Chemo- and regioselective homogeneous rhodium-catalyzed hydroamidomethylation of terminal alkenes to N-alkylamides

Article abstract of DOI:10.1002/cssc.201300484

A rhodium/xantphos homogeneous catalyst system has been developed for direct chemo- and regioselective mono-N-alkylation of primary amides with 1-alkenes and syngas through catalytic hydroamidomethylation with 1-pentene and acetamide as model substrates. For appropriate catalyst performance, it appears to be essential that catalytic amounts of a strong acid promoter, such as p-toluenesulfonic acid (HOTs), as well as larger amounts of a weakly acidic protic promoter, particularly hexafluoroisopropyl alcohol (HOR^F) are applied. Apart from the product N-1-hexylacetamide, the isomeric unsaturated intermediates, hexanol and higher mass byproducts, as well as the corresponding isomeric branched products, can be formed. Under optimized conditions, almost full alkene conversion can be achieved with more than 80 % selectivity to the product N-1-hexylamide. Interestingly, in the presence of a relatively high concentration of HOR^F, the same catalyst system shows a remarkably high selectivity for the formation of hexanol from 1-pentene with syngas, thus presenting a unique example of a selective rhodium-catalyzed hydroformylation-hydrogenation tandem reaction under mild conditions. Time-dependent product formation during hydroamidomethylation batch experiments provides evidence for aldehyde and unsaturated intermediates; this clearly indicates the three-step hydroformylation/condensation/hydrogenation reaction sequence that takes place in hydroamidomethylation. One likely role of the weakly acidic protic promoter, HOR^F, in combination with the strong acid HOTs, is to establish a dual-functionality rhodium catalyst system comprised of a neutral rhodium(I) hydroformylation catalyst species and a cationic rhodium(III) complex capable of selectively reducing the imide and/or ene-amide intermediates that are in a dynamic, acid-catalyzed condensation equilibrium with the aldehyde and amide in a syngas environment. Taking control: A rhodium/xantphos homogeneous catalyst system has been developed for direct chemo- and regioselective mono-N-alkylation of primary amides with 1-alkenes and syngas through the new catalytic hydroamidomethylation reaction (see picture). Copyright

Full text of DOI:10.1002/cssc.201300484

CHEMSUSCHEM
FULL PAPERS
DOI: 10.1002/cssc.201300484
Chemo- and Regioselective Homogeneous Rhodium-
Catalyzed Hydroamidomethylation of Terminal Alkenes to
N-Alkylamides
Saeed Raoufmoghaddam, Eite Drent, and Elisabeth Bouwman*[a]
A rhodium/xantphos homogeneous catalyst system has been
developed for direct chemo- and regioselective mono-N-alkyla-
tion of primary amides with 1-alkenes and syngas through cat-
alytic hydroamidomethylation with 1-pentene and acetamide
as model substrates. For appropriate catalyst performance, it
appears to be essential that catalytic amounts of a strong acid
promoter, such as p-toluenesulfonic acid (HOTs), as well as
larger amounts of a weakly acidic protic promoter, particularly
hexafluoroisopropyl alcohol (HORF) are applied. Apart from the
product N-1-hexylacetamide, the isomeric unsaturated inter-
mediates, hexanol and higher mass byproducts, as well as the
corresponding isomeric branched products, can be formed.
Under optimized conditions, almost full alkene conversion can
be achieved with more than 80% selectivity to the product N-
1-hexylamide. Interestingly, in the presence of a relatively high
concentration of HORF, the same catalyst system shows a re-
markably high selectivity for the formation of hexanol from 1-
pentene with syngas, thus presenting a unique example of
a selective rhodium-catalyzed hydroformylation–hydrogenation
tandem reaction under mild conditions. Time-dependent prod-
uct formation during hydroamidomethylation batch experi-
ments provides evidence for aldehyde and unsaturated inter-
mediates; this clearly indicates the three-step hydroformyla-
tion/condensation/hydrogenation reaction sequence that takes
place in hydroamidomethylation. One likely role of the weakly
acidic protic promoter, HORF, in combination with the strong
acid HOTs, is to establish a dual-functionality rhodium catalyst
system comprised of a neutral rhodium(I) hydroformylation
catalyst species and a cationic rhodium(III) complex capable of
selectively reducing the imide and/or ene–amide intermediates
that are in a dynamic, acid-catalyzed condensation equilibrium
with the aldehyde and amide in a syngas environment.
Introduction
The formation of carbon–nitrogen bonds is of great interest to
synthetic chemists because nitrogen-containing molecules are
important in bulk and fine-chemical building blocks, solvents,
surfactants, dyes, pharmaceuticals, agrochemicals, and biologi-
cally active compounds.[1] Among the various catalytic meth-
ods known for the synthesis of amines, the hydroaminomethy-
lation of alkenes is highly atom-economical and efficient. This
cascade reaction consists of initial hydroformylation followed
by reductive amination (Scheme 1),[2] and was originally discov-
ered by Reppe and Vetter in the early 1950s at BASF by using
[Fe(CO)5] in nearly stoichiometric amounts.[3] Research on this
reaction up to the mid-1990s revealed that relatively harsh
conditions (>1508C) were required to give the desired amines
in good yield.[4] The critical step in this sequence is the hydro-
genation of intermediate imino compounds, which is generally
hampered by the presence of carbon monoxide, but this can
be overcome by using alcoholic and polar solvents.[5] However,
hydroaminomethylation of alkenes is mostly limited to the use
Scheme 1. Hydroaminomethylation versus the hydroamidomethylation
reaction.
of secondary amines; for primary amines and ammonia, the se-
lectivity is generally low because of over-alkylation.
In analogy with hydroaminomethylation, the thus far un-
known hydroamidomethylation reaction would comprise of
a cascade reaction of hydroformylation and catalytic reductive
amidation (Scheme 1). The development of this new method
for the synthesis of N-alkylamides using an amide as the sub-
strate instead of an amine avoids the formation of over-alkylat-
ed side products due to the low nucleophilicity and steric en-
cumbrance at the amide nitrogen atom of the initially formed
mono-N-alkylamide. The low nucleophilicity of the primary
[a] S. Raoufmoghaddam, Prof. Dr. E. Drent, Prof. Dr. E. Bouwman
Leiden Institute of Chemistry, Gorlaeus Laboratories
Leiden University, P.O. Box 9502
2300 RA Leiden (The Netherlands)
Fax: (+31)71-5274761
E-mail: bouwman@chem.leidenuniv.nl
Supporting Information for this article is available on the WWW under
http://dx.doi.org/10.1002/cssc.201300484.
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemSusChem 2013, 6, 1759 – 1773 1759

CHEMSUSCHEM
FULL PAPERS
www.chemsuschem.org
amide substrate generally makes the development of active
and selective catalysts for the alkylation of a (primary) amide
through hydroamidomethylation even more challenging than
alkylation of (secondary) amines by hydroaminomethylation.
Inspired by our recent achievements concerning the catalyt-
ic reductive amidation of aldehydes,[6] we have undertaken an
investigation into the catalytic synthesis of N-alkylamides,
using acetamide and 1-pentene as example substrates,
through a rhodium-catalyzed hydroamidomethylation reaction
(Scheme 2).
Figure 1. Products of higher molecular mass (5) observed in the hydroami-
domethylation of 1-pentene with acetamide. Corresponding branched prod-
ucts are not shown.
distribution was determined by GC analysis. The conversion
calculated for the reactions was based on the amount of 1-
pentene found after the reaction. The amounts of linear and
branched products 3L and 3B, unsaturated products 2L and
2B, hydroformylation products 1L and 1B, and alcohols 4L
and 4B were determined by using calibration lines.
In Tables 1–4, the selectivity for all products and intermediates
is reported, as well as the linearity (%) of the aldehyde and N-
hexylacetamide (3L). It is worth noting that the individual
amounts of both types of unsaturated compounds 2 are gen-
erally small and present in an approximate 1:1 ratio; for clarity
reasons, we have only indicated the total amount of com-
pounds 2 in the product composition. The remainder consists
of the higher mass products 5, which were not individually
quantified by GC, but rather lumped together and calculated
from the mass balance of 1-pentene. Analytical product com-
position data of all experiments are given in the Supporting In-
formation.
Scheme 2. Hydroamidomethylation of 1-pentene with acetamide to form N-
hexylacetamide.
Whereas we previously reported an efficient three-compo-
nent homogeneous catalyst system consisting of rhodium/
xantphos/p-toluenesulfonic acid (HOTs) for the selective reduc-
tive amidation of aldehydes under a pure H2 atmosphere,[6] in
the present study we aim to develop hydroamidomethylation
catalyst systems that are not only effective in the hydroformy-
lation of the alkenes, but also efficient in the reductive amida-
tion of aldehydes formed in situ under a CO-containing atmo-
sphere.
Results
General considerations
The products found after a typical hydroamidomethylation re-
action of 1-pentene with acetamide are shown in Scheme 3
(full experimental details are given in the Supporting Informa-
tion). Apart from the desired product, N-1-hexylacetamide (3L),
depending on the efficiency and selectivity of the catalytic
system, the reaction mixture may consist of the isomeric unsa-
turated intermediate compounds N-(1-hexylidene)acetamide
and/or N-1-hexenylacetamide (2L), hexanal (1L), and hexanol
(4L), as well as all corresponding branched products (1B, 2B,
3B, and 4B). Additionally, various products with higher mass 5
(Figure 1), comprising self- or cross-aldol condensation prod-
ucts and the disubstituted product with two molecules of acet-
amide, may be formed.
Catalytic hydroamidomethylation of 1-pentene with acet-
amide: Initial screening studies
Of the various rhodium precursors [{RhCl(cod)}2] (cod=1,5-cy-
clooctadiene),
[Rh(acac)(CO)2]
(acac=acetylacetone),
[Rh(cod)2]BF4, [Rh(CO)(H)(PPh3)3], and [Rh(cod)2]OTf in combina-
tion with the 9,9-dimethyl-4,5-bis(diphenylphosphanyl)xan-
thene (xantphos) ligand, [Rh(acac)(CO)2] gave the highest se-
lectivity for 1L, with only low amounts of condensation prod-
ucts 5 (Table S1 in the Supporting Information). HOTs was in-
troduced as a cocatalyst to promote the coupling reaction of
1L with acetamide;[6] however, next to a higher yield of 3L,
this resulted in strongly increased formation of products 5
(Table 1, entries 1 and 2). Although the use of trifluorometha-
nesulfonic acid (HOTf) resulted
The reactions were performed starting with 5 mmol of 1-
pentene and an equivalent amount of acetamide; the product
in higher selectivity for 3L, large
amounts of higher molecular
weight products 5 were still pro-
duced (Table 1, entry 3). Increas-
ing the amount of HOTs also led
to some increased selectivity in
favor of 3L, but also resulted in
high amounts of aldol condensa-
tion and di-coupled products
(Table 1, entry 4). Finally, carrying
Scheme 3. Hydroamidomethylation of 1-pentene with acetamide: observed intermediates, products, and unde-
sired byproducts.
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemSusChem 2013, 6, 1759 – 1773 1760

