Calix[4]arene-diphosphite rhodium complexes in solvent-free hydroaminovinylation of olefins

Article abstract of DOI:10.1039/c0gc00098a

Under solvent-free conditions rhodium complexes containing hemispherical diphosphites based on a calix[4]arene skeleton catalyse efficiently the hydroaminovinylation of α-olefins, thereby leading to high proportions of linear enamines/amines (when startin

Full text of DOI:10.1039/c0gc00098a

View Article Online / Journal Homepage / Table of Contents for this issue
PAPER
www.rsc.org/greenchem | Green Chemistry
Calix[4]arene-diphosphite rhodium complexes in solvent-free
hydroaminovinylation of olefins†
Laure Monnereau, David Se´meril* and Dominique Matt*
Received 11th May 2010, Accepted 14th July 2010
DOI: 10.1039/c0gc00098a
Under solvent-free conditions rhodium complexes containing hemispherical diphosphites based
on a calix[4]arene skeleton catalyse efficiently the hydroaminovinylation of a-olefins, thereby
leading to high proportions of linear enamines/amines (when starting from secondary amines) or
imines (when starting from primary amines). When applying a Rh/olefin ratio of 1 : 5000, the
reaction turned out to be ca. 15 times faster than when operating in toluene at the same Rh/olefin
ratio and at an olefin concentration of 6.6 mol L-1. For example, in the hydroaminovinylation of
1-octene with piperidine using 5,11,17,23-tetra-tert-butyl-25,27-dipropyloxy-26,28-
bis(1,1¢binaphthyl-2,2¢dioxyphosphanyloxy)calix[4]arene, TOFs up to 4640 mol(converted
olefin).mol(Rh)-1.h-1 were observed (l/b ratio of 24.1).
ligand bite angle and the confinement of the catalytic centre in a
Introduction
tight molecular pocket. Both features favour formation of Rh-
alkyl intermediates with a linear structure, so as to minimise
the steric interactions between the alkyl ligand formed and
the surrounding pocket. In other terms the ligand shapes the
outcoming product. It should be remembered that the ability
to envelope the catalytic centre is much more pronounced in
the above hemispherical diphosphites than in other diphos-
phanes possessing a large bite angle, such as e.g. Transphos,14
Bisbi,15 Xantphos,16,17 Triptiphos18,19 as well as biphenol-based
diphosphites.20,21 Pursuing our efforts to develop environmen-
tally friendly systems for olefin functionalisation,22 we now
describe the use of two neutral [Rh(acac)(1,3-calix-diphosphite)]
complexes (Fig. 1) in solvent-free hydroaminovinylation of 1-
octene and styrene. Remarkably, both complexes were found
to be highly soluble in these olefins. This work complements
a recent study in which we have shown that under such
conditions these complexes act as efficient hydroformylation
catalysts.23 To our knowledge, only one example of solvent-
free hydroaminovinylation has been reported in the literature.24
The term “1,3-calix[4]arene-diphosphites” used in this study
designates calix[4]arenes in which two distal phenolic oxygen
atoms are substituted by a P(OR)2 moiety.
Olefin hydroaminovinylation is a valuable atom-economical
domino reaction combining terminal alkene hydroformylation
with in situ formation of enamine/imine, the firstly generated
aldehyde reacting in a second step with an amine.1,2 When car-
rying out the reaction with secondary amines, hydroaminoviny-
lation is often followed by another reaction, namely the forma-
tion of amines through catalytic hydrogenation.3–7 A current
industrial challenge is to stop the reaction at the formation
of the enamine.8–12 It is worth mentioning here that the linear
selectivity in enamine mainly depends upon the regioselectivty
of the hydroformylation step.
Recently, we developed highly regioselective olefin hydro-
formylation catalysts based on hemispherical diphosphites
constructed on a calix[4]arene skeleton (two examples of such
ligands are shown in Fig. 1).12,13 The high selectivities observed
with these catalysts mainly rely on a combination of a wide
Results and discussion
Hydroaminovinylation with secondary amines
For the study concerning secondary amines, piperidine and
dibutylamine were used. Each experiment was carried out in the
presence of 3 or 4 and 9 equiv. of the corresponding diphosphite.
