Enantioselective functionalisation of the C-2′ position of 1,2,3,4,5-pentamethylazaferrocene via sparteine mediated lithiation: Potential new ligands for asymmetric catalysis

Article abstract of DOI:10.1039/b715323f

The s-BuLi-sparteine base combination deprotonated the C-2′ position of 1,2,3,4,5-pentamethylazaferrocene 1 and subsequent reaction with a range of electrophiles gave C-2 substituted products in 76-93% yield and ～80% ee. The products could be recrystallised to enrich ee's to >90%. Resubjection of the initial addition products (～80% ee) to the deprotonation conditions led to a kinetic resolution to give products with >90% ee and superior overall yields compared to recrystallisation for the cases where the electrophiles were Ph₂CO, MeI and Ph₂S₂. Transmetallation of the 2-lithiopentamethylazaferrocene (～80% ee) with ZnCl₂ allowed palladium catalysed cross coupling with a variety of C-2 haloaryl, heteroaryl and vinyl groups to give some novel C-2′ substituted pentamethylazaferrocene derivatives in 61-77% yield in 80% ee. Potential N,N-chelate ligands were recrystallised to >95% ee. A novel C ₂-symmetric bis-pentamethylazaferrocene 10 could be synthesised by an iron catalysed oxidative coupling of the enatioenriched C-2 lithio derivative and in the presence of a PhMe-Et₂O solvent mixture proceeded in 97% ee. This journal is The Royal Society of Chemistry.

Full text of DOI:10.1039/b715323f

View Article Online / Journal Homepage / Table of Contents for this issue
PAPER
www.rsc.org/obc | Organic & Biomolecular Chemistry
Enantioselective functionalisation of the C-2ꢀ position of
1,2,3,4,5-pentamethylazaferrocene via sparteine mediated lithiation:
potential new ligands for asymmetric catalysis
James C. Anderson,*a James D. Osbornea and Thomas J. Wolteringb
Received 5th October 2007, Accepted 6th November 2007
First published as an Advance Article on the web 29th November 2007
DOI: 10.1039/b715323f
The s-BuLi–sparteine base combination deprotonated the C-2ꢀ position of
1,2,3,4,5-pentamethylazaferrocene 1 and subsequent reaction with a range of electrophiles gave C-2
substituted products in 76–93% yield and ∼80% ee. The products could be recrystallised to enrich ee’s
to >90%. Resubjection of the initial addition products (∼80% ee) to the deprotonation conditions led
to a kinetic resolution to give products with >90% ee and superior overall yields compared to
recrystallisation for the cases where the electrophiles were Ph2CO, MeI and Ph2S2. Transmetallation of
the 2-lithiopentamethylazaferrocene (∼80% ee) with ZnCl2 allowed palladium catalysed cross coupling
with a variety of C-2 haloaryl, heteroaryl and vinyl groups to give some novel C-2ꢀ substituted
pentamethylazaferrocene derivatives in 61–77% yield in 80% ee. Potential N,N-chelate ligands were
recrystallised to >95% ee. A novel C2-symmetric bis-pentamethylazaferrocene 10 could be synthesised
by an iron catalysed oxidative coupling of the enatioenriched C-2 lithio derivative and in the presence of
a PhMe–Et2O solvent mixture proceeded in 97% ee.
Introduction
The ferrocenyl motif is a sturdy platform for the synthesis of
planar chiral ligands for asymmetric catalysis. The dominance
of the planar chiral element in ferrocenyl derived ligands above
centres of chirality for the transmission of asymmetry in chemical
reactions has been well documented.1–5 Purely planar chiral
ferrocenes have been less well investigated, possibly due to the
challenges associated with enantioselective syntheses compared
Fig. 1 Design concept.
to the multitude of diastereoselective approaches, for example
to the catalytically active metal centre would give ligand systems
strategies that start with the ubiquitous Ugi’s N,N-dimethyl-1-
ferrocenylethylamine.6,7 The pioneering work of Fu et al., although
relying on resolution to synthesise planar chiral ferrocenes,
demonstrated the power of planar chiral metallocene DMAP
derivatives in the field of nucleophilic catalysis.8 These results
stimulated other work in the ferrocene field and routes to similar
systems9 and chelate ligands were developed.2,10–13 In our own
work we have used the Sniekus approach of sparteine mediated
enantioselective directed ortho metallation of N,N-diisopropyl
ferrocenecarboxamide14 to access a series of N–P, N–N, N–O and
N–S planar chiral ferrocene chelate ligands (Fig. 1).15,16 These were
shown to be less efficient than non-ferrocene N–X chiral ligand
systems that rely upon centres of chirality in palladium catalysed
asymmetric allylic alkylation.16 Although the 1,2-disubstituted
ferrocene systems we surveyed have a sterically large planar
chiral element, the nature of the substituents placed the steric
bulk of the metallocene some way away from the reaction centre
(Fig. 1, 1st generation design). Movement of the ferrocene closer
with potentially more efficient chirality transfer. This could be
most easily achieved by either synthesising 1,2-bidentate ferrocene
systems containing both heteroatoms bonded to one of the
cyclopentadienyl rings, or take advantage of a heterometallocene,
such as an azaferrocene (Fig. 1, 2nd generation design). Without
an obvious asymmetric synthesis to 1,2-heteroatom disubstituted
ferrocenes we focussed our attention on azaferrocenes, which can
be made planar chiral by substitution at either the C-2ꢀ or C-3ꢀ
position
The investigation of the versatility of planar chiral het-
erometallocenes in asymmetric catalysis has been dominated
by Fu et al. who have looked at 2-substituted pentamethy-
5
lazaferrocenes,8 pentamethylphosphaferrocenes17 and (g -1,2-
azaborolyl)iron derivatives.18 Their syntheses of planar chiral aza-
ferrocenes have relied upon preparative chiral HPLC to resolve the
two enantiomers.8c,19 We sought an efficient and direct multi-gram
enantioselective synthesis of C-2ꢀ substituted azaferrocenes. At the
outset of this work there was only one general route to enantiopure
2ꢀ-substituted pentamethylazaferrocenes that had been disclosed
by Johannsen et al.20 In that work, in a similar manner to the use
of chiral sulfoxides by Kagan et al. for the diastereoenrichment of
planar chiral ferrocenes,21 regioselective deprotonation at C-2ꢀ of
pentamethylazaferrocene and addition of an Andersen reagent22
aSchool of Chemistry, University of Nottingham, Nottingham, UK NG7
2RD. E-mail: j.anderson@nottingham.ac.uk; Fax: +44(0)115 9513564;
Tel: +44(0)115 9514194
bPharma Division, Preclinical CNS Research, F. Hoffmann-La Roche AG,
CH-4070 Basel, Switzerland
330 | Org. Biomol. Chem., 2008, 6, 330–339
This journal is
The Royal Society of Chemistry 2008
©

View Article Online
Table 1 Optimisation of enantioselective deprotonationa
Entry
s-BuLi (equiv.)
(−)-Sparteine (equiv.)
Solvent
Temp/◦C
Yield 2 (%)
eeb (%)
1
2
3
4
5
6
7
8
9
10
11
12c
13c
2.2
2.2
2.2
2.2
2.2
2.2
2.2
2.2
3.2
2.2
1.2
1.2
1.2
2.2
2.2
2.2
2.2
3.2
4.2
5.2
6.2
3.2
1.2
1.2
1.2
1.2
Et2O
i-Pr2O
PhMe
Hexane
Et2O
Et2O
Et2O
Et2O
Et2O
Et2O
Et2O
Et2O
Et2O
−78
−78
−78
−78
−78
−78
−78
−78
−78
−78
−78
−90
−100
84
60
70
66
68
82
70
84
5
77
84
82
85
77
77
77
77
77
77
70
77
77
77
77
80
83
a 4 h then addition of I2 (1 equiv. to s-BuLi). b Determined by HPLC on a Chiralcel OD-H column. c 6 h.
gave separable diastereomeric sulfoxides (Scheme 1). Treatment
of either diastereoisomer with BuLi led to an enantiomerically
in THF for 4 hours, followed by the addition of I2 (2.2 equiv.) gave
(−)-(Sp)-2-iodopentamethylazaferrocene20b 2a in a moderate 46%
yield and 10% ee [eqn (1)].24 Preliminary optimisation quickly
revealed that using the slightly less coordinating solvent Et2O
and the more reactive s-BuLi gave the optimum mixture of rate
of reaction and expression of chiral information by competitive
coordination of (−)-sparteine25 to give iodide 2a in 84% yield and
77% ee (Table 1, entry 1). Less coordinating solvents did not alter
the extent of enantioselection, but did have a detrimental effect on
the rate of reaction as shown by the isolated yield after 4 h (entries
2–4). Excess (−)-sparteine did not alter the enantioselectivity or
the yield to any significant extent (entries 5–8). Although an excess
of s-BuLi to (−)-sparteine was tolerated (entry 10), 3.2 equiv. of
both s-BuLi and (−)-sparteine were deleterious to the chemical
yield (entry 9). A decrease in temperature gave slightly better yields
and enantioselectivites, but at the expense of reaction times and
convenience (entries 12 and 13). Based on these results we◦decided
that 1.2 equiv. of both s-BuLi and (−)-sparteine at −78 C were
the most practical conditions (entry 11).