CHEMSUSCHEM
FULL PAPERS
www.chemsuschem.org
HOTs) resulted in low acetal for-
mation, but still only gave about
15% selectivity for the desired
product 3 (Table 1, entry 15). Re-
markably, a similar selectivity to
4L (ꢀ15%) was observed with
the latter solvent combination,
both in the absence and pres-
ence of HOTs (Table 1, entries 14
and 15).
Table 1. Effect of the solvent system on the hydroamidomethylation of 1-pentene with acetamide.[a]
Entry Solvent
(10 mL)
t
[h]
Additive
(0.05 mmol) [%]
Conversion
Selectivity [%][c]
Linearity [%]
1
It is worth noting that the
overall regioselectivity for the
linear products derived from 1-
pentene (1L, 3L, and 4L) in
nearly all cases is above 90%;
this is in agreement with the in-
trinsically high linearity for hy-
droformylation of terminal al-
kenes with rhodium/xantphos
catalysts reported in the litera-
ture.[7]
1
2
3
4
acetal
5
3
1
2
3
4
5
6
7
8
diglyme
diglyme
diglyme
diglyme
diglyme
diglyme[b]
MeOH
MeOH
1:1 MeOH/toluene
1:1 MeOH/diglyme
1:1 EtOH/diglyme
1:1 iPrOH/diglyme
1:1 TFE/diglyme
1:1 HORF/diglyme
1:1 HORF/diglyme
4
4
4
4
8
–
79
84
89
89
88
88
89
91
89
90
82
85
90
88
85
89
6
0
0
1
3
1
1
3
0
1
0
0
0
2
8
–
–
–
–
–
5
97
–
HOTs
HOTf
HOTs (0.1)
HOTs
36 12 13
26 11 19
24 12 23
31
25
81
26
35 19
31 17
45 22
43 20
47 12
38 97
41 92
40 94
40 96
49 86
16 96
>99
>99
>99
>99
>99
–
>99
>99
>99
–
9
8
3
9
19
15
0
1
2
1
0
7
1
8+4 HOTs
–
0
8
8
8
8
8
8
8
8
8
–
HOTs
HOTs
HOTs
HOTs
HOTs
HOTs
–
56
34
40
20
16
12
0
7
9
96
92
9
10
11
12
13
14
15
11 93
13 93
12 92
21 93
8
6
>99
>99
>99
99
71
52
3
6
1
17
96
96
HORF as a cosolvent
HOTs
14 16
6
[a] Reaction conditions: 0.01 mmol [Rh(acac)(CO)2], 0.02 mmol xantphos, 5 mmol 1-pentene, 5 mmol acetamide
Intrigued by the significantly de-
viating selectivity to alcohols in
the experiment with a diglyme/
HORF solvent mixture, we further
examined the effects of HORF on
the reaction characteristics. Thus,
we investigated the effects of an
increasing HORF/diglyme (v/v)
(Rh/xantphos/1-pentene/acetamide=1:2:500:500); PCO/H =50(1:2) bar; T=1008C; t=4–12 h; solvent: 10 mL
2
bis(2-methoxyethyl)ether (diglyme) or solvent/diglyme (v/v). [b] One-pot sequential reaction conditions; the re-
actor was charged with substrates in a similar fashion from the beginning of the reaction and then: 8 h at
1008C hydroformylation under syngas PCO/H =50 (1:2) followed by depressurizing syngas- 3 pressurizing-de-
2
pressurizing cycles with H2- pressurizing with H2-4 h, PH =80 bar at 808C reductive amidation. [c] The selectivity
2
was determined by GC analysis with decane as an internal standard.
out the hydroamidomethylation reaction sequentially by first
applying a syngas pressure followed by exposure to pure H2
did not improve the yield of 3L (Table 1, entry 6); this reaction
sequence also resulted in the formation of large amounts of 5.
A screening of various reaction solvents demonstrated that
the product selectivity was
ratio (at a constant total reaction volume) on the product com-
position. The results are depicted in Figure 2 (see also Ta-
bles S3 and S4 and Figures S2 and S3 in the Supporting Infor-
mation).
highly dependent on solvent
(Table 1, entries 7–15). The use
of alcoholic solvents such as
methanol, ethanol, isopropanol,
n-butanol, isobutanol, and tri-
fluoroethanol (TFE), as well as
their combination with diglyme
or toluene and with an acid co-
catalyst, resulted mainly in acid-
catalyzed acetal formation of 1L
with very low selectivity toward
the desired product 3 (Table 1,
entries 7–13; see also Table S2 in
Figure 2. The effect of the ratio of HORF/diglyme on the product composition in the hydroamidomethylation of 1-
pentene with acetamide catalyzed by the Rh/xantphos/HOTs system. Reaction conditions: 0.01 mmol [Rh-
(acac)(CO)2], 0.02 mmol xantphos, 0.05 mmol HOTs, 5 mmol 1-pentene, 5 mmol acetamide (Rh/xantphos/HOTs/1-
the Supporting Information).
The use of the more acidic sol-
vent
1,1,1,3,3,3-hexafluoroiso-
pentene/acetamide=1:2:5:500:500); PCO/H =50(1:2) bar; T=1008C; t=8 h; constant volume of reaction solvent:
propyl alcohol (HORF, pKa =9.3)
in combination with diglyme
(1:1; even in the presence of
2
&
10 mL; decane was used as an internal standard. =1-pentene, =aldehyde (1), =unsaturated intermediate
&
(2), =desired product (3), =alcohol (4), =other products (5); for more detailed information, see Table S3
and Figure S2 in the Supporting Information.
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemSusChem 2013, 6, 1759 – 1773 1761

CHEMSUSCHEM
FULL PAPERS
www.chemsuschem.org
Gradually increasing the ratio of HORF/diglyme from 0:1
(pure diglyme) to 1:4 (v/v), did not considerably change the
conversion of 1-pentene; the selectivity for the desired product
3, however, increased spectacularly by almost a factor of five
over this range of HORF concentration. Increasing the HORF
concentration further to about 50% (HORF/diglyme=1:1 v/v)
surprisingly resulted in a drop of the selectivity to 3 with an in-
crease in selectivity to aldehyde 1. Moving to a HORF-richer
regime of 5:1 resulted in a high selectivity (>80%) to alcohol
4, which was clearly at the cost of the aldehydes. The use of
pure HORF resulted in a decrease in the conversion of 1-pen-
tene with a steep decrease in selectivity to alcohol 4; the main
products were aldehyde 1 and condensation products 5. Thus,
it appears that the 1:4 ratio of HORF/diglyme (ꢀ20 mmol of
HORF) gives the highest selectivity to the desired product 3.
As shown in Table S3 in the Supporting Information, acetal for-
mation of 1L with HORF appeared to be undetectable, when
the applied HORF/diglyme ratio was smaller than 1:2.
Figure 3. Selected ligands used in the hydroamidomethylation of 1-pentene
with acetamide.
resulted in nearly full conversion of 1-pentene after 4 h; how-
ever, again the linearity of the aldehyde (ꢀ50%) and the selec-
tivity to the desired product was very low (ꢀ3%). Remarkably,
the use of DPEphos (Table 2, entry 3), which is considered to
be a more flexible analogue of xantphos, gave a low-perfor-
mance catalyst with low chemoselectivity to 3 (ꢀ35%). The
high regioselectivity for linear product 3L (97%) combined
with a low regioselectivity of aldehyde clearly indicates that
the subsequent reductive amidation of the linear aldehyde
occurs preferentially over its branched isomer.
Optimization and scope of the hydroamidomethylation re-
action in HORF/diglyme (1:4 v/v)
Effect of temperature and ligands
It appeared that the highest selectivity (ꢀ70%) for product 3
was obtained at a reaction temperature of 1008C (Figure S3
and Table S7 in the Supporting Information). At 1208C, the se-
lectivity for hydroamidomethylation decreased, particularly due
to the formation of higher amounts of higher-mass products 5
(Figure S4 and Table S7 in the Supporting Information). Upon
lowering the reaction temperature to 60–808C, the conversion
of 1-pentene dropped with the build up of a significant con-
centration of unsaturated intermediates 2; this indicated a pro-
gressively lower rate of reduc-
Among the various different xantphos-type ligands, the best
results were obtained with xantphos (Table 2, entry 5), Si-xant-
phos (Table 2, entry 7), and tBu-substituted xantphos (Table 2,
entry 6) with selectivity to 3 of 53–59% (at 4 h reaction time)
tion of intermediates 2 relative
to their rate of formation.
Table 2. Effect of various ligands on the hydroamidomethylation of 1-pentene with acetamide.[a]
A selection of ligands with dif-
ferent stereoelectronic proper-
ties as potential substitutes for
xantphos were also tested under
the same optimal reaction condi-
tions. An overview of the ligands
is shown in Figure 3. The catalyt-
ic results obtained with in situ
generated rhodium/ligand/HOTs
catalysts are summarized in
Table 2.
Entry
Ligand
Conversion
[%]
Selectivity [%][b]
Linearity [%]
1
2
3
4
5
1
3
1
2
3
4
5
6
7
8
none
PPh3
DPEphos
triPhos
99
96
70
10
83
80
77
11
82
46
28
97
46
42
37
67
27
26
31
71
52
44
17
50
23
20
13
0
10
3
34
0
1
0
3
0
4
3
5
0
0
3
4
1
21
36
14
33
4
4
4
17
19
5
30
51
67
70
95
96
72
58
53
97
70
28
71
>99
97
–
xantphos
9
8
7
8
57
59
53
3
>99
>99
99
>99
>99
98
tBu-xantphos
Si-xantphos
oMeO-xantphos
benzoxantphos
DBFphos
9
13
25
12
24
16
23
61
9
10
11
12
homoxantphos
xantphos(tBu)2
6
16
95
82
In the absence of any phos-
phane ligand (Table 2, entry 1)
full conversion of 1-pentene was
observed, but with a low chemo-
selectivity (ꢀ10%) to 3L. Most
striking is the low regioselectivi-
ty (ꢀ70%) for 3L, and only 30%
for aldehyde product 1L. The
use of monodentate ligands,
such as PPh3 (Table 2, entry 2),
[a] Reaction conditions: 0.01 mmol [Rh(acac)(CO)2], 0.02 (bidentate) or 0.04 mmol (monodentate) ligand,
0.05 mmol HOTs, 5 mmol 1-pentene, 5 mmol acetamide (Rh/L/HOTs/1-pentene/acetamide=1:2 (bidentate)
or
4
(monodentate):5:500:500); T=1008C; t=4 h; PCO/H =50 bar (1:2); solvent: 10 mL HORF/diglyme
2
(1:4). DPEphos=bis(2-diphenylphosphinophenyl)ether, triPhos=bis(2-phenylphosphinoethyl)phenylphosphane,
tBu-xantphos=2,7-di-tert-butyl-9,9-dimethyl-4,5-bis(diphenylphosphanyl)xanthene, Si-xantphos=4,6-bis(diphenyl-
phosphanyl)-10,10-dimethyl-10H-dibenzo[b,e][1,4]oxasilane,
oxyphenylphosphanyl)xanthene, benzoxantphos=6,8-bis(diphenylphosphino)benzo[k,l]-xanthene, DBFphos=
1,1’-(4,6-dibenzofurandiyl)bis(1,1-diphenylphosphane), homoxantphos=4,6-bis(diphenylphosphino)-10,11-di-
oMeO-xantphos=9,9-dimethyl-4,5-bis(di-ortho-meth-
hydrodibenzo[b,f]oxepine, xantphos(tBu)2 =9,9-dimethyl-4,5-bis(di-tert-butylphosphino)xanthene. [b] The selec-
tivity was determined by GC analysis with decane as an internal standard.
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemSusChem 2013, 6, 1759 – 1773 1762

