Regio- and stereoselectivity in long-range asymmetric induction controlled by carbonyliron groups: Crystal structures of dicarbonyl-L-[(1,2,3,4,5-η)-1-R-4-methoxycyclohexadienyl]iron(1+) hexafluorophosphate(1-) [(R=isopropenyl, L=CO), (R=1-styryl, L=CO), (R=1-prop-1-enyl, L=PPh3)]

Article abstract of DOI:10.1016/S0022-328X(01)00734-3

The synthesis of alkenyl-extended cyclohexadienyliron complexes is described and reactions are reported demonstrating that regio- and stereoselective functionalisation can be controlled by the judicious choice of nucleophile, alkali metal counterion and ligand set at iron. The crystal structures of tricarbonyl[(1,2,3,4,5-η)-1-(1′-methylethenyl)-4- methoxycyclohexadienyl]iron(1+) hexafluorophosphate(1-), tricarbonyl[(1,2,3,4,5-η)-1-(ethenyl-1′-phenyl)-4- methoxycyclohexadienyl]iron(1+) hexafluorophosphate(1-), (E)-dicarbonyl[(1,2,3,4,5-η)-1-(2′-methylethenyl)-4- methoxycyclohexadienyl]triphenylphosphineiron(1+) hexafluorophosphate(1-), tricarbonyl[methyl(2,3,4,5-η)-2-carbomethoxy-2-(4-methoxy-1-(1′- methylethenyl)cyclohexadien-1-yl)ethanoate]iron(0), are reported. The stereochemical course of a long-range relay of chirality controlled by the carbonyliron group is proven.

Full text of DOI:10.1016/S0022-328X(01)00734-3

Journal of Organometallic Chemistry 630 (2001) 88–103
www.elsevier.com/locate/jorganchem
Regio- and stereoselectivity in long-range asymmetric induction
controlled by carbonyliron groups: crystal structures of
dicarbonyl-L-[(1,2,3,4,5-h)-1-R-4-methoxycyclohexadienyl]iron(1+)
hexafluorophosphate(1−) [(R=isopropenyl, L=CO),
(R=1-styryl, L=CO), (R=1-prop-1-enyl, L=PPh3)] and
tricarbonyl[methyl(2,3,4,5-h)-2-carbomethoxy-2-
(4-methoxy-1-(1%-methylethenyl)cyclohexadien-1-yl)ethanoate]iron(0)
Richard D.A. Hudson a,*, Christopher E. Anson b, Mary F. Mahon c,
G. Richard Stephenson d,1
a Department of Chemistry, Uni6ersity College Dublin, Belfield, Dublin 4, Ireland
b Institut fu¨r Anorganische Chemie, Uni6ersita¨t Karlsruhe, Engesserstrasse Geb 30.45, 76131 Karlsruhe, Germany
c Department of Chemistry, Uni6ersity of Bath, Bath BA2 7AY, UK
d School of Chemical Sciences, Uni6ersity of East Anglia, Norwich NR4 7TJ, UK
Received 16 January 2001; received in revised form 6 February 2001
This paper is dedicated to Professor Myron Rosenblum on the occasion of his 75th birthday. One of us, R.D.A.H. had the pleasure and
privilege to undertake a period of postdoctoral work in Professor Rosenblum’s laboratories at Brandeis University, Waltham, MA, USA,
from 1997 to 1999 [1]
Abstract
The synthesis of alkenyl-extended cyclohexadienyliron complexes is described and reactions are reported demonstrating that
regio- and stereoselective functionalisation can be controlled by the judicious choice of nucleophile, alkali metal counterion and
ligand set at iron. The crystal structures of tricarbonyl[(1,2,3,4,5-h)-1-(1%-methylethenyl)-4-methoxycyclohexadienyl]iron(1+)
hexafluorophosphate(1−), tricarbonyl[(1,2,3,4,5-h)-1-(ethenyl-1%-phenyl)-4-methoxycyclohexadienyl]iron(1+) hexafluorophos-
phate(1−), (E)-dicarbonyl[(1,2,3,4,5-h)-1-(2%-methylethenyl)-4-methoxycyclohexadienyl]triphenylphosphineiron(1+) hexafluoro-
phosphate(1−), tricarbonyl[methyl(2,3,4,5-h)-2-carbomethoxy-2-(4-methoxy-1-(1%-methylethenyl)cyclohexadien-1-yl)ethanoate]-
iron(0), are reported. The stereochemical course of a long-range relay of chirality controlled by the carbonyliron group is proven.