t
pure carbanion that could be quenched with a range of elec-
trophiles. However the overall yield of the process was moderate,
ranging between 15–40%. In view of the success of Snieckus’
use of enantioselective metallation mediated by sparteine on
dialkyl ferrocenecarboxamides14 we investigated the possibility
of using a (−)-sparteine–butyllithium system to differentiate the
enantiotopic 2ꢀ,5ꢀ-pentamethylazaferrocene protons. During our
preliminary investigations a communication by Iwao et al. also
verified that s-BuLi–sparteine could be used to enantioselectively
metallate pentamethylazaferrocene at the C-2 position and that
it reacted with a limited set of electrophiles (Scheme 1).23 The
results we present in this paper considerably extend the preliminary
results of Iwao et al. in that a broader range of electrophiles were
investigated and we show that enantioenrichment can be achieved
by kinetic resolution or traditional recrystallisation. We also report
our investigations into the synthesis of planar chiral analogues of
some privileged ligand structures.
(1)
In order to establish the generality of the reaction the optimised
procedure was assayed with a variety of electrophiles [eqn (2),
Table 2]. The products were isolated in around 90% yield and 80%
ee (by chiral HPLC) with the exception of the iodo derivative 2a,
formed in 80% yield, and the diphenylphosphine derivative 2f that
was formed in 57% optical purity. Iodide 2a was formed in slightly
lower yield because of the tendency for it to convert back to 1 in
the presence of excess s-BuLi. The optical purity of phosphine 2f
is treated as a conservative estimate and, in view of the fact that
some of the optical rotations we measured for other derivatives
2 gave lower optical purities than ee’s determined by HPLC, the
ee of 2f could be considerably higher. We noticed no erosion of
ee in contrast to compound 13 (vide infra). We were unable to
isolate boronic acid derivative 2g by standard chromatography
or recrystallisation. The enantioselectivities of 2c and 2d were
extrapolated from derivatives. The enantiopurity of the methyl
Scheme 1 Direct precedent.
Results and discussion
Treatment of 1,2,3,4,5-pentamethylazaferrocene 1 with n-BuLi
(2.2 equiv.) in the presence of (−)-sparteine (2.2 equiv.) at −78 ◦C
This journal is
The Royal Society of Chemistry 2008
Org. Biomol. Chem., 2008, 6, 330–339 | 331
©

View Article Online
Table 2 Scope of reaction
2
E+
E
Yield (%)
ee (%)
Recryst. yield (%)f
ee (%)
a
b
c
d
e
f
I2
I
80
86
86
93
93
93
77b
82b
81c
80c
79d
57e
—
52
26
50
90b
97b
Ph2CO
TMSCl
MeI
Ph2S2
Ph2PCl
B(OMe)3
Ph2COH
TMS
Me
SPh
PPh2
B(OH)2
>99%a
—
g
—
48
64
—
96d
86e
a
g
—
—
a Not isolable. b Determined by HPLC on a Chiralcel OD-H column.20b c Determined by HPLC of a derivative (see text). d Determined by HPLC on a
Chiralcel AD column.20b e Optical purity based upon polarimetry data from Ref. 20b. f Et2O–hexane mixtures. g Soluble in hexane.
derivative 2d was extrapolated from that of 3, which was formed
by a second s-BuLi deprotonation in the presence of N,N,Nꢀ,Nꢀ-
tetramethyl-1,2-ethylenediamine (TMEDA) followed by addition
of benzophenone [eqn (3)].
place and lead to enantioenrichment of the initial product. In
addition, deprotonation of the minor enantiomer (Rp)-2 followed
by electrophilic quenching would lead to a meso compound which
should be readily separable from (Sp)-2.
(2)
(3)
In a similar manner the enantiopurity of 2c was extrapolated
from alcohol 4. One of the drawbacks of using (−)-sparteine is
that it is only available in one enantiomeric form, although there
are some excellent (+)-sparteine-like diamines now available.26
However Dai et al. demonstrated the protection of prochiral
protons with a TMS group and subsequent deprotection with
tetra-n-butylammonium fluoride (TBAF) in ferrocene systems.27
Deprotonation of a sample of 2c [from eqn (2)] with s-BuLi–
TMEDA and addition of benzophenone gave alcohol 5 in 64%
yield (Scheme 2). Subsequent treatment with TBAF in refluxing
THF gave (−)-(Rp)-alcohol 4 in 56% yield and 81% ee by chiral
HPLC.24
Scheme 3 Kinetic resolution studies.
Optimum conditions were based upon resubjection of purified
2, ∼80% ee (Table 2) to sub-stoichiometric quantities of s-BuLi–
(−)-sparteine (0.25 equiv.). Theoretically around 0.11 equiv. of
s-BuLi–(−)-sparteine could consume virtually all the minor (Rp)-
2 isomer, but we found 0.25 equiv. to be better both in terms of
yield and ee. Substrates in which the ee could be determined easily
were examined and an enhancement of ee was observed for 2b,
d and e (Table 3). The kinetic resolution gives higher material
throughput than upgrading ee by recrystallisation. The virtually
identical ee and isolated yield obtained for 2a (Table 3) after
attempted kinetic resolution, suggests a non-selective halogen–
lithium exchange reaction rather than any deprotonation.29 It was
Scheme 2
In most cases a single recrystallisation of the products enriched
the ee’s of the 2ꢀ-substituted pentamethylazaferrocenes to 90%
or more (Table 2). We thought a conceptually more appealing
method of enriching the initial enantioselectivity could be by
kinetic resolution.28 Once the products 2 have been formed there
exists a preponderance of the Sp product arising from kinetic
deprotonation of the pro-S proton of 1 (Scheme 3). Assuming
that the rate of a second proton abstraction by further treatment
with a calculated amount of s-BuLi–(−)-sparteine would be faster
for the pro-S proton of (Rp)-2 rather then the pro-R proton
of (Sp)-2 (k2 > k1), then a kinetic resolution of 2 could take
Table 3 Kinetic resolution of products 2
2
E
I
Yield (%) Yield meso-6 (%) Overall yield (%) ee (%)
a
b
d
e
80
—
—
—
19
61
68
64
69
79a
90a
Ph2COH 79
Me
SPh
69
74
>99b
93c
a Determined by HPLC on a Chiralcel OD-H column. b Determined by
HPLC of derivative 3 (see text). c Determined by HPLC on a Chiralcel AD
column.
332 | Org. Biomol. Chem., 2008, 6, 330–339
This journal is
The Royal Society of Chemistry 2008
©

View Article Online
only possible to isolate meso-6e, but we believe this supports
the postulated mechanism for this kinetic resolution. Many
attempts were made to optimise a one pot process whereby sub-
stoichiometric quantities of s-BuLi–(−)-sparteine and/or E+ were
added after the initial reaction. We found the reactions with E+
were temperature dependent and we believe byproducts formed
from the first excess of s-BuLi reacting with E+ also reacted with
subsequent additions of reagents. These factors, amongst others,
greatly complicated the development of a one pot reaction. A
process of optimisation did lead to a one pot protocol being
developed for the synthesis of 2e in 60% yield and 90% ee [eqn
(4)]. Ultimately the two step protocol provides a very efficient
route to highly enantioenriched 2-substituted azaferrocenes and
may be a viable route to more complex derivatives.
(5)
Table 4 Cross coupled products
7
R–X
Ph–I
Time/h
Yield (%)
eea (%)
c
c
a
b
5
6
77
68
c
2
4
70
68
80
80
d
(4)
e
f
2
2
62
77
80
To increase the structural diversity of potential ligand systems
and other derivatives that could be generated by this approach
we speculated that certain enantioenriched 2ꢀ-substituted pen-
tamethylazaferrocene derivatives could be used as precursors
for transition metal catalysed cross coupling reactions. Some
preliminary experiments with iodide 2a revealed that although
insertion of Pd(0) was possible, it required microwave conditions
and stoichiometric Pd(PPh3)4. The presumed Pd(II) intermediate
(by MS and TLC) was found to be reluctant to undergo any
standard coupling reaction. We then speculated that the enan-
tioenriched 2ꢀ-lithiopentamethylazaferrocene intermediate could
undergo transmetallation in order to participate in a tran-
sition metal catalysed cross coupling reaction with a 2-halo
heteroaromatic.30 Isolation of a boron derivative had proven
difficult (Table 2), which immediately ruled out investigation of
the popular Suzuki cross coupling reaction. However, there are
several literature precedents for performing Negishi couplings
on ferrocenes.31 We began a similar study by transmetallating
the enantioenriched 2-lithiopentamethylazaferrocene with ZnCl2
(1.2 equiv.) in Et2O, but quickly found we needed an excess of
ZnCl2 (4 equiv.) and PhMe as solvent so we could conduct the
reaction at higher temperatures. We speculate that excess ZnCl2 is
necessary to sequester (−)-sparteine which could otherwise chelate
to the zincated pentamethylazaferrocene and deactivate towards
the palladium catalyst. We were able to conduct coupling with
aryl, heteroaryl and alkenyl halides in high yield and identical
enantioselectivity [eqn (5) and Table 4] to the direct reaction
with electrophiles (Table 1). Slightly longer reaction times were
necessary for unactivated halides 7b,d,g. A larger excess of ZnCl2
(6 equiv.) was needed for the 2-quinoline derivative 7d as this
product may play a similar chelating and deactivating role as
mentioned for (−)-sparteine as it is formed in the reaction mixture.