CHEMSUSCHEM
FULL PAPERS
www.chemsuschem.org
and a combined selectivity, including potential pre-
cursors 1 and 2, of up to 90–95%. In particular, cata-
lytic systems with these ligands show a relatively low
accumulation of unsaturated intermediates 2, thus
showing the importance of the rigid PÀOÀP back-
bone for selective hydrogenation.[6] The use of oMeO-
xantphos (Table 2, entry 8) gave a low-activity catalyst
with low chemoselectivity to 3 and surprisingly low
regioselectivity (ꢀ60%) to the aldehyde. Apparently,
the bulky oMeO-phenyl substituents not only prevent
binding of the aldehydes and intermediates 2, but
also that of 1-pentene.
Figure 4. The effect of the syngas ratio on the hydroamidomethylation of 1-pentene
Remarkably, low selectivity (ꢀ15%) for 3 is also with acetamide catalyzed by the Rh/xantphos/HOTs system: A) effect of H2 and B) effect
of CO. Reaction conditions: 0.01 mmol [Rh(acac)(CO)2], 0.02 mmol xantphos, 0.05 mmol
obtained with benzoxantphos (Table 2, entry 9). This
HOTs, 5 mmol 1-pentene, 5 mmol acetamide (Rh/xantphos/HOTs/1-pentene/acet-
ligand gave high alkene conversion, but low regiose-
amide=1:2:5:500:500); T=1008C; t=4 h; solvent: 10 mL HORf/diglyme (1:4 v/v); decane
lectivity (50%) for linear aldehyde 1L, and thus, ex-
~
*
as an internal standard. =1-pentene, =aldehyde (1), *=unsaturated intermediates
&
^
(2), =desired product (3), =alcohol (4).
emplifies that low regioselectivity for hydroformyla-
tion can be accompanied by low selectivity for reduc-
tive amidation of the aldehydes. This is also reflected
by a significant accumulation (ꢀ15%) of unsaturated
intermediates 2, which eventually may result in increased
amounts of 5 (Table S11 in the Supporting Information). It is
surprising that the presence of a naphthyl fragment in this
ligand can have such a decisive stereoelectronic influence at
the rhodium center. A similar observation was made with
DBFphos (Table 2, entry 10) as the ligand with a build up of in-
termediate 2 of up to 25%; this is indicative of a low hydroge-
nation activity for these intermediates, even though heavy-end
formation remained low. This latter observation suggests that
not only the Brønsted acid component of the catalyst system
is responsible for the formation of aldol-type higher mass
products, but that the metal complex can also play an activat-
ing role for the formation of these products.
CO) from 10 to 40 bar (Figure 4A) resulted in an increased con-
version of 1-pentene as well as a steep increase in selectivity
to 3, with only a slight increase in the yield of 1 and 4. The
use of pressures higher than 40 bar of H2 did not further im-
prove the reaction efficiency to 3. The use of CO pressures in
excess of a hydrogen pressure of 20 bar resulted in a significant
drop in the conversion of 1-pentene and yields of both 3 and
1 (Figure 4B).
Alternative promoter compounds as a substitute for HORF
The finding that HORF acted as a promoter for hydroamidome-
thylation catalysis prompted us to search for other compounds
with similar or better performance. We hypothesized that the
acidic properties of HORF, perhaps in addition to polarity, was
a decisive parameter for its effect on catalysis. Thus, a selected
number of compounds characterized by acid strength were
screened in diglyme in the molar quantity that proved opti-
mally effective with HORF. The results of this screening study
are summarized in Table 3 (see also Table S13 in the Support-
ing Information). Some phenol-type compounds, such as
phenol, 2-fluorophenol, 3-fluorophenol, 4-fluorophenol, 2-
chlorophenol, and 4-chlorophenol, show similar catalyst pro-
moter performance to that of HORF (see Table 3, entries 2, 4, 6,
7, 8, 9, and 11). In particular, with 3-fluorophenol, 4-fluorophe-
nol, and 4-chlorophenol (Table 3, entries 7, 6, and 9, respective-
ly), the reactions seem to show comparable product composi-
tions as those observed with HORF, with low accumulation of 2
and low heavy-ends 5. All compounds that induce selectivity
for 3 above 50% have a pKa value between ꢀ8.5 and 10,
whereas the probed compounds with pKa >10 generally have
poor promoter performance. It is however clear from the gen-
eral features displayed in Table 3 that acid strength as a sole
criterion for promoter performance is not well validated. Un-
fortunately, insufficient data on polarity/dielectric constant/
dipole moment are available to identify a correlation with pro-
moter performance and a more complete survey needs to be
It is interesting to note that the highest selectivity for the
formation of 3L (>60%) was obtained with homoxantphos as
a ligand (Table 2, entry 11), albeit at a relatively low overall ac-
tivity. This catalyst combines moderate hydrogenation activity
of the intermediates 2 with low heavy-end formation.
Finally, the results with xantphos(tBu)2 (Table 2, entry 12) are
remarkable: high activity for hydroformylation was observed,
with almost full 1-pentene conversion, but, at the same time,
this ligand afforded a catalyst with almost the worst perfor-
mance for hydroamidomethylation with a selectivity for 3 of
less than 10% and highest accumulation (ꢀ25%) of the inter-
mediates 2; this indicates very low activity for the hydrogena-
tion of these intermediates. The very similar product distribu-
tion to that obtained when no ligand is applied (Table 2,
entry 1) suggests that this ligand fails to form stable complexes
with rhodium.
Effect of H2/CO ratio on the hydroamidomethylation of 1-
pentene
In the next set of experiments, the effect of the H2/CO ratio on
the hydroamidomethylation reaction was investigated
(Figure 4 and Table S5 in the Supporting Information). Increas-
ing the pressure of H2 gas (with a constant pressure of 20 bar
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemSusChem 2013, 6, 1759 – 1773 1763

CHEMSUSCHEM
FULL PAPERS
www.chemsuschem.org
sired product 3* (Table 4,
entry 4) due to the accumulation
of more stable unsaturated
linear intermediate 2*. Whereas
hydroformylation produces sig-
Table 3. Effect of various additives in combination with diglyme on the hydroamidomethylation of 1-pentene
with acetamide.[a]
Entry
Cosolvent
(20 mmol)
Dielectric
constant
pKa
Conversion
[%]
Selectivity [%][b]
Linearity [%]
1
2
3
4
5
1
3
1
2
3
4
5
6
7
8
–
–
16.8
27.7
2.95–3.1
–
–
–
–
11.2
6.26
7.4
–
–
–
–
–
–
9.2
88
90
31
19
16
23
9
19
68
34
46
1
5
2
4
4
4
5
2
6
0
3
2
2
2
0
2
1
2
1
1
1
40
5
42
19
25
9
4
32
4
16
21
55
46
60
54
52
56
21
42
7
96
93
98
88
59
92
91
93
90
56
92
>99
90
94
98
96
98
95
95
97
99
>99
>99
>99
>99
91
nificant
quantities
of
the
HORF
2
6
8
branched aldehyde, this alde-
hyde is more reluctant to under-
go reductive amidation. The use
of amides with more electron-
donating carbonyl groups, such
as propanamide, 2-methylpropa-
namide and pentanamide, result-
ed in higher selectivity to the N-
hexylamides (Table 4, entries 1, 5,
6, and 8). Applying an amide
with a bulkier carbonyl group,
such as 2,2-dimethylpropana-
mide (pivalamide), resulted in
only slightly lower selectivity to
the desired product (Table 4,
entry 7). Application of an amide
with an electron-withdrawing
TFE
PhOH
12.37 89
9.99 82
8.68 95
9.89 86
9.29 90
8.73 88
4-CF3-PhOH
4-F-PhOH
3-F-PhOH
2-F-PhOH
4-Cl-PhOH
3-Cl-PhOH
2-Cl-PhOH
4-Br-PhOH
2,6-di-Me-PhOH
2,6-di-iPr-PhOH
2,6-di-tBu-PhOH
2,6-di-Me-PhCOOH
AcOH
34 15 22
25
22
19
21
3
3
4
3
59
66
42
67
8
52
9
30
9
>99
>99
>99
>99
>99
>99
>99
>99
>99
>99
>99
>99
>99
>99
–
9
–
91
10
11
12
13
14
15
16
17
18
19
20
21
9.12 90
8.56 88
9.37 84
10.36 88
11.1
13.6
3.25 88
4.76 89
13.4
61 15
20
23
17
22
23
34
10
3
11
4
7
9
89
86
14
7
25
5
9
6.2–6.5
–
–
80
4.95
di-Ph-CHOH
di-iPr-CHOH
H2O
87
89
89
41 14 22
32 10 15
88
16
ꢀ14.5
15.7
3
8
0
12
2-naphthol
9.63 85
62
>99
[a] Reaction conditions: 0.01 mmol [Rh(acac)(CO)2], 0.02 mmol xantphos, 0.05 mmol HOTs, 5 mmol 1-pentene,
5 mmol acetamide (Rh/xantphos/HOTs/1-pentene/acetamide=1:2:5:500:500); T=1008C; t=8 h; PCO/H =50 bar
(1:2); solvent: 10 mL diglyme; 20 mmol of cosolvent. [b] The selectivity was determined by GC analysis with
decane as an internal standard.
fluoromethylcarbonyl
group
2
caused a drastic drop in the se-
lectivity to N-hexylfluoroaceta-
mide (Table 4, entry 9); the main
product of this reaction was 1L.
accomplished to shed more light on this matter. This and pos-
sible further optimization studies on some promising com-
pounds are subjects of further study.
The use of benzamides resulted in lower selectivity to 3* com-
pared with acetamide. Applying a benzamide with an electron-
donating p-methoxy substituent again gave significantly
higher selectivity to 3*, whereas p-trifluoromethylbenzamide
with an electron-withdrawing group again resulted in lower se-
lectivity. These results show a correlation between the efficien-
cy of hydroamidomethylation and the nucleophilic character of
Hydroamidomethylation product development with time
Figure 5 shows the product-development curves with time for
the rhodium/xantphos/HOTs catalytic system in HORF/diglyme
(1:4). These curves were constructed from reaction product
compositions of separate identical experiments carried out
during the respective residence times. The product composi-
tion development clearly shows that aldehydes and unsaturat-
ed compounds 2 exist as reaction intermediates to ultimately
form 3L as well as the other final products, 4L and heavy
ends. Clearly, it can be deduced from Figure 5 that the rate-
and selectivity-determining step in the overall hydroamidome-
thylation reaction with the present catalytic system involves
the reductive amidation of aldehydes.
Scope of the hydroamidomethylation reaction
Various olefins and a number of different amides were used as
substrates in the hydroamidomethylation reaction under the
optimized reaction conditions (Table 4 and Table S12 in the
Supporting Information). The use of 1-hexene or 1-octene in
the reaction with acetamide gave the desired N-alkylamide
with similar conversion, selectivity, and linearity as that for 1-
pentene (Table 4, entries 1–3). The use of styrene resulted in
high conversion, however, with decreased selectivity to the de-
Figure 5. Product development with time for the hydroamidomethylation of
1-pentene with acetamide. Reaction conditions: 0.01 mmol [Rh(acac)(CO)2],
0.02 mmol xantphos, 0.05 mmol HOTs, 5 mmol 1-pentene, 5 mmol acet-
amide (Rh/xantphos/HOTs/1-pentene/acetamide=1:2:5:500:500); T=1008C;
PCO/H =50 bar (1:2); solvent: 10 mL (v/v) HORF/diglyme (1:4); decane as an in-
2
~
&
*
^
ternal standard. =1-pentene, =1L, =3L, =4L, *=other products
(5). For clarity only the linear products are depicted (details are given in Fig-
ure S8 and Table S10 in the Supporting Information).
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemSusChem 2013, 6, 1759 – 1773 1764

