Green organometallic chemistry: CpFe+ -induced amination of chloroarenes in water

Article abstract of DOI:10.1016/S0022-328X(01)01121-4

Reaction of NH₃, primary amines and non-bulky secondary amines with the complexes [Fe^IICp(η⁶-ArCl)][X] (Ar = C₆H₅ or 0-C₆H₄Cl, X = Cl, BF₄ or PF₆) in water

Full text of DOI:10.1016/S0022-328X(01)01121-4

Journal of Organometallic Chemistry 643–644 (2002) 125–129
www.elsevier.com/locate/jorganchem
Green organometallic chemistry: CpFe+-induced amination of
chloroarenes in water
Franc¸oise Moulines, Mahmoud Kalam-Alami, Victor Martinez, Didier Astruc *
Laboratoire de Chimie Organique et Organome´tallique, UMR CNRS 5802, Uni6ersite´ Bordeaux I, 33405 Talence Cedex, France
Received 18 June 2001; accepted 13 July 2001
Dedicated to our distinguished colleague Professor Franc¸ois Mathey at the occasion of his 60th birthday
Abstract
Reaction of NH3, primary amines and non-bulky secondary amines with the complexes [FeIICp(h6-ArCl)][X] (Ar=C6H5 or
o-C6H4Cl, X=Cl, BF4 or PF6) in water at room temperature gives the nucleophilic substitution products [FeIICp(h6-
ArNR1R2)][X] isolated as PF6 salts. © 2002 Elsevier Science B.V. All rights reserved.
Keywords: Green chemistry; Water; Amination; Chloloarene; Iron
such as the nitro groups or some organometallic frag-
1. Introduction
ments facilitate this reaction [9–11]. In particular, the
isolobal 12-electron moieties Cr(CO)3, Mn(CO)+3 and
Biphasic catalysis involving an aqueous phase has
CpFe+ are activators for these reactions [11,12]. For
been a popular subject during the last two decades for
instance, nucleophilic substitution of chloride in the
monocationic complexes [FeIICp(h6-ArCl)][PF6] by a
variety of nucleophiles [13–23] including NH3 [13] and
amines [13–15,19,23,24] has been carried out under
mild conditions in organic solvents. We have re-investi-
gated these reactions using water as the solvent, and
now report that amination of chloroaromatics in the
complexes [FeIICp(h6-ArCl)][X] (X=Cl or PF6) can be
achieved at room temperature in water.
catalyst recovery, in particular with the development of
the water-soluble triply-sulfonated triphenylphosphine
P(m-C6H4SO−3 Na+)3 (TPPTS) by Kunz for the hydro-
formylation of propene to butyraldehyde, a process
industrialized by Rhoˆne-Poulenc [1–4]. The more re-
cent concern for environment-safe chemistry or ‘‘green
chemistry’’ now encourages chemists to adopt a slightly
different angle that consists in completely avoiding
organic solvents whenever possible and to use water as
the solvent [5,6]. The amination of chloroaromatics
using palladium catalysis in organic solvents has been a
challenge for some time. A key step in these reactions is
the oxidative addition of the aryl–halide bond [7,8].
This bond is all the more difficult to cleave as it is
stronger (bond strengths: ArIBArBrBArCl), e.g. the
chloroarenes, which are the only economically attrac-
tive halogenoarenes, are the most difficult to activate.
Nucleophilic substitution of halides in halogenoaromat-
ics usually only proceeds at high temperatures (typically
300 °C), but electron-withdrawing arene substituents
2. Results
The complexes [FeIICp(h6-ArCl)][X] (X=Cl, BF4 or
PF6) are easily accessible by ligand substitution between
a ring of ferrocene and the chloroaromatic in the
presence of AlCl3. These reactions are currently carried
out in refluxing heptane or any other inert alcane
around 100 °C for a few hours provided Ar does not
contain a heteroatom which is incompatible with the
presence of AlCl3 [13,25]. Aluminum powder should
not be added, unlike in the reaction with aromatics
lacking the chloro substituent, because of the partial
reduction of the chloroarene ligand by aluminum in the
* Corresponding author. Tel.: +33-556-846271; fax: +33-556-
846274.
E-mail address: d.astruc@lcoo.u-bordeaux.fr (D. Astruc).
0022-328X/02/$ - see front matter © 2002 Elsevier Science B.V. All rights reserved.
