Rhodium-catalyzed homogeneous reductive amidation of aldehydes

Article abstract of DOI:10.1002/adsc.201200837

The catalytic reductive amidation of an aldehyde (hexanal) with an amide (acetamide) is reported. Apart from the desired N-hexylacetamide, the two isomeric unsaturated intermediates as well as hexanol are produced together with higher mass products that arise from aldol condensation and diamide coupling of the aldehyde. Screening of different catalyst precursor salts, ligands and reaction conditions led to the finding that the catalytic system based on the (cyclooctadiene)rhodium chloride dimer, [Rh(cod)Cl]₂, in combination with the ligand xantphos and an acid co-catalyst results in high selectivity for the desired product. Under optimized conditions nearly full conversion is reached with high selectivity to the desired N-alkylamide and with a very high N-alkylamide/alcohol ratio, while producing only small amounts of by-products. The scope of the reaction has been investigated using different amides as well as aldehydes; the results show the general applicability of this novel reaction, but with electron-withdrawing amides the selectivity to N-alkylamide is lower. NMR studies showed that the nucleophilic addition of acetamide to hexanal is acid catalyzed, forming N-(1-hydroxyhexyl)acetamide in equilibrium with both hexanal and the dehydrated unsaturated imides. A catalytic mechanism is proposed in which a strong acid such as HOTs acts as a co-catalyst by establishing a rapid chemical equilibrium between the aldehyde, acetamide and the intermediates. Furthermore, it is proposed that the presence of acid causes a change in catalytic species, enabling a cationic Rh/xantphos hydrogenation catalyst to selectively hydrogenate the intermediates to N-hexylacetamide in the presence of hexanal. Copyright

Full text of DOI:10.1002/adsc.201200837

FULL PAPERS
DOI: 10.1002/adsc.201200837
Rhodium-Catalyzed Homogeneous Reductive Amidation of
Aldehydes
Saeed Raoufmoghaddam,a Eite Drent,a and Elisabeth Bouwmana,*
a
Leiden Institute of Chemistry, Gorlaeus Laboratories, Leiden University, P.O. Box 9502, 2300 Leiden, The Netherlands
Fax : (+31)-71527-4671; e-mail: bouwman@chem.leidenuniv.nl
Received: September 17, 2012; Published online: February 22, 2013
Supporting information for this article is available on the WWW under http://dx.doi.org/10.1002/adsc.201200837.
Abstract: The catalytic reductive amidation of an al- plicability of this novel reaction, but with electron-
dehyde (hexanal) with an amide (acetamide) is re- withdrawing amides the selectivity to N-alkylamide
ported. Apart from the desired N-hexylacetamide, is lower. NMR studies showed that the nucleophilic
the two isomeric unsaturated intermediates as well addition of acetamide to hexanal is acid catalyzed,
as hexanol are produced together with higher mass forming N-(1-hydroxyhexyl)acetamide in equilibrium
products that arise from aldol condensation and di- with both hexanal and the dehydrated unsaturated
ACHTUNGTRENNUNGamide coupling of the aldehyde. Screening of differ- imides. A catalytic mechanism is proposed in which
ent catalyst precursor salts, ligands and reaction con- a strong acid such as HOTs acts as a co-catalyst by
ditions led to the finding that the catalytic system establishing a rapid chemical equilibrium between
based on the (cyclooctadiene)rhodium chloride the aldehyde, acetamide and the intermediates. Fur-
dimer, [RhACHTUNGTRENNUNG(cod)Cl]2, in combination with the ligand thermore, it is proposed that the presence of acid
xantphos and an acid co-catalyst results in high selec- causes a change in catalytic species, enabling a cation-
tivity for the desired product. Under optimized con- ic Rh/xantphos hydrogenation catalyst to selectively
ditions nearly full conversion is reached with high se- hydrogenate the intermediates to N-hexylacetamide
lectivity to the desired N-alkylamide and with a very in the presence of hexanal.
high N-alkylamide/alcohol ratio, while producing
only small amounts of by-products. The scope of the
reaction has been investigated using different amides Keywords: homogeneous catalysis; phosphane li-
as well as aldehydes; the results show the general ap- gands; reductive amidation; rhodium
Introduction
challenging methods are hydroamination of alkenes
and alkynes with amines,[2] hydroamidation of alkenes
The reductive amidation of aldehydes with amides af- and alkynes[2f,3] and the reductive amination of car-
fording alkylated amides is a reaction with tremen- bonyl compounds.[4] In principle, these reactions can
dous synthetic potential for application in academia be performed with 100% atom efficiency, without or
as well as industry. The amide bond is one of the with very limited waste formation, fulfilling the re-
most significant functionalities found in many natural quirement of green chemistry.
products. The catalytic formation of carbon-nitrogen
bonds is highly attractive in organic chemistry, as well alkynes is defined as the addition of an H NH2, H
as in the bulk and fine chemical industries for the pro- NHR or H NR R bond across an alkene or alkyne
duction of solvents, pharmaceutical intermediates and providing a next higher substituted alkyl- or alkenyl-
The intermolecular hydroamination of alkenes or
ꢀ
ꢀ
1
1
2
ꢀ
also in the synthesis of bioactive compounds such as
ACHTUNGTRENaNGNU mine, respectively [Scheme 1, reaction (1) a-b]. This
amino acids.[1] In this respect, catalytic conversions to reaction, which can be catalyzed both by transition
N-alkylamides definitely offer potential advantages metal complexes or by strong base, has been mostly
ꢀ
over conventional methods of C N coupling reac- studied with secondary amines rather than ammonia
tions, where large quantities of salts are produced or primary amines; due to the subsequent reactivity
stoi
G
of the primary reaction product, doubly and triply
Among the different methods of carbon-nitrogen alk(en)ylated products are also formed when the
bond formations (Scheme 1), the most studied and latter substrates are being used.[2f,5] The catalytic hy-
Adv. Synth. Catal. 2013, 355, 717 – 733
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
717

Saeed Raoufmoghaddam et al.
FULL PAPERS
Scheme 1. Hydroamination; hydroamidation; hydroaminomethylation and reductive amination reactions.
droamination reaction of olefins is hampered by low cess. Separation of the reaction products is usually dif-
rates and low catalyst turnover numbers, while the ficult due to small differences in the boiling points of
thermodynamic driving force for the intermolecular the products. In attempts to reach higher selectivity
version of hydroamination is close to zero or even toward the desired product, the reductive amination
negative at the elevated temperatures generally re- reaction is often carried out using a large excess of
quired for the reaction to proceed.[5b]
the amine substrate; however, in a commercial appli-
The related hydroamidation reaction, involving the cation this would require costly (energy intensive)
[2f]
ꢀ
addition of an amideꢁs N H bond to an alkene or amine recycle processes.
alkyne[3] [Scheme 1, reaction (2) a-b] has also been
The development of a new method for the synthesis
studied; application of this reaction in an intermolec- of N-alkylamides using an amide as the substrate in-
ular fashion suffers from very similar problems as stead of an amine could be attractive as we reasoned
mentioned for intermolecular hydroamination, in par- that such a reductive amidation most likely would
ticular with alkenes as substrate.[2f]
avoid the formation of over-alkylated amide side-
As a possible atom-economical efficient synthesis products, due to a combination of relatively low nu-
of amines the hydroaminomethylation of alkenes has cleophilicity of the amide and increased steric encum-
also attracted considerable attention [Scheme 1, reac- brance in the N-alkylated amide. This lower reactivity
tion (3)]. The (exothermic) hydroaminomethylation would allow application of an aldehyde/amide sub-
of alkenes is a tandem reaction consisting of three strate ratio close to unity. Moreover, subsequent hy-
steps: initial hydroformylation of alkene, followed by drolysis of the resulting N-alkylamide may yield a pri-
the condensation of the aldehyde with a primary or mary amine; the recovered carboxylic acid can be re-
secondary amine to form an enamine or imine, and cycled to the corresponding amide with ammonia.
a final hydrogenation to give the saturated secondary Thus, the catalytic reductive amidation reaction
or tertiary amine.[1,6]
The most challenging step of hydroaminomethyla- the difficulties of reductive amination with ammonia.
tion, i.e., the metal-catalyzed reductive amination of Herein, we report our studies towards the catalytic
would be a means to make primary amines, avoiding
aldehydes with ammonia, primary amines or secon- reductive amidation reaction of hexanal with acet-
dary amines, has been extensively studied.[4,7] Because amide (Scheme 2). The ultimate goal of our research
of the high reactivity of the imine intermediates, re- is to combine the reductive amidation reaction with
ductive amination of aldehydes generally is accompa- in situ formation of the aldehyde by hydroformylation
nied by significant build-up of heavy, oligomeric by- of alkenes (hydroamidomethylation).
products, in particular with catalysts and under condi-
tions such that hydrogenation of the imine intermedi-
ates is too slow. For ammonia and primary amines as
substrates, over-alkylation of the desired primary or
secondary amines is a general issue of concern. This is
a consequence of the high reactivity of the primary or
secondary amines, which will act as competing sub-
strates in the course of the reductive amination pro- to form N-hexylacetamide.
Scheme 2. Reductive amidation of hexanal with acetamide
718
asc.wiley-vch.de
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2013, 355, 717 – 733