CHEMSUSCHEM
FULL PAPERS
www.chemsuschem.org
glyme mixture was generally
lower than that in pure diglyme.
The lower conversion was partic-
ularly distinct when HOTs was
present as a catalyst component
[HOTs/Rh=5 (mol/mol)], giving
about 90% conversion in pure
diglyme, but only 30% in the
HORF/diglyme 1:4 (v/v) mixture.
Table 4. Hydroamidomethylation reaction with different amides and alkenes.[a]
Remarkably,
hydroformylation
Entry
Alkene
(5 mmol)
Amide
(5 mmol)
Conversion
[%]
Selectivity [%][b]
Linearity [%]
1*
2*
3*
4*
5*
1*
3*
with HOTs as a catalyst compo-
nent resulted in both solvent
systems in detectably higher
amounts of 4L as well as heavy-
end byproducts. To simulate the
presence of the weakly basic
amide under actual hydroamido-
methylation conditions, hydro-
formylation was also carried out
in the presence of DMA (in an
equimolar amount to 1-pen-
tene). For the pure diglyme sol-
vent system this resulted in
lower conversion (from 70 to
55%), whereas hardly any effect
of DMA was discernible for the
HORF/diglyme solvent system;
the weak basicity of DMA seems
1
2
3
4
5
6
7
8
1-pentene
1-hexene
1-octene
acetamide
acetamide
acetamide
acetamide
propanamide
2-Me-propanamide
pivalamide
valeramide
F-acetamide
benzamide
91
89
88
86
90
89
87
93
90
69
90
57
17
20
24
20
14
18
25
10
77
38
38
29
4
3
2
36
5
4
6
2
12
8
3
69
67
62
33
74
71
61
83
4
6
6
6
4
3
3
3
4
4
0
1
0
4
4
7
8
4
4
5
1
3
92
90
93
51
96
88
91
90
96
99
99
90
>99
>99
>99
>99
>99
>99
>99
>99
>99
>99
>99
>99
styrene
1-pentene
1-pentene
1-pentene
1-pentene
1-pentene
1-pentene
1-pentene
1-pentene
9
10
11
12
24
48
21
30
10
18
p-MeO-benzamide
p-CF3-benzamide
31
[a] Reaction conditions: 0.01 mmol [Rh(acac)(CO)2], 0.02 mmol xantphos, 0.05 mmol HOTs, 5 mmol alkene,
5 mmol amide (Rh/xantphos/HOTs/alkene/amide=1:2:5:500:500); T=1008C; t=8 h; PCO/H =50 bar (1:2); sol-
2
vent: 10 mL (v/v) HORF/diglyme (1:4). [b] The selectivity was determined by GC analysis with decane as an inter-
nal standard.
the N atom of the amide; the same correlation was observed
earlier with the reductive amidation of aldehydes with pure H2
and is a clear indication of the tandem character of hydrofor-
mylation–reductive amidation in the hydroamidomethylation
reaction.[6]
to be neutralized by the excess (by about factor 4) of HORF.
When both DMA and HOTs [DMA/HOTs =100 (mol/mol)] were
applied in the catalytic reaction mixture, the catalyst activation
effect of HOTs in pure diglyme appeared to be virtually unaf-
fected, albeit with reduced alcohol formation. In contrast, the
activity-reducing effect in the HORF/diglyme mixture was virtu-
ally absent; this suggested a neutralizing effect between DMA
Influence of HORF in the separate steps: Hydroformylation
and reductive amidation
Because we observed that the
reductive amidation of 1L in
pure diglyme was severely inhib-
ited by the presence of CO,[6] the
separate reaction steps, hydro-
formylation and reductive ami-
dation with the rhodium/xant-
phos catalyst system, were stud-
ied in more detail.
Hydroformylation of 1-pentene
The hydroformylation of 1-pen-
tene with the rhodium/xantphos
catalytic system in pure diglyme,
as well as in a mixture of HORF/ Figure 6. The effect of acid and base on the hydroformylation of 1-pentene catalyzed by the Rh/xantphos catalytic
system: A) in diglyme and B) in HORF/diglyme 1:4 (v/v). Reaction conditions: 0.01 mmol [Rh(acac)(CO)2], 0.02 mmol
diglyme (1:4 v/v), was studied.
xantphos, 0.05 mmol HOTs (if acid was used), 5 mmol 1-pentene, 5 mmol N,N-dimethylacetamide (DMA; if DMA
The results given in Figure 6A
was used); Rh/xantphos/1-pentene=1:2:500; PCO/H =50 bar (1:2); t=2 h; solvent: 10 mL diglyme or 10 mL HORF/
2
and B show that the conversion
&
&
diglyme=1:4 (v/v); decane as the internal standard. =1-pentene, =aldehyde (1), =alcohol (4), =heavy-
of 1-pentene in the HORF/di- end products (5).
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemSusChem 2013, 6, 1759 – 1773 1765

CHEMSUSCHEM
FULL PAPERS
www.chemsuschem.org
and HOTs in HORF. Small amounts of 4L were still formed in
both cases (selectivity ꢀ3 and ꢀ7% for, pure diglyme and
HORF/diglyme mixed solvent systems, respectively (Table S8-
1 and Figure S5 in the Supporting Information).
The effect of different amounts of HORF on the hydroformy-
lation reaction in the presence of HOTs was also investigated.
The amounts of DMA and HOTs were kept constant in the
series of experiments depicted in Figure 7A to simulate the
presence of amide under hydroamidomethylation conditions.
a high alkene conversion of about 90%. When HOTs was pres-
ent, both the alkene conversion and selectivity to 4L de-
creased dramatically. However, in the presence of DMA [DMA/
HOTs =100 (mol/mol)], HOTs seemed to be at least partly neu-
tralized and restoration of the hydroformylation activity as well
as selectivity to 4L became apparent (Figure 7B and Table S8-2
in the Supporting Information). Again, this observation is con-
sistent with the observations depicted in Figure 2 at the same
HORF/diglyme ratio under real hydroamidomethylation condi-
tions, thus giving confidence
that the addition of DMA (in
amounts equimolar to 1-pen-
tene) in the individual reaction
steps truly approaches hydrofor-
mylation conditions that prevail
in hydroamidomethylation reac-
tions.
Reductive amidation of hexanal
The effects of the presence of
HORF as a cosolvent in rhodium/
xantphos/HOTs-catalyzed reduc-
tive amidation of 1L with acet-
amide are shown in Figure 8.
When the reductive amidation of
Figure 7. The effect of HORF in the absence and presence of acid on the hydroformylation of 1-pentene catalyzed
by the Rh/xantphos system: A) different ratio (v/v) of HORF/diglyme in the presence of HOTs and DMA, and B) in
HORF/diglyme 5:1 (v/v). Reaction conditions: 0.01 mmol [Rh(acac)(CO)2], 0.02 mmol xantphos, 5 mmol 1-pentene;
1L with acetamide was carried
out in diglyme, the selectivity to
3L dropped from about 40% in
pure H2 to only about 10% with
the addition of only 5 bar of CO
(45 bar H2). At higher CO pres-
Rh/xantphos/1-pentene=1:2:500; 0.05 mmol HOTs (if acid was used), 5 mmol DMA (if DMA was used); PCO/
F
&
=50 bar (1:2); t=8 h; solvent: 10 mL HOR /diglyme(v/v); decane as an internal standard. =1-pentene, =al-
H2
&
dehyde (1), =alcohol (4), =heavy-end products (5).
Increasing ratios of HORF/diglyme (v/v) in this series of experi-
sure, this selectivity further decreased to below 10%, while an
increased amount (>15%) of unsaturated 2 became visible
(Figure 8A).
ments revealed that, the higher the HORF/diglyme ratio, the
higher selectivity to 4L observed (Figure 7A). However, the
catalytic activity dropped gradually; the 1-pentene conversion
fell from >95% in pure diglyme to about 50% in pure HORF.
Remarkably, when applying pure
Amazingly, a strongly positive effect of CO on the selectivity
of the reductive amidation of 1L was revealed when a HORF/
HORF as a solvent, the selectivity
to 4L dropped again to about
10% from 55% at 80 vol% of
HORF. This observation is consis-
tent with those observations
under real hydroamidomethyla-
tion conditions depicted in
Figure 2, which show a quantita-
tively very similar effect at high
HORF/diglyme ratios.
In the experiments shown in
Figure 7B, a constant HORF/di-
glyme ratio of 5:1 (v/v) was ap-
plied, both in the absence and
presence of HOTs and DMA. In
this HORF-rich reaction medium
in the absence of HOTs, the se-
lectivity to 4L was boosted dra-
Figure 8. The effect of CO/H2 partial pressures on the reductive amidation of 1L with acetamide catalyzed by the
catalytic system Rh/xantphos/HOTs: A) in pure diglyme and B) in HORF/diglyme 1:4 (v/v). Reaction conditions:
0.01 mmol [Rh(acac)(CO)2], 0.02 mmol xantphos, 5 mmol 1L, 5 mmol acetamide; Rh/xantphos/HOTs/1L/acet-
amide=1:2:5:500:500; P=50 bar; t=2 h; solvent: 10 mL diglyme or 10 mL (v/v) HORF/diglyme=1:4; decane as
&
matically to 90%, while reaching an internal standard. =1L, =unsaturated imide or enamide (2L), =3L, =4L, =other products (5).
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemSusChem 2013, 6, 1759 – 1773 1766

