Decreasing Side Products and Increasing Selectivity in the Tandem Hydroformylation/Acyloin Reaction

Article abstract of DOI:10.1002/cctc.201500727

A highly selective catalyst system was developed for the recently discovered tandem hydroformylation/acyloin reaction by systematic investigations and changes of reaction conditions. This new catalyst system is characterized by an excellent selectivity of the desired reaction pathway with negligible amounts of side products. A successful application of the tandem hydroformylation/acyloin reaction to a variety of olefins is enabled with comparable excellent selectivities up to >99% for the first and second reaction step, therefore a general synthesis for the conversion of olefins into acyloins is found. Furthermore, very good to excellent yields for the intermediates and final acyloin products were observed within two catalysed reactions in one preparative step. The acyloin product was applied as a nonpolar precursor for surfactants. After attaching a polar head group to the acyloin and determination of tensiometric data, the molecule showed industrial relevant surface-active properties. Jointly successful: New catalyst systems for the tandem hydroformylation/acyloin reaction display excellent selectivities within two catalyzed reactions in one preparative step. A variety of olefins can be converted efficiently, and the acyloin product is applied successfully as a nonpolar precursor for surfactants.

Full text of DOI:10.1002/cctc.201500727

DOI: 10.1002/cctc.201500727
Full Papers
Decreasing Side Products and Increasing Selectivity in the
Tandem Hydroformylation/Acyloin Reaction
[
a]
Karoline A. Ostrowski, Thiemo A. Faßbach, Dennis Vogelsang, and Andreas J. Vorholt*
A highly selective catalyst system was developed for the re-
cently discovered tandem hydroformylation/acyloin reaction
by systematic investigations and changes of reaction condi-
tions. This new catalyst system is characterized by an excellent
selectivity of the desired reaction pathway with negligible
amounts of side products. A successful application of the
tandem hydroformylation/acyloin reaction to a variety of ole-
fins is enabled with comparable excellent selectivities up to
al synthesis for the conversion of olefins into acyloins is found.
Furthermore, very good to excellent yields for the intermedi-
ates and final acyloin products were observed within two cata-
lysed reactions in one preparative step. The acyloin product
was applied as a nonpolar precursor for surfactants. After at-
taching a polar head group to the acyloin and determination
of tensiometric data, the molecule showed industrial relevant
surface-active properties.
>
99% for the first and second reaction step, therefore a gener-
Introduction
Tandem-catalysed reactions are a challenging field in chemistry.
They merge two or more different catalyses in one preparative
step, saving effort, time and resources in a manifold manner.
Especially, the direct conversion of the intermediates skips pu-
rification processes leading to less costs and a more economi-
droformylation reactions, which play an important role in dif-
ferent organocatalysis, applying them as reactive and versatile
substrates. Hence, different tandem concepts using enamine
catalysis for conversion of the intermediate aldehydes can be
[
8b,14]
found in the literature.
The tandem hydroformylation/aldol
[
1]
cal route. These advantages meet important points, which
reaction is one of the most explored tandem sequences con-
[
2]
[15]
are described as requirements for sustainable chemistry. De-
veloping new tandem catalyses, wherein different catalysts are
taining a metal catalyst and an organocatalyst.
The development of a successful combination of a transi-
tion-metal catalyst with an N-heterocyclic carbene (NHC) as or-
ganocatalyst in a tandem system is a huge challenge, because
[3]
applied simultaneously, is not trivial at all. Additional re-
agents and different conditions increase the potential of cata-
lyst deactivation leading to decreased conversion or increased
potential for undesired side-products.
[
13b]
NHCs can act as ligands leading to potential self-quenching
of both catalysts. Therefore, only scattered examples of a suc-
cessful tandem reaction combining transition metal and car-
The homogenous transition-metal-catalysed hydroformyla-
[4]
[16]
tion is long-time known and quite a well investigated reac-
bene catalysis have been reported.
[
5]
[6]
tion. However, there are still important developments, and
hydroformylation describes a quite popular field in tandem
A successful tandem concept merging the hydroformylation
with a carbene catalysis was unexplored till our recent publica-
[
3,7]
[17]
catalysis.
Besides some specialised and rare hydroformyla-
tion (Scheme 1). Therein, we published the very first exam-
[8]
tion tandem sequences, there are only a few main concepts
investigated by different research groups. Recent publications
deal with the following tandem reactions: hydroformylation/
[
9]
[10]
hydrogenation, isomerisation/hydroformylation,
hydrofor-
[11]
[12]
mylation/acetalisation and the hydroaminomethylation.
Combining transition-metal catalysis with organocatalysis
enables unique reaction pathways, which are not accessible if
[13]
only a metal catalyst or an organocatalyst is used. Particular-
ly, the combination of the hydroformylation with organocata-
Scheme 1. The concept of the tandem hydroformylation/acyloin reaction.