© 2001 Elsevier Science B.V. All rights reserved.
Keywords: Tricarbonyliron; Cyclohexadienyliron; Chiral; Stereoselective; Regioselective
1. Introduction
exerts complete stereocontrol and for direct addition
the nucleophile approaches trans to the metal. In sto-
chiometric systems, while it is desirable in terms of
efficiency for the metal to be used several times to
control a sequence of bond-forming reactions, in prac-
tice (particularly with cyclic ligands) there are severe
limitations to the scope of reaction sequences in which
the metal moves through the ligand by a series of
h5–h4 (nucleophile addition) and h4–h5 (reactivation
Electrophilic transition metal complexes have power-
ful control effects when employed in combination with
nucleophiles in organic synthesis [2]. The metal centre
* Corresponding author.
E-mail address: richard.hudson@ucd.ie (R.D.A. Hudson).
1 *Also corresponding author.
0022-328X/01/$ - see front matter © 2001 Elsevier Science B.V. All rights reserved.
PII: S0022-328X(01)00734-3

R.D.A. Hudson et al. / Journal of Organometallic Chemistry 630 (2001) 88–103
89
sired substitution pattern of the corresponding section
of the target structure. The analysis of the reactivity
properties (particularly regio- and stereoselectivity) of
the working ligand is thus an important stage in the
design process during the planning of a synthetic route.
Even a simple sequence of two metal-mediated steps
directly at the working ligand can offer versatile pat-
terns for controlled sequential bond-formation (Fig. 1),
with 1,1- [6,7], 1,2- [8], 1,3- [9] and 1,4- [10] structures
all accessible. The challenge in recent years has been to
develop processes that allow selective access to each of
these forms and now in many cases, the required pat-
tern of reactivity has been demonstrated in actual target
oriented routes. The 1,1-pattern has been employed in
the synthesis of O-methyl joubertiamine [6] and a for-
mal total synthesis of lycoramine [7]. The key is to
exploit the directing effect of C-1 alkoxy substituents to
ensure a leaving group is in place at the site of the first
nucleophile addition, so that this again becomes a
terminus of the dienyl moiety when this is reformed in
preparation for a reaction with the second nucleophile.
Access to the 1,2-pattern is achieved, simply by moving
the position of the leaving group from C-1 to C-6 and
this approach has allowed the construction of the ABC
ring portion of hippeastrine [8]. Rearranging the substi-
tution pattern to the 1,2-dimethoxycyclohexadienyl iso-
mer gives access to the 1,3-pathway, which has been
demonstrated in a series of reactions that place C2 and
aryl substituents in positions corresponding to the main
structural features of the lycorine skeleton [9]. A
methoxy group at C-2 is used for the 1,4-pattern [10]
(Fig. 1).
Fig. 1. (a) 1,1-, (b) 1,2-, (c) 1,3- and (d) 1,4-regioisomeric substitution
patterns from reaction sequences that employ one iteration of h5-salt
formation and nucleophile addition with diene and dienyl working
ligands. Examples of natural product synthesis illustrating these
substitution patterns.
of electrophilicity) steps. Recent progress with acyclic
structures [3], however, illustrates the potential of these
procedures to effect repeated stereocontrol to form a
series of chiral centres in reactions that exploit leaving
groups to reform the h5-series. Similar approaches us-
ing h4 and h5 complexes in six membered rings [4] are
clearly limited to a maximum of two chiral centres. In
applications such as these, one of the ligands becomes
incorporated into the final target structure. This ligand
is termed the ‘working ligand’ [5] and, in an efficient
synthetic route, its substituents should match the de-
These concepts may be extended by introducing a
vinyl group at the terminus of the dienyl system which
is activated towards nucleophilic attack in a remote
sense by the working ligand (Fig. 2). In this paper, we
report results that demonstrate long-range control ef-
fects, establishing iron-controlled relative stereochem-
istry at three chiral centres. This approach allows the
metal to reach out from the hapto-bound portion of the
ligand to promote electrophilicity and impart stereo-
Fig. 2. (a) Illustration of the concepts of ‘working’ and auxiliary ligands in synthesis design. (b) and (c) An unsymmetrical working ligand forms
a chiral complex which can influence stereochemistry at nearby prochiral organic functional groups. (d) The relative positions in a reaction
sequence.