CHEMSUSCHEM
FULL PAPERS
www.chemsuschem.org
diglyme [1:4(v/v)] solvent mix-
ture was applied (Figure 8B).[8]
Application of syngas (CO/H2 =
1) even led to an increase in the
selectivity to 3 by an order of
magnitude from less than 5% at
50 bar of pure H2 to 50% at
50 bar of syngas. The yield of 3
(and to a lesser extent also of
4L) increased gradually with in-
creasing CO pressure, whereas
Scheme 4. Catalytic hydroamidomethylation of 1-pentene: a one-pot synthesis of 3L through the formation of 1L
and intermediate 2L; 4L and heavy-ends 5 are undesired byproducts (for clarity only the linear products are de-
picted). R in 5a can be H (CHO) or OH (COOH).
the amount of intermediate product 2 and heavy-end products
5 simultaneously decreased, thus indicating not only an in-
crease in the general hydrogenation activity of the catalyst,
but also indicating an increase in substrate specificity for the
hydrogenation of intermediates 2 over 1L.
mediates results in the formation of the desired product 3 in
the third step. The formation of 4 by the direct hydrogenation
of 1L constitutes a competing reaction of 1L, resulting in
lower selectivity for 3.
The unsaturated intermediates (1 as well as 2) may undergo
various side reactions, most pronounced of which are aldol
condensation and double addition reactions, resulting in the
higher mass products 5 that are indeed observed in varying
amounts. Other reported reactions of intermediates 2 are hy-
droformylation and hydrocarbonylation reactions, which are
also shown in Scheme 4;[9] however, products 5a (aldehyde or
carboxylic acid) were not observed in our reactions (see Figur-
es S1 and S23 in the Supporting Information).
It is worth noting that the use of different amounts of HORF
for the rhodium/xantphos/HOTs-catalyzed reductive amidation
of 1L with acetamide in a pure hydrogen atmosphere showed
that HORF strongly retarded reductive amidation by inhibiting
the hydrogenation of unsaturated 2, which was shown to ac-
cumulate in the reaction mixture (Figure S6 in the Supporting
Information). The presence of progressively more HORF result-
ed in lower conversion of 1L (Figure S6 and Table S9 in the
Supporting Information). In the absence of HOTs, the rhodium/
xantphos catalyst system in pure HORF led to almost exclusive
hydrogenation of 1L (Figure S7 in the Supporting Information),
a result similar to that in diglyme.[6] However, when HOTs was
applied in the rhodium/xantphos catalytic system, a significant
difference between the role of the solvents diglyme and HORF
again became apparent. Whereas in diglyme both the reduc-
tive amidation product 3 and 4L were coproduced, in HORF
hardly any 4L or product 3 was formed, thus revealing that
the catalytic system in HORF under a pure H2 atmosphere had
very poor hydrogenation activity (Figure S7 and Table S9 in the
Supporting Information). All observations reported in this sec-
tion thus suggest that different rhodium/xantphos complexes
are operative as imide/enamide or aldehyde hydrogenation
catalysts under different conditions, for example, in different
solvent systems; with different reducing gas environments;
and in the presence or absence of potential catalyst promoter
compounds, such as HOTs and HORF.
Finding active and selective alkene hydroamidomethylation
catalysts will thus involve developing catalyst systems that are
not only active for the hydroformylation of alkenes, but, under
the prevailing syngas conditions, also possess the ability to se-
lectively hydrogenate the unsaturated intermediates 2 in the
presence of aldehyde 1; compounds that are mutually involved
in the condensation equilibrium shown in Scheme 4. In this
way, intermediates 2 are removed selectively from the conden-
sation equilibrium and the reaction is drawn to completion by
catalytically forming the N-alkylamide product. If, on the other
hand, the catalyst system has preference for hydrogenation of
the aldehyde substrate 1 over 2, the condensation equilibrium
can be drawn to completion to alcohol 4.
Considerations on the roles of catalyst promoters HOTs and
HORF in Rh/xantphos-catalyzed hydroamidomethylation of
alkenes
Our studies on the catalytic reductive amidation of 1L with
acetamide by using pure H2 revealed that the optimal catalytic
system comprised a rhodium precursor with a xantphos-type
ligand and a catalytic amount of a strong acid promoter, such
as HOTs, in diglyme as the reaction solvent. The addition of
a catalytic amount of strong acid [HOTs/Rh ꢀ5 (mol/mol)] is
necessary to catalyze the addition equilibrium reaction of the
amide with the aldehyde.[6] In this respect, the mechanism is
different from that of reductive amination of an aldehyde, in
which case the addition of an amine NÀH bond to the alde-
hyde occurs spontaneously, without the involvement of any
acid.[5a–d] However, a strong acid such as HOTs also modifies
the catalytic system.
Discussion
General considerations
All experimental results concerning the hydroamidomethyla-
tion reaction, but, in particular, the development of the reac-
tion products with time (Figure 5) prominently indicate that
the hydroamidomethylation of 1-pentene with acetamide com-
prises three sequential steps, which are shown schematically in
Scheme 4. The first step is hydroformylation of 1-pentene to al-
dehyde 1. The second step consists of two equilibrium reac-
tions: the nucleophilic addition of acetamide to 1L forming 2a
followed by dehydration to form the two isomeric unsaturated
intermediates 2L.[6] Hydrogenation of these unsaturated inter-
The first acid–base reaction that would take place is proto-
nation of [Rh(acac)(CO)2] to form the [RhI(CO)2(xantphos)]OTs
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemSusChem 2013, 6, 1759 – 1773 1767

CHEMSUSCHEM
FULL PAPERS
www.chemsuschem.org
species with the release of Hacac; this rhodium/xantphos spe-
cies then can be feasibly converted into the well-known alkene
hydroformylation catalyst [RhI(CO)(H)(xantphos)]. Additionally,
the presence of HOTs may induce equilibria between neutral
RhI hydroformylation and cationic RhIII hydrogenation species.
One of our most prominent observations, however, is that the
additional presence of a weakly acidic promoter compound,
such as HORF [in solvent-like quantities, typically HORF/Rh
ꢀ2000 (mol/mol) at HORF/diglyme=1:4 (v/v)], also has
a strong impact on the course of the catalytic hydroamidome-
thylation reaction, as shown in Figure 2, creating an active hy-
drogenation catalyst in the CO-containing atmosphere. Similar
effects are observed in the separate hydroformylation and re-
ductive amidation reactions (Figures 6–8).
availability of ORFÀ anions due to protonation. This stronger
electrophilicity would cause O-coordination of the aldehyde to
the RhIII center rather than p-carbonyl coordination required
for hydrogenation. When DMA is added, ORFÀ anions are gen-
erated by the deprotonation equilibrium. These ORFÀ anions
coordinate more strongly to the RhIII center and the original
situation is substantially restored; the electrophilicity of the
RhIII metal center is reduced, as diagnosed by increasing alco-
hol formation.
Reductive amidation
The reductive amidation of 1L with acetamide in the presence
of HOTs and with pure H2 as the reductant is strongly inhibited
by the addition of HORF (Figure S6 in the Supporting Informa-
tion). Not only the hydrogenation of intermediates 2, but also
hydrogenation of aldehydes is severely suppressed by the
presence of HORF. This could suggest that the [RhIII(H)2-
(xantphos)](OTs) species proposed to be the active Rh hydro-
genation species in reductive amidation under pure H2 pres-
sure[6] is converted by protonation at the hydride with HORF to
the species [RhIII(OTs/ORF)3(xantphos)], which has no intrinsic
activity for hydrogenation under pure H2 pressure.
The effects of HORF and HOTs on reductive amidation with
the rhodium/xantphos catalyst system under pure H2 pressure
are further illustrated in Figure S7 in the Supporting Informa-
tion; in diglyme and diglyme/HORF (ꢀ1:4 v/v), the introduction
of HOTs is required for the establishment of the aldehyde–
amide addition equilibrium.[6] The acidity of HORF is not suffi-
ciently strong to catalyze this equilibrium under the applied re-
action conditions. It also appears that, in the absence of HOTs,
but in the presence of HORF, the rhodium/xantphos catalyst is
highly active for the hydrogenation of 1L to 4L (Figure S7 in
the Supporting Information). This could be a consequence of
the fact that aldehyde coordination to a metal center only re-
quires one free coordination site, whereas a catalyst for hydro-
genation of compounds 2 is likely to have two adjacent posi-
tions to allow for the selective bidentate binding of 2 in the
presence of aldehyde.
Hydroformylation
Clearly, the addition of HORF leads to a lower hydroformylation
rate of 1-pentene (Figure 6). Interestingly, the rate of hydrofor-
mylation even further decreases when HOTs is added to this
system; the formation of 4L becomes visible. A remarkable ob-
servation is the recovery of catalyst performance to a level ach-
ieved before HOTs addition upon the addition of DMA in a stoi-
chiometric quantity relative to 1-pentene. However, a decrease
in hydroformylation activity due to the presence of HORF re-
mains, even if DMA is added. This is likely to be due to the
large (fourfold molar ratio) excess of HORF over DMA under
the given conditions.
The high acidity of HOTs (pKa ꢀÀ2.7) combined with a large
excess of weakly basic DMA (DMA/HOTsꢀ100) accounts for
a likely complete deprotonation of HOTs and formation of the
salt. The resulting protonated HDMA+ cation of course still be-
haves as quite a strong acid and is responsible for the forma-
tion of cationic RhIII species, as suggested from the formation
4L (Figure 6B), lowering the concentration of the neutral RhI
hydroformylation catalyst. It is thought that the distinct de-
crease in hydroformylation activity due to the addition of the
weakly acidic HORF (pKa in water of ꢀ9) is due to its very large
excess over Rh, which thus may contribute to the generation
and stabilization of RhIII species, similar to HOTs. Again a neu-
tralization effect is observed when DMA [HORF/DMAꢀ4–20
(mol/mol)] is added to the reaction (DMA+HORFQHDMA+ +
ORFÀ; for which the equilibrium constant would be much
lower than that for the strong acid HOTs). The decreasing rate
of hydroformylation with increasing concentration of HORF,
while the product composition changes from 1L to 4L (Fig-
ure 7A), is fully consistent with a dual catalyst species model,
in which cationic RhIII hydrogenation species gradually replace
neutral RhI hydroformylation species.
The effect of CO on the reductive amidation in diglyme is as
expected: with increasing CO pressure, a strongly negative
effect is observed on the yield of 3L as well as 4L (Figure 8A),
and consequently, the yields of unsaturated compounds 2 and
higher condensation products increase. In contrast, when the
same experiments are carried out in diglyme/HORF, the yield of
3L and 4L strongly increases with CO pressure, thus a strongly
increased general hydrogenation activity is manifested by the
presence of CO. This implies that, in the presence of HORF,
a catalytic hydrogenation system other than the one present
under pure H2 pressure must be operative. We thus arrive at
the conclusion that the cationic [RhIII(H)2(xantphos)]+ species is
likely to operate as a hydrogenation catalyst under pure H2 in
an aprotic solvent such as diglyme,[6] but does not exist in the
presence of HORF under syngas conditions. Instead, a cationic
[RhIII(CO)x(H)(OTs/ORF)(xantphos)]+ catalyst species with a mix-
ture of ORFÀ and OTsÀ anions will arise in addition to neutral
RhI hydroformylation species. The observed increase in reduc-
Remarkably, hydroformylation in pure HORF not only results
in a low hydroformylation rate, but also in a low selectivity for
alcohol formation (Figure 7A). Similarly, the addition of HOTs at
a high concentration of HORF leads to a severe drop in hydro-
formylation and hydrogenation activity (Figure 7B). This could
be a consequence of the stronger electrophilic nature of the
RhIII center in pure HORF, due to relatively easy dissociation of
the associated OTsÀ anions in this polar medium and the un-
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemSusChem 2013, 6, 1759 – 1773 1768