[3]
lytic reactions is quite popular. Aldehydes are formed in hy-
ple of an orthogonal tandem hydroformylation/acyloin reac-
tion, in which two different catalysts were applied for each
[
18]
[
a] K. A. Ostrowski, T. A. Faßbach, D. Vogelsang, Dr. A. J. Vorholt
TU Dortmund, Bio- and Chemical Engineering
Emil-Figge Straße 66, 44227 Dortmund (Germany)
E-mail: andreas.vorholt@bci.tu-dortmund.de
transformation: A rhodium catalyst and a carbene catalyst. It
was also the first example of applying a carbene as organoca-
talyst under hydroformylation conditions.
The acyloin reaction links two aldehydes upon formation of
Supporting Information for this article is available on the WWW under
http://dx.doi.org/10.1002/cctc.201500727.
[
19]
a linear CÀC bond by carbene catalysis. Other syntheses of
ChemCatChem 2015, 7, 2607 – 2613
2607
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Papers
acyloins are often not atom effi-
[
20]
cient, undergo three prepara-
[
21]
tive steps,
or use oxidizing
[
22]
agents
.
The synergy of the metal cat-
alysis and organocatalysis sup-
ports the reaction pathway yield-
ing linear a-hydroxy ketones di-
rectly from various olefins. The
first step is described by the rho-
dium-catalysed hydroformylation
[17]
Scheme 3. Direct transfer of the optimised conditions to 1-decene (1).
of a 1-alkene obtaining n-alde-
hydes with high n/bra (=n-alde-
hydes/iso- and branched alde-
hydes) ratio. Two equivalents of the intermediate n-aldehydes
are converted into the corresponding acyloin. This conversion
is catalysed by the in situ formed carbene C1a catalyst from
the carbene precursor thiamine/vitamin B1 (C1) with base
Owing to the complex reaction mixture, the definition of
each variable used in this manuscript is given in Equation (1).
P
n1
n1;0
k¼2;3;6nk
n1
Y2
1
2
: step : X ¼ 1 À ; Y ¼
; S ¼ X
1
2
2
1
(
Scheme 2). We recently showed the general feasibility of this
n2
n3
n1;0
Y3
Y2ÁX2
P
ð1Þ
: step : X ¼ 1 À
; Y ¼ ; S ¼
2
3
3
k¼2;3;6nk
n
Yl ¼ n with l ¼ 5, 6, 7, 8
1
;0
Equation (1) gives the calculations of conversion (X), yield (Y)
i
i
and selectivity (S) with i=respective substance according
i
Scheme 4; n =moles at start for i; with respect to the stoichi-
i,0
Scheme 2. Deprotonation of thiamin/vitamin B
bene C1a.
1
(C1) obtaining the free car-
ometry, n and n must be equalised by the factor 2 needing
3
6
two equivalents of 2.
Starting from our model substrate 1-decene (1), different re-
action pathways could be observed: the desired reaction path-
way and the undesired reaction pathways are displayed in
Scheme 4. The undesired pathways must be avoided with pre-
cisely executed changes regarding the conditions and used re-
agents. The desired reaction pathway starts with the hydrofor-
new tandem catalysis, although it suffered from low selectivity
towards the desired reaction pathway: from the alkene via the
aldehyde to the acyloin. This led to relatively poor yields of
3
1% for the acyloin and moderate selectivities for the inter-
mediate n-aldehyde of 44%, which was a serious drawback.
Therefore, we anticipated the key to success by overcoming
isomerisation issues and progressed this tandem hydroformyla-
tion/acyloin reaction to a new level establishing a highly selec-
tive reaction pathway for both steps, the hydroformylation and
the acyloin reaction, through systematic investigations. To
enable the relevance of acyloin products in real-life applica-
tions, we demonstrate an application of the acyloins as build-
[
23]
ing blocks for surfactants next to anti-skimming agents and
[
24]
natural products.
Results and Discussion
The aforementioned low selectivity resulted from undesired
side reactions, which were almost exclusively isomerisation re-
actions. These reactions were favoured, after applying the opti-
[17]
mised conditions to long-chain molecules. Herein, we ap-
plied the conditions to 1-decene (1, Scheme 3). A yield of 57%
of isomer 4 was observed and only 31% n-aldehyde 2 with
a poor selectivity of 35% was formed leading to 24% acyloin 3
after two steps. Owing to this observed high isomerisation ac-
tivity paired with low selectivity in the first reaction step, we
chose 1 as model substrate to overcome the side reaction.
Scheme 4. Desired and undesired reaction pathways.
ChemCatChem 2015, 7, 2607 – 2613
www.chemcatchem.org
2608
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Papers
mylation of 1 yielding n-undecylenic aldehyde (2). Two equiva-
lents of 2 were converted into the desired final acyloin prod-
uct 3. Further undesired products besides the isomer 4 were:
the alkane 5, the aldol condensation product 6 and the iso- (7)
and branched (8) aldehydes.
tested systems the alkane 5 was observed throughout in negli-
gible amounts of <1%. Xantphos showed similar conversion
of 1 (X =88%) in hydroformylation yielding 2 (Y =34%) in
1
2
similar amounts to Biphephos, whereas the results for 2. step
revealed less conversion of the aldehyde 2 (X =50%, no. 1.2).