90
R.D.A. Hudson et al. / Journal of Organometallic Chemistry 630 (2001) 88–103
Scheme 1. Reagents and conditions: (1) vinyllithium, THF, −100°C; (2) HPF6, Ac2O, 0°C; (3) Me3NO, PPh3, acetone; (4) HPF6, Ac2O, 0°C.
control. In this way, metals moving within cyclohexa-
diene/cyclohexadienyl ligands can control many chiral
centres as far as two atoms out from the working ligand
itself. There is considerable precedence for the use of
electrophilic metal complexes to activate alkenes. Early
work by Rosenblum established that (1,2-h)-butadiene
complexes of Fe(CO)2Cp+ could be electrophilic at C-4
[11]. Cuprates or Grignards were shown to add regiose-
lectively to the remote terminus in good yields and in
an extension of this concept both enamines and dialkyl
malonates were found to react similarly [12]. He also
demonstrated that the uncomplexed double bond in the
h5-cycloheptatrienyliron system is also activated to-
wards nucleophilic attack [13].
Selected bond lengths and angles are presented in Table
1 (see Table 3 for crystallographic data and structure
refinement). In general, the features of the structures do
not differ greatly from other substituted tricarbonyliron
cyclohexadienyl compounds reported in the literature
[15]. The carbonyl ligands do not differ in bond lengths
or angles within or between complexes 2a and 2d
(representative examples are: 2a Fe(1)ꢀC(2)=1.818(9);
,
C(1)ꢀO(1)=1.121(9) A; C(1)ꢀFe(1)ꢀC(2)=96.5°; 2d
,
A;
Fe(1)ꢀC(2)=1.830(8);
C(1)ꢀO(1)=1.127(11)
C(1)ꢀFe(1)ꢀC(3)=96.1°) and are similar to the litera-
ture values [15]. Dicarbonyl complex 4d exhibits slightly
shortened Fe to carbonyl bond lengths (Fe(1)ꢀC(1)=
,
1.781(6); Fe(1)ꢀC(2)=1.779(7) A) and a slightly wider
angle between them (C(1)ꢀFe(1)ꢀC(2)=99.9°) with re-
spect to 2a and 2d, indicating more of the back bonding
from the metal to phosphine ligand than steric com-
pression. Of course, this balance of p-bonding effects is
also seen in the carbonyl region of the IR spectra
(cm−1): w(CO) 2a 2108, 2067, 2038; 2d 2107, 2058,
2035; 4b 2032, 1984. The Fe(CO2)L groups are orien-
tated such that one of the carbonyl ligands eclipses the
methylene group and the tripod adopts a staggered
2. Results and discussion
2.1. Synthesis of 2 and 4
Complexes 2 were synthesised by the addition of
alkenyllithium reagents to the dimethoxy-substituted
complex 1 followed by the removal of the allylic OMe
group from the partially purified adducts to afford the
salts in good yield on a multigram scale. The corre-
sponding phosphine analogues 4 were prepared by a
ligand exchange reaction (CO to PPh3) on the neutral
methoxy adducts followed by salt formation as before.
Complexes 2b and 4b were formed in stereoconvergent
syntheses, by addition of the stereo-undefined E/Z mix-
ture of lithiopropenes (formed from the corresponding
mixture of bromopropenes) to 1 followed by serendipi-
tous equilibration of the alkene adducts during the
demethoxylation step (Scheme 1).
2.2. X-ray structures of 2a, 2d and 4b
Compounds 2a, 2d and 4b afforded single crystals
suitable for X-ray analysis and Ortex [14] diagrams of
the molecular structures are presented in Figs. 3–5.
Fig. 3. Ortex plot of cation in 2a. Ellipsoids represented at the 30%
probability level.

