
are available commercially, although the racemic
analogues could undoubtedly be used. Using the dual
catalyst system, the hydroformylation activity comes
entirely from the ligand pre-mixed with
[Rh(acac)(CO)2], and hence in the best cases,
regioselectivity, e.e. or d.r. levels are significantly
higher than what could be realised using the original
ATHF protocol for a range of substrates identified as
problematic. It is hoped this not only provides some
syngas-free protocols for the specific targets
described, but is enabling for other workers who may
wish to use syngas-free hydroformylations.
1
vial. A small sample was taken and analysed by H
NMR spectroscopy (C6D6) to calculate the conversion
of the resulting aldehydes. The products were
generally reduced as described in the ESI to deliver
pure
products
after
chromatography.
The
enantioselectivity of the reaction was determined by
measuring e.e. on these corresponding primary
alcohols using chiral HPLC. Full experimental details,
analytical data, NMR and HPLC spectra can be found
in the ESI (for compounds not previous
communicated). Underpinning data (NMR files) are
also available.[20]
Acknowledgements
Acknowledgements: The EPSRC doctoral training grant
supported RP (DTG grant: 1518070)). The EPSRC
(EP/M003868/1) is also acknowledged for funding (JAF&PD).
We thanks all the technical staff in St Andrews Chemistry for
their assistance, and thank Dr Martin Fox (Dr Reddys (UK) for a
fruitful discussion, and Dr Reddys (UK) for a gift of substrate 28.
Experimental Section
CAUTION: We have never had any incidents with
vials cracking due to excess pressure. In experiments
carried out where pressure was measured in smaller
10 mL vessels, but using similar amount of
formaldehyde, the largest pressure measured was 7
bar (observed in an example where no
hydroformylation occurred to use up the gas).
However, the pressure will be proportional to the
References
[1] a) P. W. N. M. van Leeuwen, in Rhodium Catalyzed
Hydroformylation, Vol. 22 (Eds.: P. W. N. M. van
7
This article is protected by copyright. All rights reserved.

10.1002/adsc.201900640
Advanced Synthesis & Catalysis
Leeuwen, C. Claver), Springer Netherlands, 2002, pp.
2014 136, 8418; d) C. Grünanger, B. Breit, Angew.
Chem. Int. Ed. 2010, 49, 967; e) A. D. Worthy, M. M.
Gagnon, M. T. Dombrowski, K. L. Tan, Org. Lett,
2009, 11, 2764; f) A. L. Watkins, B. G. Hashiguchi C.
L. Landis, Org. Lett. 2008, 10, 4553; g) C. You, X. Li,
Y. Yang, Y-S Yang, X. Tan, S. Li, B. Wei, H. Lv, L-W
Chung, X. Zhang, Nat. Commun. 2018, 2045.
1-13; b) R. Franke, D. Selent, A. Börner, Chem. Rev.
2012, 112, 5675-5732, c) A. Börner, R. Franke, in
Hydroformylation. Fundamentals, Processes, and
Applications in Organic Synthesis; Wiley-VCH Verlag
GmbH & Co. KGaA, 2016; d) L. Gonsalvi, A.
Guerriero, E. Monflier, F. Hapiot, M. Peruzzini, in
Hydroformylation for Organic Synthesis, Vol. 342
(Eds.: M. Taddei, A. Mann), Springer Berlin
Heidelberg, 2013, pp. 1-47.
[9] a) N. E. Leadbetter, B. J. Tominack, Tetrahedron Lett.
2003, 44, 8653; b) See also: M. Alami, F. Ferri G.
Linstrumelle, Tetrahedron Lett. 1993, 34, 6403
[2] a) T. Morimoto, K. Kakiuchi, Angew. Chem. Int.
Ed.,2004, 43, 5580. b) T. Okano, T. Kobayashi, H.
Konishi, J. Kiji, Tetrahedron Lett.,1982, 23, 4967. c) H.
S. Ahn, S. H. Han, S. J. Uhm, W. K. Seok, H. N. Lee,
G. A. Korneeva, J. Mol. Catal. A.,1999, 144, 295. d) L.
Wu, Q. Liu, R. Jackstell, M. Beller, Angew. Chem. Int.
Ed., 2014, 53, 6310. e) M. Rosales, A. González, B.
González, C. Moratinos, H. Pérez, J. Urdaneta, R. A.
Sanchez-Delgado, J. Organomet. Chem., 2005, 690,
3095. f) M. Rosales, F. Arrieta, P. Baricelli, A.
González, B. González, Y. Guerrero, C. Moratinos, I.
Pacheco, H. Pérez, J. Urdaneta, Catal. Lett., 2008, 126,
367; g) M. Uhlemann, S. Doerfelt, and A. Börner,
Tetrahedron Lett. 2013, 54, 2209; h) E. Cini, E. Airau,
N. Giraud, A. Mann, J. Salvadore, M. Taddei, Synlett,
2011, 199.
[10] Y. Obara, H. Moriya, M. Tokunaga, Y. Tsuji, J.