2
To achieve our goal developing a highly selective tandem
catalysis, it was necessary to undertake detailed investigations
regarding all observed products, intermediates and side prod-
ucts during a stepwise change of the applied conditions. As
a carbene was necessary for the acyloin step as catalyst and in-
fluenced the hydroformylation step as ligand, both steps
needed to be investigated together in a tandem catalysis.
For increasing the selectivity towards the hydroformylation,
we tested other bidentate phosphorous ligands (Figure 1) at
first. These ligands are known for their diminished activity in
These results suggest a deactivation of the carbene catalyst,
see below.
If BISBI was used, higher selectivity in hydroformylation (S =
2
53%) along with less isomers (Y =32%) was observed
4
(Table 1, no. 1.3). The least isomerisation activity with 18% iso-
mers was detected by applying Nixantphos (no. 1.4). This led
to a relatively good selectivity of 74% for 2. Unfortunately, the
Rh/Nixantphos system had a lower reaction rate in hydrofor-
mylation leading to less conversion of 50% n-aldehydes 2, al-
though the yield of the overall formed aldehydes was the
highest with 58%. In summary, Rh/Biphephos was more active
regarding the hydroformylation rate and Rh/Nixantphos led to
less isomerisation activity.
[25]
double-bond isomerisations (Table 1), whereas the isomerisa-
tion of the double bond is favoured especially during low tem-
[26]
peratures if applying Biphephos (no. 1.1). Noteworthy, for all
If higher temperatures were applied, the hydroformylation
was enhanced and the yield of the undesired isomers was de-
creased. As seen in Table 1, no. 1.5, we observed less isomerisa-
tion with 24% isomers and an increased selectivity of 56% for
2
. Unfortunately, the carbene precursor C1, which was added
as hydrochloride salt, is not stable at higher temperatures.
Hence, the derivative C2 of thiamine C1, which was more
[
19a]
stable at higher temperatures,
was applied (Scheme 5). Ad-
Scheme 5. Deprotonation of thiamine derivative C2 obtaining the free car-
bene C2a.
Figure 1. Chemical structures of used ligands
vantageously, less base was added, as a consequence of the
absence of one equivalent hydrochloride.
Table 1. Application of different ligands in the tandem hydroformylation/
acyloin reaction.
[a]
The addition of C2a instead of C1a led to higher selectivity
in hydroformylation (S ), thus in general indicating the influ-
2
ence on both reaction steps. In the presence of Biphephos,
less isomer 4 was obtained increasing the overall selectivity to
moderate 46% (Table 1, no. 1.6). A very good selectivity of S =
2
9
2% was achieved by applying C2a along with Nixantphos
1
. Step
2. Step
Undesired
leading to only 6% of isomer 4 (no. 1.7). Apparently, the car-
No.
Lig.
X
1
Y
2
S
2
X
2
Y
3
S
3
Y
7
Y
4
Y
6
bene was active in both ways, as catalyst in the acyloin reac-
[
27]
1
1
1
1
1
1
1
.1
.2
.3
.4
.5
.6
.7
L1
L2
L3
L4
L1
L1
L4
88 31 35
88 34 39
70 37 53
78 58 74
57 32 56
89 41 46
83 76 92
77 24 >99
<1 57 <1
tion and as ligand in the hydroformylation reaction. This was
supported by the different yields for the aldehyde 2 with 58%
if applying C1 (no. 1.4) or 76% if applying C2 (no. 1.7). Unfortu-
nately, still poor yields of 38% for the final acyloin product 3
(no. 1.7) were obtained.
50 10
38
50 20
22
78 30
51 38
59
57
69
71
94
97
2
1
2
1
1
1
52
32
18
24
47
6
7
6
9
2
2
1
8
[b]
[c]
[c]
5
Even though high yields of isomer 4 were obtained (Table 1,
no. 1.1, 1.2, 1.6), remarkably, no considerable amounts of
branched aldehydes 8 were detected (yields under the limit of
determination, that is, !1%). To prove no branched aldehydes
were formed in the hydroformylation step, we considered
[
a] Conditions: 6 mmol 1-decene, 0.03 mmol Rh(CO)
2
acac, 0.09 mmol
2
(1:1), 5 mL DMF,
ligand, 0.6 mmol C1·HCl, 2.4 mmol Et N, 30 bar CO/H
3
6
1
08C, 16 h. Results determined by GC. [b] 908C, [c] 0.6 mmol C2 and
.2 mmol Et N.
3
ChemCatChem 2015, 7, 2607 – 2613
www.chemcatchem.org
2609
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Papers
reaction parts and other conditions that influence the interac-
tion between both reaction steps. MeCN (acetonitrile), g-butyr-
olactone, propylene carbonate or toluene were not beneficial
solvents in this tandem catalysis. The moderate conversion of
X1 =42% led to poor yields in MeCN as solvent (Table 3,
no. 3.2). g-Butyrolacton effected the hydroformylation by ob-
taining poor n/bra ratio of 2:1. In propylene carbonate we ob-
Table 2. Investigation of the hydroformylation with internal double
bonds.
[a]
served very poor conversion X of the aldehydes and toluene
2
No.
Ligand
C2
Et
3
N
X
9
Y
10
showed insufficient solubility of the components.