All runs were performed under solvent-free conditions at 130 ◦C
in a stainless steel autoclave, using an olefin/Rh ratio of 5000 : 1.
The products which may form in these reactions are shown in
Scheme 1.
We began the catalytic study by assessing complex 4 in the
hydroaminovinylation of styrene with piperidine. Using a 1 : 1
CO/H2 mixture and applying a pressure of 20 bar for 4 h
Fig. 1 Hemispherical ligands and rhodium complexes used in this
study.
Laboratoire de Chimie Inorganique Mole´culaire et Catalyse, UMR 7177
CNRS, Universite´ de Strasbourg, 1 rue Blaise Pascal, F-67008,
Strasbourg, Cedex, France.
E-mail: dmatt@chimie.u-strasbg.fr, dsemeril@chimie.u-strasbg.fr
† Electronic supplementary information (ESI) available: Typical chro-
matograms obtained after hydroformylation and hydroaminovinylation
experiments. See DOI: 10.1039/c0gc00098a
1670 | Green Chem., 2010, 12, 1670–1673
This journal is
The Royal Society of Chemistry 2010
©

View Article Online
Table 1 Rhodium-catalysed hydroaminovinylation of styrene using complexes 3 and 4 in the presence of piperidinea
Product distributionc
P(CO/H2)/bar
V(CO)/V(H2)
Entry
[Rh]
Time/h
Conv. (%)
TOFb
Aldehydes (%) l:b
Enamines (%) l:b
Amines (%) l:b
PhEt (%)
Solvent-free reaction
1
2
3
4
5
6
4
4
4
3
3
4
20 1/1
20 1/2
20 1/2
20 1/2
10 1/1
10 1/1
4
4
8
8
4
4
64.4
77.0
100
100
68.8
60.7
800
920
630
630
860
760
2.7 62.2 : 31.8
3.7 55.2 : 44.8
0.8 51.2 : 48.8
Traces
7.7 73.8 : 26.2
8.4 65.8 : 34.2
59.5 63.2 : 36.8
53.1 59.2 : 40.8
39.4 55.2 : 44.7
41.2 63.0 : 37.0
52.0 82.8 : 17.2
57.7 85.2 : 24.8
35.6 98.7 : 1.3
43.1 97.3 : 2.7
58.0 88.8 : 12.2
55.9 80.6 : 19.4
39.9 84.0 : 16.0
33.7 96.4 : 3.6
1.1
1.2
1.8
2.9
0.4
0.2
Reaction in toluene (20 mL), 130 ◦
20 1/2
C
7
4
4
48.5
62.2
0.8
610
780
10
5.1 81.9 : 18.1
4.8 76.4 : 23.6
/
14.9 85.1 : 14.9
30.3 93.3 : 6.7
/
13.3 87.4 : 12.6
2.4 100:traces
/
15.2
24.7
0.8
Reaction in cyclohexane (20 mL), 100 ◦
20 1/2 4
C
8
4
Reaction in THF (20 mL), 85 ◦
20 1/2
C
9
4
4
a Styrene (10 mmol), amine (12 mmol), [Rh] (2 mmol, styrene/Rh = 5000), ligand (18 mmol), 130 ◦C. b Mol(converted styrene).mol(Rh)-1.h-1.
c Determined by GC using decane (0.5 mL) as standard and 1H NMR spectroscopy.
applying a CO/H2 mixture of 1 : 1 for 4 h led, for both complexes,
to an increase of the enamine and a decrease of the amine
proportion (Table 1, entries 5 and 6). For example, using 4 gave
57.7% of enamines, 33.7% of amines, the corresponding linear
selectivities being 85.2% and 96.4%, respectively. It is worth
mentioning here that carrying out the hydroaminovinylation
reactions in toluene or cyclohexane gave lower conversions,
the linear selectivities remaining high (Table 1, entries 7 and
8). We further observed that under these conditions, high
Scheme 1 Rhodium-catalyzed hydroaminovinylation of a-olefins.
proportions ofethylbenzene were produced. The catalyst showed
no hydroformylation activity when THF was used (Table 1,
entry 9).