The ee of the heteroaryl derivatives 7c–f were all measured by chiral
HPLC and were consistent with previous results. We therefore
think it is not unreasonable to assume that aryl 7a,b and alkenyl
7g derivatives are also formed in 80% ee. In addition the ee of
pyridyl 7c and 2-quinoline 7d derivatives could be enriched to 95%
72b
c
g
4
61
a Determined by HPLC on a Chiralcel OD-H column. b ZnCl2 (6 equiv.).
c Assumed ∼80% ee, see text.
by a single recrystallisation from Et2O. Although the overall yield
after recrystallisation from azaferrocene for 7c was disappointing
(9%), that for 7d was much better (36%). The nature of this
palladium coupling seems to be robust to a variety of activated and
unactivated coupling partners and demonstrates that it is possible
to synthesise a range of N,N- and N,S-ligands in good yield and
enantiomeric excess.
With an efficient cross coupling protocol developed we in-
vestigated the synthesis of planar chiral ferrocene analogues of
privileged ligand systems. Fu et al. have also published the synthe-
sis and utility of methylene bridged bis-pentamethylazaferrocene
88c,32 and we were interested in synthesising an analogue 9 using
the s-BuLi–(−)-sparteine chemistry developed above (Fig. 2).
Both of these bis-azaferrocene skeletons have similarities with
the popular and very effective bis-oxazoline type ligands.33
The synthesis of dimer 9 by classic Ullmann chemistry was
found to be inappropriate because of the thermal instability of
Fig. 2 Bis-pentamethylazaferrocenes.
This journal is
The Royal Society of Chemistry 2008
Org. Biomol. Chem., 2008, 6, 330–339 | 333
©

View Article Online
pentamethylazaferrocene toward the high temperatures involved
in such couplings.34 Stoichiometric Ni coupling failed despite
its use in electron rich pyridine dimerisations.35 During the
development of the cross coupling chemistry above we had some
success in preparing the para-chloro derivative of 7a from iodopen-
tamethylazaferrocene 2a via iodine–magnesium exchange with i-
PrMgCl36 followed by transmetallation with ZnCl2. Coupling with
1-chloro-4-iodobenzene using Pd(PPh3)4 as a catalyst in refluxing
THF gave coupled product in 49% yield.37 Attempting an identical
procedure but with 2a also as the electrophile led to no reaction,
as did the corresponding Kumada type coupling between the
ferrocenyl Grignard, generated via halogen–magnesium exchange
as above, and 2a with stoichiometric Ni. Eventually we found
we could use an iron catalysed oxidative homocoupling of the
ferrocenyl Grignard following a procedure reported by Hayashi
and Nagano,38 but with some optimisation of reaction temper-
ature. The required Grignard reagent could be generated di-
rectly from the sparteine derived 2-lithiopentamethylazaferrocene
by transmetallation with freshly prepared MgBr2. Warming to
−40 ◦C followed by treatment with catalytic FeCl3 (5 mol%)–
ClCH2CH2Cl in Et2O and warming to room temperature gave the
homocoupled product 9 which was quickly purified by column
chromatography on silica in 50% yield [eqn (6)]. The structure of
(7)
The C2-symmetric pyridine-bis-azaferrocene 11 (Fig. 3) is
analogous to the successful Pybox type ligands 12.33 At-
tempted Negishi type coupling of zincated pentamethylazafer-
rocene (2 equiv.) with 2,6-dibromopyridine under our optimised
conditions [eqn (5), ZnCl2 (6 equiv.)] led to consumption of
starting materials in 2 hours. The 1H NMR spectrum of the
product oil revealed peaks consistent with the formation of 11, but
contaminated with ∼10% of 1. The appearance of a small amount
of starting material was apparent in our other coupling reactions
(Table 4), but was easily separated by column chromatography.
Unfortunately purification of crude 11 by chromatography or
recrystallisation was unsuccessful and pure 11 could not be
isolated.
1
9 was characterised by H NMR and mass spectroscopy, but 9
proved to be considerably unstable in solution or its solid form.
(6)
Fig. 3 Chiral planar ferrocene analogues of privileged ligands.
The success of phosphine-containing ligands across a variety
of asymmetric catalytic processes encouraged us to investigate the
possibility of a P,N-azaferrocene chelate structure 13 which bears
significant similarities to ligand systems 14 independently devel-
oped by Sprinz and Helmchen, von Matt and Pfaltz, and Williams
et al. (Fig. 3).39 Coupling of zincated pentamethylazaferrocene
with 2-bromo-1-iodobenzene proceeded in an unoptimised 44%
yield and required a palladium(0) tri-ortho-furyl phosphine cata-
lyst (Scheme 4). The aryl bromide substituent was then subjected
to halogen–lithium exchange and quenched with Ph2PCl to give
ligand 13 in 64% yield. Unfortunately during recrystallisation, in
From qualitative observations we had noted that the 2ꢀ-
phenylpentamethylazaferrocene (7a) was significantly more stable
than 1. We attempted the synthesis of a 2ꢀ-phenyl analogue of
9 from 7a (presumed 80% ee) in anticipation that this would be
more stable. Initial attempts to synthesise dimer 10 using Et2O as
a solvent were unsuccessful and after some deuterium quenching
studies we found the optimum conditions for deprotonation
required the combination of s-BuLi and TMEDA in THF. Under
the optimised coupling conditions we obtained 10 in 66% yield
and 77% ee [eqn (7)]. As predicted, 10 proved to be significantly
more stable than 9, which bodes well for its future investigation
as a chiral ligand in asymmetric reactions. Deprotonation of 7a
under identical conditions, but in PhMe, and continuation of the
coupling reaction in Et2O [as in eqn (6)] for solubility reasons
led to the isolation of 10 in 34% yield and 97% ee. This dramatic
increase in ee indicated a possible kinetic resolution was operating
in this coupling reaction. Although no meso product was isolated,
we propose that the formation of this product, or intermediates
to it, is fast in Et2O–PhMe and consumes all minor enantiomer
of 7a.
Scheme 4
334 | Org. Biomol. Chem., 2008, 6, 330–339
This journal is
The Royal Society of Chemistry 2008
©

View Article Online
an attempt to upgrade the ee, the compound was found to racemise
(+)-(Sp)-2ꢀ-Diphenylhydroxymethyl-1,2,3,4,5-
pentamethylazaferrocene 2b
◦
in solution. After 2 days in Et2O at −18 C the enantiopurity of
13 had decreased from 78% to 62% ee as measured by chiral
HPLC. A similar phenomenon has been observed for several
aromatic phosphine and sulfur substituted ferrocenes by Snieckus
et al.14a and is suggested to be due to Cp–aryl exchange.
Synthesised in an identical manner to 2a using benzophenone as an
electrophile. Purification by flash column chromatography (15%
EtOAc–5% Et3N–pet. ether) gave 2a (221 mg, 86%) as yellow
crystals, mp = 120 ◦C (50% hexane–50% Et2O), in 82% ee as
determined by HPLC analysis: Daicel Chiralcel OD-H column,
1% i-PrOH–hexane, 0.5 mL min−1; tr (minor) 10.8 min, tr (major)
12.3 min; [a]D = +120 (benzene, c 0.33, 22 ◦C) [lit20b [a]D = +159
(benzene, c 0.39)]. All other data was in exact agreement with those
of the literature.20b
Conclusions
The s-BuLi–(−)-sparteine mediated deprotonation of pen-
tamethylazaferrocene and addition of a range of electrophiles led
to the enantioselective functionalisation of the C-2ꢀ position in
around 80% ee for the majority of cases examined. A novel kinetic
resolution was developed that enriched the ee of the isolated
products (∼80% ee) to 90–99% ee in 64–69% overall yield for
the cases where the electrophiles Ph2CO, MeI or Ph2S2 were
used. This method was higher yielding than upgrading the ee
by recrystallisation and may find use for more complex or non-
crystalline substrates.