CHEMSUSCHEM
FULL PAPERS
www.chemsuschem.org
tive amidation product, with associated decrease in intermedi-
ates 2, is approximately first-order in CO pressure (Figure 8B),
which is consistent with necessary binding of CO to a cationic
RhIII center as the catalyst in the overall rate-determining step
of hydroamidomethylation, that is, hydrogenation of the unsa-
turated intermediates 2. The relative occupation of coordina-
tion sites in this RhIII species by CO and the ORFÀ/OTsÀ anions
will depend on the relative quantities of HORF, HOTs, and base
(amide and N-alkylamide) as well as CO/H2 pressure, as sug-
gested by the separate results on hydroformylation and reduc-
tive amidation.
Synthesis of the overall mechanistic proposal
In our study concerning the reductive amidation of 1L with
acetamide in a pure hydrogen atmosphere, we postulated the
formation of a neutral RhIÀhydride species A to be responsible
for the hydrogenation of aldehyde to alcohol, whereas a cation-
ic RhIIIÀdihydride species B would be active in the hydrogena-
tion of the unsaturated intermediates 2 (Scheme 5a).[6] These
two species are related through an acid/base equilibrium; the
addition of acid resulted in higher selectivity for 3L, whereas
the addition of base resulted in the formation of 4L.
As implied by the effects of
added amide (such as DMA) on
hydroformylation with the rhodi-
um/xantphos/HOTs
catalyst
system, we may conclude that
under hydroamidomethylation
conditions the amide reactant
(and N-alkylamide product) also
plays an important controlling
role in the observed catalytic
performance. It is likely that par-
tial deprotonation of HOTs and
HORF occurs, thus providing
anions of which the reaction-
medium-dependent coordinat-
ing properties to RhIII may con-
trol some of the catalyst’s attri-
butes, such as specificity of hy-
drogenation of unsaturated sub-
strates.
Scheme 5. Generation of various hydrogenation catalysts under reductive amidation conditions (xant=xantphos):
a) in pure diglyme with a H2 atmosphere; b) in diglyme with a H2/CO atmosphere; c) in HORF/diglyme with a H2
atmosphere; and d) in HORF/diglyme with a H2/CO atmosphere.
When reductive amidation is carried out in diglyme with
a partial pressure of CO, both the hydrogenation of aldehyde
and intermediate 2 is poisoned (Figure 8A); remarkably the hy-
drogenation of 1L seems to be the most sensitive to the pres-
ence of CO. This observation can be rationalized by the coordi-
Hydroaminomethylation versus hydroamidomethylation
nation of CO to both species A and B, forming C and D, re-
spectively (Scheme 5b). Species C is a hydroformylation cata-
lyst that is known to have low hydrogenation activity. Carbon
monoxide binds more strongly to RhI than to RhIII, making it
plausible that the hydrogenation activity of species B is partial-
ly retained.
The reaction conditions for hydroamidomethylation can be
compared to those for well-known hydroaminomethylation of
alkenes with amines. Under the strongly basic conditions with
amines as substrates, acetal formation cannot occur. Several
studies of hydroaminomethylation applied simple alcohols
(such as methanol) in this reaction.[5a–d] In general, higher hy-
droaminomethylation reaction rates are observed in the pres-
ence of alkanols, thus also indicating a promoting effect of
protic cosolvents in hydroaminomethylation. Although a mech-
anistic study into the nature of this promoting effect has not
been published, it has been postulated that the protic solvent
could somehow serve to stabilize cationic rhodium hydrogena-
tion species.[10] We wish to suggest that the simple alcohols in
hydroaminomethylation serve a similar purpose to that of
HORF in hydroamidomethylation.[11] The difference is that asso-
ciated alkoxide anions are formed due to deprotonation of the
alcohol by the strongly basic amine substrate and product. Be-
cause hydroamidomethylation involves much weaker basic
amide substrates and products, a stronger acidic alcohol needs
to be applied to successfully generate alkoxide anionic species
to stabilize the cationic RhIII center.[12]
The effect of the addition of HORF on the hydrogenation ac-
tivity of the catalytic system in reductive amidation in a pure
hydrogen atmosphere is even more detrimental (Figure 8B); in
HORF/diglyme (1:4) only a small amount of 3L is formed. This
may be rationalized by protonation of both species A and B,
forming the inactive species [(xantphos)RhIII(OTs/ORF)3] (E;
Scheme 5c). Remarkably, the hydrogenation activity of the
system is restored with the addition of a partial pressure of
CO; apparently carbon monoxide is able to compete with ORFÀ
anions for coordination sites at the RhIII center in ‘neutral’ spe-
cies E, resulting in the formation of cationic species F. This
allows for the activation of hydrogen by heterolytic splitting,
resulting in the formation of HORF and the cationic RhIII–hy-
dride species G (Scheme 5d), which again may be an active hy-
drogenation catalyst.
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemSusChem 2013, 6, 1759 – 1773 1769

CHEMSUSCHEM
FULL PAPERS
www.chemsuschem.org
In the hydroformylation of 1-pentene in diglyme, the equi-
librium shown in Scheme 5b is also operative; by the addition
of HOTs, hydroformylation catalyst C is partially converted into
the RhIII–hydride species D; this is apparent from the formation
of 4L (Figure 6A). This conversion is reversed by the addition
of the base DMA, resulting in the formation of species C to
give 1L as the major product. The catalytic hydroformylation
system in a mixture of HORF/diglyme (1:4) behaves almost the
same, indicating that similar catalytic species are operative,
albeit with lower overall activity (Figure 6B). The activity of the
hydroformylation catalyst is inversely related to the amount of
HORF added (Figure 7A); with increasing amounts of HORF the
hydroformylation activity decreases, whereas the hydrogena-
tion activity increases up to a HORF/diglyme ratio of 5:1 (Fig-
ure 7B). This phenomenon again may be attributed to the for-
mation of species D by the weakly acidic HORF, lowering the
concentration of hydroformylation catalyst C. In the presence
of HOTs, but in the absence of base (DMA), the hydroformyla-
tion catalyst C is virtually nonexistent (Figure 7B) and most
likely species E and/or F are prevalent.
Now we are ready to consider the overall picture of the hy-
droamidomethylation reaction by considering the results most
prominently illustrated in Figure 2. From these results, five dif-
ferent regimes seem to be present upon changing from a pure
diglyme to a pure HORF solvent system. The overall picture is
complicated by the different acid/base equilibria induced by
the presence of base (acetamide and 3L, at a constant concen-
tration), strong acid (HOTs, only in relatively small amounts),
and weak acid (HORF, in increasing amounts).
aldehyde–amide condensation equilibrium that involves both
aldehyde and imide substrates. This equilibrium is coupled
through two hydrogenation cycles that involve either the
imide or aldehyde as the substrate, both catalyzed by a cationic
RhIII species. Competition between the last two substrates in-
volved in the respective catalytic cycles, thought to the gov-
erned by anion coordination to the cationic RhIII center, is the
basis for the chemoselectivity of the overall hydroamidomethy-
lation process.[13]
In pure diglyme, an active hydroformylation catalyst is
formed, but the catalytic system in the presence of CO lacks
hydrogenation activity; the most prominent catalytic com-
pound present in solution is most likely to be species C. Any
RhIII–hydride species present in solution is possibly blocked by
coordination of CO, forming species D (Scheme 5). Addition of
more of the strong acid HOTs is not an option because it re-
sults in the formation of more aldol condensation products.
However, the addition of weakly acidic and polar HORF as a co-
solvent results in the formation of an active hydrogenation cat-
alyst (such as species G) that is capable of hydrogenating the
intermediate products 2 or aldehyde 1, depending on the rela-
tive amount of HORF. A HORF/diglyme ratio of 1:4 (v/v) gener-
ates a more polar environment, in which the OTsÀ anions
become less coordinating, and thus, are likely to provide two
binding sites for imide intermediate 2 (species G2À1), resulting
in an active catalytic system for hydroamidomethylation with
the highest selectivity for N-alkylamides. Upon increasing the
HORF/diglyme ratio to 1:1, the hydroformylation activity is re-
tained, but the hydrogenation activity of the catalytic system is
severely inhibited; this may be due to the formation of
a higher concentration (relative to [Rh]) of relatively strongly
coordinating ORFÀ anions. Upon increasing the HORF/diglyme
Our view of the general mechanism of hydroamidomethyla-
tion is shown in Scheme 6. Hydroformylation of alkenes cata-
lyzed by a neutral RhI species is followed by an acid-catalyzed
Scheme 6. Proposed overall mechanism for the hydroamidomethylation of 1-alkenes.
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemSusChem 2013, 6, 1759 – 1773 1770