Repeating the above reactions with the slightly more basic
dibutylamine increased the enamine hydrogenation rate and
therefore led to a lower proportion of enamines (Table 2).
Complex 4, for example, gave 29.9% of enamines with a
regioselectivity towards the linear product of 89.8% (Table 2,
entry 4).
In a second series of reactions, 1-octene was subjected to
hydroaminovinylation with piperidine and dibutylamine. For
both diphosphites, the proportion of enamines was higher
with piperidine than with dibutylamine (Table 3, entries 3,7
and 4,8). This result can again be explained by the different
reactivities towards hydrogenation of the enamines generated,
produced the corresponding enamines and amines in 59.5% and
35.6%, respectively (Table 1, entry 1). As expected, the propor-
tion of linear product increased in the order aldehyde < enamine
< amine, this variation corresponding to the higher reactivity of
the linear product both in the condensation with amine and
in the hydrogenation step. Increasing the partial pressure of
hydrogen by 33% and doubling the reaction time raised both
the conversion and the proportion of amines (58.0%, Table 1,
entry 3), the latter consisting then of 88.8% of linear product.
As previously observed in styrene hydroformylation experiments
carried out either in toluene or water,12,22 replacement of the
benzyl substituents of 4 by propyl groups (leading to 3), resulted
in lower proportions of linear products (Table 1, entry 4), as
a consequence of a somewhat wider pocket surrounding the
catalytic centre. Reducing the CO/H2 pressure to 10 bar and
Table 2 Solvent-free rhodium-catalysed hydroaminovinylation of styrene using the rhodium complexes 3 and 4 in the presence of dibutylaminea
P(CO/H2)/bar
V(CO)/V(H2)
Entry
[Rh]
Time/h
Conv. (%)
TOFb
Product distribution
Aldehydes (%) l:b
5.7 62.1 : 37.9
4.5 61.2 : 38.8
2.2 82.3 : 17.7
7.7 84.0 : 16.0
Enamines (%) l:b
13.7 66.4 : 33.6
23.8 69.2 : 30.8
22.4 81.7 : 18.3
29.9 89.8 : 10.2
Amines (%) l:b
79.7 75.7 : 24.3
71.7 79.3 : 20.7
75.1 86.4 : 13.6
62.1 90.9 : 9.1
PhEt (%)
0.9
0.6
0.3
0.3
1
2
3
4
3
4
3
4
20 1/2
20 1/2
10 1/1
10 1/1
4
4
4
4
98.1
100
89.9
87.7
610
630
1120
1100
a Styrene (10 mmol), amine (12 mmol), [Rh] (2 mmol, styrene/Rh = 5000), ligand (18 mmol), 130 ◦C. b Mol(converted styrene).mol(Rh)-1.h-1.
c Determined by GC using decane (0.5 mL) as standard and 1H NMR spectroscopy.