Enantiomerically enriched 2-lithiopentamethylazaferrocene
(∼80% ee) was used in Negishi cross couplings with a variety of
aryl, vinyl and heteroaryl halides and led to the enantioselective
synthesis of some C1-symmetric N,N (7c,d) and N,S (7e,f) chelate
ligands in one step. The N,N-bidentate ligands 7c and 7d were
recrystallised to a synthetically useful >95% ee.
The homodimerisation of pentamethylazaferrocene was
achieved by an iron catalysed oxidative coupling to give the novel
C2-symmetric ligand 10. Coupling in a PhMe–Et2O mixture as
opposed to THF resulted in a 34% yield of 10 in 97% ee. The
methodology described will be useful for the synthesis of a variety
of potential chelate systems that may be useful as chiral ligands
in enantioselective catalysis. We are currently investigating the
effectiveness of 7c, d and 10 as chiral ligands in asymmetric
processes.
(−)-(Sp)-2ꢀ-Trimethylsilyl-1,2,3,4,5-pentamethylazaferrocene 2c
Synthesised in an identical manner to 2a using TMSCl as an
electrophile. Purification by flash column chromatography (15%
EtOAc–5% Et3N–pet. ether) gave 2c (167 mg, 86%) as a red
solid, mp = 110 ◦C (hexane); [a]D −70 (benzene, c 0.27, 22 ◦C)
[lit20b [a]D = −108 (benzene, c 0.37)]. All other data was in exact
agreement with those of the literature.20b
(−)-(Sp)-2ꢀ-Methyl-1,2,3,4,5-pentamethylazaferrocene 2d
Synthesised in an identical manner to 2a using MeI as an
electrophile. Purification by flash column chromatography (30%
EtOAc–5% Et3N–pet. ether) gave 2d (150 mg, 93%) as an orange
solid, mp = 54 ◦C; [a]D −17 (benzene, c 0.30, 22 ◦C); IR mmax
(solid) 2967, 2907, 2359, 2339, 1380, 1031, 816 cm−1; H NMR
1
d 1.95 (15H, s, Cp*H), 2.21 (3H, s, Me) 3.95 (1H, s, py-H), 4.08
(1H, s, py-H), 4.89 (1H, s, py-H) ppm; 13C NMR d 10.8, 13.9, 72.7,
74.8, 80.5, 91.1, 102.3 ppm; m/z (EI+) 271 (100% M+), 190 (51%),
123 (25%); HRMS C15H21FeN calcd 271.1017, found 271.1023.
(−)-(Sp)-2ꢀ-Thiophenyl-1,2,3,4,5-pentamethylazaferrocene 2e
Synthesised in an identical manner to 2a using Ph2S2 as an
electrophile. Purification by flash column chromatography (30%
EtOAc–5% Et3N–pet. ether) gave 2e, (150 mg, 93%) as an
orange solid, mp = 114 ◦C (Et2O), in 79% ee as determined
by HPLC analysis: Diacel Chiralcel AD column, 1% i-PrOH–
hexane, 0.2 mL min−1; tr (minor) 8.5 min, tr (major) 12.3 min;
Experimental
Our general experimental details have been published previously.15
Commercial (−)-sparteine was distilled under reduced pressure
˚
and stored over 4 A molecular sieves and ZnCl2 was dried by
◦
fusing under vacuum with a Bunsen burner.
[a]D +8 (benzene, c 0.3, 22 C); IR mmax (solid) 2953, 2909, 1582,
1382, 1025, 897 cm−1; 1H NMR d 1.95 (15H, s, Cp*H), 4.32–4.34
(1H, m, py-H), 4.37 (1H, dd, J 2.0, 1.0 Hz, py-H), 5.06 (1H, s,
py-H), 7.05–7.10 (3H, m, Ar-H), 7.15–7.21 (2H, m, Ar-H) ppm;
13C NMR d 10.9, 77.7, 80.0, 82.3, 94.7, 96.7, 125.3, 127.0, 128.6,
138.9 ppm; m/z (EI+) 365 (100%), 255 (16%), 134 (16%), 119
(22%); HRMS C20H24NSFe calcd 366.0979, found 366.0947; anal.
calcd for C20H24NSFe: C, 65.58; H, 6.60; N, 3.82, found C, 65.67;
H, 6.43; N, 3.84%.
(−)-(Sp)-2ꢀ-Iodo-1,2,3,4,5-pentamethylazaferrocene 2a
To a solution of 1,2,3,4,5-pentamethylazaferrocene 1 (150 mg,
0.58 mmol) in Et2O (2.5 mL) was added (−)-sparteine (0.16 mL,
0.70 mmol, 1.2 equiv.). The reaction was cooled to −78 ◦C and s-
BuLi (0.70 mL of 1 N in hexanes, 0.70 mmol, 1.2 equiv.) was added
◦
dropwise and the reaction was stirred for 4 hours at −78 C. A
solution of I2 (178 mg, 0.70 mmol, 1.2 equiv.) in Et2O (7 mL)
was added dropwise. The reaction was then warmed to room
temperature and left to stir for 15 minutes. A small amount of
silica gel was added and the mixture was concentrated in vacuo.
Purification by flash column chromatography (10% EtOAc–5%
Et3N–pet. ether) gave 2a (180 mg, 80%) as an orange solid, mp =
(−)-(Sp)-2ꢀ-Diphenylphosphine-1,2,3,4,5-pentamethylazaferrocene
2f
Synthesised in an identical manner to 2a using Ph2PCl as an
electrophile. Purification by flash column chromatography (15%
EtOAc–5% Et3N–pet. ether) gave 2f (239 mg, 93%) as yellow
crystals, mp = 161 ◦C (50% hexane–50% Et2O), in 57% ee; [a]D =
−20 (benzene, c 0.26, 22 ◦C) [lit20b [a]D = −35 (benzene, c 0.26)]. All
other data was in exact agreement with those of the literature.20b
◦
84 C (hexane), in 77% ee as determined by HPLC analysis20b
(Daicel Chiralcel OD-H column); [a]D −20 (benzene, c 0.35, 22 ◦C)
[lit20b [a]D = −88 (benzene, c 0.12)]. All other data was in exact
agreement with those of the literature.20b
This journal is
The Royal Society of Chemistry 2008
Org. Biomol. Chem., 2008, 6, 330–339 | 335
©

View Article Online
(−)-(Rp)-2ꢀ-Diphenylhydroxymethyl-5-methyl-1,2,3,4,5-
(−)-Sparteine–s-BuLi mediated kinetic resolution of 2ꢀ-substituted
pentamethylazaferrocene 3
pentamethylazaferrocences 2
To a solution of 2d (100 mg, 0.36 mmol) in Et2O (2.5 mL) was
To a solution of 2ꢀ-substituted-1,2,3,4,5-pentamethylazaferrocene
2 (1 equiv.) in Et2O (5 mL mmol−1) was added (−)-sparteine
(0.25 equiv.). The reaction was cooled to −78 ◦C, s-BuLi
(0.25 equiv.) was added dropwise and the reaction was stirred
for 4 h at −78 ◦C. The electrophile (0.25 equiv.) was then
added dropwise either as a neat liquid or in an ether solution.
The reaction was then warmed to rt and left to stir for 15
minutes. A small amount of silica gel was then added and the
mixture was concentrated in vacuo in preparation for flash column
chromatography.
added TMEDA (49 mg, 0.43 mmol, 1.2 equiv.). The reaction
◦
was cooled to −78 C and s-BuLi (358 lL of 1.2 N in hexanes,
0.43 mmol, 1.2 equiv.) was added dropwise and the reaction was
stirred for 4 h at −78 ◦C. A solution of benzophenone (78 mg,
0.43 mmol, 1.2 equiv.) in Et2O (2 mL) was added dropwise, the
reaction was then warmed to room temperature and left to stir
for 15 minutes. A small amount of silica gel was added and the
mixture was concentrated in vacuo. Purification by flash column
chromatography (10% EtOAc–5% Et3N–pet. ether) gave 3 (64 mg,
39%) as a dark orange oil in 80% ee as determined by HPLC
analysis: Diacel Chiralcel OD-H column, 1% i-PrOH–hexane,
Products 2a, 2b, 2d and 2e gave identical physical data as to that
reported above or in Ref. 20b. Magnitude of ee was measured as
above.
0.15 mL min−1; tr (major) 22.3 min, tr (minor) 23.9 min; [a]D
=
−78 (benzene, c 0.74); IR mmax (film) 3457, 2909, 1658, 1319, 1281,
942 cm−1; 1H NMR d 1.74 (15H, s, Cp*H), 1.90 (3H, s, Me), 4.10
(1H, d, J 2.2 Hz, py-H), 4.18 (2H, s, py-H and OH), 6.92–6.97 (2H,
m, Ar-H), 7.04–7.16 (2H m, Ar-H), 7.27–7.32 (2H, m, Ar-H), 7.41
(2H, t, J 7.9 Hz, Ar-H), 7.77–7.80 (2H, m, Ar-H) ppm; 13C NMR
d 11.0, 14.5, 46.7, 73.0, 74.8, 77.9, 81.4, 102.4, 112.9, 126.8, 126.9,
127.0, 127.4, 127.4, 128.0, 145.0, 150.8 ppm; m/z (ES+) HRMS:
C28H31NOFe calcd 454.1833, found 454.1824.