Rhodium-Catalyzed Homogeneous Reductive Amidation of Aldehydes
Results
tion mixtures after the reaction. Amounts of desired
product 3, unsaturated products N-(1-hexylidene)acet-
ACHTUNGTRENaNUNG mide or N-(1-hexenyl)acetamide 2, and hexanol 4
General Considerations
were determined using calibration lines.
The products found in a typical reductive amidation
The remainder consisted of the higher mass prod-
experiment with hexanal 1 and acetamide as the sub- ucts 5, which were not generally individually quanti-
strates, are shown in Scheme 3. Apart from the de- fied from GC, but rather calculated as a lumped-to-
gether-number from the uneven mass balance of
hexanal-derived GLC-measurable products. In the
tables the selectivity for 2 as well as for the sum of
3+4 are reported. The specificity of the catalysts for
reductive amidation versus hydrogenation of hexanal
1 is given as the ratio 3/4. Full analytical data of all
reaction mixtures are given in the Supporting Infor-
mation.
Catalytic Reductive Amidation: Initial Screening
Studies
To investigate reductive amidation of aldehydes, we
started our exploratory studies by testing various rho-
dium precursors. As shown in Table 1, the precursors
Scheme 3. Reductive amidation of hexanal with acetamide:
observed products.
Table 1. Reductive amidation of hexanal with acetamide.[a]
Rhodium
precursor
Conv.
[%]
2
3/4
3+4
Sel. [%] Ratio Sel. [%]
sired N-hexylacetamide 3, the two isomeric unsaturat-
ed compounds N-(1-hexylidene)acetamide or N-(1-
hexenyl)acetamide 2 as well as hexanol 4 were ob-
served with GC analysis. Additionally, formation of
various products with higher mass (Figure 1), com-
prising aldol condensation products 5a and the con-
densation product of hexanal with two molecules of
acetamide 5b was confirmed by both 1H and
13C NMR and mass spectroscopic analysis.
The reductive amidation experiments were per-
formed starting with 5 mmol of hexanal 1 and a small
excess of acetamide (6 mmol); a quantitative product
distribution was determined from GC analysis. The
1
2
3
4
5
6
[Rh
Rh(acac)(CO)2
[Rh(cod)2]BF4
RhCl3·xH2O
Rh(CO)(H)
[Rh(cod)2]OTf
(cod)Cl]2
56
25
47
53
35
42
29
29
6
14.7
3.1
1.9
8.7
0.0
2.7
30
12
46
20
57
54
AHCTUNGTRENNUNG
AHCTUNGTRENNUNG
AHCTUNTGREG(NNUN PPh3)3 27
G
58
16
[a]
Reaction conditions: 0.01 mmol Rh precursor, 5 mmol
hexanal, 6 mmol acetamide (Rh:hexanal:acetamide=
1:500:600); PH =80 bar; T=1008C; t=4 h; solvent:
25 mL diglyme;2decane as internal standard.
hexanal conversions shown in the tables were calcu- [RhACTHNUGTRNE(NNGU cod)Cl]2, [RhACHUTNTRGEG(NNUN cod)2]BF4 and [RhACHTGUNTREN(NUGN cod)2]OTf
lated from the amount of hexanal found in the reac- turned out to yield the highest hexanal 1 conversion
with only moderate differences in the 3/4 ratio and
3+4 selectivity. It appeared, however, that neither of
the tested rhodium complexes showed a high hydro-
genation activity nor good selectivity to the desired
N-hexylacetamide 3. High specificity for reductive
amidation versus hydrogenation of hexanal 1 is lack-
ing, as is apparent from the relatively low 3/4 ratio. In
addition, a significant conversion of hexanal to heavi-
er products 5 (~35% selectivity) is apparent from the
uneven hexanal-derived product balance as derived
from GLC analysis. Because of its low sensitivity to
air and moisture and because this precursor showed
the highest product specificity
[Rh
for further studies.
3
versus 4,
AHCTUNGTREN(GNUN cod)Cl]2 was selected as the catalyst precursor
Figure 1. Products of higher molecular mass (5) observed in
the reductive amidation reaction.
Adv. Synth. Catal. 2013, 355, 717 – 733
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
719

Saeed Raoufmoghaddam et al.
FULL PAPERS
Figure 2. Ligands used in reductive amidation studies of hexanal.
In the following screening studies [Rh
combined in situ with a large number of monodentate the bidentate phosphane ligands dppe and dppp, how-
or bidentate phosphane ligands. ever, gave only moderate overall results with a low
ACHTUNGTRNE(NUGN cod)Cl]2 was generally gave higher hexanal 1 conversion.The use of
Variation in the stereoelectronic properties of the selectivity to 3 (entries 11–14), while the addition of
ligands was achieved by changing the P-substituent acid again gave rise to higher conversion, but with
for monodentate ligands (aryl, alkyl, aryloxy or alkyl- only a slightly higher 3/4 product ratio and formation
ACHTUNGTRENNUNG
The catalytic results of the screening studies using product together with small amounts of 5; the addi-
the different ligands are summarized in Table 2. The tion of acid merely resulted in formation of more 5.
addition of various monodentate ligands to Better results were obtained when ligands with larger
[RhACHTUNGTRENNUNG(cod)Cl]2 resulted in lower conversion of the sub- bite angle are used (entries 17–26). With the ligands
strate relative to that obtained with the metal com- 1,2-dppmb and 1,3-dppmb fairly high 3/4 product
plex precursor alone (cf. entries 1, 3, 5, 7, 9). More ratios were achieved, but with poor total hydrogena-
importantly, these catalytic systems are quite active in tion selectivity to 3+4. On the other hand, use of
hexanal 1 hydrogenation as shown by the low 3/4 dppf or bisbi resulted in catalytic systems with higher
ratio. Introducing a 5-fold molar excess on Rh of p- selectivity to 3+4, but with only moderate 3/4 prod-
toluenesulfonic acid (HOTs) as additive appeared to uct ratios. Much better yields of 3 were achieved with
have a pronounced impact on conversion and selectiv- the xantphos ligand in combination with an added
ity; however, in most cases the selectivity to the de- catalytic quantity of the acid HOTs (cf. entries 25 and
sired product remained low, as for example for the 26). The addition of the acid dramatically improved
ligand PPh3 (entries 3 and 4). The use of the more the 3/4 product ratio from 0.3 to 17.5 with 80% com-
basic ligand PACHTUNGTRENNUNG(n-Bu)3 resulted in a catalytic system bined hydrogenation selectivity to 3+4 at a high con-
with high selectivity for hexanol 4 formation (80%) version of hexanal 1 (90%). At a lower reaction tem-
while here addition of acid did not have a significant perature (808C), the 3/4 product ratio and 3+4 selec-
effect on the 3/4 ratio (cf. entries 5 and 6). A higher tivity increased to, respectively, ~21 and ~95% with
product ratio 3/4 was observed with the catalytic sys- only a slightly lower hexanal 1 conversion (entry 27).
tems containing the acid additive and the monoden-
In a next set of catalytic experiments in the pres-
tate ligands P(ad)2A(n-Bu) or P(O-di-t-BuPh)3 (en- ence of HOTs, the rhodium/xantphos ratio was varied
CTHUNGTRENNUNG
tries 7–10), but the selectivity for the total amount of (Table 3); the results show that increasing the rhodi-
hydrogenated products 3+4 is still modest. The use um/xantphos ratio from 1:1 to 1:3 leads to lower con-
of catalytic systems based on the bidentate ligands versions but only slightly lower selectivity for hydro-
720
asc.wiley-vch.de
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2013, 355, 717 – 733