R.D.A. Hudson et al. / Journal of Organometallic Chemistry 630 (2001) 88–103
91
Fig. 4. Ortex plot of cation in 2d. Ellipsoids represented at the 30% probability level.
conformation so that the pendant alkene does not
eclipse any of the three carbonyl vectors in 2a and 2d
and lies between the two carbonyls of 4b. The angle
between the carbonyls straddling the central methines
of the dienyls in 2a and 2d (C(1)ꢀFe(1)ꢀC(3)=91.9°
and C(1)ꢀFe(1)ꢀC(32)=92.6°) is compressed with re-
spect to the other angles in the Fe(CO3) groups (95.9–
96.5°). All the structures show that the h5-portions of
the cyclohexadienyl systems are completely flat and
they lie in an almost coplanar (2a is more distorted)
pseudo-trans arrangement with respect to the alkenes
which are inclined away from the metal-bound face of
the working ligand (dihedral angles: 2a C(8)ꢀC(9)ꢀ
C(11)ꢀC(12)=161.5(7); 2d C(7)ꢀC(8)ꢀC(11)ꢀC(18)=
173.2(8); 4b C(24)ꢀC(25)ꢀC(28)ꢀC(29)=170.1(7)°). The
bond lengths of the alkenes do not differ significantly
from one another (2a C(11)ꢀC(12)=1.356(11); 2d
the distances between the iron to dienyl termini (D
,
,
0.084 A). This is smaller than 2d (D 0.159 A) and 4b (D
,
0.127 A) due to the added steric bulk of the phenyl
groups on the alkene and phosphine, respectively. In
the case of tricarbonyl(2-methoxycyclohexadienyl)-
,
iron(1+) (2.186(8); 2.172(6) A) [17] the bond lengths of
the iron to dienyl termini are only very slightly different
,
(D 0.014 A) and this may not be crystallographically
significant. Molecular orbital calculations show that
p-donor substituents induce a subtle slipping of the
metal away from the far terminal carbon, partially
decomplexing it and causing it to become more elec-
trophillic. The well-known para-directing ability of the
2-methoxy substituent is related to this subtle distortion
and it has been noted that small distortions within a
molecule may have large consequences on its
,
C(11)ꢀC(18)=1.346(11); 4b C(28)ꢀC(29)=1.297(9) A)
or from the ‘normal’ sp2–sp2 double bond length (1.32
,
A) [16]. The single bond which appends the alkenes to
the dienyl system is also not significantly different
between these examples (2a C(9)ꢀC(11)=1.486(9); 2d
,
C(8)ꢀC(11)=1.484(10); 4b C(25)ꢀC(28)=1.464(10) A)
2
2
,
or from a normal sp –sp single bond (1.48 A) [16]. In
all cases, the iron atoms do not lie directly below the
centroids of the unsymmetrical dienyl systems with the
iron to alkene-substituted carbon distances (2a
Fe(1)ꢀC(9)=2.230(4); 2d Fe(1)ꢀC(8)=2.290(7); 4b
,
Fe(1)ꢀC(25)=2.282(6) A) being significantly longer
than the iron to unsubstituted-termini distances of the
dienyls (2a Fe(1)ꢀC(5)=2.146(6); 2d Fe(1)ꢀC(4)=
,
2.131(7); 4b Fe(1)ꢀC(21)=2.155(7) A). Comparison
with the literature reveals that tricarbonyl(1-methyl-4-
methoxycyclohexadienyl)iron(1+) [17] exhibits a simi-
lar distortion with the 1-methyl-substituted carbon to
iron bond being significantly lengthened (2.245(8)) with
respect to the far end of the dienyl (adjacent to the
,
,
methoxy group) to iron bond (2.165(7) A; D 0.080 A).
These differences may be largely due to steric hindrance
[17] and indeed compound 2a has a similar difference in
Fig. 5. Ortex plot of cation in 4b. Ellipsoids represented at the 30%
probability level.