The usage of dimethylformamide (DMF) as solvent and Bi-
phephos as phosphorous ligand at higher temperatures led to
a very good selectivity of 93% for 2, however, the yield of the
acyloin 3 was only moderate with 46% (Table 3, no. 3.1). On
the other hand the combination of N-methyl-2-pyrrolidone
2
2
2
2
.1
.2
.3
.4
+
+
+
+
–
+
–
–
–
+
+
63
<1
1
63
<1
1
+
—
—
[
a] Conditions: 6 mmol 9, 0.03 mmol Rh(CO)
.6 mmol C2, 1.2 mmol Et N, 30 bar CO/H
Results determined by GC.
2
acac, 0.09 mmol Biphephos,
(1:1), 5 mL DMF, 608C, 16 h.
0
3
2
(
NMP) and Biphephos were not fruitful for the hydroformyla-
tion, owing to lower yield of 2 with 76% (no. 3.3). As the appli-
cation of DMF and Nixantphos at lower temperatures led to
very good selectivity of S =92% (Table 1, no. 1.7), it was quite
2
a more detailed investigation with cyclooctene 9 as a model
substrate (Table 2). Under typical hydroformylation conditions,
that is, in the absence of C2 and base, 63% aldehydes 10 were
formed (Table 2, no. 2.1). In the presence of either C2 (no. 2.2)
or base (no 2.3), only small amounts of aldehydes 10 were ob-
tained. If the carbene C2a was formed in situ, by deprotona-
tion of C2 with base, no conversion and no acyloins were ob-
served (no. 2.4). This suggests a highly regioselective hydrofor-
mylation, by converting only terminal double bonds, if the
astonishing why a lower conversion of 52% for 1 was ob-
served at 908C (Table 3, no. 3.4). However, this could be ex-
plained by undesired interactions between both reaction steps
and conditions. Nevertheless, through the systematic change
of conditions and reagents and the synergy of NMP, Nixant-
phos and C2a at 908C, an excellent conversion of X = >99%,
1
selectivity of S =96% with 95% formed 2 and good yield of
2
61% for the acyloin product 3 were achieved (no. 3.5). Only
2% of isomer 4 were obtained. Under comparable conditions
we applied imidazolium salts as carbene precursors as well
(Figure 2), because Rh/imidazole NHCs are well known for their
[
rhodium/phosphorous–ligand/carbene–ligand] catalyst com-
[
17]
plex was applied. This verified our preliminary results that
no branched aldehydes were observed if 1-octene was applied
and internal double bonds were formed thereof.
[
27,28]
activity in the hydroformylation.
These imidazole NHCs dis-
played comparable high yields up to 92% for 2 in the hydro-
formylation step. However, very poor yields of 0–6% for the
acyloin 3 were obtained (Table 3, no. 3.6, 3.7, 3.8), indicating
poor activity in the acyloin reaction. These results showed the
importance of matching the parameters on each reaction step
in this tandem catalysis. These adjusted parameters decided
about desired and undesired interactions between
both reactions, wherefore systematic investigations
regarding this tandem catalysis were necessary.
As seen in Table 1, higher temperature increased the selec-
tivity for 2 (Table 1, no. 1.5), however, unfortunately, C1 is not
stable under these conditions. Thereupon, we added C2 at
higher temperature (Table 3). Furthermore, other solvents were
used, because the application of new catalyst systems in both
[a]
Table 3. Investigation of different solvents at higher temperature.
For a better understanding of the reaction prog-
ress of both reaction steps in the tandem hydrofor-
mylation/acyloin reaction, we investigated the reac-
tion progress over time. This investigation regarding
the optimised conditions from Table 3, no. 3.5, is
shown in Figure 3. After 3 h, the hydroformylation re-
1. Step
2. Step
Undesired
No.
Solv.
Ligand
X
1
Y
2
S
2
X
2
Y
3
S
3
Y
7
Y
4
Y
6
Y
5
action was finished, and a full conversion of 1-decene
was obtained. The acyloin reaction was completed
after a maximum of 16 h giving 61% yield of acy-
loins. This yield was not increasing further even after
[
b]
1
3
3
3
3
3
3
3
3
.1
.1
.2
.3
.4
.5
.6
.7
.8
DMF
DMF
MeCN L1
NMP L1
DMF
NMP L4
DMF
DMF
DMF
L1
L1
88 31 35
92 86 93
42 35 83
90 76 84
52 46 88
77 24 >99
58 46
17
47 34
65 28
66 61
<1 57 <1 <1
<1
2
92
83
94
93
97
5
4
10
4
2
5
4
1
2
1
1
1
5
3
2
2
2
2
4
2
4 h. The yield of the isomers was constantly less
L4
2
<1
2
than 2% and not displayed in the diagram. Further
conversion of the n-aldehyde 2 after 16 h led only to
aldol side-products. Detailed investigations revealed
that, if more aldehyde was added after 16 h, no fur-
ther conversion into the acyloin was observed. There-
fore, the active carbene species was not recovered
after 16 h, proving a deactivation of the carbene. Fur-
>99 95 96
99 92 93
>99 90 90
>99 81 81
<1
<1 <1
<1 <1
[c]
[d]
[e]
L4
L4
L4
2
2
0
6
>99
–
>99
<1
8
7
15 <1 <1
[
a] Conditions: 6 mmol 1-decene, 0.03 mmol Rh(CO)
C2, 1.2 mmol Et N, 30 bar CO/H (1:1), 5 mL solvent=Solv., 908C, 16 h. Results deter-
mined by GC. [b] 608C; [c] 0.6 mmol C3. [d] 0.6 mmol C4. [e] 0.6 mmol C5.