Data for meso-6e. Subjection of 2e (150 mg) to the resolution
conditions above gave enantioenriched 2e (111 mg, 74%) in 93%
ee and meso-6e (53 mg, 19%) as yellow crystals, mp = 137 ◦C
1
(hexane); IR mmax (solid) 3078, 2910, 1479, 1023, 735 cm−1; H
NMR d 2.55 (15H, s, Cp*H), 4.71 (2H, s, py-H), 6.85–7.01 (10H,
m, Ar-H) ppm; 13C NMR d 10.6, 83.0, 83.7, 99.6, 125.7, 126.4,
127.3, 128.7, 129.1, 138.3 ppm; m/z (EI+) 473 (24%, M+), 147
(32%), 109 (27%), 95 (45%), 83 (52%), 73 (100%), 69 (81%), 57
(98%), 55 (100%). HRMS C26H27NS2Fe calcd 473.0934, found
473.0936.
(−)-(Sp)-2ꢀ-Trimethylsilyl-5-diphenylhydroxymethyl-1,2,3,4,5-
pentamethylazaferrocene 5
One pot (−)-sparteine–s-BuLi mediated kinetic resolution to
prepare (−)-(Sp)-2ꢀ-thiophenyl-1,2,3,4,5-pentamethylazaferrocene
Synthesised in an identical manner to 3, 2c (136 mg, 0.41 mmol)
gave after purification by flash column chromatography (15%
EtOAc–5% Et3–pet. ether) 5 (135 mg, 64%) as a dark orange
2e
To a solution of 1,2,3,4,5-pentamethylazaferrocene 1 (150 mg,
0.58 mmol) in Et2O (2.5 mL) was added (−)-sparteine (160 mg,
0.70 mmol, 1.2 equiv.). The reaction was cooled to −78 ◦C and s-
BuLi (0.70 mL of 1 N in hexanes, 0.70 mmol, 1.2 equiv.) was added
dropwise and the reaction was stirred for 4 h at −78 ◦C. A solution
of Ph2S2 (152 mg, 0.70 mmol, 1.2 equiv.) in Et2O (2 mL) was then
added dropwise and the reaction was warmed to rt and left to stir
for 15 minutes. After this time (−)-sparteine (34 mg, 0.15 mmol,
◦
◦
solid, mp = 41 C (Et2O); [a]D = −72 (benzene, c 0.86, 22 C);
IR mmax (solid) 2956, 2902, 1246, 1032, 838, 760 cm−1; H NMR
1
d 0.39 (9H, s, TMS), 1.90 (15H, s, Cp*H), 4.28 (1H, d, J 2.1 Hz,
py-H), 4.43 (1H, d, J 2.1 Hz, py-H), 4.56 (1H, s, OH), 6.79 (2H,
d, J 6.4 Hz, Ar-H), 7.12–7.15 (2H, m, Ar-H), 7.34–7.41 (2H,
m, Ar-H), 7.50 (2H, t, J 7.7 Hz, Ar-H), 7.85 (2H, d, J 7.7 Hz,
Ar-H) ppm; 13C NMR d −0.4, 11.3, 15.4, 73.9, 77.7, 80.8, 81.1,
98.0, 117.2, 126.4, 126.7, 126.8, 127.1, 127.3, 127.4, 127.6, 128.1,
144.7, 151.3 ppm; m/z (EI+) 511 (50%), 494 (29%), 493 (47%), 119
(22%), 73 (100%). HRMS: C30H37SiNOFe calcd 511.1994, found
511.1990; anal. calcd for C30H37SiNOFe: C, 70.43; H, 7.29; N, 2.74,
found C, 70.28; H, 7.39; N, 2.58%.
◦
0.25 equiv.) was added, the reaction was re-cooled to −78 C, s-
BuLi (0.15 mL of 1 N in hexanes, 0.15 mmol, 0.25 equiv.) was
added dropwise and the reaction was stirred for 3 h at −78 ◦C. A
solution of Ph2S2 (32 mg, 0.15 mmol, 0.25 equiv.) in Et2O (0.5 mL)
was then added dropwise and the reaction was warmed to rt and
left to stir for 15 minutes. A small amount of silica gel was added
and the mixture was concentrated in vacuo. Purification by flash
column chromatography (10–20% EtOAc–5% Et3N–pet. ether)
gave 2e (127 mg, 60%) in 90% ee (determined by HPLC as before)
and meso-6e (47 mg, 17% yield).
(−)-(Rp)-2ꢀ-Diphenylhydroxymethyl-1,2,3,4,5-
pentamethylazaferrocene 4
To a solution of 5 (49 mg, 0.10 mmol) in THF (1 mL) was added
TBAF (0.19 mL of 1 N in THF, 0.19 mmol, 1.9 equiv.). The
reaction was stirred for 18 h at reflux. The solution was cooled
to rt and diluted with Et2O (10 mL), then washed with H2O
(10 mL) and dried (MgSO4). The crude filtrate was concentrated
in vacuo to yield an orange oil. Purification by flash column
chromatography (10% EtOAc–5% Et3N–pet. ether) gave 5 (22 mg,
(−)-(Sp)-2ꢀ-Phenyl-1,2,3,4,5-pentamethylazaferrocene 7a
To a solution of 1,2,3,4,5-pentamethylazaferrocene 1 (1.00 g,
3.90 mmol) in toluene (20 mL) was added sparteine (1.08 mL,
4.68 mmol, 1.2 equiv.). The reaction was cooled to −78 ◦C and s-
BuLi (4.68 mL of 1 N in hexanes, 4.68 mmol, 1.2 equiv.) was added
dropwise, the reaction was then stirred for 4 hours at −78 ◦C. After
this time a solution of pre-dried ZnCl2 in THF (15.6 mL of 1 N
in THF, 15.6 mmol) was added dropwise. The reaction was then
warmed to room temperature and left to stir for 15 minutes and a
◦
54%) as yellow crystals, mp = 124 C, in 81% ee as determined
by HPLC analysis: Diacel Chiralcel OD-H column, 1% i-PrOH–
hexane, 0.5 mL min−1; tr (major) 10.8 min, tr (minor) 12.3 min;
[a]D = −112 (benzene, c 0.33) (lit20b [a]D = −146). All other data
was in exact agreement with those of the literature.20b
336 | Org. Biomol. Chem., 2008, 6, 330–339
This journal is
The Royal Society of Chemistry 2008
©

View Article Online
(+)-(Sp)-2ꢀ-(8-Quinoline)-1,2,3,4,5-pentamethylazaferrocene 7d
bright orange precipitate was observed. A solution of Pd(PPh3)4
(0.23 g, 0.20 mmol, 5 mol%) and PhI (1.19 g, 5.85 mmol, 1.5 equiv.)
in THF (15 mL) was stirred at rt for 15 mins and then added to
the reaction. The reaction was heated to reflux and followed by
monitoring the disappearance of the aromatic halide by TLC. The
mixture was diluted with CH2Cl2 (40 mL), poured onto a satd
aq. solution of NH4Cl (40 mL), the layers were separated and the
combined aqueous layers were washed with CH2Cl2 (2 × 20 mL).