92
R.D.A. Hudson et al. / Journal of Organometallic Chemistry 630 (2001) 88–103
Table 1
Selected bond lengths and angles for structures 2a, 2d and 4b
Portion of
Structure
2a
2d
4b
,
,
,
Atoms
Length (A) or
angle (°)
Atoms
Length (A) or
angle (°)
Atoms
Length (A) or
angle (°)
Pendant
Alkene
C(9)ꢀC(11)
1.486(9)
1.356(11)
1.460(11)
120.7
C(8)ꢀC(11)
1.484(10)
1.346(11)
1.486(11)
120.8
C(25)ꢀC(28)
C(28)ꢀC(29)
C(29)ꢀC(30)
C(25)ꢀC(28)ꢀC(29)
1.464(10)
1.297(9)
1.511(17)
125.5
C(12)ꢀC(11)
C(11)ꢀC(13)
C(9)ꢀC(11)ꢀC(12)
C(11)ꢀC(18)
C(11)ꢀC(12)
C(8)ꢀC(11)ꢀC(18)
Iron to diene
Fe(1)ꢀC(9)
Fe(1)ꢀC(8)
Fe(1)ꢀC(7)
Fe(1)ꢀC(6)
Fe(1)ꢀC(5)
C(5)ꢀFe(1)ꢀC(9)
2.230(4)
2.216(6)
2.091(7)
2.198(6)
2.146(6)
66.5
Fe(1)ꢀC(8)
Fe(1)ꢀC(7)
Fe(1)ꢀC(6)
Fe(1)ꢀC(5)
Fe(1)ꢀC(4)
C(4)ꢀFe(1)ꢀC(8)
2.290(7)
2.107(8)
2.097(7)
2.205(7)
2.131(7)
64.7
Fe(1)ꢀC(25)
Fe(1)ꢀC(24)
Fe(1)ꢀC(23)
Fe(1)ꢀC(22)
Fe(1)ꢀC(21)
C(21)ꢀFe(1)ꢀC(25)
2.282(6)
2.133(6)
2.108(6)
2.179(6)
2.155(7)
64.9
Auxiliary ligands
Fe(1)ꢀC(1)
Fe(1)ꢀC(2)
Fe(1)ꢀC(3)
1.803(9)
1.819(9)
1.808(9)
Fe(1)ꢀC(1)
Fe(1)ꢀC(2)
Fe(1)ꢀC(3)
1.830(8)
1.821(9)
1.836(10)
Fe(1)ꢀC(1)
Fe(1)ꢀC(2)
1.781(6)
1.779(7)
Fe(1)ꢀP(1)
C(1)ꢀO(1)
C(2)ꢀO(2)
2.264(3)
1.129(6)
1.139(7)
C(1)ꢀO(1)
C(2)ꢀO(2)
C(3)ꢀO(3)
1.127(11)
1.125(11)
1.127(11)
C(1)ꢀO(1)
C(2)ꢀO(2)
C(3)ꢀO(3)
1.121(9)
1.109(9)
1.128(9)
C(1)ꢀFe(1)ꢀC(2)
C(1)ꢀFe(1)ꢀC(3)
C(2)ꢀFe(1)ꢀC(3)
Fe(1)ꢀC(1)ꢀO(1)
Fe(1)ꢀC(2)ꢀO(2)
Fe(1)ꢀC(3)ꢀO(3)
96.5
91.9
96.2
178.4
177.1
177.6
C(1)ꢀFe(1)ꢀC(2)
C(1)ꢀFe(1)ꢀC(3)
C(2)ꢀFe(1)ꢀC(3)
Fe(1)ꢀC(1)ꢀO(1)
Fe(1)ꢀC(2)ꢀO(2)
Fe(1)ꢀC(3)ꢀO(3)
92.6
96.1
95.9
178.5
177.3
178.0
C(1)ꢀFe(1)ꢀC(2)
C(1)ꢀFe(1)ꢀP(1)
C(1)ꢀFe(1)ꢀC(2)
Fe(1)ꢀC(1)ꢀO(1)
Fe(1)ꢀC(2)ꢀO(2)
99.9
88.1
91.2
177.7
177.6
Scheme 2. Reagents and conditions: CH(CO2Me)2M, THF, 0°C.
sis and so investigation of the mode of attack of various
types of nucleophiles must be addressed. Cuprates are
known to add in a conjugate sense to a,b-unsaturated
ketones or esters and this has been established for these
complexes with diasteroselectivities as high as 1:8 and
reported elsewhere [18]. Stabilised enolates are often
used as building blocks in synthesis and can add in a
variety of modes, so we chose to examine the regio-
chemical outcome of dimethyl malonate addition to 2
and 4. Compounds 5–7 (Scheme 2) are all possible
products from these reactions and their distributions
are presented in Table 2. (The overall isolated yields for
these transformations were low (30–40%) but the ratio
of products was deduced from the NMR of the crude
material after workup and so is expected to reflect the
reactivity [17]. The sizeable distortions evident in struc-
tures 2a, 2d and 4b are probably due in large part to
steric hindrance between the alkenes and Fe(CO2)L
groups but a small amount of p-orbital overlap from
the vinyl group may also contribute and cannot be
ruled out since the alkenes all have a similar preferred
orientation. This may in turn facilitate directed conju-
gate addition to the far end of the alkene (see Section
2.3).