Rhodium-Catalyzed Homogeneous Reductive Amidation of Aldehydes
Table 2. Reductive amidation of hexanal with acetamide catalyzed by [RhACHTNUGRTNEUNG
(cod)Cl]2 in combination with different ligands.[a]
Ligand
HOTs
[mmol]
Conversion
[%}
2
3/4
Ratio
3+4
Selectivity [%]
G
Selectivity [%]
1
2
3
4
5
6
7
8
–
–
–
0.05
–
0.05
–
0.05
–
0.05
–
0.05
–
0.05
–
0.05
–
0.05
–
0.05
–
0.05
–
0.05
–
0.05
–
0.05
0.05
56
80
14
79
36
52
24
78
28
82
21
73
55
80
100
100
47
75
48
78
53
89
64
81
82
90
84
35
15
57
16
0
0
8
3
7
2
21
7
42
5
0
14.7
4.3
0.0
1.0
0.0
0.3
2.5
20.0
2.3
6.5
0.4
10.7
0.6
1.9
0.0
0.0
7.6
5.3
14.4
18.1
1.5
3.0
0.4
1.1
0.3
17.5
20.6
30
30
6
PPh3
PPh3
P
16
78
73
58
43
71
37
49
22
47
37
92
78
35
36
33
22
74
60
40
43
85
82
93
G
E
P
P(O-di-t-BuC6H3)3
P(O-di-t-BuC6H3)3
P
9
N
G
N
R
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
P
dppe
dppe
dppp
dppp
bcope
bcope
1,2-dppmb
1,2-dppmb
1,3-dppmb
1,3-dppmb
dppf
dppf
bisbi
bisbi
xantphos
xantphos
xantphos[b]
0
34
13
37
18
14
5
35
12
2
0
1
[a]
Reaction conditions: 0.005 mmol [RhACTHNUTRGEN(UNG cod)Cl]2, 5 mmol hexanal, 6 mmol acetamide [Rh:L:hexanal:acetamide=1:1.25 (bi-
dentate) or 2.5 (monodentate):500:600]; PH =80 bar; T=1008C; t=4 h; solvent: 25 mL diglyme; decane as internal stan-
2
dard.
T=808C
[b]
genated products. The use of 1.25 equivalents of xant-
phos on rhodium appears to give the best overall
yield of N-hexylacetamide 3 (entry 2).
Table 3. The effect of Rh/xantphos ratio on the reductive
amidation of hexanal with acetamide.[a]
For the Rh/xantphos/HOTs catalytic system the use
of different ratios of acetamide/hexanal (1/1 up to 2/
1) did not result in large changes in hexanal conver-
sion; it appeared that a small excess of acetamide
(ratio of 1.2) gave the highest conversion of hexanal
and selectivity to 3 and no over-alkylation was ob-
served (see the Supporting Information, Table S4).
Screening of various reaction solvents revealed that
for the Rh/xantphos/HOTs catalytic system diglyme is
the most suitable solvent; the use of tetrahydrofuran
(THF) gave comparable results with slightly lower se-
lectivity to 3. Use of other solvents like toluene, mesi-
tylene, dichloromethane and N-methylpyrrolidone re-
Rh/L
Ratio
Conversion
[%]
3/4
ratio
3+4
Sel. [%]
3
Sel. [%]
1
2
3
4
5
1/1
82
84
76
74
70
17.5
20.6
21.4
22.3
22.5
90
93
82
81
80
85
88
78
78
77
1/1.25
1/1.5
1/2
1/3
[a]
Reaction conditions: 0.005 mmol [RhACTHNUTRGEN(UNG cod)Cl]2, 5 mmol
hexanal, 6 mmol acetamide (Rh:hexanal:acetamide=
1:500:600); HOTs (0.05 mmol); PH =80 bar; T=808C;
t=4 h; solvent: 25 mL diglyme; decane as internal stan-
2
dard.
Adv. Synth. Catal. 2013, 355, 717 – 733
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
721

Saeed Raoufmoghaddam et al.
FULL PAPERS
sulted in significantly lower specificity to 3 as well as
lower combined hydrogenation selectivity 3+4, thus
leading to rather mediocre overall selectivity to 3.
When methanol was used as solvent, full conversion
of hexanal 1 was obtained, however hexanal 1 was
mainly (~90%) converted to the dimethyl acetal and
3 was not obtained (see the Supporting Information,
Table S5).
The effect of the reaction temperature on the hexa-
nal conversion and product distribution is shown in
Figure 3. Lowering of the reaction temperature from
1208C to 808C leads to a pronounced increase in se-
lectivity for 3, from 70% up to 90%. At 808C, the for-
mation of the heavy aldol condensation and diamide
coupled products is suppressed. Upon further lower-
ing of the reaction temperature, however, the conver-
sion of hexanal 1 becomes lower, while the relative
amount of unsaturated intermediates 2 and other
heavier products 5 increases, thus indicating that hy-
drogenation activity to 3+4 decreases with tempera-
ture below 808C. It thus appears that a reaction tem-
perature of 808C is the optimal temperature for this
catalytic system.
Figure 3. The effect of reaction temperature on product dis-
tribution in the reductive amidation reaction using the Rh/
xantphos/HOTs catalytic system. Reaction conditions:
0.005 mmol [RhACHTUNRGTNEUNG(cod)Cl]2, 0.0125 mmol xantphos, 0.05 mmol
HOTs, 5 mmol hexanal, 6 mmol acetamide (Rh:xantphos:H-
OTs:hexanal:acetamide=1:1.25:5:500:600); PH =80 bar; t=
2
4 h; solvent: 25 mL diglyme; decane as internal standard.
&=hexanal (1), &=N-hexylacetamide (3), =unsaturated
N-(1-hexylidene)acetamide or N-(1-hexenyl)acetamide (2),
=hexanol(4), =other products (5).
The effect of the hydrogen pressure on the reduc-
tive amidation was also determined. When lowering
the hydrogen gas pressure from 100 to 20 bar the
overall hydrogenation activity of the catalytic system
As it seemed that the xantphos ligand could play
significantly decreases, thus resulting in increased for- a unique role in the hydroamidation of an aldehyde
mation of 5. For the remaining studies a pressure of to produce N-alkylamides in good selectivity, a variety
80 bar was applied, as this resulted in relatively high of other xantphos-type ligands shown in Figure 4
hexanal 1 conversion combined with good selectivity were applied under the same acidic conditions as ap-
to 3 (see the Supporting Information, Table S7 and plied with xantphos. Catalytic results are collected in
Figure S1).
Table 4.
Figure 4. Selected xantphos-type ligands used in the reductive amidation of hexanal (An=ortho-anisole).
722
asc.wiley-vch.de
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2013, 355, 717 – 733

Rhodium-Catalyzed Homogeneous Reductive Amidation of Aldehydes
Table 4. Reductive amidation of hexanal with acetamide cat-
Table 5. Effect of the addition of various acids or bases on
alyzed by [Rh
(cod)Cl]2 in combination with different xant-
the reductive amidation of hexanal with acetamide using
phos-type ligands.[a]
Rh/xantphos catalytic system.[a]
[b]
Ligand
Conv.
[%]
2
3/4
3+4
Additive
(0.05 mmol)
pKa
Conv. 3/4
3+4
3
Sel. [%] Ratio Sel. [%]
[%]
Ratio Sel. [%] Sel. [%]
1
2
3
4
5
6
7
8
9
xantphos
t-Bu-xantphos
Si-xantphos
oMeO-xantphos 60
xantphos
DPEphos
DBFphos
homoxantphos
benzoxantphos
84
80
78
1
4
20.6
16.9
4.1
6.2
10.2
6.3
8.8
1.1
5.9
93
88
74
12
19
34
27
39
26
1
2
3
4
5
6
7
8
9
no acid
HOTs[c]
HOTs
76
0.3
2.4
20.6
24.5
30.8
4.4
0.1
0.4
0.7
1.6
84
84
93
87
84
88
86
76
76
78
82
80
87
86
91
21
60
88
83
81
72
9
21
32
49
76
76
84
3
ꢀ2.7 78
ꢀ2.7 84
ꢀ2.7 92
ꢀ2.7 98
ꢀ2.7 86
11
31
28
28
29
24
28
HOTs[d]
HOTs[e]
HOTs[f]
AcOH
G
59
64
50
60
47
4.8
2
1.8
70
68
68
H3PO4
PhH2PO3
10 TFA
11 HBF4
12 HCl
13 HOTf
14 NMP
15 NEt3
ꢀ0.3 74
[a]
Reaction
conditions:
0.005 mmol
[RhACTHNGUTERNN(UG cod)Cl]2,
ꢀ4
ꢀ7
66
70
12.5
20.8
26.0
0.0
0.0125 mmol ligand, 0.05 mmol HOTs, 5 mmol hexanal,
6 mmol acetamide (Rh:L:hexanal:acetamide=
1:1.25:500:600); PH =80 bar; T=808C; t=4 h; solvent:
ꢀ5.1 90
2
72
90
25 mL diglyme; decane as internal standard.
0.0
0
[a]
Reaction
conditions:
0.005 mmol
[RhACTHNGUTERNN(UG cod)Cl]2,
0.0125 mmol xantphos, 5 mmol hexanal, 6 mmol acet-
amide Rh:xantphos:hexanal:acetamide=1:1.25:500:600);
The best results were obtained with xantphos and t-
Bu-xantphos (Table 4, entries 1 and 2). Remarkably,
even relatively small modifications, such as the use of
Si-xantphos or Benzo-xantphos instead of xantphos,
in the catalytic system already resulted in a very sig-
nificantly lower 3/4 ratio and lower selectivity to 3
(entry 3). Using the other xantphos-type ligands also
resulted in significant suppression of the 3/4 ratio and
selectivity to 3+4. Apparently this is mainly due to
a lower hydrogenation activity of these catalytic sys-
tems as also reflected by a significant increase in for-
PH =80 bar; T=808C; t=4 h; solvent: 25 mL diglyme;
2
decane as internal standard.
The pKa values are determined in water as solvent (see
ref.[8]).
0.025 mmol HOTs.
0.1 mmol HOTs.
[b]
[c]
[d]
[e]
[f]
0.2 mmol HOTs.
HOTs 12% in acetic acid.
mation of unsaturated intermediates 2; accumulation stronger acids leads to progressively higher selectivity
of these intermediates also results in increased to 3 (entries 10–13); with triflic acid (HOTf) excellent
amounts of 5 (entries 4–9).
conversion with high 3/4 ratio and high selectivity is
obtained (entry 13). Conversely, the use of the weak
base N-methylpyrrolidone (NMP) or stronger base
triethylamine (NEt3) both result in the formation of
hexanol 4 as the exclusive hydrogenation product, re-
Effect of Acid and Base Additives
As shown above the addition of HOTs to the reduc- markably with a similar hexanal 1 conversion rate as
tive amidation reaction mixture has a significant obtained under acidic conditions (entries 14 and 15).
effect on the selectivity of the catalytic system based
on rhodium/xantphos. This prompted us to investigate
the effect of acids or bases as additives in the reaction Effect of Acid on Hydrogenation of Aldehyde
in more detail (Table 5).
Increasing the amount of HOTs from 0.025 to To elucidate the selectivity-improving role of acid ad-
0.05 mmol positively affects the 3/4 product ratio and dition in reductive amidation as apparent from the
the selectivity in favour of 3 (entries 2 and 3). Addi- spectacular increase of the 3/4 product ratio with the
tion of a larger amount of HOTs (e.g., 8 mM; HOTs/ rhodium/xantphos catalytic system the influence of
Rh~20) leads only to a slight decrease in selectivity acid addition on the hydrogenation of hexanal 1 to
predominantly due to the formation of more aldol hexanol 4 was separately investigated, both in the ab-
condensation products 5 (entries 4 and 5). With the sence and presence of an inert amide model com-
addition of the same amounts of significantly weaker pound, N,N’-dimethylacetamide (Table 6).
acids such as acetic acid, phosphoric acid and phenyl-
In the absence of HOTs and acetamide, the rhodi-
phosphonic acid, hexanol 4 becomes the dominant hy- um/xantphos catalytic system hydrogenates hexanal
drogenation product (entries 7–9), quite similar to the 1 to hexanol 4 with about 50% conversion after 4 h at
experiment in which no acid is added. The use of 808C (entry 1). Addition of a catalytic amount of
Adv. Synth. Catal. 2013, 355, 717 – 733
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
723

