Diphosphines of dppf-type incorporating electron-withdrawing furyl moieties substantially improve the palladium-catalysed amination of allyl acetates

Article abstract of DOI:10.1002/adsc.200505050

Highly active Pd/diphosphine catalytic systems incorporating new, air-stable ferrocenyl-furylphosphines allow nucleophilic allylic amination at room temperature with unprecedented turnover frequencies. For instance, in the presence of 0.01 mol % catalyst the coupling of aniline to allyl acetate occurs at a TOF of more than 10,000 h^-1; even the addition of the less nucleophile morpholine to allyl acetate is observed with a TOF of 4250 h ^-1. The amination of the sterically demanding geranyl acetate, a monoterpene derivative of interest in the flavour industry, at low catalyst loadings demonstrates the scope of this methodology, which provides in addition noticeable advantages in terms of economical (resource- and energy-saving) and sustainable chemistry (high selectivity, no additive, low metal content, and thus easier purification).

Full text of DOI:10.1002/adsc.200505050

COMMUNICATIONS
Diphosphines of dppf-Type Incorporating Electron-Withdrawing
Furyl Moieties Substantially Improve the Palladium-Catalysed
Amination of Allyl Acetates
Aziz Fihri,a Jean-Cyrille Hierso,a,* Anthony Vion,a Duc Hanh Nguyen,b
b,
b
a
a
´
Martine Urrutigoïty, * Philippe Kalck, Regine Amardeil, Philippe Meunier
a
`
`
´
´
Laboratoire de Synthese et Electrosynthese Organometalliques UMR-CNRS 5188, Universite de Bourgogne 9 avenue
Alain Savary, 21078 Dijon, France
Phone: (þ33)-38-039-6106, Fax: (þ33)-38-039-3682, e-mail: jean-cyrille.hierso@u-bourgogne.fr
b
´
´
LCCFP, Ecole Nationale Superieure des Ingenieurs en Arts Chimiques et Technologiques, 31077 Toulouse, France
Phone: (þ33)-56-288-5698, Fax: (þ33)-56-288-5600; e-mail: Martine.Urrutigoity@ensiacet.fr
Received: January 25, 2005; Accepted: April 6, 2005
Supporting Information for this article is available on the WWW under http://asc.wiley-vch.de/home/.
Abstract: Highly active Pd/diphosphine catalytic sys-
tems incorporating new, air-stable ferrocenyl-furyl-
phosphines allow nucleophilic allylic amination at
room temperature with unprecedented turnover fre-
quencies. For instance, in the presence of 0.01 mol %
catalyst the coupling of aniline to allyl acetate occurs
at a TOF of more than 10,000 hÀ1; even the addition
of the less nucleophile morpholine to allyl acetate is
observed with a TOF of 4250 hÀ1. The amination of
the sterically demanding geranyl acetate, a monoter-
pene derivative of interest in the flavour industry, at
low catalyst loadings demonstrates the scope of this
methodology, which provides in addition noticeable
advantages in terms of economical (resource- and
energy-saving) and sustainable chemistry (high selec-
tivity, no additive, low metal content, and thus easier
purification).
reactions in the presence of low amounts of metal cata-
lyst to achieve high TONs and high TOFs.[4]
Ferrocenylphosphines were chosen to combine stabil-
ity from the robustness of the ferrocenyl backbone with
enhanced activity from steric and electronic effects of
tunable phosphineꢀs substituents.[5] The synthesis and
catalytic properties of dppf-type phosphines (dppf¼
1,1’-bis[diphenylphosphino]ferrocene) built with phos-
phorus bearing heterocyclic substituents have been no-
tably less studied than the chemistry of ferrocenyl heter-
oannular diphosphines bearing more classical aryl or al-
kyl substituents.[6] The new ligands 1,1’-bis[di(5-methyl-
2-furyl)phosphino]ferrocene (1) and 1-[di(5-methyl-
2-furyl)phosphino]-1’-(diphenylphosphino)-ferrocene
(2) are obtained, after work-upprocedures, in high to
moderate yields (80% and 25%, respectively) from the
reaction of lithiated ferrocenes with the bis(5-methyl-
2-furyl)bromophosphine BrP(FuMe
)
(Scheme 1).[7]
2
À
Keywords: allylic amination; catalysis; C N coupling;
While 1 is isolated as a crystalline orange powder that
is easy to handle, 2 is obtained as a sticky orange-brown
oil. Both compounds shows stability to air and moisture,
even in solution, and no decomposition is observed upon
heating them in toluene for several hours. In the solid
state 1 is indefinitely stable at room temperature under
ferrocene; furylphosphine; palladium
The discovery of the palladium-catalysed allylic alkyla- air.