The organics were combined and washed with NaHCO3 solution
(40 mL), brine (40 mL), dried over MgSO4) and concentrated in
vacuo. Purification by flash column chromatography (10% EtOAc–
5% Et3N–pet. ether) gave 7a (1.00 g, 77% yield) as orange crystals,
mp = 109 ◦C (Et2O); [a]D −8 (CH2Cl2, c 0.82, 22 ◦C); IR mmax
(solid) 2971, 2908, 1602, 1452, 1380, 1066, 1033, 801, 762 cm−1;
1H NMR d 1.71 (15H, s, Cp*H), 4.34 (1H, d, J 1.6 Hz, py-H),
4.58 (1H, d, J 2.4 Hz, py-H), 5.13 (1H, s, py-H), 7.26 (1H, d, J
7.6 Hz, Ar-H), 7.39 (2H, t, J 7.8 Hz, Ar-H), 7.70 (2H, d, J 7.2 Hz,
Ar-H) ppm; 13C NMR d 10.3, 69.9, 76.3, 81.2, 92.5, 102.1, 125.3,
126.4, 128.4, 135.7 ppm; m/z (EI+) 333 (100%, M+), 331 (32%),
190 (27%), 143 (41%). HRMS C20H23NFe calcd 333.1180, found
333.1171; anal. calcd for C20H23NFe: C, 72.11; H, 6.96; N, 4.20,
found C, 72.22; H, 7.02; N, 4.17%
Synthesised in an identical manner to 7a using 8-bromoquinoline
as a coupling partner. Purification by flash column chromatog-
raphy (15% EtOAc–5% Et3N–pet. ether) gave 7d (425 mg, 71%
yield) as red crystals, mp = 172 ◦C (Et2O) in 80% ee as determined
by HPLC analysis: Diacel Chiralcel OD-H column, 10% i-PrOH–
hexane, 0.5 mL min−1; tr (major) 7.8 min, tr (minor) 12.4 min.; [a]D
◦
+140 (CHCl3, c 0.30, 22 C); [a]D +614 (CHCl3, c 0.60); IR mmax
(solid) 2970, 2901, 1502, 1372, 1069, 1054, 800 cm−1; 1H NMR d
1.62 (15H, s, Cp*H), 4.49 (1H, d, J 2.4 Hz, py-H), 5.16 (1H, s,
py-H), 6.23 (1H, d, J 2.3 Hz, py-H), 7.44 (1H, dd, J 8.0, 4.0 Hz,
Ar-H), 7.56 (1H, t, J 8.0 Hz, Ar-H), 7.74 (1H, d, J 7.6 Hz, Ar-H),
8.21 (1H, d, J 8.4 Hz, Ar-H), 8.48 (1H, d, J 7.2 Hz, Ar-H), 9.03
(1H, d, J 2.0 Hz, Ar-H) ppm; 13C NMR d 10.3, 77.7, 81.2, 91.8,
120.6, 126.9, 126.5, 127.7, 128.8, 134.7, 136.4, 138.1, 148.7 ppm;
m/z (ES+) HRMS C23H25N2Fe calcd 385.1373, found 385.1367;
anal. calcd for C23H24N2Fe: C, 71.85; H, 6.30; N, 7.29, found C,
71.82; H, 6.29; N, 7.30%.
(−)-(Sp)-2ꢀ-(2-Thiophene)-1,2,3,4,5-pentamethylazaferrocene 7e
Synthesised in an identical manner to 7a using 2-bromothiophene
as a coupling partner. Purification by flash column chromatogra-
phy (15% EtOAc–5% Et3N–pet. ether) gave 7e (123 mg, 62% yield)
(−)-(Sp)-2ꢀ-(3,5-Difluorophenyl)-1,2,3,4,5-
◦
pentamethylazaferrocene 7b
as yellow crystals, mp = 106 C (Et2O) in 77% ee as determined
by HPLC analysis: Diacel Chiralcel OD-H column, 10% i-PrOH–
hexane, 0.5 mL min−1; tr (major) 7.8 min, tr (minor) 12.4 min;
[a]D −167 (acetone, c 0.52, 22 ◦C); IR mmax (CHCl3) 2966, 2905,
1381, 1065, 1034, 841, 806 cm−1; 1H NMR d 1.75 (15H, s, Cp*H),
4.27–4.30 (1H, m, py-H), 4.43 (1H, d, J 2.4 Hz, py-H), 5.07 (1H, s,
py-H), 7.04–7.09 (1H, m, Ar-H), 7.20–7.24 (2H, m, Ar-H) ppm;
13C NMR d 10.1, 69.9, 75.9, 81.4, 92.1, 99.2, 122.2, 123.4, 127.5,
139.6 ppm; m/z (ES+) HRMS C18H22NSFe calcd 340.0822, found
340.0804; anal. calcd for C18H21NSFe: C, 63.70; H, 6.24; N, 4.13,
found C, 63.33; H, 6.23; N, 3.93%
Synthesised in an identical manner to 7a using 3,5-difluoro-
iodobenzene as a coupling partner. Purification by flash col-
umn chromatography (10% EtOAc–5% Et3N–pet. ether) gave 7b
(145 mg, 68% yield) as yellow crystals, mp = 143 ◦C (Et2O); [a]D
−35 (CHCl3, c 0.28, 22 ◦C); IR mmax (solid) 2972, 2910, 1621, 1587,
1117, 1083, 1033, 813 cm−1; 1H NMR d 1.72 (15H, s, Cp*H), 4.40
(1H, dd, J 2.2, 0.8 Hz, py-H), 4.52 (1H, d, J 2.5, 0.6 Hz, py-H),
5.15 (1H, s, py-H), 6.71 (1H, tt, J 8.8, 2.2 Hz, Ar-H), 7.20 (2H, dd,
J 9.0, 2.3 Hz, Ar-H) ppm; 13C NMR d 10.3, 70.6, 81.6, 93.2, 101.4
(t, JC-F 25.4 Hz), 107.8 (dd, JC-F 26.1, 7.1 Hz), 164.5, 164.7 ppm;
m/z (ES+) HRMS C20H22NF2Fe calcd 370.1070, found 370.1043;
anal. calcd for C20H21NF2Fe: C, 65.02; H, 5.73; N, 3.79, found C,
65.01; H, 5.68; N, 3.78%.
(+)-(Sp)-2ꢀ-(2-Benzothiophene)-1,2,3,4,5-pentamethylazaferrocene
7f
Synthesised in an identical manner to 7a using 2-bromobenzo-
thiophene as a coupling partner. Purification by flash column chro-
matography (20% EtOAc–5% Et3N–pet. ◦ether) gave 7f (150 mg,
72% yield) as yellow crystals, mp = 118 C (Et2O) in 80% ee as
determined by HPLC analysis: Diacel Chiralcel OD-H column;
[a]D +35 (CHCl3, c 0.45, 22 ◦C); IR mmax (CHCl3) 2953, 2908, 1450,
1382, 1307, 1131, 1072, 1030, 881 cm−1; 1H NMR d 1.83 (15H, s,
Cp*H), 4.38 (1H, d, J 1.5 Hz, py-H), 4.63 (1H, d, J 2.0 Hz, py-H),
5.21 (1H, s, py-H), 7.30–7.41 (2H, m, Ar-H), 7.43 (1H, s, Ar-
H), 7.82 (1H, d, J 7.7 Hz, Ar-H), 7.89 (1H, d, J 7.9 Hz, Ar-H)
ppm; 13C NMR d 10.2, 70.9, 76.6, 81.6, 92.9, 98.4, 118.2, 122.3,
122.8, 123.6, 124.3, 139.3, 140.5, 140.8 ppm; m/z (ES+) HRMS
C22H24NSFe calcd 390.0979, found 390.0954.
(+)-(Sp)-2ꢀ-(2-Pyridyl)-1,2,3,4,5-pentamethylazaferrocene 7c
Synthesised in an identical manner to 7a using 2-bromopyridine as
a coupling partner. Purification by flash column chromatography
(10% EtOAc–5% Et3N–pet. ether) gave 7c (395 mg, 68% yield)
◦
as orange crystals, mp = 99 C (Et2O) in 80% ee as determined
by HPLC analysis: Diacel Chiralcel OD-H column, 10% i-PrOH–
hexane, 0.5 mL min−1; tr (major) 9.5 min, tr (minor) 24.0 min;
[a]D +140 (CHCl3, c 0.30, 22 ◦C); IR mmax (solid) 2966, 2911, 2855,
1
2358, 2344, 1587, 1493, 1033, 810, 792 cm−1; H NMR d 1.69
(15H, s, Cp*H), 4.41 (1H, dd, J 2.5, 0.9 Hz, py-H), 5.05 (1H, dd,
J 2.5, 0.8 Hz, py-H), 5.13 (1H, s, py-H), 7.13 (1H, ddd, J 7.6, 4.9,
1.3 Hz, Ar-H), 7.67 (1H, td, J 7.7, 1.8 Hz, Ar-H), 7.79 (1H, dt, J
8.0, 1.1 Hz, Ar-H), 8.60 (1H, ddd, J 5.0, 2.0, 0.9 Hz, Ar-H) ppm;
13C NMR d 10.3, 72.3, 77.2, 81.5, 101.4, 120.0, 120.9, 136.0, 149.2,
156.3 ppm; m/z (EI+) 334 (92%, M+), 200 (59%), 144 (100%),
117 (56%), 105 (75%). HRMS C19H22N2Fe calcd 334.1124, found
334.1132; anal. calcd for C19H22N2Fe: C, 68.24; H, 6.64; N, 8.38,
found C, 67.97; H, 6.57; N, 8.10%.
(+)-(Sp)-2ꢀ-(2-Propene)-1,2,3,4,5-pentamethylazaferrocene 7g
Synthesised in an identical manner to 7a using 2-bromopropene as
a coupling partner. Purification by flash column chromatography
(10% EtOAc–5% Et3N–pet. ether) gave 7g (531 mg, 61% yield)
as orange crystals, mp = 95 ◦C (Et2O); [a]D −46 (CHCl3, c 0.88,
This journal is
The Royal Society of Chemistry 2008
Org. Biomol. Chem., 2008, 6, 330–339 | 337
©

View Article Online
22 ◦C); IR mmax 3562, 3246, 2949, 2909, 1711, 1627, 1455, 1382,
894 cm−1; 1H NMR d 1.84 (15H, s, Cp*H), 2.06 (3H, s, Me), 4.21
Repeating the reaction, but initially deprotonating in PhMe
and then proceeding with MgBr2 and FeCl3 in Et2O, in a similar
manner as for the preparation of 9, led to the isolation of 10 in
34% yield and 97% ee by HPLC.