Saeed Raoufmoghaddam et al.
FULL PAPERS
Table 6. Hydrogenation and reductive amidation of hexanal
Catalyst Systems other than those Based on Rh
with and without acetamide.[a]
Under the optimized reaction conditions shown
above, we tested various other transition metal pre-
cursors (based on Ru, Pd, Pt, Ir and Co) in combina-
tion with xantphos – with or without HOTs – in the
reductive amidation reaction (see the Supporting In-
formation, Table S13). None of the used complexes
showed a performance comparable with the Rh cata-
lysts. The catalytic system containing Rh/xantphos/
HOTs turned out to be by far the most active and se-
lective catalytic system for reductive amidation.
Amide
(6 mmol)
HOTs
ACHTUNGTRENNUNG
Conv. 3/4
3+4
4
G
Ratio Sel. [%] Sel. [%]
1
2
3
4
5
6
–
–
52
14
54
26
76
84
–
–
–
–
0.3
20.6
–
–
–
–
84
93
85
9
81
62
63
4
0.05
0.05
DMA[b]
DMA[b]
acetamide
acetamide 0.05
[a]
Reaction
conditions:
0.005 mmol
[RhACTHNGUTERNN(UG cod)Cl]2,
0.0125 mmol xantphos, 5 mmol hexanal, in some cases
6 mmol amide (Rh:xantphos:hexanal=1:1.25:500); PH
=
80 bar; T=808C; t=4 h; solvent: 25 mL diglyme; deca2ne
as internal standard.
Product Evolution in Time
[b]
N,N’-Dimethylacetamide.
The product development over the time for the Rh/
xantphos/HOTs catalytic system is depicted in
HOTs causes a significant drop in conversion; assum- Figure 5. After ten hours at 808C nearly full conver-
ing first order reaction kinetics one can calculate a de- sion (98%) (TON~500) is reached with about 88%
crease in hydrogenation rate constant of about selectivity to 3. It is shown that the unsaturated inter-
a factor 5 by the presence of acid.[9] The effect on the mediates 2 do not build up to any significant extent in
selectivity of hydrogenation is even more dramatic, this experiment, while the aldehyde concentration
due to a competing acid-catalyzed aldol condensation monotonously decreases according to an approxi-
of hexanal 1 giving 5 (entry 2). The use of N,N’-dime- mately first-order reaction. This could indicate that
thylacetamide, added to the reaction mixture to simu- formation of the unsaturated amide adducts is rate-
late a similar reaction medium as applied in actual re-
ductive amidation, gives a qualitatively similar result
(cf. entries 3 and 4). However, likely due to the weak
basicity of N,N’-dimethylacetamide the addition of
HOTs now leads to a decrease in hydrogenation rate
constant only by a factor of about 2.5, while maintain-
ing a reasonably good selectivity for hydrogenation to
4. For comparison, the data for the hexanal reduction
in the presence of acetamide are summarized (en-
tries 5 and 6): without acid almost exclusively 4 is
formed as the reduction product, while in the pres-
ence of a catalytic quantity of HOTs, 3 is the almost
exclusive reduction product.
Influence of Added Water
As in the reductive amidation of an aldehyde one
equivalent of water is formed, the effect of added
water in the reaction media was also investigated. Ad-
dition of 2.5 mmol of water, theoretically correspond-
ing to about 50% conversion of hexanal 1, already
Figure 5. Product development in time in the reductive ami-
has a significant impact on the reaction; still ~75%
dation of hexanal with acetamide; Reaction conditions:
0.005 mmol [RhACHTUNRGTNEUNG(cod)Cl]2, 0.0125 mmol xantphos, 0.05 mmol
conversion is reached, but the selectivity to 3 drops to
about 70%. Increasing the amount of water to
5 mmol has a more dramatic effect, in particular on
selectivity to 3. The addition of molecular sieves
(3 ꢂ) to the reaction does not have a significant
effect on the conversion, but seems to have a slightly
positive effect on the 3/4 ratio (see the Supporting In-
formation, Table S12).
HOTs, 5 mmol hexanal, 6 mmol acetamide (Rh:xantphos:
HOTs:hexanal:acetamide=1:1.25:5:500:600); PH =80 bar;
T=808C; solvent: 25 mL diglyme; decane as inte2rnal stan-
~
&
^
dard. =hexanal (1), =N-hexylacetACTHNUTRGNEUNGamide (3), =unsatu-
rated N-(1-hexylidene) acetamide or N-(1-hexenyl)aceta-
mide (2), ꢃ =hexanol (4), *=other products (5) (the exact
data are given in the Supporting Information, Table S14)
724
asc.wiley-vch.de
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2013, 355, 717 – 733

Rhodium-Catalyzed Homogeneous Reductive Amidation of Aldehydes
determining under the prevailing conditions, and that Applying p-methoxybenzamide with an electron-do-
hydrogenation of these intermediates is a relatively nating group gives higher selectivity and conversion,
fast reaction.
whereas p-trifluoromethylbenzamide with an elec-
The robustness of the catalytic system in the opti- tron-withdrawing group again results in considerably
mized reaction conditions was investigated using lower selectivity and conversion. Clearly, the nucleo-
10 mmol of substrate with 0.0025 mmol of rhodium philicity of the N atom in the amide plays a decisive
precursor in combination with xantphos and HOTs at role in a successful reductive amidation of aldehyde.
808C. After 10 h 75% conversion of hexanal 1 was
The branched aldehyde 2-methylpentanal and the
reached with 60% overall selectivity to 3. The cumu- electronically different benzaldehyde, (entries 10 and
lative turnover number thus obtained was about 1800, 11) are considerably less reactive in reductive amida-
i.e., quite an acceptable number considering the scale, tion with acetamide; not only the conversion is lower,
water build-up and low catalyst concentration applied but also the hydrogenation activity and substrate spe-
in this experiment.
cificity (ratio 3*/4*) quite drastically decreases.
Scope of Reductive Amidation; Applying Different
Amide and Aldehyde Substrates
NMR Study of the Hexanal/Acetamide Mixture in
the Absence of Rh Catalyst
A number of different N-nucleophiles were applied in The adduct formation of hexanal 1 with acetamide
the reductive amidation of hexanal 1 (Table 7). The was studied by following the time evolution of
use of amides with one electron-donating alkyl group 1H NMR spectra of a 1:1 mixture hexanal/acetamide
at the alpha C of the amide such as propanamide and in THF-d8 at 608C in the absence of the rhodium/
pentanamide, results in higher conversion and selec- ligand catalytic system. It appears that condensation
tivity (entries 1, 2, 5). The use of amides with two or of hexanal 1 with acetamide in THF-d8 does not pro-
three alpha-C alkyl substituents leading to bulkier al- ceed in the absence of HOTs (see the Supporting In-
kylamides, such as isopropylamide and tert-butyl- formation). However, as shown in Figure 6, in the
ACHTUNGTRENNUNGamide, results in a somewhat lower efficiency of the presence of only 1 mol% of HOTs already after
reaction (entries 3 and 4). For comparison, application 10 min adduct formation between hexanal 1 and acet-
of the electron-withdrawing fluoroacetamide results amide is observed, as indicated by the simultaneous
in a very significant drop in the selectivity to 3* formation of proton resonances at 5.25 ppm attributed
(entry 6). The use of benzamide also results in lower to -CH(OH) (2a*) and at 7.65 ppm attributed to NH
conversion and selectivity compared to acetamide. of 2a (Figure 6).
Table 7. Scope of reductive amidation using various amides and aldehydes.[a]
Aldehyde
Amide
Conversion[b]
[%]
3*/4*
Ratio
3*+4*
Sel. [%]
3*
Sel. [%][c]
1
2
3
4
5
6
7
8
hexanal
hexanal
hexanal
hexanal
hexanal
hexanal
hexanal
hexanal
acetamide
propanamide
84
90
77
75
92
32
58
72
38
57
42
21
24
13
12
31
2
8
14
4
92
93
85
87
95
51
81
85
73
41
63
88
90
79
81
92
35
72
80
58
31
9
isopropylamide
tert-butylamide
pentanamide
fluoroacetamide
benzamide
para-methoxybenzamide
para-trifluoromethylbenzamide
acetamide
9
10
11
hexanal
2-methylpentanal
benzaldehyde
3
0.1
acetamide
[a]
[b]
[c]
Reaction conditions: 0.005 mmol [RhACTHNURTGNENG(U cod)Cl]2, 0.0125 mmol xantphos, 0.05 mmol HOTs, 5 mmol aldehyde, 6 mmol
amide (Rh:xantphos:HOTs:hexanal:amide=1:1.25:5:500:600); PH =80 bar; T=808C; t=4 h; solvent: 25 mL diglyme;
2
The conversion was determined by GC analysis based on the amount of aldehyde found in the reaction mixtures after
the reaction.
The selectivity was determined by GC analysis using decane as an internal standard.
Adv. Synth. Catal. 2013, 355, 717 – 733
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
725