2
acac, 0.09 mmol ligand, 0.6 mmol
3
2
ChemCatChem 2015, 7, 2607 – 2613
www.chemcatchem.org
2610
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Papers
ther investigations of the acyloin reaction starting from the al-
dehyde under hydroformylation conditions were also per-
formed. The same activities in the presence and absence of
the hydroformylation catalyst system were observed, indicating
no influence of the hydroformylation catalyst system on the
acyloin reaction.
We then explored the transferability of these optimised con-
ditions (Table 3, no. 3.5) by using the same substrates as in our
[
17]
recent publication,
because a more accurate comparison
was given this way (Table 4). This should reveal whether a sup-
pression of numerous side reactions can be realised by apply-
ing other substrates as well.
Figure 2. Applied imidazolium salts as carbene precursors.
All substrates showed much lower isomerisation activity
under optimised conditions with yields of only YIsom =2–4% for
isomers (Table 4, no. 4.1–4.6). Fortunately, other side-products
such as iso-aldehydes Yiso-Ald, hydrogenated substrates YHydr and
aldol products YAldol were also formed consistently with yields
in the range of 1–2% (no. 4.1–4.5). This resulted in good selec-
tivities for the desired intermediate/products and thus also in
an excellent selectivity for the whole tandem catalysis for all
applied substrates. The only exception was methyl 10-undece-
noate, which corresponding n-aldehydes tend to perform aldol
condensation under the chosen tandem conditions (no. 4.6).
Excellent selectivities of SAld =94–>99% and very good to ex-
cellent yields of Yn-Ald =88–96% for the formed n-aldehydes
were obtained (no. 4.1–4.6). Good yields of YAcyl =61–79% for
the final acyloin products were achieved as well (no. 4.1–4.6).
Figure 3. Reaction–time plot of the reaction progress in tandem hydrofor-
mylation/acyloin reaction. Conditions: 6 mmol 1, 0.03 mmol Rh(CO)
2
acac,
0
9
.09 mmol L4, 0.6 mmol C2, 1.2 mmol Et
08C, 16 h. Results determined by GC.
N, 30 bar CO/H (1:1), 5 mL NMP,
3 2
[
a]
Table 4. Comparison of the new and previous catalyst system.
1. Step
2. Step
Undesired
No.
Product
X
Olefin
Y
n-Ald
S
n-Ald
X
n-Ald
Y
Acyl
S
Acyl
Y
iso-Ald
Y
Isom
Y
Hydr
Y
Aldol
4
5
.1
.1
89
98
89
98
>99
>99
81
78
72
76
>99
>99
–
–
–
–
<1
<1
<1
<1
[b]
4
5
.2
.2
>99
74
96
49
96
66
79
84
76
41
>99
>99
1
1
3
25
<1
<1
<1
<1
[b]
4
5
.3
.3
100
100
95
82
95
82
79
89
75
73
>99
>99
1
<1
4
18
<1
<1
<1
<1
[
[
[
[
b]
b]
b]
b]
4
5
.4
.4
99
88
95
31
96
35
66
77
61
24
97
>99
2
<1
2
57
<1
<1
2
<1
4
5
.5
.5
93
81
88
36
95
44
90
86
79
31
>99
>99
2
1
3
45
<1
<1
<1
<1
4
5
.6
.6
>99
>99
94
45
94
45
96
87
66
34
73
87
1
1
4
53
1
3
24
5
[
a] Conditions: 6 mmol olefin, 0.03 mmol Rh(CO)
2
acac, 0.09 mmol Nixantphos, 0.6 mmol C2, 1.2 mmol Et
3
N, 30 bar CO/H
2
(1:1), 5 mL NMP, 908C, 16 h. Re-
sults determined by GC. [b] 6 mmol olefin, 0.03 mmol Rh(CO)
2
acac, 0.09 mmol Biphephos, 0.6 mmol C1, 2.4 mmol Et
3
N, 30 bar CO/H (1:1), 5 mL DMF,
2
[17]
6
08C, 16 h.
ChemCatChem 2015, 7, 2607 – 2613
www.chemcatchem.org
2611
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Papers
In summary, we improved the selectivity and yield for the n-
aldehyde and acyloin of every applied substrate, obtaining
a general method for the direct conversion of olefins into acy-
loins in one preparative step. This orthogonal tandem catalysis
is advantageous towards a one-pot reaction wherein the sub-
sequent addition of the carbene precursor and base favour the
aldol reaction, which led to low selectivities.
kenes. It is conceivable that ethoxylation of 3 will directly lead
to a surfactant. Thereby, a surfactant can be synthesised within
only two preparative steps.