CHEMSUSCHEM
FULL PAPERS
www.chemsuschem.org
ratio to 5:1, the hydroformylation activity is retained, but now
4L is the major hydrogenation product. In this HORF-rich
medium, the ORFÀ anions become more readily solvated to
allow binding of the monodentate substrate aldehyde, via spe-
cies G1À1. Finally, it is important to note that we believe that
the hydroformylation reaction concerns a neutral catalytic spe-
cies that activates H2 by oxidative addition, shuttling between
RhI and RhIII. The hydrogenation cycles proceed via cationic
RhIII species; the intermediate products are protonated in the
acidic medium to form species F. Through the CO-assisted het-
erolytic splitting of hydrogen proposed above, species G is
then restored.
species.[14] In this proposed mechanism, the intermediate RhÀ
acyl species formed in hydroformylation is directly hydrogenat-
ed (by a proton-assisted isomerization to a RhÀhydroxycarbe-
noid) without intermediate formation of an aldehyde. However,
we have shown that with the rhodium/xantphos/HOTs catalytic
system, the hydroamidomethylation and related alcohol forma-
tion reactions occur through sequential steps that involve al-
dehyde–amide adducts and free aldehyde, respectively, as indi-
cated by the build up of intermediates 2 and aldehydes, de-
pending on the reaction conditions. A high selectivity for one
over the other can be obtained by adjusting the competitive
hydrogenation of the respective molecules. We believe that
one interesting aspect of the present work is the revelation of
the selectivity-controlling ability of the amide substrate, pre-
sumably by acting as a weak base, on the amount and type of
anions associated with the cationic rhodium species as part of
the overall catalyst system.
Both HORF and HOTs can provide anions to the cationic
RhIII–hydride species, by protonation of the weakly basic amide
reactant. Depending on their relative amounts, the relative
acidic strength, and coordinative properties to the Rh center, it
can thus be rationalized that the immediate anionic environ-
ment around a cationic [(xantphos)RhIIIH]2+ species plays an
important role in the hydrogenation substrate specificity of the
intermediate 2 relative to the aldehyde, which are both in-
volved in a dynamic condensation equilibrium. Such a model
can rationalize the dramatic change in selectivity from 3L as
a product (>80%) at a relatively low HORF concentration to
4L as the product (>90%) at a high concentration of HORF.
At low HORF concentrations, the weakly coordinating OTsÀ
anions render the RhIII species receptive to binding of the bi-
dentate substrate 2, even in the presence of the monodentate
aldehyde. The aldehyde substrate is less hindered by the
anions around the rhodium center, which is likely to be be-
cause of its monodentate binding: even with the intrinsically
more strongly coordinating ORFÀ anions, the aldehyde’s car-
bonyl functionality succeeds in approaching the RhIII center
and can become hydrogenated, in particular, when immersed
in a high concentration of polar HORF molecules. The relatively
sharp decline in hydrogenation activity, now accompanied
with a significant decrease in hydroformylation activity at very
high concentration of HORF (>90%) may be attributed to an
increased presence of species E, but now associated almost ex-
clusively with the relatively strongly coordinating ORFÀ anions,
thus behaving more as a neutral RhIII species. It is thought that
this process may proceed to the extent that heterolytic activa-
tion of H2 (producing active cationic species F and G) becomes
rate determining and relatively slow.
Conclusions
We have successfully developed a novel catalytic system that
comprised of rhodium/xantphos/HOTs in the presence of HORF
as a cosolvent for the atom-economic hydroamidomethylation
of terminal alkenes to form N-alkylamides. It appeared that the
presence of both strong acid (HOTs) and a polar acidic solvent
(HORF) was crucial in determining the reaction selectivity and
efficiency. The strong acid was necessary to establish the equi-
librium addition/condensation reaction between the aldehyde
formed in situ and the amide. The presence of the polar,
weakly acidic cosolvent was necessary for the generation and
solvation of cationic RhIII–hydride species that were active as
hydrogenation catalysts in the presence of carbon monoxide.
By choosing the right circumstances, this catalytic system
could be fine-tuned to make either an alcohol or an N-alkyla-
mide from terminal alkenes. The novel hydroamidomethylation
reaction has potential in the synthesis of a wide range of sec-
ondary amides, as shown by its applicability to different olefins
and different aliphatic and aromatic amides. We are currently
seeking to utilize this catalytic system for the hydroamidome-
thylation reaction of internal alkenes with an amide to form
linear N-alkylamides, for which the isomerization step is, of
course, highly challenging.
Herein, we have proposed plausible catalytic events occur-
ring in the hydroamidomethylation (hydroformylation–reduc-
tive amidation) reaction that can rationalize most of our obser-
vations. To the best of our knowledge, in studies reporting the
related hydroaminomethylation (hydroformylation–reductive
amination) reaction, no clear mechanistic proposals have been
provided specifically for the second step (reductive amination)
of the reaction. However, it has been proposed that the hydro-
genation step benefits from the presence of protic solvents be-
cause of the formation of cationic rhodium species.[4c,5c,d] The
role of alcohol as a protic cosolvent in the tandem hydrofor-
mylation–hydrogenation reaction, in which alcohols are
formed directly from alkenes, has also been explained by
a mechanism based on the formation of a rhodium–carbenoid
Experimental Section
Chemicals: The phosphane ligands PPh3, DPEphos, bis(diphenyl-
phosphinoethyl)phenylphosphane, and xantphos were purchased
from Strem Chemicals, Germany. The bidentate phosphane ligands
benzoxantphos, homoxantphos, and DBFphos were purchased
from Innovative Catalyst Technologies (InCatT B.V.), the Nether-
lands. Other phosphorous ligands, such as tBu-xantphos, Si-xant-
phos, oMeO-xantphos, and 4,5-bis[di(tert-butyl)phosphanyl]-9,9-di-
methylxanthene (di-tBu-xantphos), were generously provided by
Shell Global Solutions Amsterdam B.V., and were synthesized ac-
cording to literature procedures.[7a,15] All other chemicals, solvents,
acids, and bases were purchased from Acros Organics or Sigma Al-
drich, the Netherlands.
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemSusChem 2013, 6, 1759 – 1773 1771