=
=
(1H, d, J 1.8 Hz, C CH2), 4.24 (1H, d, J 1.8 Hz, C CH2), 5.00
(1H, s, py-H), 5.14 (1H, s, py-H), 5.29 (1H, s, py-H) ppm; 13C
NMR d 10.3, 20.6, 70.5, 76.2, 80.6, 92.4, 103.8, 109.8, 138.4 ppm;
m/z (ES+) HRMS C17H24NFe calcd 298.1258, found 298.1233.
(−)-(Sp)-2ꢀ-(2-Bromobenzene)-1,2,3,4,5-pentamethylazaferrocene
15
(Sp,Sp)-2ꢀ-Bis-1,2,3,4,5-pentamethylazaferrocene 9
Synthesised in an identical manner to 7a, using 1-iodo-2-
bromobenzene as coupling partner and Pd(dba)2 (2.5 mol%)
premixed with tri-ortho-furylphosphine (10 mol%) in THF as
a catalyst. Purification by flash column chromatography (10%
EtOAc–5% Et3N–pet. ether) gave 15 (1.05 g, 44% yield) as an
orange solid, mp = 113 ◦C (Et2O); [a]D +99 (CHCl3, c 0.55, 22 ◦C);
IR mmax 2952, 2908, 1651, 1622, 1589, 1484, 1453, 1382, 1340, 1030,
To a solution of 1,2,3,4,5-pentamethylazaferrocene 1 (150 mg,
0.58 mmol) in Et2O (3 mL) was added sparteine (162 mg,
0.70 mmol, 1.2 equiv.), the reaction was cooled to −78 ◦C and
s-BuLi (538 lL of 1.3 N in hexanes, 0.70 mmol, 1.2 equiv.) was
added dropwise. The reaction was then stirred for 4 h at −78 ◦C
◦
before a freshly prepared solution of MgBr2 etherate at −78 C
1
[0.81 mmol, 1.4 equiv., prepared by adding 1,2-dibromoethane
(152 mg, 0.81 mmol) to Mg (21 mg, 0.87 mmol, 1.07 equiv.) in Et2O
(2 mL) and stirring for 30 mins at rt] was added dropwise and the
reaction was warmed to −40 ◦C over 15 min. A solution of FeCl3
(5 mg, 0.03 mmol, 5 mol%) in Et2O (2 mL) and 1,2-dichloroethane
(62 mg, 0.87 mmol, 1.5 equiv.) was added sequentially to the
reaction via cannula and the reaction was warmed to rt. The
reaction was then diluted with Et2O (15 mL), washed with water
(15 mL), dried (MgSO4) and concentrated in vacuo. Purification by
flash column chromatography (25% EtOAc–5% Et3N–pet. ether)
gave 9 (75 mg, 50% yield) as a yellow solid which decomposed over
a few hours. 1H NMR d 1.77 (30H, s, Cp*H), 4.13 (2H, s, py-H),
4.15 (2H, s, py-H), 5.12 (2H, s, py-H) ppm; m/z (ES+) HRMS
C28H36N2Fe2 calcd 513.1656, found 513.1665.
1008, 893 cm−1; H NMR d 1.74 (15H, s, Cp*H), 4.34 (1H, dd,
J 1.2, 0.4 Hz, py-H), 4.58 (1H, s, py-H), 5.13 (1H, d, J 2.0 Hz,
py-H), 7.10 (1H, td, J 6.4, 1.6 Hz, Ar-H), 7.37 (1H, td, J 6.4,
0.8 Hz, Ar-H), 7.67 (1H, dd, J 6.4, 0.8 Hz, Ar-H), 8.24 (1H, dd,
J 6.4, 1.2 Hz, Ar-H) ppm; 13C NMR d 10.5, 73.9, 77.3, 81.4, 92.2,
120.8, 127.1, 127.4, 130.5, 134.6 (10 observed signals); m/z (ES+)
HRMS C20H23NBrFe calcd 412.0363, found 412.0378; anal. calcd
for C20H22NBrFe: C, 58.39; H, 5.39; N, 3.41, found C, 58.26; H,
5.28; N, 3.34%.
(−)-(Sp)-2ꢀ-(2-Diphenylphosphine)phenyl-1,2,3,4,5-
pentamethylazaferrocene 13
To a solution of 15 (1.00 g, 2.43 mmol) in Et2O (15 mL) at −78 ◦C
was added n-BuLi (2.43 mL of 2.5 N in hexanes, 6.07 mmol,
2.5 equiv.) and the reaction was warmed to −40 ◦C over a period
of 0.5 h. The reaction was cooled to −78 ◦C, Ph2PCl (1.61 g,
7.28 mmol, 3.0 equiv.) was added and the reaction was warmed
to rt over 15 mins. After addition of H2O (0.5 mL) the mixture
was poured onto neutral alumina and concentrated in vacuo.
Purification by flash column chromatography (neutral alumina;
15% EtOAc–5% Et3N–pet. ether) gave 13 (813 mg, 64% yield) as
a red solid, mp = 104 ◦C (Et2O); [a]D +164 (CHCl3, c 0.75, 22 ◦C);
IR mmax 2940, 1710, 1587, 1454, 1382, 1122, 1110, 1092, 1070, 1029,
893 cm−1; 1H NMR d 1.64 (15H, s, Cp*H), 4.23 (1H, s, py-H), 4.83
(1H, s, py-H), 5.07 (1H, s, py-H), 6.94–6.96 (1H, br m, Ar-H),
7.13–7.19 (7H, br m, Ar-H), 7.42–7.43 (5H, br m, Ar-H), 8.19–
8.22 (1H, br m, Ar-H) ppm; 13C NMR d 10.5, 74.6, 75.0, 76.9,
81.2, 126.4, 128.3, 128.3, 128.4, 128.8, 128.8, 128.8, 130.1, 133.7,
133.8, 134.4, 134.5, 134.9 ppm; m/z (ES+) HRMS C32H33NPFe
calcd 518.1700, found 518.1704.
(+)-(Rp,Rp)-5ꢀ-Bis-(2ꢀ-phenyl-1,2,3,4,5-pentamethylazaferrocene)
10
To a solution of 7a (200 mg, 0.60 mmol) in THF (3 mL) was
added TMEDA (84 mg, 0.72 mmol, 1.2 equiv.), the reaction was
cooled to −78 ◦C and s-BuLi (762 lL of 1.3 N in hexanes,
0.99 mmol, 1.65 equiv.) was added dropwise. The reaction was
then stirred for 4 h at −78 ◦C before a freshly prepared solution of
MgBr2 at −78 ◦C [1.14 mmol, 1.4 equiv., prepared by adding 1,2-
dibromoethane (214 mg, 1.14 mmol) to Mg (29 mg, 1.22 mmol,
1.07 equiv.) in THF (2 mL) and stirring for 30 mins at rt] was
added dropwise and the reaction was warmed to −40 ◦C over
15 min. A solution of FeCl3 (10 mg, 0.06 mmol, 5 mol%) in THF
(2 mL) and 1,2-dichloroethane (158 mg, 0.90 mmol, 1.5 equiv.) was
added sequentially to the reaction via cannula and the reaction was
warmed to rt. The reaction was then diluted with Et2O (15 mL),
washed with water (15 mL), dried (MgSO4) and concentrated in
vacuo. Purification by flash column chromatography (10% EtOAc–
5% Et3N–pet. ether) gave 10 (132 mg, 66% yield) as an orange
solid, mp = 121 ◦C (Et2O) in 77% ee as determined by HPLC
analysis: Diacel Chiralcel OD-H column, 5% i-PrOH–hexane,
Notes and references
1 T. Hayashi, T. Hagihara, M. Katsuro and M. Kumada, Bull. Chem.
Soc. Jpn., 1983, 56, 363.
2 G. Argouarch, O. Samuel and H. B. Kagan, Eur. J. Org. Chem., 2000,
2885.
3 X.-L. Hou, X.-W. Wu, L.-X. Dai, B.-X. Cao and J. Sun, Chem.
Commun., 2000, 1195.
4 H.-Y. Jang, H. Seo, J. W. Han and Y. K. Chung, Tetrahedron Lett.,
2000, 41, 5038.
5 M. Li, K. Yuan, Y.-Y. Li, B.-X. Cao, J. Sun and X.-L. Hou, Tetrahedron:
Asymmetry, 2003, 14, 3347.
6 (a) D. Marquarding, H. Klusacek, G. Gokel, P. Hoffman and I.