Saeed Raoufmoghaddam et al.
FULL PAPERS
1
Figure 6. 1H NMR time evolution of part of the H NMR spectra relevant for the condensation reaction of hexanal with
acetamide; hexanal:acetamide:PTSA=1:1:0.01 ratio in THF-d8; T=608C.
The intensity of these resonances keeps growing up
From the time evolution of the proton resonances
to about 150 min; after this time resonances attribut- representative for hexanal, acetamide, 2a and 2 the
ed to the dehydrated product 2 appear [the peak at time-dependent concentration of these compounds
6.95 ppm can be assigned to the imide functionality- has been constructed and is graphically depicted in
CH=N (2**)].[10] The evolution of the proton resonan- Figure 7. From this it is immediately clear that the
ces is thus consistent with a two-stage addition pro- two stages in the acetamide–aldehyde addition pro-
ꢀ
cess of the acetamideꢁs N H bond to hexanal 1, ini- cess both involve equilibrium processes. Two separate
tially forming the nucleophilic addition product 2a equilibrium constants can be calculated from these
followed by dehydration to give 2. Small proton reso- data. The first equilibrium constant (K1 =2.701Mꢀ1)
nances at 5.10 and 6.70 ppm can be assigned to the shows that in the presence of a catalytic amount of
other isomeric dehydration product, N-(1-hexenyl)- HOTs, addition of acetamide to hexanal forming 2a is
ACHTUNGTRENNUNG
acetamide 2;[10] however, the integrals of these reso- an equilibrium reaction that proceeds under the pre-
nances remain relatively small under our condi- vailing conditions (hexanal:acetamide=1:1; THF sol-
tions.[11]
vent; 608C) to slightly less than 50% hexanal conver-
sion. The subsequent dehydration reaction forming 2
726
asc.wiley-vch.de
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2013, 355, 717 – 733

Rhodium-Catalyzed Homogeneous Reductive Amidation of Aldehydes
ly spontaneous addition of the amine to aldehyde,
forming imines (or carbinolamines) under very mild
conditions. Clearly, a reduced nucleophilicity of the N
atom of amides relative to amines must be responsi-
ble for this difference.
Discussion
The reductive amidation of hexanal 1 with acetamide
comprises two sequential steps as shown in Scheme 4.
The first step consists of two consecutive equilibrium
reactions: the nucleophilic addition of acetamide to
hexanal 1 yielding the intermediate 2a which may un-
dergo dehydration to form the two isomeric unsatu-
rated intermediates 2, i.e-, N-(1-hexylidene)acetamide
and to a lesser extent N-(1-hexenyl)acetamide. Irre-
versible hydrogenation of these unsaturated inter-
mediates will subsequently result in the formation of
desired product 3. Direct hydrogenolysis of 2a to give
3 may also occur, as has also been suggested for inter-
mediate N-carbinolamines in the reductive amination
of aldehyde with secondary amines.[4a,12] Formation of
4 by the hydrogenation of hexanal 1 constitutes a loss
of starting material, thus resulting in lower selectivity
Figure 7. Development of the relative intensities of resonan-
ces assigned to hexanal (9.72 ppm), N-(1-hydroxyhexyl)acet-
ACHTUNGTRENNUNG
ACHTUNGTRENNUNG
1:1:0.01 ratio in THF-d8; T=608C. Measuring in the first
^
hour every 10 min and then every 60 min until 20 h.
hexanal (1), =2a, =2.
=
~
&
proceeds only to a limited extent, as shown by the ap- for reductive amidation.
parent low value of the second equilibrium constant
An efficient catalytic system for the reductive ami-
(K2 =0.042Mꢀ1) which implies that only relatively low dation reaction thus comprises a catalyst that has a sig-
concentrations of 2 can be present in the reaction nificantly higher rate of hydrogenation of the unsatu-
mixture at equilibrium.
rated intermediates 2 (kim) (and/or hydrogenolysis of
However, it must be noted that for this model 2a) relative to hydrogenation of the aldehyde (kald) in
study, a low reaction temperature of 608C was applied combination with rapid establishment of the equilibri-
for practical reasons. One may expect the dehydration um addition reaction of the aldehyde with the amide.
equilibrium (K2) to shift to higher conversion at the The water formed in the dehydration step in the
higher temperatures used in the catalytic experiments second equilibrium eventually will have an adverse
(80–1008C). We conclude from this model study that effect on this equilibrium. Applying an active and se-
ꢀ
the addition of an acetamideꢁs N H to the aldehyde lective catalytic system for the hydrogenation of the
proceeds in two consecutive stages catalyzed by unsaturated intermediates 2 (or hydrogenolysis of 2a)
(strong) acid. In this respect the addition of acetamide not only advances the reductive amidation reaction
strongly deviates from addition of a more nucleophilic by irreversibly removing these intermediates from the
amine to aldehyde which generally involves a thermal- equilibria, but also avoids accumulation of the unsatu-
Scheme 4. Catalytic reductive amidation of hexanal; formation of N-hexylacetamide 3 via N-(1-hydroxyhexyl)acetamide 2a
and imide or enamide intermediates 2. Hexanol 4 is a by-product of the reaction.
Adv. Synth. Catal. 2013, 355, 717 – 733
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
727

Saeed Raoufmoghaddam et al.
FULL PAPERS
Scheme 5. Catalytic reductive amidation; formation of other products characterized by NMR, GC and GC-MS.
rated intermediates that eventually may form heavier reaches equilibrium at 608C in about 10 h. At a practi-
by-products. Possible routes to the higher-mass cou- cal temperature of 80–1008C of the catalytic reductive
pling products are shown in Scheme 5. In the absence amidation reaction, this process will be faster. The
of an active hydrogenation catalyst the intermediates proposed mechanism is consistent with the acid-cata-
2 may react with a second molecule of acetamide to lyzed nucleophilic addition of inactivated amines to
form the di-coupled product 5b. The other competing a carbonyl functionality as described in the litera-
side-reactions proceed via the aldol condensation of ture.[13] The dehydration of the intermediate N-1-car-
hexanal 1 yielding product 5a; this aldol product can binolamide 2a to form the unsaturated acetamides 2
also undergo nucleophilic attack of acetamide to form should also be acid catalyzed, but only occurs to a rel-
5a’ and the di-coupling reaction with acetamide giving atively small extent (K2 =0.042Mꢀ1) at 608C.
5a’’.
Influence of Ligands, Reagents and Reaction
Conditions
Influence of Acid
For nearly all catalytic systems that have been The use of different rhodium precursor complexes in
screened in our study, the addition of a catalytic the absence of ligand results in catalytic systems with
amount of acid results in higher conversion of hexanal reasonable conversion of hexanal 1, but low selectivi-
1 and improved selectivity to 3. The formation of ty. Notably, the use of rhodium precursors with car-
aldol condensation and di-coupling products is also bonyl ligands results in lower conversions. The addi-
promoted by the introduction of acid, as is clearly ob- tion of various monodentate or bidentate phosphorus
served for those catalytic systems that have low hy- ligands to [Rh
ACHTUNGTNER(NUNG cod)Cl]2 generally leads to higher con-
drogenation activity. version of hexanal. The use of electron-donating al-
The results of the NMR study show that the nucle- kylphosphane ligands results in catalytic systems that
ꢀ
ophilic addition of acetamideꢁs N H to hexanal clear- are highly selective for hydrogenation of hexanal to 3.
ly is promoted by acid (Scheme 6); in absence of acid The use of xantphos results in a catalytic system that
(and hydrogenation catalyst) the addition reaction shows high conversion to 4 and only a very small
does not proceed at all, whereas in the presence of amount of 3, while the intermediates 2 are observed
acid the reaction to 2a and the dehydrated products 2 only in trace amounts. Remarkably, upon the addition
Scheme 6. Proposed mechanism for the acid-catalyzed condensation reaction of hexanal with acetamide.
728
asc.wiley-vch.de
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2013, 355, 717 – 733