2.3. Nucleophile additions to 2 and 4
The regiochemistry of nucleophillic addition to these
alkenyl-extended cyclohexadienyl salts is of paramount
importance if these complexes are to be used in synthe-

R.D.A. Hudson et al. / Journal of Organometallic Chemistry 630 (2001) 88–103
93
Table 2
Results for the addition of the sodium and lithium enolates of dimethyl malonate to 2 and 4
Entry
R1
R2
L
M
Ratio (compound)
5
6
7
1
2
Me
H
Me
Ph
Me
H
Me
Ph
Me
H
H
CO
CO
CO
CO
CO
CO
CO
CO
Na
Na
Na
Na
Li
Li
Li
Li
B1 (5a)
B1 (5b)
B1 (5c)
65 (5d)
\99 (6a)
\99 (6b)
\99 (6c)
35 (6d)
30 (6a)
75 (6b)
80 (6c)
B1 (6d)
B1 (6aPPh3)
B1 (6bPPh3)
B1 (6dPPh3)
B1
B1
B1
B1
B1
B1
B1
B1
B1
B1
B1
Me
Me
H
3 a
4
5
6
H
70 (5a)
Me
Me
H
H
Me
H
25 (2:3) b (5b)
20 (2:3) b(5c)
\99 (5d)
7 a
8
9
10
11
PPh3
PPh3
PPh3
Li/Na
Li/Na
Li/Na
\99 (5aPPh3)
\99 (1:3) b (5bPPh3)
\99 (5dPPh3)
Ph
a E/Z stereochemistry could not be unambiguously assigned for the 2-butenyl salts.
b Ratio of diastereoisomers at the b-position was determined by 1H-NMR.
4b address this issue and malonate additions to these
salts afforded diastereomeric mixtures in the ratio 2:3
for the products 5b and 5c and 1:3 for 5bPPh3 reflecting
the greater steric bulk of the phosphine ligand leading
to a higher degree of facial selectivity. Notably, the
ratio of diasteroeisomers is independent of the counte-
rion used in the malonate enolate. The third stereo-
chemical issue arises during the formation of the
products 6 (cf. Fig. 2d, 3-addition) which were all
expected to exhibit trans addition of the malonyl group
by analogy with overwhelming literature precedence.
An X-ray diffraction study of 6a and an Ortex [15]
diagram of its molecular structure are shown in Fig. 6
together with selected bond lengths and angles (see
Table 3 for crystallographic data and structure refine-
ment). Only poorly diffracting crystals could be grown
of 6a and so the structure is not highly refined (R=
13%), however, it can be seen that addition is indeed
from the M-exo face. The other structural features of
the crystal are unremarkable and are in agreement with
the literature precedents for bond lengths and angles
within other unsymmetrically substituted h4-tricar-
bonyliron complexed dienes [19].
true product distribution.) Compound 7 (cf. Fig. 2d,
1-addition), was not observed due to the methoxy
substituent at C2. Products 5 and 6 (cf. Fig. 2d, 2%- and
3-addition, respectively) were obtained and the ratio of
their distributions was found to be dependent on the
nature of the malonyl counterion and the ligand set at
iron. Addition of sodium enolates to 2 afforded the
3-addition (cf. Fig. 2d) products 6 (Table 2, entries
1–4), but on addition of lithium enolates there was a
tendency to form significant amounts of the 2%-adducts,
5 (Table 2, entries 5–8). Ligand exchange to form the
triphenylphosphine salts 4 followed by addition of ei-
ther the sodium or lithium enolates afforded exclusively
the 2%-adducts, 5. Therefore, almost total 3- or 2%-regio-
control may be obtained by the judicious choice of
metal counterion (dependant on its coordinating abil-
ity) and ligand set (dependant on its size). (Note that in
entry 4, Table 2, incomplete control arises here for
steric reasons.)