This journal is
The Royal Society of Chemistry 2010
Green Chem., 2010, 12, 1670–1673 | 1671
©

View Article Online
Table 3 Solvent-free rhodium-catalysed hydroaminovinylation of 1-octene using complexes 3 and 4a
Product distributionc
Entry [Rh]
Piperidine
V(CO)/V(H2)
Time/h Conv. (%) TOFb Aldehydes (%) (l/b)
Enamines (%) (l/b)
Amines (%) (l/b) Isomer (%)
1
2
3
4
3
4
3
4
1/1
1/1
1/2
1/2
1
1
4
4
92.9
78.0
98.5
89.4
4640
3900
1230
1120
7.9 (15.2)
0.2 (nd)
3.3 (11.7)
5.1 (17.5)
68.7 (24.2)
75.1 (25.0)
48.3 (22.8)
47.2 (23.5)
11.4 (43.8)
8.9 (56.3)
36.1 (37.1)
29.6 (40.4)
12.0
15.8
12.3
18.1
Dibutylamine
5
6
7
8
9
3
4
3
4
4
1/1
1/1
1/2
1/2
1/1
1
1
4
4
4
86.7
89.7
98.1
98.7
97.5
4340
4490
1230
1230
1220
2.5 (2.8)
0.7 (nd)
2.5 (3.0)
2.8 (4.4)
3.0 (4.6)
75.1 (28.9)
74.8 (30.9)
28.9 (20.8)
42.1 (24.8)
57.9 (28.9)
8.8 (39.8)
13.6
14.4
14.1
17.2
21.3
10.1 (41.4)
54.5 (12.5)
37.9 (15.5)
17.8 (29.5)
a 1-Octene (10 mmol), amine (12 mmol), [Rh] (2 mmol, 1-octene/Rh = 5000), ligand (18 mmol), P(CO/H2) = 20 bar, 130 ◦C. b Mol(converted
1-octene).mol(Rh)-1.h-1. c Determined by GC using decane (0.5 mL) as standard and 1H NMR spectroscopy.
the enamine formed with piperidine being somewhat more
difficult to hydrogenate than the one formed with dibutylamine.
On the other hand, decreasing the partial pressure of hydrogen
reduced, as expected, the rate of hydrogenation, hence leading
to a higher proportion of enamines. For example, with dibuty-
lamine and complex 4 (4 h run) the proportion of enamines
increased from 42.1% to 57.9% when the CO/H2 ratio passed
from 1 : 2 to 1 : 1 (Table 3, entries 8 and 9). By reducing the
reaction time from 4 h to 1 h, the proportion of enamines
increased to 74.8% (Table 3, 6). The highest linear enamine
selectivity (CO : H2 = 1 : 1) was obtained with diphosphite 2, the
linear/branched ratios being then 25.0 and 30.9 with piperidine
and dibutylamine, respectively. In these latter experiments, the
corresponding proportions of enamine were 75.1% (piperidine)
and 74.8% (dibutylamine) (Table 3, entries 2 and 6).
produce imines (Scheme 2). The following investigations, in
which benzylamine was employed, were again performed under
solvent-free conditions.
Scheme 2 Formation of imines via tandem reactions.
Our expectations were first confirmed with styrene. Thus, in
a typical run carried out with complex 4 at a pressure of 20 bar
(CO/H2 = 1 : 2), styrene was fully converted into N-benzylimines
within 8 h, no trace of amine being detected. The linear imine
selectivity was 82.1% (Table 4, entry 2). Reducing the pressure to
10 bar and using a 1 : 1 CO/H2 mixture only slightly increased
the proportion of linear imine (85.1% with complex 4; Table 4,
entry 4). Consistent with the observations previously made for
these ligands in hydroformylation studies,23 the highest TOF,
Hydroaminovinylation with secondary amines (formation of
imines)
Primary amines are known to react with aldehydes thereby
forming imines. Since the hydrogenation of imines with rhodium
is unfavourable, we anticipated that carrying out the hy-
droaminovinylation with a primary amine would selectively
Table 4 Solvent-free rhodium-catalysed formation of imines using the rhodium complexes 3 and 4a
Product distributionc
Aldehydes (%) Imines (%) l% : b% = l/b
Entry
[Rh]
P(CO/H2)/bar V(CO)/V(H2)
Time/h
Conv. (%)
TOFb
Isomer (%)
Styrene
1
2
3
4
3
4
3
4
20 (1/2)
20 (1/2)
10 (1/1)
10 (1/1)
8
8
4
4
99.6
99.6
98.6
92.1
620
620
1230
1180
Traces
Traces
Traces
Traces
100 76.9 : 33.1 = 2.3
100 82.1 : 21.9 = 3.7
100 81.3 : 18.7 = 4.3
100 85.1 : 14.9 = 5.7
/
/
/
/
1-Octene
5
6
7
8
3
4
3
4
20 (1/1)
20 (1/1)
20 (1/2)
20 (1/2)
1
1
4
4
88.7
89.1
99.5
96.7
4440
4450
1240
1210
1.7
1.3
2.2
1.5
83.7 95.5 : 4.5 = 21.4
83.2 96.9 : 3.1 = 30.9
81.8 91.3 : 8.7 = 10.8
82.2 92.2 : 7.8 = 11.9
14.6
15.5
16.0
16.3
a Olefin (10 mmol), H2NBn (12 mmol), [Rh] (2 mmol, styrene/Rh = 5000), ligand (18 mmol), 130 ◦C. b Mol(converted olefin).mol(Rh)-1.h-1.