Rhodium-Catalyzed Homogeneous Reductive Amidation of Aldehydes
of a catalytic amount of HOTs the desired reductive
amidation product 3 is obtained with high selectivity
of about 90% and a 3/4 product ratio > 20. This
shows that the addition of HOTs not only leads to
a more rapid establishment of the equilibrium be-
tween aldehyde and enamide/imide as shown by our
NMR study, but apparently also leads to significant
changes of the relative rates of hydrogenation of the
unsaturated amides 2 (giving 3) and aldehyde (giving
4, Scheme 4).
Generally, the conversion is strongly affected by
the choice of the N-substrates, which supports the
idea that the coupling reaction is dependent on the
nucleophilicity and bulkiness around nitrogen of the
Scheme 7. Proposed pathways for the formation of the dif-
ferent catalyst precursors for the reductive amidation reac-
tion in the presence or absence of acid; &=open site or sol-
vent molecule.
It is our hypothesis that species B is a selective cat-
substrates. Lowering of the reaction temperature alyst for the hydrogenation of hexanal (Scheme 8).
from 1208C to 808C results in formation of less aldol The neutral, low-valent Rh(I) compound would
condensation products, and thus in higher overall (3+ favour the binding of the aldehyde via its p electrons,
4) hydrogenation selectivity. However, at tempera- with subsequent migration of the hydride resulting in
tures lower than 808C the hydrogenation activity of
the catalysts is also drastically decreased, which
lowers the overall hexanal 1 conversion as well as the
hydrogenation selectivity; a higher aldehyde concen-
tration profile during the reaction period will general-
ly lead to higher amounts of heavy aldol products.
Also, lower hydrogen gas pressures in the reaction
lead to lower hydrogenation activity, consequently re-
sulting in the accumulation of intermediates that
eventually yields higher amounts other products 5.
Mechanistic Considerations
The catalytic system based on rhodium/xantphos in
the absence of added acid, but in the presence of the
amide, predominantly hydrogenates hexanal 1 to 4;
intermediates 2 are observed only in trace amounts.
However, upon the addition of a catalytic amount of
HOTs the desired product 3 is obtained with high se-
Scheme 8. Proposed catalytic cycle for hydrogenation of
hexanal.
lectivity.
One reason for the high specificity in the hydroge-
nation of unsaturated amides (or hydrogenolysis of the formation of the Rh(I) alkoxide species D. In the
2a) lies in the intrinsically lower activity for aldehyde final step, heterolytic activation of dihydrogen yields
hydrogenation of the Rh/xantphos species generated 4 with the regeneration of mono-hydride species B. In
in the presence of acid as shown in Table 7 (cf. en- this mechanism, the xantphos ligand may bind in
tries 1 vs. 2 and 3 vs. 4).
In Scheme 7 a proposed pathway is given for the in its Rh(I) state throughout the catalytic cycle.
formation of the different catalytic species in the pres- On the other hand, the cationic, high-valent species
ence or absence of acid. Starting from [Rh(I)(cod)Cl]2 E, [Rh(III)
(xantphos)(H)2]+ would favour binding of
and xantphos, the complex [Rh(I)(xantphos)Cl] A is the imido intermediate 2 through both the carbonyl
a meridional fashion, stabilizing the catalytic species
A
R
ACHTUNGTRENNUNG
AHCTUNGTRENNUNG
initially formed. In the absence of acid, heterolytic oxygen and the imide double bond over the p-bond
splitting of dihydrogen may result in the formation of of the aldehydeꢁs carbonyl, forming species [Rh-
the neutral mono-hydride species [Rh(I)(xant-
phos)H], B. In the presence of acid the cationic dihy- soning could hold for possible N-(1-hexylidene)aceta-
dride species [Rh(III)
(xantphos)(H)2]+ E may be mide. Then, hydride migration to the imide carbon
(imide)]+ F (Scheme 9). A similar rea-
ACHTUGTNRNENUG(xantphos)(H)2ACHTUNGTRENNUGN
ACHTUNGTRENNUNG
A
ACHTUNGTRENNUNG
formed by protonation of the monohydride Rh(I) atom results in the formation the N-hexylacetamido
compound.[14]
species G. Finally, N-hexylacetamide 3 is released via
reductive elimination to form the Rh(I) species H;
Adv. Synth. Catal. 2013, 355, 717 – 733
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
729

Saeed Raoufmoghaddam et al.
FULL PAPERS
Scheme 9. Proposed catalytic cycle for the hydrogenation of intermediate N-(hexenylidene)acetamide 2 or N-(1-hydroxyhex-
yl)acetamide 2a; &=open site or solvent molecule.
subsequent oxidative addition of dihydrogen recovers
species E. In this mechanism, the rhodium center is ence of an acid (and at relatively low temperatures),
cationic and shuttling between the Rh(I) and Rh(III) instead of one of the unsaturated intermediates 2 the
From the NMR studies it is clear that in the pres-
AHCTUNGTRENNUNG
oxidation state. The selective binding and subsequent addition product, carbinolamide 2a, is predominantly
H-migration involving the N-(1-hexylidene)acetamide present in solution. Thus, alternatively, this bifunc-
and N-(1-hexenyl)acetamide intermediates
2
is tional carbinolamide may also bind selectively to spe-
thought to contribute to the unusually high substrate cies E in the presence of aldehyde, effectively giving
specificity for hydrogenation of 2 that at all times selective direct dehydrative hydrogenolysis of the al-
during the reaction is present in only relatively low cohol group via a species F’.
concentrations in equilibrium with the aldehyde. An
The POP-pincer structure in xantphos-type ligands
additional factor disfavouring the hydrogenation of has been shown to be important for the stabilization
aldehyde could be that, in contrast with the Rh(I) of catalytic intermediates in different reactions such
center in the absence of acid as proposed in as intramolecular hydroamination,[15] methanol car-
Scheme 8, the now strongly electrophilic cationic di- bonylation,[16] and intermolecular hydroacylation of
hydride RhACHTUNGTRENNUNG
(III) center would favour coordination of alkenes and alkynes.[17] Structural studies have also
the hard carbonyl oxygen over p-coordination of the shown the importance of the flexibility of this ligand,
carbonyl bond. This carbonyl oxygen binding mode is with the possible trans bidentate and fac or mer tri-
not a suitable configuration for hexanal 1 hydrogena- dentate coordination modes.[17,18] Moreover, the crys-
tion, as in this case hydride migration to the carbonyl tal structure of a Rh
carbon is not easily possible.
ACHTUNGTNER(NUNG III) compound containing a 4-
730
asc.wiley-vch.de
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2013, 355, 717 – 733