Three stereochemical issues arise from the formation
of 5 or 6. In the first case, e.g. 5, either configuration of
the exocyclic triene could be formed and it was estab-
lished by n.O.e. experiments that only a single alkene
was observed, assigned as the E configuration (Scheme
2). Since equilibration of the alkene after nucleophilic
addition is unlikely, as the presence of an a-methyl or
phenyl substituent (5a,c,d and 5aPPh3, cPPh3) reduces
the likelihood of total selective interconversion of the
exocyclic trienes, we favour the explanation based on
kinetic control during approach of the nucleophiles.
This suggests that the orientation adopted in the solid
state for each cationic complex (Figs. 3–5) is probably
the reacting conformer. The further stereochemical con-
sequence of the conjugate addition pathway occurs if
the vinyl terminus is substituted in some way, introduc-
ing a prochiral group, which may undergo attack from
either face (cf. Fig. 2b, R2"H). Compounds 2b,c and
Inspection of complexes 5a, c and d reveals that a
new prochiral group (cf. Fig. 2c) has been formed
adjacent to the working ligand. Exocyclic trienes of this
Scheme 3. Reagents and conditions: (1) R2CuLi, THF, −30°C; (2)
HPF6, CH2Cl2, 0°C.

94
R.D.A. Hudson et al. / Journal of Organometallic Chemistry 630 (2001) 88–103
triphenylphosphine adduct 15 [21]. This replacement of
an auxiliary ligand does not interfere with the stere-
ocenters in the working ligand.2 The product 15 was
crystallised to afford single crystals, which proved suit-
able for X-ray analysis [21]. The protonation event was
clearly seen to have taken place from the M-exo face of
the molecule and the relative stereochemistry between
C5 and C1%, formed by the long-range chirality relay,
was assigned as 5S,1%S,5R*,1%R* [21,22].
type may undergo protonation at the 1%-site leading to
reformation of the cationic working ligand and we have
investigated the stereochemical outcome of such reac-
tions in the series of trienes, 8, shown in Scheme 3.
Thus, cuprate additions to 2a proceeded with complete
regio- and stereocontrol to afford 8, which were proto-
nated with HPF6 to reform the h5-cyclohexadienyl moi-
ety. Protonation could proceed from either side of the
prochiral group leading to a mixture of diasteroeiso-
mers, but inspection of the 1H- and 13C-NMR indicated
that only one diastereoisomer had been formed in each
case but the relative stereochemistry could not be deter-
mined from these data. No crystals could be obtained
from any of the products 9 in order resolve this issue
and so a second sequence of reactions was designed in
which a series of three organoiron-mediated reactions
were combined to implement long-range stereocontrol
[21] (Scheme 4).
Starting from 10 [20] nucleophile addition as before
followed by removal of the OH group formed the
cationic product 11 [21]. This was reacted with dimethyl
cuprate to afford the single isomer of the exocyclic
triene 12 [21] as shown. Salt formation by protonation
with HPF6 reformed the h5-dienyl system 13 [21] which
unfortunately did not provide us with X-ray quality
crystals and so this was further elaborated by enolate
addition to the far terminus of the dienyl system fol-
lowed by ligand exchange to form the dicarbonyl
3. Conclusions
We have shown that judicious choice of nucleophile,
counter ion and ligand set, may promote complete
regio- and significant stereocontrol in conjugate addi-
tion to the remote terminus of alkenyl-extended cyclo-
hexadienyl salts and that the acid-induced reactivation
of the emergent exocyclic trienes to reform the h5-salts
is completely stereoselective. The reaction sequence
from 11 to 14 is illustrative of these synthetically useful
steps and makes use of three organoiron-promoted
nucleophile additions during which the combined 1,3/
1%,2% sequence, including the protonation step, creates
two chiral centres four atoms apart. The planar chiral-
ity of the metal complex controls the formation of both
centres and so we have demonstrated in this reaction
sequence a long-range relay of chirality via the metal.
4. Experimental
4.1. General methods
All reactions were performed under a nitrogen atmo-
sphere. Diethyl ether and THF were freshly distilled
from sodium benzophenone ketyl. The dimethoxy cy-
clohexadienyl complex 1 was prepared according to the
literature procedure [4]. All other chemicals and
reagents were used as received without further purifica-
1
tion. H-NMR spectra were measured on a JEOL EX
270 spectrometer operating at 270 MHz or a Varian
Unity 400 spectrometer operating at 400 MHz. 13C-
NMR spectra were obtained on a JEOL EX 270 spec-
trometer operating at 67 MHz. FTIR spectra were
obtained on a Perkin–Elmer 1720X as a thin film.