Chem. Soc., Chem. Commun., 2003, 2820.
[11] A. Giraud, O. Provot, A. Hamze, J.-D. Brion, M.
Alami, Tetrahedron. Lett., 2008, 49, 1107.
[12] R. A. Baber, M. L. Clarke, K. Heslop, A. Marr, A. G.
Orpen, P. G. Pringle, A. M. Ward, D. A. Zambrano-
Williams, Dalton Trans.,2005, 1079.
[13] a) J. M. Heemstra, S. A. Kerrigan, D. R. Doerge, W.
G. Helferich, W A. Boulanger, Org. Lett., 2006, 8,
5441 and ref’s therein. b) Y. Takashima, Y. Kobayashi,
Tetrahedron Lett., 2008, 49, 5156. c) S. Yang, S. Zhu,
C. Zhang, S. Song, Y. Yu, S. Li, Q. Zhou, Tetrahedron.,
2012, 68, 5172.
[14] T. P. Clark, C. R. Landis, S. L. Freed, J. Klosin, K. A.
[3] G. Makado, T. Morimoto, Y. Sugimoto, K. Tsutsumi,
N. Kagawa, K. Kakiuchi, Adv. Synth. Catal., 2010,
352, 299.
Abboud, J. Am. Chem. Soc., 2005, 127, 5040.
[15] a) G. N. Noonan, J.A. Fuentes, C. J. Cobley M. L.
Clarke, Angew. Chem. Int. Ed., 2012, 51, 2477. b) P.
Dingwall, J. A. Fuentes, L. Crawford, A. M. Z. Slawin,
M. Bühl, M. L. Clarke, J. Am. Chem. Soc. 2017, 139,
15921-15932; c) (S,S,S)-BOBPHOS is commercially
available from Strem Chemicals.
[4] J. A. Fuentes, R. Pittaway, M. L. Clarke, Chem.
Eur. J., 2015, 21, 10645.
[5] T. Morimoto, T. Fujii, K. Miyoshi, G. Makado, H.
Tanimoto, Y. Nishiyama, K. Kakiuchi, Org. Biomol.
Chem., 2015, 13, 4632.
[16] R. Pittaway, J. A. Fuentes, M. L. Clarke, Org. Lett.
2017, 19, 2845.
[6] A. T. Axtell, C. J. Cobley, J. Klosin, G. T. Whiteker, A.
Zanotti-Gerosa, K. A. Abboud, Angew. Chem. Int. Ed.
2005, 44, 5834.
[17] a) A. Arjona, Drugs Future., 2009, 3, 196. b) H.
Ming-chu, R. K. Chi-Hsin, C. Shu-Jen, L. Lin, U.S.
Patent US8211909 B2, 2012.
[7] I.A. Tonks., R. D. Froese., C. R. Landis., ACS
Catal. 2013, 3, 12, 2905.
[18] C. J. Cobley, K. Gardner, J. Klosin, C. Praquin, C.
Hill, G. T. Whiteker, A. Zanotti-Gerosa, J. L. Petersen,
K. A. Abboud, J. Org. Chem. 2004, 69, 4031.
[8]For
selected
examples
of
regioselective
hydroformylation of internal alkenes: a) M. K. T.
Soltner, P. W. N. M. van Leeuwen, J. N. H. Reek, J.
Am. Chem. Soc. 2006 128, 11344; b) T. Gadzikwa, R.
Bellini, H. L. Dekker, J. N. H. Reek, J. Am. Chem.
Soc. 2012 134, 2860; c) P. Dydio, R. J. Detz, J. N. H.
Reek, J. Am. Chem. Soc 2013 135, 10817; P. Dydio, R.
J. Detz, B. de Bruin, J. N. H. Reek, J. Am. Chem. Soc.
[19] G. M. Noonan, C. J. Cobley, T. Lebl M. L. Clarke,
Chem. Eur. J. 2010, 16, 12788.
[20] The research data underpinning this publication can
be accessed at https://doi.org/10.17630/e491713d-
9de2-4fa4-8853-4d7ce2656cfd
.
8
This article is protected by copyright. All rights reserved.

10.1002/adsc.201900640
Advanced Synthesis & Catalysis
FULL PAPER
optimised Rh catalyst for
[CH2O]n decomposition
CO-free Enantioselective Hydroformylation of
Functionalised Alkenes: Using a Dual Catalyst
System to give Improved Selectivity and Yield
R2
R1
R2
R1
CHO
optimised Rh catalyst for
asymmetric hydroformylation
* high yield
* optimised
regioselectivity
* up to 96% e.e.
[CH2O]n
Adv. Synth. Catal. Year, Volume, Page – Page
Two Rh catalysts with different roles are
compatible with each other, enabling several
different chiral ligands to be used in CO-free
hydroformylation
Rachael Pittaway, a Paul Dingwall,b José. A.
Fuentes, a Matthew. L. Clarke *
9
This article is protected by copyright. All rights reserved.