Rhodium-Catalyzed Homogeneous Reductive Amidation of Aldehydes
thyl)benzene (1,2-dppmb), 1,3-bis(diphenylphosphanylme-
membered imidocarbonyl chelate similar to our pro-
posal for species G has recently been reported.[19]
thyl)benzene (1,3-dppmb), 9,9-dimethyl-4,5-bis(diphenyl-
phosphanyl)xanthene (xantphos) and 2,2’-bis(diphenylphos-
phanylmethyl)biphenyl (bisbi) were purchased from Strem
Chemicals, Germany. The other phosphorus ligands such as
1,2-bis(cyclooctylphosphanyl)ethane (bcope) and 2,7-di-tert-
butyl-9,9-dimethyl-4,5-bis(diphenylphosphanyl)xanthene (t-
Bu-xantphos), 4,6-bis(diphenylphosphanyl)-10,10-dimethyl-
Conclusions
We have successfully developed a highly selective cat-
alytic system comprising rhodium/xantphos/HOTs for
the reductive amidation of an aldehyde with an amide
to form N-alkylamides in high yield and under mild
conditions. We have achieved a promising TON of
1800 with use4 of only a small excess of amide sub-
strate. The use of an acid co-catalyst appears to be
important for both the nucleophilic addition of the
amide to the aldehyde as well as the substrate specif-
icity in the hydrogenation step. The selective reduc-
tive amidation of aldehydes bears resemblance with
the well-known transition metal-catalyzed reductive
amination of aldehydes, but which often suffers from
the formation of substantial amounts of side-products
originating from (base-catalyzed) aldol condensation
and over-alkylation.
We are currently seeking to utilize our reductive
amidation catalytic system for a ꢄhydroamidomethyla-
tionꢁ reaction, comprising a tandem hydroformyla-
tion–reductive amidation process of an alkene with
syngas and an amide. For this purpose the reductive
amidation catalyst must of course be able to tolerate
the presence of carbon monoxide necessary for the
hydroformylation step. Preliminary investigations
have shown that the presence of traces CO in the re-
ductive amidation reaction results in quenching of the
hydrogenation activity and selectivity of the catalyst.
Further studies are directed at ways to reduce or
eliminate the CO sensitivity of the catalytic system.
10H-dibenzoACTHNUTRGENN[UG b,e]ACHTUNGTERNN[UGN 1,4]oxasilane (Si-xantphos), 9,9-dimethyl-
4,5-bis(di-ortho-methoxyphenyl-phosphanyl)xanthene
(oMeO-xantphos) and 4,5-bis[di(tert-butyl)phosphanyl]-9,9-
dimethylxanthene (di-t-Bu-xantphos) were generously pro-
vided by Shell Global Solutions Amsterdam b.v., where they
were synthesized according to literature procedures.[20]
Apparatus
The stainless steel autoclave reactors (100 mL) of HEL Lim-
ited, UK, were equipped with magnetic stirrer, pressure
transducer and temperature controlling thermocouple. A
Hewlett Packard HP6890 Series auto-sampler GC system
was used for regular GC analysis. GC-MS analyses were car-
ried out on an Agilent technologies 7820 A GC system
series coupled with an Agilent technologies 5975 series GC-
MSD system. Nuclear magnetic resonance spectra were re-
corded on a Bruker DPX300 (300 MHz) or a Bruker
DMX400 (400 MHz) spectrometer. A glovebox of M. Braun
Inertgas-System GmbH, Germany, was used for storing and
handling of air-sensitive phosphane ligands.
Procedures
All preparations and manipulations were performed using
standard Schlenk techniques under an argon atmosphere.
The solvent bis(2-methoxyethyl) ether (diglyme) was dis-
tilled from CaH2, deoxygenated and then saturated with
argon and used immediately after the purification process.
The catalytic reactions were carried out under varying hy-
drogen pressures and reaction temperatures. For all the cat-
alytic experiments the catalyst precursor was formed in situ
in the autoclave by transferring the metal precursor and the
selected phosphane ligands into the reactor.
Experimental Section
Chemicals
Hexanal, acetamide, propionamide, isopropylamide, tert-bu-
tylamide, fluoroacetamide, valeramide, benzamide, p-me-
thoxybenzamide, p-trifluoromethylbenzamide, 2-methylpen-
tanal, benzaldehyde, decane (internal standard), hexanol,
para-toluenesulfonic acid, chlorido(1,5-cyclooctadiene)rho-
dium(I) dimer, bis(1,5-cyclooctadiene)rhodium(I) trifluoro-
methanesulfonate, (acetylacetonato)dicarbonylrhodium(I),
carbonylhydridotris(triphenylphosphane)rhodium(I), bis(cy-
cloocta-1,5-diene)rhodium(I) tetrafluoridoborate, bis(2-me-
thoxyethyl) ether (diglyme) and other solvents, acids and
bases were purchased from Acros Organics and Sigma Al-
drich, the Netherlands. The monodentate ligands triphenyl-
In the preparation of a typical catalytic reaction mixture
containing solid ingredients that are not air sensitive,
0.005 mmol of [RhACTHNUTRGNE(UNG cod)Cl]2 (2.46 mg, 0.01 mmol of Rh) and
0.0125 mmol of bidentate phosphane ligand or 0.025 mmol
of monodentate phosphane ligand were weighed in air and
transferred into an autoclave. The autoclave was closed and
subsequently filled with argon using a Schlenk line connect-
ed to one of the valves of the autoclave. Through another
valve under a continuous flow of argon subsequently were
added: 25 mL of dried and degassed diglyme as solvent,
3.125 mmol (0.605 mL) decane as an internal standard,
5 mmol (0.615 mL) hexanal and 6 mmol (0.354 g) acetamide
(for selected experiments acid or another additive was
added, dissolved in the diglyme). Then the reactor was in-
serted into the heating block and connected to the gas line
of the reactor block with a continuous flow of N2 gas
through the gas line of autoclave to remove the air inside
the gas line. The autoclave reactor was flushed three times
phosphane (PPh3), tri-n-butylphosphane [PACHTUNGTRENN(UNG n-Bu)3], n-butyl-
di-1-adamantylphosphane [P(n-Bu)(1-ad)2], tris(2,4-di-tert-
A
ACHTUNGTRENNUNG
butylphenyl)phosphite [P(O-di-t-BuC6H3)3], the bidentate li-
gands 1,2-bis(diphenylphosphanyl)ethane (dppe), 1,3-bis(di-
phenylphosphanyl)propane (dppp), 1,1’-bis(diphenylphos-
phanyl)ferrocene (dppf), 1,2-bis(diphenylphosphanylme-
Adv. Synth. Catal. 2013, 355, 717 – 733
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
731