c Determined by GC using decane (0.5 mL) as standard and 1H NMR spectroscopy.
1672 | Green Chem., 2010, 12, 1670–1673
This journal is
The Royal Society of Chemistry 2010
©

View Article Online
1230 mol(converted styrene).mol(Rh)-1.h-1, was obtained with
complex 3 (corresponding to a conversion of 98.6%; Table 4,
entry 3).
General procedure for the hydroaminovinylation experiments
The hydroaminovinylation experiments were carried out in
a glass-lined, 100 mL stainless steel autoclave containing a
magnetic stirring bar. In a typical run, the autoclave was
charged under nitrogen with [Rh(acac)(1,3-calix-diphosphite)]
(0.002 mmol), the corresponding “1,3-calix-diphosphite”, olefin
(10 mmol), amine (12 mmol) and the internal standard (decane,
V = 0.5 mL). Once closed, the autoclave was flushed twice
with syngas (CO/H2 1 : 1 v/v), pressurised with the appropriate
CO/H2 mixture and heated. At the end of each run, the autoclave
was cooled to room temperature before being depressurised. A
sample was taken and analysed by GC and 1H NMR.
Performing the hydroaminovinylation with 1-octene (20 bar;
CO/H2 = 1 : 1) led to conversions of 88.7% (complex 3) and
89.1% (complex 4), respectively, after 1 h reaction time. As in
the case of styrene, no amine was produced. We noted that
the activities (TOF = 4440 (3) and 4450 (4) mol(converted
1-octene).mol(Rh)-1.h-1) were ca. 4 times higher than those
obtained with styrene. The linear imine selectivities (l/b = 21.4
(3); l/b = 30.9 (4); Table 4, entries 5 and 6) compare with
those observed for the enamines obtained in the reactions with
piperidine and dibutylamine (see above). Increasing the reaction
time and the partial pressure of hydrogen led to lower l/b ratios.
It is well known that higher hydrogen pressures favour formation
of internal octenes, which then may lead to branched imines,
so as to result in a regioselectivity decrease (Table 4, entries 7
and 8).
References
1 P. Eilbracht, L. Ba¨rfacker, C. Buss, C. Hollmann, B. E. Kitsos-
Rzychon, C. L. Kranemann, T. Rische, R. Roggenbuck and A.
Schmidt, Chem. Rev., 1999, 99, 3329.
2 P. Eilbracht and A. Schmidt, Top. Organomet. Chem., 2006, 18, 65.
3 S. Castillo´n and E. Ferna´ndez, in Rhodium Catalysed Hydroformyla-
tion (ed.: P. W. N. M. van Leeuwen, C. Claver), Kluwer, Dordrecht,
2000, pp 145-187.
Conclusion
4 M. Ahmed, R. P. J. Bronger, R. Jackstell, P. C. J. Kamer, P. W. N. M.
van Leeuwen and M. Beller, Chem.–Eur. J., 2006, 12, 8979.
5 M. Ahmed, C. Buch, L. Routaboul, R. Jackstell, H. Klein, A.
Spannenberg and M. Beller, Chem.–Eur. J., 2007, 13, 1594.
6 B. Hamers, P. S. Ba¨uerlein, C. Mu¨ller and D. Vogt, Adv. Synth. Catal.,
2008, 350, 332.