tion[1] has initiated an intense research effort directed to-
Following our catalysis objective, the coupling of var-
wards the synthesis of useful allylic building blocks.[2] ious amines and allyl acetates (Scheme 2) by palladium
Amination of allyl halides, acetates or malonates has complexes stabilised by 1 and 2 was investigated and
proven to be efficient when conducted in the presence representative results are summarised in Table 1. It rap-
of 1–10 mol % of palladium catalysts and phosphine idly appeared that highly active catalysts usable in very
auxiliaries.[2,3] A critical improvement in this field, con- low concentrations were formed, underlining the tre-
nected to both sustainable chemistry and economic con- mendous influence of the electron-withdrawing phos-
cerns, is the necessity to provide catalytic systems mini- phine ligands on the rate of the catalysed reaction. To
mising the consumption of depletive resources (i.e., pal- gain more information on the coordination chemistry
ladium combined with sophisticated ligands). We initiat- of these new catalytic auxiliaries, well-defined palladi-
ed a program aiming to develop air-stable, robust cata- um complexes were synthesised. The reaction in equi-
lytic auxiliaries that could allow accelerated catalytic molar quantities of 1 (or 2) with PdCl2(PhCN)2 gives
1198
ꢁ 2005 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
DOI: 10.1002/adsc.200505050
Adv. Synth. Catal. 2005, 347, 1198–1202

Palladium-Catalysed Amination of Allyl Acetates
COMMUNICATIONS
Scheme 1. Reagents and conditions: (i) 0.5 equivs. FeCp2Li2-
TMEDA, hexane, room temperature, 2 h; (ii) 1 equiv.
CpFe{Cp(PPh2)}Li, THF, À508C, 12 h; (iii) 1 equiv. PdCl2
(PhCN)2, CH2Cl2, room temperature, 20 min.
Figure 1. Molecular structure of 3. Ellipsoids set at 50%
probability, H atoms are omitted for clarity. Selected distan-
À
À
ces (Š) and angles (8): Pd P1 2.2633(8), Pd P2 2.2622(8),
À
À
À
À
Pd Cl1 2.3366(8), Pd Cl2 2.3333(8); Cl2 Pd Cl1 89.40 (3),
À
À
À À
P2 Pd P1
96.98(3),
Ct1 Fe Ct2
178.05
(13),
À
À
À
P1 Ct1 Ct2 P2 37.6 (2).
We tested the reaction of aniline (7) with allyl acetate
(5) at room temperature to compare the efficiency as
catalytic auxiliaries of the ferrocenylphosphines 1 and
2 with the related diphosphine dppf (Scheme 2 and Ta-
ble 1, entries 1–5). After 1 h, in the presence of
Scheme 2.
the chelate complex 3, PdCl2{Fc[P(FuMe)2]2} (or 4, 1 mol % catalysts a total conversion is obtained from
PdCl2{Fc[P(FuMe)2(PPh2)]},
respectively),
see the sterically congested dppf and as well from 1, with a
1
Scheme 1. Compound 3 has been characterised by H, good selectivity in monoallylamine (96–94%, entries 1
13C, 31P NMR spectroscopy, and by single crystal X-ray and 3). In the presence of 0.01 mol % low concentration
analysis (Fig. 1).[8]
catalysts, using 1 a total conversion is obtained in 1 h at
The 31P NMR chemical shifts observed for the ligands room temperature with a 96% selectivity in monoallyla-
and their complexes, and compared to the ones reported niline (entry 4), leading us thus, in non-optimised condi-
for dppf and analogous compounds, confirm the elec- tions, to estimate a TOF of at least 10,000 hÀ1. In the
tron-withdrawing influence exerted by the furyl sub- same conditions, using 2 the total conversion obtained
stituents (for 1, a singlet at d¼ À63 ppm is noted, after 2 h is slightly less selective in monoallylaniline
against À17 ppm for dppf). The molecular structure in- (91%, entry 5). Under identical conditions, dppf as aux-
dicates staggered Cprings with an average dihedral an-
iliary shows its limits since a maximum 60% conversion
À
À
gle of 33.9(4)8. The P Pd P angle of 96.98(3)8, more is obtained after 40 h at room temperature (entry 2,
closed than the angle values reported for PdCl2(dppf) TOF 150 hÀ1). At such low concentrations and temper-
structures [99.07(5)[9]], opens for chemical reactions at atures monophosphines are ineffective.