Saeed Raoufmoghaddam et al.
FULL PAPERS
with N2 and one time with H2 and finally pressurized with
hydrogen gas. The reaction mixture was stirred at 500 rpm
for 20–30 min to ensure complex formation.
In the set of experiments using air-sensitive ligands such
as PACHTUNGTRENNUNG(n-Bu)3, PACHTUNGTRENNUNG(n-Bu)ACHTNUGTNER(NUGN 1-ad)2, bcope and di-t-Bu-xantphos, in
a glove box the metal precursor complex and ligand (for
some experiments with other additives like acid), were
weighed into a Schlenk flask and dissolved in 10 mL of
dried and degassed diglyme; dissolution generally was com-
plete in about 2–3 min as was visible by the formation of
a transparent yellowish solution. The flask was then con-
nected to a Schlenk line and the solution was transferred
through the valve of a reactor under a continuous flow of
argon into the 100 mL stainless steel autoclave reactor. The
procedure for gas intake in the autoclave was carried out as
described above.
[2] a) I. Kadota, A. Shibuya, L. M. Lutete, Y. Yamamoto,
J. Org. Chem. 1999, 64, 4570; b) L. M. Lutete, I.
Kadota, Y. Yamamoto, J. Am. Chem. Soc. 2004, 126,
1622; c) M. Utsunomiya, J. F. Hartwig, J. Am. Chem.
Soc. 2003, 125, 14286; d) M. Utsunomiya, J. F. Hartwig,
J. Am. Chem. Soc. 2004, 126, 2702; e) M. Utsunomiya,
R. Kuwano, M. Kawatsura, J. F. Hartwig, J. Am. Chem.
Soc. 2003, 125, 5608; f) T. E. Muller, K. C. Hultzsch, M.
Yus, F. Foubelo, M. Tada, Chem. Rev. 2008, 108, 3795.
[3] a) M. Arndt, K. S. M. Salih, A. Fromm, L. J. Goossen,
F. Menges, G. Niedner-Schatteburg, J. Am. Chem. Soc.
2011, 133, 7428; b) S. Obika, Y. Yasui, R. Yanada, Y.
Takemoto, J. Org. Chem. 2008, 73, 5206; c) N. T. Patil,
Z. B. Huo, G. B. Bajracharya, Y. Yamamoto, J. Org.
Chem. 2006, 71, 3612; d) Y. S. Salprima, Y. Kuninobu,
K. Takai, Org. Lett. 2007, 9, 5609; e) Y. Yu, G. A. Ste-
phenson, D. Mitchell, Tetrahedron Lett. 2006, 47, 3811.
[4] a) S. Gomez, J. A. Peters, T. Maschmeyer, Adv. Synth.
Catal. 2002, 344, 1037; b) V. I. Tararov, R. Kadyrov,
T. H. Riermeier, A. Borner, Chem. Commun. 2000,
1867.
The reaction mixtures were heated up to 808C, 1008C or
1208C (within 30 min) under stirring at 500 rpm. All reac-
tion conditions of the catalytic process were controlled by
computerized software panels. After standing for two, four
or ten hours at this temperature, the autoclave was cooled
down to room temperature over about one hour. The auto-
clave was then slowly vented to atmospheric pressure.
After each catalytic run the reaction mixture was taken
from the reactor and immediately analyzed with gas chro-
matography. Calibration lines for each analyte were used to
determine the conversion of the substrates and yields of the
various products. The assignments of the products were con-
firmed with GC-MS and comparison with authentic and
pure commercial samples. The products in Table 7 compris-
ing N-hexylacetamide,[21] N-hexylpropanamide,[21b,22] N-
[5] a) M. Beller, C. Breindl, M. Eichberger, C. G. Hartung,
J. Seayad, O. R. Thiel, A. Tillack, H. Trauthwein, Syn-
lett 2002, 1579; b) J. Seayad, A. Tillack, C. G. Hartung,
M. Beller, Adv. Synth. Catal. 2002, 344, 795.
[6] a) M. Ahmed, R. P. J. Bronger, R. Jackstell, P. C. L.
Kamer, P. van Leeuwen, M. Beller, Chem. Eur. J. 2006,
12, 8979; b) M. Ahmed, A. M. Seayad, R. Jackstell, M.
Beller, J. Am. Chem. Soc. 2003, 125, 10311; c) G. Ange-
lovski, P. Eilbracht, Tetrahedron 2003, 59, 8265; d) C.
Buch, R. Jackstell, D. Buhring, M. Beller, Chem. Ing.
Tech. 2007, 79, 434; e) W. H. Chiou, G. H. Lin, C. C.
Hsu, S. J. Chaterpaul, I. Ojima, Org. Lett. 2009, 11,
2659; f) P. Dubon, A. Farwick, G. Helmchen, Synlett
2009, 1413; g) B. Gall, M. Bortenschlager, O. Nuyken,
R. Weberskirch, Macromol. Chem. Phys. 2008, 209,
1152; h) C. S. Graebin, V. L. Eifler-Lima, R. G. da
Rosa, Catal. Commun. 2008, 9, 1066; i) B. Hamers, E.
Kosciusko-Morizet, C. Muller, D. Vogt, ChemCatChem
2009, 1, 103; j) H. Klein, R. Jackstell, M. Kant, A.
Martin, M. Beller, Chem. Eng. Tech. 2007, 30, 721;
k) A. Martin, M. Kant, R. Jackstell, H. Klein, M.
Beller, Chem. Ing. Tech. Chem. Ing. Techn. 2007, 79,
891; l) A. Seayad, M. Ahmed, H. Klein, R. Jackstell, T.
Gross, M. Beller, Science 2002, 297, 1676; m) T. O.
Vieira, H. Alper, Chem. Commun. 2007, 2710; n) T. O.
Vieira, H. Alper, Org. Lett. 2008, 10, 485; o) Y. Y.
Wang, J. H. Chen, M. M. Luo, H. Chen, X. J. Li, Catal.
Commun. 2006, 7, 979; p) Y. Y. Wang, M. M. Luo, Q.
Lin, H. Chen, X. J. Li, Green Chem. 2006, 8, 545.
hexylACHTUNGTRENNUNG
isopropylamide,[21b,23] N-hexyl-tert-butylamide,[21c,23,24] N-
hexylpentanamide,[25] N-hexylbenzamide,[21b,26] N-hexyl-4-
methoxybenzamide,[21b] N-hexyl-4-trifluoromethylbenzam-
A
N-hexyl-2-fluoroacetamide,[27]
N-(2-
ACHTUNGTRENNUNG
amide and N-benzylacetamide[21a,c] are
1
known compounds and were identified by H and 13C NMR
and GC-MS (see the Supporting Information Figure S15 to
Figure S36).
Acknowledgements
This research was performed within the framework of the
CatchBio program. We gratefully acknowledge the support of
the SmartMix Program of the Netherlands Ministry of Eco-
nomic Affairs and the Netherlands Ministry of Education,
Culture and Science. NWO ASPECT (ACTS) is kindly ac-
knowledged for financial support of the initial studies. Shell
Global Solutions International B.V. is kindly acknowledged
for generously donating phosphane ligands. We thank Dr. J.
Brandts (BASF) and Dr. R. Parton (DSM) for stimulating
and fruitful discussions.
[7] a) S. Enthaler, ChemCatChem 2010, 2, 1411; b) S.
Fleischer, S. L. Zhou, K. Junge, M. Beller, Chem. Asian
J. 2011, 6, 2240; c) A. W. Heinen, J. A. Peters, H. van
Bekkum, Eur. J. Org. Chem. 2000, 2501; d) P. B.
Quynh, T. H. Kim, Tetrahedron Lett. 2011, 52, 5004;
e) B. C. Ranu, A. Majee, A. Sarkar, J. Org. Chem.
1998, 63, 370; f) T. Gross, A. M. Seayad, M. Ahmad, M.
Beller, Org. Lett. 2002, 4, 2055.
References
[8] R. Stewart, Proton: applications to organic chemistry,
[1] a) M. Ahmed, C. Buch, L. Routaboul, R. Jackstell, H.
Klein, A. Spannenberg, M. Beller, Chem. Eur. J. 2007,
13, 1594; b) J. R. Briggs, J. Klosin, G. T. Whiteker, Org.
Lett. 2005, 7, 4795.
Academic Press Inc, 1985.
[9] ln(1ꢀconv/100)=ꢀkt!k (without acid)=0.18, while k
(with acid)=0.038, assuming first-order kinetics in
732
asc.wiley-vch.de
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2013, 355, 717 – 733

Rhodium-Catalyzed Homogeneous Reductive Amidation of Aldehydes
hexanal. In the presence of N,N’-dimethylacetamide
(entries 3 and 4) these numbers are respectively, 0.19
and 0.075.
shov, Organometallics 2012, 31, 1315; b) R. Dallanegra,
A. B. Chaplin, A. S. Weller, Organometallics 2012, 31,
2720.
[10] The NMR assignment and chemical shift were deter-
mined by compering with NMR predictors of Chem-
[19] M. E. Tauchert, C. D. Incarvito, A. L. Rheingold, R. G.
Bergman, J. A. Ellman, J. Am. Chem. Soc. 2012, 134,
1482.
ACHTUNGTRENNUNGOffice and NMRDB simulator.
[11] Unlike NMR analysis of the equilibrium hexanal/acet-
amide mixtures at low temperatures, the intermediate
2a was never observed in GLC analysis. It is expected
that any amount present in the reaction medium would
be detected in the GLC diagram as unsaturated inter-
mediates 2, as dehydration is likely a rapid thermal
process at 2508C prevailing in the injection port of the
GLC.
[20] a) R. P. J. Bronger, P. C. J. Kamer, P. W. N. M van Leeu-
wen, Organometallics 2003, 22, 5358; b) M. Kranen-
burg, Y. E. M. Vanderburgt, P. C. J. Kamer, P. W. N. M
van Leeuwen, K. Goubitz, J. Fraanje, Organometallics
1995, 14, 3081; c) 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, Organo-
metallics 2000, 19, 872.
[12] a) H. A. Edward, J. Schwoegler, J. Am. Chem. Soc.
1939, 61, 3499; b) I. J. Pachter, G. Suld, J. Org. Chem.
1960, 25, 1680.
[13] a) F. A. Carey, Advanced Organic Chemistry: Part A:
Structure and Mechanisms, Springer, New York, 2003;
b) E. J. Corey, Name Reactions in Heterocyclic Chemis-
try II, Wiley, New York, 2011.
[21] a) K. P. Dhake, Z. S. Qureshi, R. S. Singhal, B. M.
Bhanage, Tetrahedron Lett. 2009, 50, 2811; b) K. Ishi-
hara, T. Yano, Org. Lett. 2004, 6, 1983; c) L. Rubio-
Perez, P. Sharma, F. J. Perez-Flores, L. Velasco, J. L.
Arias, A. Cabrera, Tetrahedron 2012, 68, 2342; d) R. M.
Waters, N. Wakabayashi, E. S. Fields, Org. Prep.
Proced. Int. 1974, 6, 53.
[14] R. R. Schrock, J. A. Osborn, J. Am. Chem. Soc. 1976,
98, 4450.
[15] L. D. Julian, J. F. Hartwig, J. Am. Chem. Soc. 2010, 132,
13813.
[16] G. L. Williams, C. M. Parks, C. R. Smith, H. Adams, A.
Haynes, A. Meijer, G. J. Sunley, S. Gaemers, Organo-
metallics 2011, 30, 6166.
[22] a) C. Chen, S. H. Hong, Org. Lett. 2012, 14, 2992; b) T.
Kametani, O. Umezawa, Chem. Pharm. Bull. 1966, 14,
369.
[23] W. Krawczyk, G. T. Piotrowski, J. Chrom. 1989, 463,
297.
[24] H. M. Meshram, G. S. Reddy, M. M. Reddy, J. S. Yadav,
Tetrahedron Lett. 1998, 39, 4103.
[17] G. L. Moxham, H. E. Randell-Sly, S. K. Brayshaw, R. L.
Woodward, A. S. Weller, M. C. Willis, Angew. Chem.
2006, 118, 7780; Angew. Chem. Int. Ed. 2006, 45, 7618.
[18] a) V. I. Bakhmutov, F. Bozoglian, K. Gomez, G. Gonza-
lez, V. V. Grushin, S. A. Macgregor, E. Martin, F. M.
Miloserdov, M. A. Novikov, J. A. Panetier, L. V. Roma-
[25] A. J. M. van Dijk, T. Heyligen, R. Duchateau, J. Meul-
dijk, C. E. Koning, Chem. Eur. J. 2007, 13, 7664.
[26] A. Nemchik, V. Badescu, O. Phanstiel, Tetrahedron
2003, 59, 4315.
[27] S. Miscevic, D. Minic, S. Petrovic, J. Fluorine Chem.
1992, 59, 239.
Adv. Synth. Catal. 2013, 355, 717 – 733
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
733