Conclusion
We described an advanced development of the tandem hydro-
formylation/acyloin reaction, which leads directly to acyloins
from olefins by suppressing numerous side products. We
found new catalyst systems for each reaction, by changing re-
agents and conditions step by step, which was necessary be-
cause of the interactions between both catalyst systems. This
tandem catalysis is finally transferable to various substrates. Ex-
cellent overall reaction selectivities up to >99% have been
achieved with excellent yields up to 96% for the desired inter-
mediate n-aldehydes and very good yields up to 79% for the
final acyloin products were obtained. For the side products,
only negligible amounts up to 5% were observed. Further-
more, we observed foaming properties and determined surfac-
tant activity after attaching a polar head group to the acyloin
starting from 1-decene, which opens up a new class of surfac-
tants.
Both acyloin products from 5-hexenenitrile (Table 4, no. 4.5)
and methyl undecylenoate (no. 4.6) are interesting precursors
for potential monomers in polymerisation, owing to their a,w-
functionalisation. A possible application of long-chain acyloins,
for example, from 1-decene (no. 4.4) is not visible at first
glance. Owing to the long carbon chain and the alcohol
moiety, the acyloin backbone can be used as a nonpolar pre-
cursor in surfactants creating a double-tail surfactant with two
linear alkyl chains. Attaching succinic acid anhydride as a linker
to the acyloin 3, which resulted in 69% of 3a, we were able to
attach poly(ethylene glycol) 600 (PEG-600) yielding 3b almost
quantitatively (Scheme 6). This synthesized double-tail mole-
Experimental Section
General procedure of the tandem hydroformylation/acyloin
reaction
Rh(CO) (acac) (0.03 mmol), L4 (0.09 mmol), C2 (0.6 mmol) and
2
Scheme 6. Three-step reaction pathway for the double-tail surfactant 3b.
1
(6 mmol) were dissolved in a 40 mL custom-made stainless-steel
autoclaves in 5 mL NMP. After adding triethylamine (1.2 mmol) the
autoclave was pressurised with 30 bar syngas (CO/H =1:1) and the
2
cule with a polar head group showed foaming activity during
the work-up process. A quantification of the surface active
properties revealed a critical micelle concentration (C.M.C.) of
mixture was stirred at 908C for 16 h magnetically. After cooling to
RT the autoclave was depressurized and flushed with argon. For
the GC measurements, a sample of the mixture was taken and ana-
lysed by GC with dodecane as internal standard and isopropyl al-
cohol as additional solvent. Otherwise the reaction mixture was
concentrated and purified by column chromatography (cyclohex-
ane/ethyl acetate 100:1!5:1) yielding white solid 12-hydroxydoco-
À1
À1
0
.13 gL , a water solubility of 0.022 gL and a surface ten-
À1
sion of 33.1 mNm (Figure 4). The value of the C.M.C. was
low, attributable to the poor water solubility. Therefore, a modi-
fication to shorter chain length should increase the water solu-
bility. In this way, a new synthetic route and a new class of sur-
factants has been opened up employing three steps from 1-al-
1
san-11-one (3). H NMR (400 MHz, CDCl3): d=0.90 (m, 6H), 1.29 (m,
2
9H), 1.45–1.64 (m, 4H), 1.83 (m, 1H), 2.46 (m, 2H), 3.50 (d, J=
13
4.8 Hz, 1H), 4.18 ppm (m, 1H); C NMR (100 MHz, CDCl3): d=14.5
2”C), 23.1 (2”C), 24.1, 25.2, 29.6, 29.7, 29.7, 29.8, 29.8, 29.9 (2”C),
0.0 (2”C), 30.0, 32.3 (2”C), 34.2, 38.3, 76.8, 213.0 ppm; HRMS
(
3
+
(
ESI) analysis calcd for C H O [M+H]
341.34141, found
22
45
2
À1
3
41.34083. IR: n˜ =3322, 2913, 2848, 1709 cm .
The characterisation data of other acyloin products can be taken
[17]
from the literature.
Procedure for the synthesis of 3b
We followed the procedure according to dl-a-tocopherol succinate
yielding 69% of 3a and TPGS-750M yielding 99% of 3b from the
[29]
1
experimental section in the literature. For 3b: H NMR (400 MHz,
CDCl ): d=0.87 (m, 6H), 1.25 (m, 30H), 1.56 (m, 2H), 1.71 (m, 2H),
3
2
4
.35 (m, 1H), 2.45 (m, 1H), 2.56 (bs, 1H), 2.68 (m, 3H), 3.64 (m,
8H), 4.24 (m, 2H), 4.98 ppm (m, 1H); C NMR (100 MHz, CDCl3):
13
d=2”14.3, 2”22.9, 23.3, 25.1, 25.4, 29.1, 29.2, 29.4, 29.5, 29.5,
29.5, 29.6, 29.6, 29.7, 29.7, 29.8, 30.7, 32.1, 34.4, 38.8, 62.0, 63.6,
64.1, 69.3, 69,4, 70.6, 70.8, 70.9, 72.7, 79.0, 172.3, 172.1, 207.7 ppm;
Figure 4. Pendency of the surface tension to the concentration of the surfac-
tant 3b.