Ugi, J. Am. Chem. Soc., 1970, 92, 5389; (b) G. Gokel, P. Hoffmann,
0.5 mL min−1; tr (major) 8.7 min, tr (minor) 18.6 min; [a]D
+
475 (CHCl3, c 0.41, 22 ◦C); IR mmax 2907, 1602, 1452, 1382, 1071,
1031, 908, 874 cm−1; 1H NMR d 1.65 (30H, s, Cp*H), 4.40 (2H, d,
J 2.2 Hz, py-H), 4.55 (2H, d, J 2.2 Hz, py-H), 7.26–7.29 (2H, m,
Ar-H), 7.38 (4H, t, J 7.7 Hz, Ar-H), 7.66 (4H, d, J 7.4 Hz, Ar-H)
ppm; 13C NMR d 9.6, 70.6, 76.7, 82.6, 97.6, 101.0, 125.4, 127.0,
128.4, 134.3 ppm; m/z (ES+) HRMS C40H45N2Fe2 calcd 665.2282,
found 665.2275.
338 | Org. Biomol. Chem., 2008, 6, 330–339
This journal is
The Royal Society of Chemistry 2008
©

View Article Online
H. Klusacek, D. Marquarding, E. Ruch and I. Ugi, Angew. Chem., Int.
Ed. Engl., 1970, 9, 64.
24 Absolute configuration determined by chiroptical comparison with
literature data in Ref. 20.
7 T. Hayashi, T. Mise, M. Fukushima, M. Kagotani, N. Nagashima, Y.
Hamada, A. Matsumoto, S. Kawakami, M. Konishi, K. Yamamoto
and M. Kumada, Bull. Chem. Soc. Jpn., 1980, 53, 1138.
8 (a) G. C. Fu, Acc. Chem. Res., 2004, 37, 542; (b) B. L. Hodous, J. C.
Ruble and G. C. Fu, J. Am. Chem. Soc., 1999, 121, 2637; (c) M. M. C.
Lo and G. C. Fu, J. Am. Chem. Soc., 1998, 120, 10270; (d) I. P. Dosa,
J. C. Ruble and G. C. Fu, J. Org. Chem., 1997, 62, 444.
9 J. G. Seitzberg, C. Dissing, I. Søtofte, P. Norrby and M. Johannsen,
J. Org. Chem., 2005, 70, 8332.
25 D. Hoppe and T. Hense, Angew. Chem., Int. Ed. Engl., 1997, 36, 2282.
26 (a) M. J. Dearden, C. R. Firkin, J.-P. R. Hernet and P. O’Brien, J. Am.
Chem. Soc., 2002, 124, 11870; (b) J.-P. R. Hernet, D. W. Porter, M. J.
Dearden, J. R. Harrison, T. Koplin, P. O’Brien, J. Parmene, V. Tyurin,
A. C. Whitwood, J. Gilday and N. M. Smith, Org. Biomol. Chem., 2003,
1, 3977.
27 L.-X. Dai, T. Tu, S.-L. You, W.-P. Deng and X.-L. Hou, Acc. Chem.
Res., 2003, 36, 659.
28 H. Pellissier, Tetrahedron, 2003, 59, 8291.
10 W. Zhang, T. Shimanuki, T. Kida, Y. Nakatsuji and I. Ikeda, J. Org.
Chem., 1999, 64, 6247.
11 L. Routaboul, S. Vincendeau, J.-C. Daran and E. Manoury, Tetrahe-
dron: Asymmetry, 2005, 16, 2685.
12 R. Kuwano, T. Uemura, M. Saitoh and Y. Ito, Tetrahedron: Asymmetry,
2004, 15, 2263.
29 An alternative interpretation could involve formation of the meso
compound from the minor isomer Rp-2a, followed by selective halogen–
lithium exchange of the pro-S iodine to give back Rp-2a by any excess
s-BuLi/–(−)-sparteine.
30 Metal Catalysed Cross-Coupling Reactions, ed. F. Diederich and
P. J. Stang, Wiley-VCH, New York, 1998.
13 O. G. Manchen˜o, R. G. Arraya´s and J. C. Carretero, J. Am. Chem. Soc.,
2004, 126, 456.
14 (a) M. Tsukazaki, M. Tinkl, A. Roglans, B. J. Chapell, N. J. Taylor
and V. Snieckus, J. Am. Chem. Soc., 1996, 118, 685; (b) R. S. Laufer,
U. Veith, N. J. Taylor and V. Snieckus, Org. Lett., 2000, 2, 629; (c) C.
Metallinos, H. Szillat, H. N. J. Taylor and V. Snieckus, Adv. Synth.
Catal., 2003, 345, 370.
15 J. C. Anderson, A. J. Blake and J. C. Arnall-Culliford, Org. Biomol.
Chem., 2003, 1, 3586.
16 J. C. Anderson and J. Osborne, Tetrahedron: Asymmetry, 2005, 16, 931.
17 (a) S. Qiao and G. C. Fu, J. Org. Chem., 1998, 63, 4168; (b) K. Tanaka,
S. Qiao, M. Tobisu, M. M. C. Lo and G. C. Fu, J. Am. Chem. Soc.,
2000, 122, 9870.
31 For example: (a) H. L. Pedersen and M. Johannsen, Chem. Commun.,
1999, 2517; (b) V. Mamane and O. Riant, Tetrahedron, 2001, 57, 2555;
(c) M. Lotz, G. Kramer and P. Knochel, Chem. Commun., 2002, 2546;
(d) H. L. Pedersen and M. Johannsen, J. Org. Chem., 2002, 67, 7982;
(e) T. Sturm, W. Weissensteiner and F. Spindler, Adv. Synth. Catal.,
2003, 345, 160; (f) V. Mamane, I. Ledoux-Rak, S. Deveau, J. Zyss and
O. Riant, Synthesis, 2003, 455; (g) H. K. Cotton, F. H. Fernando and
J.-E. Ba¨ckvall, Eur. J. Org. Chem., 2003, 2756; (h) A. Mamane and Y.
Fort, J. Org. Chem., 2005, 70, 8220; (i) H. Li, H.-S. Cheng, A.-H. Seow
and T.-P. Loh, Tetrahedron Lett., 2007, 48, 2209.
32 (a) M. M. C. Lo and G. C. Fu, Tetrahedron, 2001, 57, 2621; (b) M. M. C.
Lo and G. C. Fu, J. Am. Chem. Soc., 2002, 124, 4572.
33 J. S. Johnson and D. A. Evans, Acc. Chem. Res., 2000, 33, 325.
34 J. Hassan, M. Se´vignon, C. Gozzi, E. Schulz and M. Lemaire, Chem.
Rev., 2002, 102, 1359.
35 F. Tre´court, M. Mallet, O. Mingin, B. Gervais and G. Quguiner,
Tetrahedron, 1993, 49, 8373.
36 P. Knochel, W. Dohle, N. Gommermann, F. F. Kneisel, F. Kopp, T.
Korn, I. Sapountzis and V. A. Vu, Angew. Chem., Int. Ed., 2003, 42,
4302.
37 J. D. Osborne, PhD thesis, University of Nottingham, 2006.
38 T. Hayashi and T. Nagano, Org. Lett., 2005, 7, 491.
39 (a) J. Sprinz and G. Helmchen, Tetrahedron Lett., 1993, 34, 1769;
(b) P. von Matt and A. Pfaltz, Angew. Chem., Int. Ed. Engl., 1993,
32, 566; (c) G. J. Dawson, G. C. Frost, J. M. J. Williams and S. J. Coote,
Tetrahedron Lett., 1993, 34, 3149.
18 (a) J. Tweddell, D. A. Hoic and G. C. Fu, J. Org. Chem., 1997, 62, 8286;
(b) S.-Y. Liu, I. D. Hills and G. C. Fu, J. Am. Chem. Soc., 2005, 127,
15352.
19 J. C. Ruble and G. C. Fu, J. Org. Chem., 1996, 61, 7230.
20 (a) J. G. Hansen, I. Søtofte and M. Johannsen, Org. Lett., 2001, 3, 499;
(b) J. G. Hansen and M. Johannsen, J. Org. Chem., 2003, 68, 1266.
21 O. Riant, G. Argouarch, D. Guillaneux, O. Samuel and H. B. Kagan,
J. Org. Chem., 1998, 63, 3511.
22 K. K. Andersen, Tetrahedron Lett., 1962, 3, 93.
23 The work also developed the enantioselective lateral lithiation of
1ꢀ,2,2ꢀ,3ꢀ,4ꢀ,5,5ꢀ-heptamethylazaferrocene with s-BuLi and an S-valine
derived bis(oxazoline). T. Fukuda, K. Imazato and M. Iowa, Tetrahe-
dron Lett., 2003, 44, 7503.
This journal is
The Royal Society of Chemistry 2008
Org. Biomol. Chem., 2008, 6, 330–339 | 339
©