7 B. Hamers, E. Kosciusko-Morizet, C. Mu¨ller and D. Vogt, Chem-
CatChem, 2009, 1, 103.
8 Y. Dong and C. A. Busacca, J. Org. Chem., 1997, 62, 6464.
9 Y.-S. Lin, B. E. Ali and H. Alper, J. Am. Chem. Soc., 2001, 123, 7719.
10 E. Teuma, M. Loy, C. Le Berre, M. Etienne, J.-C. Daran and P. Kalck,
Organometallics, 2003, 22, 5261.
11 M. Ahmed, A. M. Seayad, R. Jackstell and M. Beller, Angew. Chem.,
Int. Ed., 2003, 42, 5615.
12 D. Se´meril, D. Matt and L. Toupet, Chem.–Eur. J., 2008, 14, 7144.
13 D. Se´meril, C. Jeunesse, D. Matt and L. Toupet, Angew. Chem., Int.
Ed., 2006, 45, 5810.
14 N. J. De Stefano, D. K. Johnson, R. M. Lane and L. M. Venanzi,
Helv. Chim. Acta, 1976, 59, 2674.
In conclusion, the results outlined in this study show that solvent-
free conditions do not modify the linear product selectivities ob-
tained with “Rh(1,3-calix-diphosphite)” catalysts with respect
to those obtained when carrying out the reaction in toluene.12
Remarkably, the solvent-free conditions led to a considerable
activity increase of the calix-diphosphite complexes, both cata-
lysts converting the substrates about 15 times faster than in the
presence of solvent, as a result of a higher olefin concentration.
Although these conditions led to a drop in enamine selectivity of
ca. 20%, the proportion of isolable enamine remains satisfactory
for an industrial process. Overall, the solvent-free conditions are
particularly adapted to a selective production of linear imines
from primary amines. Finally this study also constitutes a rare
illustration of the potential of hydroformylation catalysts that
are soluble in the usual olefins.
15 C. P. Casey, G. T. Whiteker, M. G. Melville, L. M. Petrovich, J. A
Gavney, Jr. and D. R. Powell, J. Am. Chem. Soc., 1992, 114, 5535.
16 L. A. van der Veen, P. C. J. Kamer and P. W. N. M. van Leeuwen,
Angew. Chem., Int. Ed., 1999, 38, 336.
Experimental
17 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 and A. L. Spek,
Organometallics, 2000, 19, 872.
18 W. Ahlers, M. Ro¨per, P. Hofmann, D. C. M. Warth and R. Paciello,
WO 01/58589 (BASF), 2001.
19 T. Schnetz, M. Ro¨der, F. Rominger and P. HofmanN, Dalton Trans.,
2008, 2238.
General methods
All syntheses were performed in Schlenk-type flasks under dry
nitrogen. Solvents were dried by conventional methods and were
distilled immediately prior to use. Routine 1H NMR spectra were
1
recorded by using a Bruker AVANCE 300. H NMR spectra
20 R. Paciello, L. Siggel and M. Ro¨per, Angew. Chem., Int. Ed., 1999,
were referenced to residual protonated solvents (7.26 ppm for
CDCl3). The catalytic solutions were analysed by using a Varian
3900 gas chromatograph equipped with a WCOT fused-silica
column (25 m ¥ 0.25 mm). The “1,3-calix-diphosphites” 1 and
2 and the [Rh(acac)(1,3-calix-diphosphite)] complexes 3 and 4
were prepared according to literature procedures.12
38, 1920.
21 J. R. Briggs and G. T. Whiteker, Chem. Commun., 2001, 2174.
22 L. Monnereau, D. Se´meril, D. Matt and L. Toupet, Adv. Synth.
Catal., 2009, 351, 1629.
23 L. Monnereau, D. Se´meril and D. Matt, Eur. J. Org. Chem., 2010,
3068.
24 M. L. Clarke and G. J. Roff, Green Chem., 2007, 9, 792.
This journal is
The Royal Society of Chemistry 2010
Green Chem., 2010, 12, 1670–1673 | 1673
©