ChemCatChem 2015, 7, 2607 – 2613
www.chemcatchem.org
2612
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Papers
+
HRMS (ESI) Elemental analysis calcd (%) for C H101O18 [M+H]
1
6
[11] a) C. G. Vieira, J. G. da Silva, C. a. a. Penna, E. N. dos Santos, E. V. Gusev-
skaya, Appl. Catal. A 2010, 380, 125–132; b) J. Norinder, C. Rodrigues, A.
Bçrner, J. Mol. Catal. A 2014, 391, 139–143.
5
2
013.69824, found 1013.70325 (valid for 13 glycol units in PEG-
00).
[12] a) A. Behr, T. Seidensticker, A. J. Vorholt, Eur. J. Lipid Sci. Technol. 2014,
116, 477–485; b) K. C. B. Oliveira, S. N. Carvalho, M. F. Duarte, E. V. Gu-
sevskaya, E. N. dos Santos, J. El Karroumi, M. Gouygou, M. Urrutigoïty,
Appl. Catal. A 2015, 497, 10–16.
Acknowledgements
[
[
13] a) Z. Shao, H. Zhang, Chem. Soc. Rev. 2009, 38, 2745–2755; b) Z. Du, Z.
Shao, Chem. Soc. Rev. 2013, 42, 1337–1378.
14] a) S. Chercheja, T. Rothenbücher, P. Eilbracht, Adv. Synth. Catal. 2009,
The authors thank Dr. D. Leinweber and Dr. S. Romanski from
Clariant AG for the measurement of the surface tension of the
surfactant 3b and UMICORE AG & Co. KG for the donation of the
3
1
51, 339–344; b) S. T. Kemme, T. Smejkal, B. Breit, Chem. Eur. J. 2010,
6, 3423–3433; c) J. Stiller, A. J. Vorholt, K. A. Ostrowski, A. Behr, M.
precursor Rh(CO) acac. Additionally, we thank T. Szigat for help-
2
Christmann, Chem. Eur. J. 2012, 18, 9496–9499.
ing with the cover picture.
[15] a) S. Chercheja, P. Eilbracht, Adv. Synth. Catal. 2007, 349, 1897–1905;
b) O. Abillard, B. Breit, Adv. Synth. Catal. 2007, 349, 1891–1895; c) S.
Chercheja, S. K. Nadakudity, P. Eilbracht, Adv. Synth. Catal. 2010, 352,
637–643; d) X. Fang, R. Jackstell, R. Franke, M. Beller, Chem. Eur. J. 2014,
20, 13210–13216.
Keywords: aldehydes · alkenes · high-pressure chemistry ·
hydroformylation · surfactants
[
16] a) T. Nemoto, T. Fukuda, Y. Hamada, Tetrahedron Lett. 2006, 47, 4365–
368; b) R. Lebeuf, K. Hirano, F. Glorius, Org. Lett. 2008, 10, 4243–4246.
17] K. A. Ostrowski, T. A. Faßbach, A. J. Vorholt, Adv. Synth. Catal. 2015, 357,
374–1380.
4
[
1] a) L. F. Tietze, Chem. Rev. 1996, 96, 115–136; b) D. E. Fogg, E. N. dos San-
tos, Coord. Chem. Rev. 2004, 248, 2365–2379; c) H. Pellissier, Tetrahedron
[
1
2013, 69, 7171–7210; d) A. Behr, A. J. Vorholt, K. A. Ostrowski, T. Seiden-
[
[
18] T. L. Lohr, T. J. Marks, Nat. Chem. 2015, 7, 477–482.
sticker, Green Chem. 2014, 16, 982–1006.
19] a) H. Stetter, R. Y. Rämsch, H. Kuhlmann, Synthesis 1976, 733–735; b) H.
Stetter, G. Dämbkes, Synthesis 1977, 403–404; c) H. Stetter, G. Dämbkes,
Synthesis 1980, 309–310; d) H. Stetter, R. Y. Rämsch, Synthesis 1981,
[
[
2] P. T. Anastas, J. C. Warner, Green Chemistry, Theory and Practice, Oxford
University Press, Oxford, 1998.
3] P. Eilbracht, L. Bärfacker, C. Buss, C. Hollmann, B. E. Kitsos-Rzychon, C. L.
Kranemann, T. Rische, R. Roggenbuck, A. Schmidt, Chem. Rev. 1999, 99,
4
5
77–478; e) D. Enders, O. Niemeier, A. Henseler, Chem. Rev. 2007, 107,
606–5655; f) N. Marion, S. Díez-Gonzµlez, S. P. Nolan, Angew. Chem. Int.
3
329–3366.
4] O. Roelen, German Patent DE 849548, 1938/1952; U.S. Patent 2327066,
943.
Ed. 2007, 46, 2988–3000; Angew. Chem. 2007, 119, 3046–3058.
20] V. L. Hansley, U.S. Patent 1996471, 1935.