PII: S0022-328X(01)01121-4

126
F. Moulines et al. / Journal of Organometallic Chemistry 643–644 (2002) 125–129
2. Add aqueous H+PF6− in order to precipitate or
further extract with methylene chloride the partly
insoluble organometallic PF−6 salt before carrying
out the nucleophilic reaction in water (method B). If
aqueous H+BF−4 is added instead of H+PF−6 , the
organometallic BF−4 salt remains water soluble but
can also be extracted from water using methylene
chloride whereas such an extraction is too tedious
with the chloride salt.
course of the reaction. After this ligand substitution
reaction, slow hydrolysis by ice-water provides an
aqueous phase which contains the cationic salt
[FeIICp(h6-ArCl)][Cl] resulting from the hydrolysis of
the water-sensitive AlCl−4 anion. Then, 1 N aqueous
ammonia is added in order to precipitate the Al3+
cations as the hydroxide Al(OH)3 which is filtered off.
Reactions of these organoiron salts with NH3 and a
variety of amines have been reported by the Nes-
meyanov and Sutherland groups to readily proceed in
methylene chloride (Scheme 1). Two strategies are pos-
sible at this stage in order to perform these reactions in
water (Scheme 2):
1. Concentrate these aqueous solutions of the water-
soluble organometallic chloride salt resulting from
hydrolysis of the ligand substitution reaction and
elimination of the Al3+ ions in order to carry out
the nucleophilic substitution with the amine
(method A). Since it is a bimolecular reaction, it
should be carried out at a reasonable concentration.
Aqueous HPF6 can be added afterwards in order to
precipitate, extract and purify the PF6 organoiron
salt.
In either case, it is essential to perform the nucle-
ophilic substitution reactions at a relatively high con-
centration in water in order to reduce the reaction time
at room temperature. The yellow salts [FeIICp(h6-
ArCl)][PF6] were also isolated, recrystallized and char-
acterized by comparison with literature data. When the
BF4 salts were involved, they usually formed as oils
because of their high hygroscopicity even if they were
free of other impurities.
Reactions of the parent compound [FeIICp(h6-
C6H5Cl)][X] (X=Cl, BF4 or PF6) were carried out at
the mmol scale in 5 ml water with benzylamine,
phenethylamine, methylbenzylamine, dimethylamine
and 7 N ammonia at room temperature and were found
1
by H-NMR to be complete after 5 h. The products
1
were readily characterized using H-NMR by compari-
son with literature data. Residual water was found by
1H-NMR and was difficult to remove completely, the
arylamine complexes being hygroscopic even as PF6
salts. The integration showed some excess of the alkyl
and aryl protons due to some [FeII(amine)6][PF6]2
formed by decomplexation. This impurity could be
removed by washing the organoiron products upon
refluxing in ether or THF in which the impurities were
extracted. The yields in complexes [FeIICp(h6-
ArNR1R2)][PF6] were found to be between 50 and 70%
after reprecipitation of methylene chloride solutions
with diethylether.
Scheme 1.
With 7 N ammonia, the reaction was found to pro-
ceed slowly at room temperature. For instance, after 1
day, only 10% of the starting chlorobenzene complex
was converted to the final aniline complex. More rapid
consumption of the starting complex can be achieved
within 1 day at 60 °C, giving 58% yield of aniline
complex and some of the red complex [FeII(amine)6]-
[PF6]2.
The
nucleophilic
reaction
of
[FeIICp(h6-
C6H5Cl)][PF6] was also attempted with the bulkier sec-
ondary amine diethylamine in water. In this case, there
was no reaction within 1 day at room temperature
whereas the reaction was complete in methylene chlo-
ride. Upon heating at 50 °C for 96 h, the nucleophilic
1
reaction occurred, but the H-NMR spectrum of the
crude reaction mixture showed an excess of ethyl pro-
tons. This was again attributed to decomplexation
yielding a minor amount of [FeII(NHEt2)6][PF6]2 which
was formed together with the aminoarene sandwich
Scheme 2.

F. Moulines et al. / Journal of Organometallic Chemistry 643–644 (2002) 125–129
127
product which limits the reaction yields; this product
can be washed out using warm ethers.
Thus, it is clear that the nucleophilic substitution
does not proceed quite as easily in water as in methyl-
ene chloride, although the fact that these reactions
proceed in water at room temperature illustrates the
impressive electron-withdrawing properties of the
cationic organoiron activator. The mechanism consists
in the rate-limiting nucleophilic attack of the amine
onto the ipso carbon atom followed by the removal of
HCl from the chlorocyclohexadienylammonium inter-
mediate by excess amine [25–28]. Thus, the reaction
stoichiometry required the use of two equivalents of
amine. Water (in large excess) gives a hydrogen bond
HꢀOꢀH···NHR2 which decreases the electron density
on the nitrogen atom of the amine and weakens its
nucleophilicity, disfavoring the nucleophilic attack.
We noticed that the complexes [FeIICp(h6-
ArNR1R2)][PF6] formed were very robust. Usually, de-
complexation of the arene ligand was obtained by
UV–vis photolysis [25,29,30], but such a procedure did
not work in this case, the complexes being recovered
after such reactions. This is due to the electron-releas-
ing character of the amino substituents, but also to the
contribution of the cyclohexadienyliminium resonance
form to the actual structure (Chart 1).
Chart 1.
complex in the ratio 1/6. The yield of isolated
[FeIICp(h6-C6H5NEt2)][PF6] under these conditions was
25%. Attempts to perform the nucleophilic substitution
with dibenzylamine did not give the desired nucle-
ophilic substitution product.
The complex [FeIICp(h6-o-C6H4Cl2)][PF6] is also ac-
cessible in 20% yield from ferrocene and o-C6H4Cl2 in
the presence of AlCl3. Reactions with benzylamine and
phenethylamine were similarly carried out on the PF6
salt in water at room temperature for 5 h and gave
exclusively the monosubstitution products as already
reported for the same reactions in dichloromethane.
The products were purified as above by reprecipitation
and the yields of the nucleophilic substitution reactions
were 65%.
3. Discussion
Theoretical calculations confirmed this contribution,
the calculated structure showing a small but significant
dihedral angle of the arene with a long distance be-
tween iron and the ipso carbon [31]. This strong influ-
ence of the amino substituent on the arene ligand also
explains why only monosubstitution occurs with
[FeIICp(h6-o-C6H4Cl2)][PF6], as already reported for
the reaction performed using methylene chloride as the
solvent [23]. The first amino substituent deactivates the
arene ligand towards the second substitution for the
reasons indicated above.
Complexation of chloroaromatics by CpFe+ leading
to yellow salts not only provides a strong activator for
nucleophilic substitution, but also allows solubilization
of these products in water provided the counter cation
is not too large (BPh−4 would be too large an anion).
The large water solubility of the chloride and BF4−
organoiron salts and the modest solubility of the PF6−
ones render reactions in water possible. This brings
flexibility to the reaction procedure since the nucle-
ophilic substitution can be achieved either before or
after addition of the aqueous solution of the acid of the
chosen anion provided the concentration is high
enough.
The nucleophilic substitution of chloride in
[FeIICp(h6-C6H5Cl)][X] and [FeIICp(h6-o-C6H4Cl2)][X]
can be achieved with NH3, primary amines and to some
extent with secondary amines provided the alkyl groups
are not too bulky (Me, Et). The reactions with NH3,
primary amines and dimethylamine can be carried out
in water at room temperature, but more forcing condi-
tions (50 °C) are necessary with diethylamine giving a
mediocre reaction yield, and dibenzylamine does not
lead to the substitution product in water. This is a
limitation because heating provokes decomplexation
with formation of [FeII(amine)6][PF6]2, probably occur-
ring before the nucleophilic substitution reaction, be-
cause the chloride substituent weakens the iron–arene
bond, whereas the amino group strengthens it. Even
reactions at room temperature give some of this side
The
exergonic
single-electron
reduction
of
[FeIICp(h6-C6H5NEt2)][PF6] using the electron-reservoir
complex [FeICp(C6Me6)] [30,32] in the presence of 1
atm CO successfully led to decomplexation forming
[FeIICp(C6Me6)][PF6] and [FeICp(CO)2]2. It is also pos-
sible to carry out this decomplexation reaction in the
absence of CO, which is more convenient, but chro-
matographic purification is then necessary in order to
isolate the free aniline derivative. Despite these prelimi-
nary attempts, an extensive study with the aim to
recover good yields of the free arylamines on the
gramme scale with all the compounds was not carried
out. Anyway, the recovery of free amines using the
temporary stoichiometric complexation by transition-
metal groups is not a promising strategy in view of the
remarkable progress of the palladium catalyzed aryl-
coupling reactions. It is hoped that this recent progress
in palladium catalysis will finally provide a viable ami-
nation route of chloroaromatics [33,34]. Attempts to

128
F. Moulines et al. / Journal of Organometallic Chemistry 643–644 (2002) 125–129
use the CpFe+ moiety in catalytic amounts to activate
amination of chloroaromatics were not successful. The
synthesis and transformation of arene complexes in
water has an interest per se, however, because of the
introduction of the redox center and the potential use
of such organometallic materials as building blocks for
molecular electronics, for instance with non-linear opti-
cal properties. Finally, it should be noted along this line
that another family, the complexes [FeCp(h6-
ArCHR1NH2)][PF6] in which the amino group is lo-
cated in b position from the arene ligand, have been
obtained by Moinet by reduction of the corresponding
oximes [35].
4.2.2. Method B
1 mmol of the complex [FeCp(h6-ArCl)][X] (X=BF4
or PF6) previously reported in the literature was dis-
solved in 5 ml water and two equivalents amine was
added. The solution was left in air at r.t. for 5 h (case
of the water-soluble BF4 salt) or stirred (case of the
partly soluble PF6 salt), then extracted with twice 10 ml
methylene chloride. The methylene chloride solution
was dried over Na2SO4, filtered, concentrated under
reduced pressure to 5 ml, and the organometallic salt
was precipitated by addition of 20 ml ether. The result-
ing complex was characterized and identified by com-
parison with the 1H-NMR data of the literature
[16–22]. The yield was similar to that obtained using
method A.
4. Experimental
4.1. General
Acknowledgements
The solvents were purified as indicated elsewhere and
bi-distilled water was used. The NMR spectra were
recorded with a Brucker AC 200 (200MHz) spectrome-
ter. All the complexes [FeCp(h6-C6H5Cl)][X], [FeCp(h6-
o-C6H4Cl2)][X] and [FeCp(h6-ArNR1R2)][PF6] were
previously described and synthesized according to the
classic known procedures [13–23]. The complexes
[FeCp(h6-ArNR1R2)][PF6] were identified by compari-
son between the 1H-NMR spectra recorded in
CD3COCD3 using Me4Si as the reference and those
reported in the literature [13–23].
Fruitful discussions with Drs Laurent Gilbert and
Serge Ratton (Rhoˆne-Poulenc) and financial support
from Rhoˆne-Poulenc, the CNRS and the University
Bordeaux I is gratefully acknowledged.
References
[1] E. Kunz, Chemtech 17 (1987) 570.
[2] For other examples of water-soluble phosphines, see for in-
stance: R.T. Smith, M.C. Baird, Organometallics 2 (1983) 1138.
[3] For a recent review, see: B. Cornils, W.A. Herrmann, in: B.
Cornils, W.A. Herrmann (Eds.), Applied Homogeneous Cataly-
sis with Organometallic Compounds, VCH, Weinheim, 1996, pp.
575–601 chapter 3.1.1.1.
4.2. Procedures for the amination of the complexes
[FeCp(p6-ArCl)][X] (X=Cl, BF4 or PF6) in water
4.2.1. Method A
The aqueous solution resulting from slow hydrolysis
with 50 g ice-water of the ligand substitution reaction
between ferrocene and the arene in the presence of
AlCl3 (two equivalents) on a 5-mmol-scale [32,36] and
addition of 1 N NH3 until pH 10 followed by filtration
and washing Al(OH)3 with five times 20 ml water was
concentrated under reduced pressure to 5 ml. Then 10
ml 7 N aq. NH3 was added, and this homogeneous
reaction mixture was left at room temperature (r.t.) in
air for 5 h. One mmol aq. HPF6 was added (assuming
the yield of the ligand exchange reaction was not higher
than 20%), which led to the precipitation of a yellow–
brown solid. Extraction with twice 5 ml methylene
chloride, drying the methylene chloride solution, filtra-
tion, concentration to 5 ml under reduced pressure
followed by addition of 20 ml Et2O led to the precipita-
tion of a yellow powdery solid. This powder was
washed by refluxing with 10 ml ether or THF for 2 h
and filtered. The 1H-NMR spectrum of the yellow
complex was compared to literature data [16–22],
which confirmed the structure and purity of the com-
plex. See text for the yields.
[4] Former review: W.A. Herrmann, C.W. Kohlpaintner, Angew.
Chem. Int. Ed. Engl. 32 (1993) 1524.
[5] Green chemistry, in: P.T. Anastas, T.C. Williamson (Eds.), ACS
Symp. Ser. 626, ACS, Washington, DC, 1996.
[6] The Rio Environmental Declaration, United Nations Conference
on Environment and Development, Rio de Janeiro, June 3–4,
1992.
[7] F. Mathey, A. Sevin, Molecular Chemistry of the Transition
Elements. An Introductory Course, Wiley, New York, 1996.
[8] L.S. Hegedus, Transition Metals in the Synthesis of Complex
Organic Molecules, University Science Books, Mill Valley, CA,
1994.
[9] J. Miller, Aromatic Nucleophilic Substitution, Elsevier, New
York, 1968.
[10] J. March, Organic Chemistry, McGraw Hill, New York, 1985,
pp. 584–618 chapter 13.
[11] L. Balas, D. Jhurry, L. Latxague, S. Grelier, Y. Morel, M.
Hamdani, N. Ardoin, D. Astruc, Bull. Soc. Chim. Fr. 127 (1990)
401.
[12] F. Rose, E. Rose, Coord. Chem. Rev. 178–180 (1998) 249.
[13] A.N. Nesmeyanov, N.A. Vol’kenau, L.S. Isaeva, Dokl. Akad.
Nauk. SSSR 167 (1967) 106.
[14] A.N. Nesmeyanov, N.A. Vol’kenau, I.N. Bolesova, Dokl. Akad.
Nauk. SSSR 175 (1967) 606.
[15] A.N. Nesmeyanov, N.A. Vol’kenau, I.N. Bolesova, Dokl. Akad.
Nauk. SSSR 183 (1968) 354.

F. Moulines et al. / Journal of Organometallic Chemistry 643–644 (2002) 125–129
129
[16] J.F. Helling, W.A. Hendrickson, J. Organomet. Chem. 168
(1979) 87.
[17] C.C. Lee, C.I. Azogu, P.C. Chang, R.G. Sutherland, J.
Organomet. Chem. 234 (1981) 181.
[18] R.G. Sutherland, P.C. Chang, C.C. Lee, J. Organomet. Chem.
234 (1982) 197.
[19] C.C. Lee, U.S. Gill, M. Iqbal, C.I. Azogu, R.G. Sutherland, J.
Organomet. Chem. 231 (1982) 151.
[27] R. Davis, L.A.P. Kane-Maguire, in: G. Wilkinson, F.G.A.
Stone, E.W. Abel (Eds.), Comprehensive Organometallic Chem-
istry, Pergamon, Oxford, 1982, pp. 953–1077 chapter 26.
[28] L.A.P. Kane-Maguire, E.D. Honig, D.A. Sweigart, Chem. Rev.
84 (1984) 525.
[29] T.P. Gill, K.R. Mann, J. Organomet. Chem. 216 (1981) 65.
[30] J.-R. Hamon, D. Astruc, P. Michaud, J. Am. Chem. Soc. 103
(1981) 758.
[20] R.M. Moriarty, U.S. Gill, Organometallics 5 (1986) 253.
[21] U.S. Gill, R.M. Moriarty, Synth. React. Inorg. Met. Org. Chem.
16 (1986) 485.
[22] U.S. Gill, Inorg. Chim. Acta 114 (1986) L25.
[23] C.C. Lee, A.S. Abd El Aziz, R.L. Chowdhurry, U.S. Gill, A.
Pioko, R.G. Sutherland, J. Organomet. Chem. 315 (1986) 79.
[24] A.J. Pearson, J.G. Park, S.H. Yang, Y.H. Chuang, J. Chem.
Soc. Chem. Commun. (1989) 1363.
[31] J. Ruiz, F. Ogliaro, J.-Y. Saillard, J.-F. Halet, F. Varret, D.
Astruc, J. Am. Chem. Soc. 120 (1998) 11693.
[32] D. Astruc, in: V. Balzani (Ed.), Electron Transfer in Chemistry,
vol. 2, in: P. Mattay, D. Astruc (Eds.), Section 2, Inorganic and
Organometallic Compounds, VCH-Wiley, New York, 2001, pp.
714–803 (chapter 4).
[33] J.F. Hartwig, Acc. Chem. Res. 31 (1998) 852.
[34] S.J. Buchwald, Acc. Chem. Res. 31 (1998) 805.
[35] C. Moinet, E. Raoult, J. Organomet. Chem. 229 (1982) C13.
[36] D. Astruc, J.-R. Hamon, M. Lacoste, M.-H. Desbois, E. Ro-
ma´n, in: R.B. King (Ed.), Organomet. Synthesis, vol. IV, El-
sevier, New York, 1988, pp. 172–187.
[25] D. Astruc, Tetrahedron 39 (1983) 4027.
[26] D. Astruc, in: F.R. Hartley, S. Patai (Eds.), The Chemistry of
the Metal–Carbon Bond, vol. IV, Wiley, New York, 1987, pp.
625–731.