À
À
palladium a larger spatial area “in front” of the P Pd P
moiety.
Next we tried to evaluate the scope and limitations of
the systems Pd/1 and Pd/2 using the same high ratio [sub-
Adv. Synth. Catal. 2005, 347, 1198–1202
asc.wiley-vch.de
ꢁ 2005 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
1199

Aziz Fihri et al.
COMMUNICATIONS
Table 1. Palladium-catalysed allylic amination.
Catalytic
system
Allyl
acetate
Amine
Conversion
Yield [%]
Ratio
(mono/di)
Ratio
substrate/catalyst
TOF [hÀ1
]
Entry
Pd/dppf
5
5
7
7
100
96/4
97/3
94/6
96/4
91/9
75/25
76/24
–
100
10000
100
100
150
100
10000
5000
1825
1071
4250
3333
5000
4900
100
480
–
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
60[a]
100
Pd/1
100
10000
10000
10000
10000
10000
10000
10000
10000
10000
1000
100
Pd/2
Pd/1
Pd/2
Pd/1
Pd/2
Pd/1
Pd/1
Pd/1
5
5
5
5
5
5
5
5
7
8
8
9
9
10
11
12
100[b]
73[c]
75[d]
85[b]
100[e]
100[b]
98[b]
40[a]
96[b, g]
7[i]
–
–
–
–
–
Pd/PCy3
Pd/PPh3
Pd/1
Pd/1
Pd/2
6
6
6
6
6
6
7
7
7
7
7
9
100/0
100/0
100/0
55/j
29[i]
100
100
100
100
100
100
–
5
50
50
60
5
13
333
–
75[f]
100[b, h]
100[b, h]
60
50/j
Pd/1
100/0
100/0
100/0
98/j
98[f]
50[a]
100[e, g]
5>
1000
1000
10000
100
traces
Pd/1
6
12
100[e]
traces/j
–
Conditions: catalysts preparation see ref.[8], toluene, room temperature, 1 h.
[a]
40 h.
2 h.
4 h.
7 h.
3 h.
20 h.
808C.
1108C.
[b]
[c]
[d]
[e]
[f]
[g]
[h]
[i]
408C.
[j]
Unidentified secondary products are formed, presumably allylic isomers.[3c]
strate/catalyst]¼10,000, at room temperature, with var- amine (12) to allyl acetate since, in the presence of
ious primary and secondary amines of different steric 0.01 mol % catalyst, only 40% conversion is observed
and nucleophilic properties, namely: benzylamine (8), after 40 h, resulting in a TOFof 100 hÀ1 (entry 12). How-
morpholine (9), piperidine (10), pyrrolidine (11) and dii- ever, using 0.1 mol % catalyst, 96% conversion of 12 is
sopropylamine (12). The addition rate is slightly de- obtained at 808C in 2 hours giving a TOFof 480 hÀ1 (en-
creased for the reaction of 8 with allyl acetate. Around try 13).
75% conversion is observed after 4 h using 1 (TOF
These promising results prompted us to investigate
1825 hÀ1, entry 6) and in 7 h using 2 (TOF 1071 hÀ1, en- the allylic amination reaction using a much more de-
try 7). This slower addition rate probably causes the de- manding allyl acetate substrate. A monoterpene deriva-
crease in selectivity mono-/diallylamine. Unexpectedly, tive [geranyl acetate (6)] was chosen, due to both the in-
excellent results are obtained in the coupling of the less dustrial interest of monoterpenes functionalisation[11]
nucleophilic cyclic morpholine:[10] after 2 hours 85% and their steric hindrance. The reaction of 6 on aniline
conversion is observed using 1 (TOF 4250 hÀ1, entry 8) (7) at 408C in the presence of 1 mol % catalysts using
and a total conversion is accessible in 3 h with 2 (TOF classical monophosphines gives, with Pd/PPh3, 29% con-
3333 hÀ1, entry 9). In line with these results the cyclic version in monoallylamine (entry 5) and only 7% con-
amines 10 and 11 are converted in high yield after 2 h version with the bulky electron-rich PCy3 (entry 4); at
(TOF 5000 hÀ1 and 4900 hÀ1 entries 10 and 11, respec- room temperature the reactivity being, as expected,
tively). A substantial, presumably steric effect, previ- even more sluggish. At room temperature using
ously noted,[10] occurs upon coupling of diisopropyl- 1 mol % Pd/1, 75% conversion in geranylaniline is ob-
1200
ꢁ 2005 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
Adv. Synth. Catal. 2005, 347, 1198–1202

Palladium-Catalysed Amination of Allyl Acetates
COMMUNICATIONS
tained after 20 h with total selectivity (entry 16). In con- Experimental Section
trast, complete conversion of 6 is observed in 2 h at
The syntheses of the palladium complexes and a detailed char-
acterisation of all the compounds are available as Supporting
Information.
elevated temperature (entries 17 and 18), but only 55–
50% of the expected geranylaniline is formed, the high
temperature being responsible for undesired side reac-
tions.
To identify the limitations of our system, morpholine
and diisopropylamine (entries 19–24) were added to
geranyl acetate. Surprisingly, the addition of 9 in the
presence of 1 mol % catalyst at room temperature give
a 60% conversion in 1 h, and an almost total conversion
after 20 h. Using 0.1 mol % Pd/1 at room temperature
the addition rate is significantly slower (entry 21). How-
ever, an excellent TOF of 333 hÀ1 is obtained at 808C
with 98% of geranylamine being obtained in 3 h (en-
try 22). At 0.01 mol % the catalyst becomes inactive.
The coupling of the two sterically demanding com-
pounds 6 and 12 did not proceed satisfactorily in pres-
ence of 1 mol % Pd/1 (entry 24). It is noteworthy that
the results presented here could probably be optimised,
especially concerning solvent and temperature condi-
tions,[3c] since room temperature and toluene were sys-
temically used.
Synthesis of Fc[P(FuMe)2]2 (1)
To a stirred suspension of 6.78 g (21.61 mmol) of FeCp2Li2-
TMEDA in 60 mL of hexane was added dropwise a solution
of 11.8 g (43.22 mmol) of BrP(FuMe)2 in 25 mL of hexane. After
two hours stirring at room temperature, 20 mL of degassed wa-
ter were added. An orange powder was obtained which was
washed three times with hexane and then dried under vacuum.
Purification was carried out by dissolution in chloroform and
filtration on silica gel; yield of 1: 9.86 g (17.30 mmol, 80%).
Synthesis of Fc[P(FuMe)2(PPh2)] (2)
To a solution of 1.25 g (3.39 mmol) of BrP(FuMe)2 in 20 mL of
THF was added at À508C a solution of 1.28 g (3.39 mmol) of
CpFe{Cp(PPh2)}Li in 40 mL of THF. The mixture was stirred
for 12 hours and then evaporated under vacuum to give a
brown oil. The crude product was purified by chromatography
(toluene/hexane, 1:1) on neutral silica. The first fraction ob-
Finally, in order to definitively demonstrate the scope
of this methodology, preliminary tests using more di-
verse examples of functionalised allyl acetate and
amines as substrates were carried out. As expected, tained is CpFeCp(PPh2) (0.520 g, 41% yield) and the 2nd frac-
tion is the dissymmetrical ligand 2 (0.420 g, 25% yield) ob-
tained as a pure orange-brown oil.
good to excellent conversions were obtained using di-
ethylamine, aniline, piperidine and morpholine with
the substituted allyl acetates cinnamyl acetate and
hex-2-en-1-yl acetate. This was not a surprise since these
substituted allyl acetates appear a priori less demanding
than geranyl acetate. From the reaction of cinnamyl ace-
tate with diethylamine, aniline and piperidine a conver-
sion of, respectively, 76%, 65% and 30% was obtained
after 20 h at 508C with a 0.01 mol % catalyst loading
of Pd/1. With a higher loading of 0.1 mol % catalyst, a
Catalytic Reactions; Typical Procedure for a Ratio
[substrate/catalyst¼10000]
The palladium/ferrocenyl furylphosphine complexes were
prepared by stirring in 10 mL of toluene, under argon for
30 min the diphosphines 1 (0.04 mmol, 22.8 mg) or 2 with
100% conversion for all the substrates is accessible at [Pd(h3-C3H5)Cl]2 (0.01 mmol, 3.65 mg). The same batch of cat-
alyst was used for more than a week without any activity de-
crease (stable in solution in the refrigerator under argon).
1 mL of the previously prepared catalyst solution (thus
0.001 mmol Pd) was added to the mixture of allyl acetate or
geranyl acetate (10 mmol) and amine (20 mmol) in 25 mL tol-
uene. Themixture was stirred atambienttemperature, or 808C,
or 1108C during 1 to 40 h. Products were obtained after addi-
tion of water, extraction with organic solvents, drying of the or-
ganic phase, and chromatography on silica gel.
room temperature in the same time. In any case, a high
regioselectivity of linear/branched product (93/7%)
was obtained. Under the latter conditions, hex-2-en-1-
yl acetate reacts with piperidine and morpholine to re-
spectively yield the corresponding allylamines in 100
and 98% conversion (ratio linear/branched 94/6%).
The global comparison of catalytic activity for the sys-
tems incorporating the related ligands dppf, 1 and 2 (of
roughly similar steric features) led us to think that the
electronic properties of the furyl group[12] might be at
the origin of the impressive TOFs obtained for the cou-
pling of allyl acetate with various primary and secondary
aliphatic and cyclic amines. Further work is underway to
explain and explore the effectiveness of our ligands in al-
lylic amination and in other attractive palladium-cata-
lysed reactions.
Acknowledgements
We are indebted to D. Poinsot and S. Chambrey (from I. Tkatch-
enkoꢀs group) for GC-MS analysis, and to H. Doucet for the
checking ofsome reactions using substituted-allyl acetates.
Thanks are due to P. Richard (X-ray analysis), E. Pousson, S.
Royer and G. Delmas for technical assistance.
Adv. Synth. Catal. 2005, 347, 1198–1202
asc.wiley-vch.de
ꢁ 2005 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
1201

Aziz Fihri et al.
COMMUNICATIONS
References and Notes
[6] a) I. R. Butler, W. R. Cullen, T.-J. Kim, S. J. Rettig, J.
Trotter, Organometallics 1985, 4, 972; b) I. R. Butler,
W. R. Cullen, T.-J. Kim, Synth. React. Inorg. Met.-Org.
Chem. 1985, 15, 109; c) B. C. Hamann, J. F. Hartwig, J.
Am. Chem. Soc. 1998, 120, 7369; d) A. L. Boyes, I. R.
Butler, S. C. Quayle, Tetrahedron Lett. 1998, 39, 7763.
[7] This synthetic pathway is alternative to the one described
to get the family compound 1,1’-bis[di(2-furyl)phosphi-
no]ferrocene, see: B. C. Hamann, J. F. Hartwig, J. Am.
Chem. Soc. 1998, 120, 3694.
[8] Syntheses, catalytic reactions, characterisation of the li-
gands, complexes and coupling products are available
as electronic supplementary information (ESI).
[9] T. Hayashi, M. Konishi, Y. Kobori, M. Kumada, T. Higu-
chi, K. Hirotsu, J. Am. Chem. Soc. 1984, 106, 158.
[10] M. Feuerstein, D. Laurenti, H. Doucet, M. Santelli,
Chem. Commun. 2001, 43.
[1] M. Morikawa, J. Takahashi, J. Tsuji, Tetrahedron Lett.
1965, 6, 4387.
[2] M. Johannsen, K. A. Jørgensen, Chem. Rev. 1998, 98,
1689.
[3] a) O. Kuhn, H. Mayr, Angew. Chem. Int. Ed. 1999, 38,
343; b) B. M. Trost, L. T. Calkins, C. Oerlet, J. Zambrano,
Tetrahedron Lett. 1998, 39, 1713; c) for rare catalytic sys-
tems operating at lower concentration, see: M. Feuer-
stein, D. Laurenti, H. Doucet, M. Santelli, J. Mol. Catal.
A 2002, 182–183, 471 and references cited therein.
[4] a) J.-C. Hierso, A. Fihri, R. Amardeil, H. Doucet, M.
Santelli, V. V. Ivanov, P. Meunier, Org. Lett. 2004, 6,
3473; b) J.-C. Hierso, A. Fihri, R. Amardeil, P. Meunier,
H. Doucet, M. Santelli, B. Donnadieu, Organometallics
2003, 22, 4490.
[11] S. Kad, Topics in Catalysis 2004, 27, 143.
[5] For a review on disubstituted ferrocenyl ligands used in
catalysis, see: R. C. J. Atkinsson, V. C. Gibson, N. J.
Long, Chem. Soc. Rev. 2004, 33, 313.
[12] a) N. G. Andersen, B. A. Keay, Chem. Rev. 2001, 101,
997; b) A. Scrivanti, V. Beghetto, E. Campagna, U. Mat-
teoli, J. Mol. Catal. A 2001, 168, 75.
1202
ꢁ 2005 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
Adv. Synth. Catal. 2005, 347, 1198–1202