CHEMSUSCHEM
FULL PAPERS
www.chemsuschem.org
Instruments: The stainless-steel autoclave reactors (100 mL) from
HEL Ltd, UK, were equipped with a magnetic stirrer, pressure trans-
ducer, and temperature controlling thermocouple. A Hewlett Pack-
ard HP6890 Series auto-sampler GC system was used for regular
GC analysis. GC–MS analysis were carried out on an Agilent tech-
nologies 7820A GC system series coupled with an Agilent technol-
ogies 5975 series GC-MSD system. A glove box from M. Braun In-
ertgas-System GmbH, Germany, was used for storing and handling
air-sensitive phosphane ligands. NMR spectra were recorded on
Bruker DPX300 (300 MHz) or Bruker DMX400 (400 MHz) spectro-
meters.
Acknowledgements
This research was performed within the framework of the Catch-
Bio program. We gratefully acknowledge the support of the
SmartMix Program of the Netherlands Ministry of Economic Af-
fairs and the Netherlands Ministry of Education, Culture and Sci-
ence. Shell Global Solutions International B.V. is kindly acknowl-
edged for generously donating phosphane ligands. We thank Dr.
J. Brandts (BASF) and Dr. R. Parton (DSM) for stimulating and
fruitful discussions.
Catalytic high-pressure reaction: All preparations and manipula-
tions were performed by using standard Schlenk techniques under
an argon atmosphere. The solvent diglyme was distilled from CaH2,
deoxygenated, and used immediately after the purification process.
The catalytic reactions were carried out under varying syngas pres-
sures and reaction temperatures. For all catalytic experiments, the
active catalyst precursor was formed in situ in the autoclave by
transferring the metal precursor and selected phosphane ligands.
Keywords: chemoselectivity
hydroamidomethylation · regioselectivity · rhodium
·
homogeneous catalysis
·
[1] a) P. Anastas, N. Eghbali, Chem. Soc. Rev. 2010, 39, 301; b) J. R. Briggs, J.
Klosin, G. T. Whiteker, Org. Lett. 2005, 7, 4795; c) P. Eilbracht, L. Barfacker,
C. Buss, C. Hollmann, B. E. Kitsos-Rzychon, C. L. Kranemann, T. Rische, R.
Roggenbuck, A. Schmidt, Chem. Rev. 1999, 99, 3329; d) F. KoÅ, M. Wys-
zogrodzka, P. Eilbracht, R. Haag, J. Org. Chem. 2005, 70, 2021.
[2] a) S. Enthaler, ChemCatChem 2010, 2, 1411; b) S. Fleischer, S. L. Zhou, K.
Junge, M. Beller, Chem. Asian J. 2011, 6, 2240; c) S. Gomez, J. A. Peters,
T. Maschmeyer, Adv. Synth. Catal. 2002, 344, 1037; d) T. Gross, A. M.
Seayad, M. Ahmad, M. Beller, Org. Lett. 2002, 4, 2055; e) A. W. Heinen,
J. A. Peters, H. van Bekkum, Eur. J. Org. Chem. 2000, 2501; f) P. B. Quynh,
T. H. Kim, Tetrahedron Lett. 2011, 52, 5004; g) B. C. Ranu, A. Majee, A.
Sarkar, J. Org. Chem. 1998, 63, 370; h) V. I. Tararov, R. Kadyrov, T. H. Rier-
meier, A. Borner, Chem. Commun. 2000, 1867.
In the preparation of
a typical catalytic reaction mixture,
0.01 mmol of [Rh(acac)(CO)2] (2.58 mg) and 0.02 mmol of bidentate
phosphane ligand or 0.04 mmol of monodentate phosphane
ligand were weighed and transferred into an autoclave. The auto-
clave was tightly closed and subsequently filled with argon
through the use of a Schlenk line that was connected to one of
the valves of the autoclave. Through another valve under a continu-
ous flow of argon, dried and degassed diglyme (8 mL) was subse-
quently added along with HORF (2 mL), decane (0.605 mL,
3.125 mmol) as an internal standard, 1-pentene (0.548 mL,
5 mmol), and acetamide (0.295 g, 5 mmol) followed by HOTs
(9.51 mg, 0.05 mmol). Then the reactor was inserted into the heat-
ing block and pressurized with 50 bar (CO/H2 =1:2) syngas. This re-
action mixture was stirred at 500 rpm for 30 min to ensure that
complex formation was complete. The reaction mixture was
heated to 1008C (within 30 min) under stirring at 500 rpm. All reac-
tion conditions of the catalytic process were controlled by compu-
terized software panels. After standing for 2, 4, or 8 h at this tem-
perature, the autoclave was cooled to room temperature over
about 1 h. The autoclave was then carefully vented to atmospheric
pressure.
[3] a) W. Reppe, H. Vetter, Liebigs Ann. Chem. 1953, 582, 133; b) W. Reppe,
Experientia 1949, 5, 93.
[4] a) G. Angelovski, P. Eilbracht, Tetrahedron 2003, 59, 8265; b) M. Beller, C.
Breindl, M. Eichberger, C. G. Hartung, J. Seayad, O. R. Thiel, A. Tillack, H.
Trauthwein, Synlett 2002, 1579; c) D. Crozet, M. Urrutigoity, P. Kalck,
ChemCatChem 2011, 3, 1102; d) C. S. Graebin, V. L. Eifler-Lima, R. G.
da Rosa, Catal. Commun. 2008, 9, 1066; e) D. S. Melo, S. S. Pereira, E. N.
dos Santos, Appl. Catal. A 2012, 411, 70; f) K. S. Mꢁller, F. Koc, S. Ricken,
P. Eilbracht, Org. Biomol. Chem. 2006, 4, 826; g) A. Seayad, M. Ahmed, H.
Klein, R. Jackstell, T. Gross, M. Beller, Science 2002, 297, 1676; h) T. O.
Vieira, H. Alper, Chem. Commun. 2007, 2710; i) Y. Y. Wang, M. M. Luo, Q.
Lin, H. Chen, X. J. Li, Green Chem. 2006, 8, 545; j) J. C. Wasilke, S. J.
Obrey, R. T. Baker, G. C. Bazan, Chem. Rev. 2005, 105, 1001.
[5] a) M. Ahmed, R. P. J. Bronger, R. Jackstell, P. C. L. Kamer, P. W. N. M. van
Leeuwen, M. Beller, Chem. Eur. J. 2006, 12, 8979; b) M. Ahmed, C. Buch,
L. Routaboul, R. Jackstell, H. Klein, A. Spannenberg, M. Beller, Chem. Eur.
J. 2007, 13, 1594; c) M. Ahmed, A. M. Seayad, R. Jackstell, M. Beller, J.
Am. Chem. Soc. 2003, 125, 10311; d) B. Hamers, E. Kosciusko-Morizet, C.
Muller, D. Vogt, ChemCatChem 2009, 1, 103; e) G. D. Liu, K. X. Huang,
C. X. Cai, B. N. Cao, M. X. Chang, W. J. Wu, X. M. Zhang, Chem. Eur. J.
2011, 17, 14559.
After each catalytic run, the reaction mixture was taken from the
reactor and immediately analyzed by gas chromatography. Calibra-
tion lines for each analyte were used to determine the conversion
of the substrates and yields of the various products. The two iso-
meric unsaturated intermediates 2a and 2L are lumped together
as 2; generally these two isomeric compounds are present in only
low amounts. The assignments of the products were confirmed by
GC–MS and compared with authentic and pure commercial sam-
ples.
[6] S. Raoufmoghaddam, E. Drent, E. Bouwman, Adv. Synth. Catal. 2013,
355, 717.
[7] a) L. A. van der Veen, P. H. Keeven, G. C. Schoemaker, J. N. H. Reek, P. C. J.
Kamer, P. W. N. M. van Leeuwen, M. Lutz, A. L. Spek, Organometallics
2000, 19, 872; b) Rhodium Catalyzed Hydroformylation (Eds.: P. W. N. M.
van Leeuwen, C. Claver), Kluwer, Dordrecht, 2000; c) P. W. N. M. van
Leeuwen, C. F. Roobeek, J. Mol. Catal. 1985, 31, 345.
[8] All experiments were performed three times and were reproducible.
[9] E. Nagy, C. Benedek, M. Heil, S. Toros, Appl. Organomet. Chem. 2002, 16,
628.
[10] B. Hamers, P. S. Baeuerlein, C. Muller, D. Vogt, Adv. Synth. Catal. 2008,
350, 332.
[11] D. Crozet, A. Gual, D. McKay, C. Dinoi, C. Godard, M. Urrutigoity, J. C.
Daran, L. Maron, C. Claver, P. Kalck, Chem. Eur. J. 2012, 18, 7128.
[12] It is worthwhile briefly commenting on the results of our search for al-
ternative catalyst promoter compounds to HORF for hydroamidomethy-
lation (Table 3). The fact that some phenolic compounds with pKa
values of around 9–10 show similar performances to HORF; this sug-
gests that the promoting effect depends upon the acidity of the pro-
The products in Table 4,[16] N-heptylacetamide,[1a,17] N-nonylaceta-
mide,[18] N-(3-phenylpropyl)acetamide,[19] N-hexylpropanamide,[16b,20]
N-hexylisopropylamide,[16b,21] N-hexyl-2,2-dimethylpropionamide (N-
hexylpivalamide),[16c,21,22] N-hexylpentanamide,[23] N-hexylbenzami-
de,[16b,24] N-hexyl-4-methoxybenzamide,[16b] N-hexyl-4-trifluorome-
thylbenzamide,[16b] N-hexyl-2-fluoroacetamide,[25] N-(2-methylpenty-
l)acetamide, and N-benzylacetamide[16a,c] are known compounds
1
and were identified by H and 13C NMR spectroscopy and GC–MS
(see Figures S11–S18 in the Supporting Information).[6]
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemSusChem 2013, 6, 1759 – 1773 1772

CHEMSUSCHEM
FULL PAPERS
www.chemsuschem.org
moter compounds. Only a rather narrow pKa range of potential promot-
ers can be applied; acids that are too strong cannot be used because
this will lead to high rates of aldol condensation, whereas weaker acids,
such as simple alkanols, cannot be used because of aldehyde acetal for-
mation.
Rubio-Pꢂrez, P. Sharma, F. J. Perez-Flores, L. Velasco, J. L. Arias, A. Cab-
rera, Tetrahedron 2012, 68, 2342; d) R. M. Waters, N. Wakabayashi, E. S.
Fields, Org. Prep. Proced. Int. 1974, 6, 53.
[17] K. Singh, Tetrahedron 2009, 65, 10395.
[18] a) L. Friedman, H. Shechter, Tetrahedron Lett. 1961, 2, 238; b) G. O. Tor-
osyan, N. K. Tagmazyan, A. T. Babayan, Zh. Org. Khim. 1984, 20, 506.
[19] a) D. Q. Gao, F. Scholz, H. G. Nothofer, W. E. Ford, U. Scherf, J. M. Wessels,
A. Yasuda, F. von Wrochem, J. Am. Chem. Soc. 2011, 133, 5921; b) A.
L’Heureux, F. Beaulieu, C. Bennett, D. R. Bill, S. Clayton, F. LaFlamme, M.
Mirmehrabi, S. Tadayon, D. Tovell, M. Couturier, J. Org. Chem. 2010, 75,
3401; c) R. Pelagalli, I. Chiarotto, M. Feroci, S. Vecchio, Green Chem.
2012, 14, 2251.
[13] To explain why neither carbonylation nor hydroformylation of inter-
mediates 2 occurs in our reaction, it is thought that hydrogenation
may proceed through insertion of one of the compounds 2, that is, via
N-hexylideneacetamide. The insertion of C=N into a cationic RhIIIÀH
species can selectively give the intermediate RhIIIÀimido species (RhÀN-
CHR), which is immediately converted into the RhIII species and 3L
through protonation in the acidic medium. If hydride migration were to
result in a RhÀC-NHR intermediate, it seems quite likely that CO inser-
tion into RhÀC might occur to give RhÀC(O)ÀC-NHR, which may then
be terminated by hydrogenolysis to give the a-amidoaldehyde.
[14] a) I. I. F. Boogaerts, D. F. S. White, D. J. Cole-Hamilton, Chem. Commun.
2010, 46, 2194; b) P. Cheliatsidou, D. F. S. White, D. J. Cole-Hamilton,
Dalton Trans. 2004, 3425; c) P. Cheliatsidou, D. F. S. White, A. M. Z.
Slawin, D. J. Cole-Hamilton, Dalton Trans. 2008, 2389.
[20] C. Chen, S. H. Hong, Org. Lett. 2012, 14, 2992.
[21] W. Krawczyk, G. T. Piotrowski, J. Chrom. 1989, 463, 297.
[22] H. M. Meshram, G. S. Reddy, M. M. Reddy, J. S. Yadav, Tetrahedron Lett.
1998, 39, 4103.
[23] A. J. M. van Dijk, T. Heyligen, R. Duchateau, J. Meuldijk, C. E. Koning,
Chem. Eur. J. 2007, 13, 7664.
[24] A. Nemchik, V. Badescu, O. Phanstiel, Tetrahedron 2003, 59, 4315.
[25] S. Miscevic, D. Minic, S. Petrovic, J. Fluorine Chem. 1992, 59, 239.
[15] a) R. P. J. Bronger, P. C. J. Kamer, P. W. N. M. van Leeuwen, Organometal-
lics 2003, 22, 5358; b) M. Kranenburg, Y. E. M. Vanderburgt, P. C. J.
Kamer, P. W. N. M. van Leeuwen, K. Goubitz, J. Fraanje, Organometallics
1995, 14, 3081.
[16] a) K. P. Dhake, Z. S. Qureshi, R. S. Singhal, B. M. Bhanage, Tetrahedron
Lett. 2009, 50, 2811; b) K. Ishihara, T. Yano, Org. Lett. 2004, 6, 1983; c) L.
Received: May 17, 2013
Part of a Special Issue on ”CatchBio“. To view the complete issue, visit:
http://onlinelibrary.wiley.com/doi/10.1002/cssc.v6.9/issuetoc
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemSusChem 2013, 6, 1759 – 1773 1773