21] N. Rieber, J. H. Teles, U.S. Patent 5576462, 1996.
22] T. Saito, H. Kumobayashi, S. Murahashi, U.S. Patent 5210315, 1993.
23] G. I. W. Weiss, W.-D. Balzer, R. Fikentscher, H. Wolf, U.S. Patent 4618371,
[
[
[
[
[
1
[
[
[
5] F. Hebrard, P. Kalck, Chem. Rev. 2009, 109, 4272–4282.
6] R. Franke, D. Selent, A. Bçrner, Chem. Rev. 2012, 112, 5675–5732.
7] a) M. Vasylyev, H. Alper, Synthesis 2010, 2893–2900; b) X.-F. Wu, X. Fang,
L. Wu, R. Jackstell, H. Neumann, M. Beller, Acc. Chem. Res. 2014, 47,
1986.
[
[
24] Z. Lin, L. Marett, R. W. Hughen, M. Flores, I. Forteza, M. Anne, G. P. Con-
cepcion, S. Espino, B. M. Olivera, G. Rosenberg, M. G. Haygood, A. R.
Light, E. W. Schmidt, Bioorg. Med. Chem. Lett. 2013, 23, 4867–4869.
25] a) M. Kranenburg, Y. E. M. van der Burgt, P. C. J. Kamer, P. W. N. M. van
Leeuwen, Organometallics 1995, 14, 3081–3089; b) P. W. N. M. van Leeu-
wen, P. C. J. Kamer, J. N. H. Reek, Pure Appl. Chem. 1999, 71, 1443–1452;
c) Z. Freixa, P. W. N. M. van Leeuwen, P. W. N. M. van Leeuwen, Dalton
Trans. 2003, 1890–1901; d) P. W. N. M. van Leeuwen, P. C. J. Kamer,
J. N. H. Reek, P. Dierkes, Chem. Rev. 2000, 100, 2741–2769.
1
041–1053.
[
8] a) B. Breit, S. K. Zahn, Angew. Chem. Int. Ed. 1999, 38, 969–971; Angew.
Chem. 1999, 111, 1022–1024; b) B. Breit, B. M. Trost, Angew. Chem. Int.
Ed. 2001, 40, 1910–1913; Angew. Chem. 2001, 113, 1964–1967; c) P.
Kçhling, A. M. Schmidt, P. Eilbracht, Org. Lett. 2003, 5, 3213–3216; d) L.
Monnereau, D. SØmeril, D. Matt, Green Chem. 2010, 12, 1670–1673;
e) C. G. Vieira, M. C. de Freitas, E. N. dos Santos, E. V. Gusevskaya, Chem-
CatChem 2012, 4, 795–801; f) D. Fuchs, M. Nasr-Esfahani, L. Diab, T.
ˇ
Smejkal, B. Breit, Synlett 2013, 24, 1657–1662; g) M. Ali, A. Gual, G. Ebel-
[
26] a) A. Behr, D. Obst, C. Schulte, T. Schosser, J. Mol. Catal. A 2003, 206,
ing, J. Dupont, ChemCatChem 2014, 6, 2224–2228; h) A. R. Almeida,
R. D. Dias, C. J. P. Monteiro, A. R. Abreu, P. M. P. Gois, J. C. Bayon, M. M.
Pereira, Adv. Synth. Catal. 2014, 356, 1223–1228.
9] a) K. Takahashi, M. Yamashita, T. Ichihara, K. Nakano, K. Nozaki, Angew.
Chem. Int. Ed. 2010, 49, 4488–4490; Angew. Chem. 2010, 122, 4590–
179–184; b) A. Behr, D. Obst, A. Westfechtel, Eur. J. Lipid Sci. Technol.
2005, 107, 213–219.
[
[
27] W. Gil, A. M. Trzeciak, Coord. Chem. Rev. 2011, 255, 473–483.
28] A. R. Almeida, A. F. Peixoto, M. J. F. Calvete, P. M. P. Gois, M. M. Pereira,
Curr. Org. Synth. 2011, 8, 764–775.
[
4
592; b) K. Takahashi, M. Yamashita, K. Nozaki, J. Am. Chem. Soc. 2012,
[
29] B. H. Lipshutz, S. Ghorai, A. R. Abela, R. Moser, T. Nishikata, C. Duplais, A.
1
34, 18746–18757; c) D. Fuchs, G. Rousseau, L. Diab, U. Gellrich, B. Breit,
Krasovskiy, R. D. Gaston, R. C. Gadwood, J. Org. Chem. 2011, 76, 4379–
Angew. Chem. Int. Ed. 2012, 51, 2178–2182; Angew. Chem. 2012, 124,
220–2224.
10] a) M. Vilches-Herrera, L. Domke, A. Bçrner, ACS Catal. 2014, 4, 1706–
724; b) J. Ternel, J.-L. Couturier, J.-L. Dubois, J.-F. Carpentier, Adv. Synth.
Catal. 2013, 355, 3191–3204.
4391.
2
[
1
Received: June 25, 2015
Published online on August 12, 2015
ChemCatChem 2015, 7, 2607 – 2613
www.chemcatchem.org
2613
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim