Copper(I)-mediated synthesis of ferrocenyl alkyl ethers

Article abstract of DOI:10.1021/om1006209

The copper(I)-mediated Ullman-type coupling of iodoferrocene and diverse alcoholates has been used for the preparation of a series of ferrocenyl alkyl ethers. In this manner oxygen-substituted ferrocenes that cannot be synthesized via the classical Williamson ether synthesis are accessible in good yields. The structure of three samples in the solid state is reported.

Full text of DOI:10.1021/om1006209

4196 Organometallics 2010, 29, 4196–4198
DOI: 10.1021/om1006209
Copper(I)-Mediated Synthesis of Ferrocenyl Alkyl Ethers†
Dieter Schaarschmidt and Heinrich Lang*
€
€
Institut fu€r Chemie, Lehrstuhl fu€r Anorganische Chemie, Fakultat fur Naturwissenschaften, Technische
Universita€t Chemnitz, Strasse der Nationen 62, 09111 Chemnitz, Germany
Received June 28, 2010
Summary: The copper(I)-mediated Ullman-type coupling of
iodoferrocene and diverse alcoholates has been used for the
preparation of a series of ferrocenyl alkyl ethers. In this
manner oxygen-substituted ferrocenes that cannot be synthe-
sized via the classical Williamson ether synthesis are accessible
in good yields. The structure of three samples in the solid state
is reported.
We report here on the copper(I)-mediated Ullman-type
coupling of iodoferrocene with diverse alcoholates to give
ferrocenyl alkyl ethers.
Results and Discussion
Recently, we have reported on novel planar chiral P,O-
ferrocenes that show unrivaled activity for the Suzuki cou-
pling of hindered substrates.9 For the synthesis of an en-
antiomerically enriched P,O-ferrocene we intended to apply
enantiomerically pure ferrocenyl ethers. Scheme 1 sum-
marizes our attempts to synthesize 3a starting from iodofer-
rocene 2.
Introduction
In the last decades ferrocene has emerged as an essential
motif in organometallic chemistry.1 It has found numerous
applications in academia and industry; the spectrum ranges
from its use as a highly selective and efficient ligand in
asymmetric homogeneous catalysis2 to fuel additives for
diesel engines.3
The field of oxygen-substituted ferrocenes is one aspect that
is still not fully developed. This may be attributed to the
sensitivity of the phenol analogue hydroxyferrocene4 or the
tendency of oxygen electrophiles, e.g. peroxides or peresters,
to irreversibly oxidize the iron center.5 Since 1959, when
Nesmeyanov and co-workers reported on the synthesis of
some ferrocenyl alkyl ethers via etherification of hydroxy-
ferrocene,6 only a few new molecules have been prepared
utilizing mostly primary alkyl halides or tosylates.7 To our
knowledge, the successful application of neither secondary nor
tertiary alkyl derivatives has been reported so far.
All our attempts to prepare 3a via the classical Williamson
ether synthesis failed; neither menthyl chloride, menthyl
methansulfonate, nor menthyl 4-nitrobenzenesulfonate was
successfully attacked by 1. As menthyl derivatives seem to be
relatively inert toward a nucleophilic attack, we decided to
use menthol as the nucleophile in a copper(I)-mediated
Ullman-type coupling with iodoferrocene. At first an O-ar-
ylation protocol (CuI, 1,10-phenanthroline, Cs2CO3, to-
luene, 90 °C) published by Buchwald et al.10 was used.
Unfortunately, this methodology did not result in the for-
mation of desired 3a. Following this, a modified reaction
protocol developed by Bolm et al. for the synthesis of nitrogen-
substituted ferrocenes (CuI, KOtBu, NMP, 70 °C)11 was
applied. This procedure allowed isolating 3a apart from un-
reacted 2 in approximately 10% yield; however, the product
was contaminated with FcOtBu (Fc = Fe(η5-C5H4)(η5-
C5H5)). This prompted us to deprotonate menthol with sodium
hydride prior to use rather than to apply menthol and a base in
the coupling reaction, which indeed improved the yield of 3a
(82%) significantly and avoided the formation of unwanted
FcOtBu.
In contrast, ferrocenyl aryl ethers are easily accessible via
copper(I)-mediated coupling of iodoferrocene and various
phenols as demonstrated by Plenio et al. recently.8
† Dedicated to Professor Stefan Spange on the occasion of his 60th
birthday.
*To whom correspondence should be addressed. E-mail: heinrich.lang@
chemie.tu-chemnitz.de.
The optimized reaction conditions (3.0 equiv of NaOR,
1.0 equiv of CuI, 0.2 equiv of 2,20-bipyridine, 70 °C, 18 h) for
the synthesis of ferrocenyl alkyl ethers were applied to
several alcohols ROH to screen the synthetic potential of
this coupling reaction (Table 1).12
As can be seen from Table 1 secondary (entries 1-6) and
primary (entries 7-10) alcohols could successfully be applied
in this Ullman-type coupling. Within this reaction, besides 3,
(1) (a) Ferrocenes: Ligands, Materials and Biomolecules; Stepnicka, P.,
Ed.; John Wiley & Sons Ltd: Chichester, 2008. (b) Ferrocenes: Homoge-
neous Catalysis, Organic Synthesis, Material Science; Togni, A., Hayashi,
T., Eds.; VCH Verlagsgesellschaft: Weinheim, 1995.
(2) Dai, L.-X.; Tu, T.; You, S.-L.; Deng, W.-P.; Hou, X.-L. Acc.
Chem. Res. 2003, 36, 659.
(3) Ferrocene is added to diesel fuel in order to improve combustion
and to reduce the emission of soot.
(4) Nesmeyanov et al. report that hydroxyferrocene decomposes
within several days when exposed to air.
(5) Schaarschmidt, D.; Lang, H. Unpublished results.
(6) (a) Nesmeyanov, A. N.; Sazanova, V. A.; Drozd, V. N. Tetra-
hedron Lett. 1959, 17, 13. (b) Nesmejanow, A. N.; Ssasonowa, W. A.; Drosd,
V. N. Chem. Ber. 1960, 93, 2717.
(7) (a) Akabori, S.; Sato, M.; Ebine, S. Synthesis 1981, 278. (b)
Akabori, S.; Ohtomi, M.; Sato, M.; Ebine, S. Bull. Chem. Soc. Jpn. 1983,
56, 1455. (c) Singewald, E. T.; Mirkin, C. A.; Stern, C. L. Angew. Chem., Int.
Ed. Engl. 1995, 34, 1624. (d) Anderson, J. C.; Blake, A. J.; Arnall-Culliford,
J. C. Org. Biomol. Chem. 2003, 1, 3586.
(9) (a) Schaarschmidt, D.; Lang, H. Eur. J. Inorg. Chem. 2010,
accepted, doi: 10.1002/ejic.201000722. (b) Schaarschmidt, D.; Lang, H.
DE Patent 10 2010 001 364.1, 2010.
(10) Shafir, A.; Lichtor, P. A.; Buchwald, S. L. J. Am. Chem. Soc.
2007, 129, 3490.
€
(11) Ozc-ubukc-u, S.; Schmitt, E.; Leifert, A.; Bolm, C. Synthesis 2007,
389.
(8) an der Heiden, M. R.; Frey, G. D.; Plenio, H. Organometallics
2004, 23, 3548.
(12) Changing the solvent to acetonitrile or dmso did not affect the
yield positively.
r
pubs.acs.org/Organometallics
Published on Web 08/31/2010
2010 American Chemical Society

Note
Organometallics, Vol. 29, No. 18, 2010 4197
Scheme 1. Synthesis of 3aa
Table 2. Comparison of Ullman-Type Coupling and Williamson
Ether Synthesis
a (a) 1. AcOH, Cu2O, CH3CN, reflux, 3 h (93%); 2. KOH, H2O,
EtOH, reflux, 20 min; 3. HCl, H2O, EtOH, ambient temperature, 10 min
(95%); (b) base, menthylX (X = Cl, OS(O)2CH3, OS(O)2-4-NO2-C6H4),
solvent, ambient temperature to 100 °C, 12-24 h; (c) CuI, 2,20-bipyr-
idine, NaOmenthyl, NMP, 70 °C, 18 h (82%).
Table 1. Ullman-Type Coupling of Iodoferrocene and Various
Alcohols to Ferrocenyl Alkyl Ethersa
a Based on isolated material. b Reaction conditions: 1.0 equiv of
iodoferrocene, 3.0 equiv of NaOR, 1.0 equiv of CuI, 0.2 equiv of 2,20-
bipyridine, NMP (5 mL/mmol iodoferrocene), 70 °C, 18 h. c Reaction
conditions: 1.0 equiv of FcOAc, 2.0 equiv of KOH, 2.0 equiv of RX (X =
Br, Cl, OS(O)2CH3, OS(O)2-4-NO2-C6H4, DMF (7 mL/mmol FcOAc),
100 °C, 18 h.
Scheme 2. Ullman-Type Coupling of 1,10-Diiodoferrocene and
Menthola
a (a) 1.0 equiv of 1,10-diiodoferrocene, 6.0 equiv of NaOmenthyl, 2.0
equiv of CuI, 0.4 equiv of 2,20-bipyridine, 10 mL of NMP, 70 °C, 18 h.
following order of reactivity can be established: secondary
> primary . tertiary alcohols. Considering the secondary
alcohols, it is obvious that steric hindrance at the β-carbons
improves the yield of the reaction (Table 1, entries 1, 2, 5, and
6 vs 4) as long as the system is not overcrowded (Table 1,
entry 3).
a Reaction conditions: 1.0 equiv of iodoferrocene, 3.0 equiv of NaOR,
1.0 equiv of CuI, 0.2 equiv of 2,20-bipyridine, NMP (5 mL/mmol
iodoferrocene), 70 °C, 18 h. b Based on isolated material. c 0.1 equiv of
CuI. d 2.0 equiv of NaOR.
A beneficial property of this Ullman-type coupling is that
chiral alcohols can be applied without any loss of stereo-
information. The reaction runs with full retention of config-
uration, which could be confirmed by NMR experiments
and X-ray diffraction studies.
We also tried to perform this reaction catalytically based
on copper(I) (Table 1, entry 1, 0.1 equiv of CuI), which
resulted in a dramatic loss of activity. Nevertheless, the yield
of this experiment revealed that this coupling reaction is at
least substoichiometric concerning copper(I).
Compared to the classical Williamson ether synthesis the
presented Ullman-type coupling is especially beneficial when
sterically demanding alcohols are applied (Table 2, entries 1
and 2 vs 3 and 4). Obviously both synthesis methodologies
are complementary.
Also 1,10-diiodoferrocene was applied in this coupling
reaction. Using excess menthol resulted in the formation of
disubstituted ferrocene 5 besides 3a (Scheme 2, ratio 1:1) as
the only products. A ferrocene proving a 2-fold reaction
only unreacted iodoferrocene 2 and ferrocene could be
identified.
While the coupling of primary and secondary alcohols is
well documented in the literature, to the best of our know-
ledge, the use of tertiary alcohols has not been reported so
far,13 and hence, 2 was reacted with NaOR (R = tBu, Ph3C,
1-adamantanyl). It was found that the coupling occurred but
that the appropriate products were formed only in poor
yields (<5%); however, in the case of R = 1-adamantyl the
formation of the ferrocenyl ether was not observed.
Obviously, secondary alcohols show better results in this
Ullman-type coupling reaction than primary ones. The
(13) (a) Naidu, A. B.; Sekar, G. Tetrahedron Lett. 2008, 49, 3147. (b)
Niu, J.; Zhou, H.; Li, Z.; Xu, J.; Hu, S. J. Org. Chem. 2008, 73, 7814. (c)
Wolter, M.; Nordmann, G.; Job, G. E.; Buchwald, S. L. Org. Lett. 2002, 4,
973. (d) Naidu, A. B.; Jaseer, E. A.; Sekar, G. J. Org. Chem. 2009, 74, 3675.
(e) Chang, J. W. W.; Chee, S.; Mak, S.; Buranaprasertsuk, P.; Chavasiri, W.;
Chan, P. W. H. Tetrahedron Lett. 2008, 49, 2018. (f) Zhang, H.; Ma, D.; Cao,
W. Synlett 2007, 243.

4198 Organometallics, Vol. 29, No. 18, 2010
Schaarschmidt and Lang
Figure 3. ORTEP diagram of 3e. Thermal ellipsoids are at 50%
˚
probability. Selected bond distances (A) and angles (deg): D1-Fe1
1.6591(2), D2-Fe1 1.6523(2), C1-O1 1.372(2), O1-C11 1.4543(18),
D1-Fe1-D2 177.80(2), C1-O1-C11 117.48(13) (D1 = centroid
C1-C5, D2 = centroid C6-C10).
All new ferrocenes have been characterized by standard
analytical methods including NMR spectroscopy, elemental
analysis, and mass spectrometry. For 3a, 3c, and 3e single
crystals could be obtained by crystallization from n-pentane
at -40 °C. Figures 1 to 3 show the results of the X-ray
Figure 1. ORTEP diagram of 3a. Thermal ellipsoids are at 50%
˚
probability. Selected bond distances (A) and angles (deg): D1-Fe1
1.6549(4), D2-Fe1 1.6529(4), C1-O1 1.365(2), O1-C11 1.442(3),
D1-Fe1-D2 179.50(2), C1-O1-C11 117.67(19) (D1 = centroid
C1-C5, D2 = centroid C6-C10).
˚
diffraction analyses. Representative bond distances (A) and
angles (deg) are given in the captions of these figures. The
ferrocenyl moiety in 3c exhibits a staggered conformation,
whereas 3a and 3e show an almost eclipsed orientation of the
two cyclopentadienyl rings.
Conclusion
Within this study the straightforward preparation of
ferrocenyl alkyl ethers by an Ullman-type coupling of iodo-
ferrocene and various alcohols has been developed, which is
in the case of chiral alcohols highly diastereoselective. Pri-
mary and secondary alcohols can successfully be applied in
this reaction, whereas tertiary alcohols are almost unreactive.
A comparison of the developed coupling reaction with the
classical Williamson ether synthesis revealed that both meth-
odologies are complementary. Especially when secondary
alcohols are applied, the respective ethers that are not accessible
via etherification of ferrocenol can be isolated in good yields,
which enriches this field of chemistry. Nevertheless, there is still
a need for a rational synthesis route that utilizes tertiary alkyl
derivatives.
Figure 2. ORTEP diagram of 3c. Thermal ellipsoids are at 50%
˚
probability. Selected bond distances (A) and angles (deg): D1-Fe1
1.6563(3), D2-Fe1 1.6560(3), C1-O1 1.360(2), O1-C11 1.447(2),
D1-Fe1-D2 179.17(2), C1-O1-C11 117.65(14) (D1 = centroid
C1-C5, D2 = centroid C6-C10).
Acknowledgment. This work was supported by the
Deutsche Forschungsgemeinschaft and the Fonds der
Chemischen Industrie. D.S. thanks the Fonds der Che-
mischen Industrie for a fellowship. We wish to thank
Mr. D. Ress for experimental assistance.
could not be identified. Unfortunately we were not able to
separate the two ferrocenyl ethers by column chromatogra-
phy. The formation of monosubstituted 3a can be explained
by hydrodehalogenation (vide supra).14
It must be mentioned that the ferrocenyl ethers (3a, 3b, and
3i) tend to decompose slowly in solution, to give a para-
magnetic species that could be removed by filtration.
Supporting Information Available: Text and figures giving
further experimental and spectroscopic details and CIF files
giving crystallographic data. This material is available free of
charge via the Internet at http://pubs.acs.org. Crystallographic
data for 3a, 3c, and 3e are also available from the Cambridge
Crystallographic Database as file nos. CCDC 781730, 781732,
and 781731.
(14) Tye, J. W.; Weng, Z.; Giri, R.; Hartwig, J. F. Angew. Chem. 2010,
122, 2231.