Relative peak heights for MS are shown in parentheses.
4.2. General procedure for the formation of salts 2
Fig. 6. Ortex diagram of 6a. Ellipsoids represented at the 50%
probability level. For clarity only major components of disorder in
,
the ester groups are shown. Selected bond lengths (A): C(1)ꢀC(8)=
Following the work of Seebach and Neuman [23] the
required alkenylbromide (two equivalents) was dis-
1.596; C(8)ꢀC(9)=1.314; C(8)ꢀC(10)=1.473; C(1)ꢀC(11)=1.587;
Fe(1)ꢀC(2)=2.104; Fe(1)ꢀC(3)=2.080; Fe(1)ꢀC(4)=2.085; Fe(1)ꢀ
C(5)=2.117; Fe(1)ꢀC(16)=1.837; Fe(1)ꢀC(17)=1.666; Fe(1)ꢀ
C(18)=1.781. Selected bond angles (°): C(10)ꢀC(8)ꢀC(1)=114.4;
C(9)ꢀC(8)ꢀC(1)=123.8; Fe(1)ꢀC(16)ꢀO(16)=178.0; Fe(1)ꢀC(17)ꢀ
O(17)=176.3; Fe(1)ꢀC(18)ꢀO(18)=175.6.
2 All the products reported in this paper have been prepared in
racemic form and stereoisomers illustrated depict only relative stereo-
chemistry.

R.D.A. Hudson et al. / Journal of Organometallic Chemistry 630 (2001) 88–103
95
Table 3
Crystal data and structure refinement for 2a, 2d, 4b and 6a
Compound
2a
2d
4b
6a
Empirical formula
Formula weight
Crystal system
Space group
C13H13O4PF6Fe
449.5
Monoclinic
P21/n
C18H15O4PF6Fe
496.12
Orthorhombic
P212121
C30H28O3P2F6Fe
668.31
Triclinic
P-1 (no. 2)
C17H20O8Fe
408.18
Triclinic
P-1
Unit cell dimensions
,
a (A)
14.223(6)
8.087(3)
14.992(5)
90
90.82(4)
90
1716.2
4
9.415(1)
14.109(2)
15.013(2)
90
90
90
1994.3
4
9.741(3)
10.852(2)
16.153(3)
87.50(2)
78.90(2)
63.37(2)
1500.1
2
10.185(7)
10.436(7)
11.142(7)
63.49(5)
77.14(5)
65.81(5)
942.4
,
b (A)
,
c (A)
h (°)
i (°)
k (°)
U (A )
3
,
Z
2
Temperature (K)
293(2)
293(2)
293(2)
293(2)
F(000)
902
1.74
1000
1.65
0.914
0.30×0.30×0.30
742
1.61
424
1.40
Dcalc. (mg m−3
)
v(Mo–Ka) (mm−1
Crystal size (mm)
Diffractometer
)
1.048
0.20×0.25×0.25
CAD4 automatic 4-circle
0.677
0.30×0.30×0.30
CAD4 automatic 4-circle
0.822
0.48×0.41×0.19
Siemens P4
CAD4 automatic 4-circle
diffractometer
0.71073
3030
1449
0.0280
2680
diffractometer
0.71073
1826
1326
0.1315
1798
diffractometer
0.71073
5017
3472
0.0110
4703
,
u(Mo–Ka) (A)
0.71073
4050
3827
0.0757
2306
Data measured
Unique data
Rint
Final R indices
[I\2|(I)]
Parameters/restraints
wR2, R1 (all data)
wR2, R1 [I\2|(I)]
S (all data)
284/0
281/0
395/0
275/23
0.1632, 0.1146
0.1387, 0.0581
1.079
0.1150, 0.0790
0.0977, 0.0415
1.062
0.1239, 0.0675
0.1159, 0.0439
1.058
0.3532, 0.1890
0.3087, 0.1320
1.181
Largest difference peak
0.486 and −0.695
0.362 and −0.327
0.581 and −0.301
2.047 and −1.206
−3
,
and hole (e A
)
solved in dry diethyl ether and added slowly to a
cooled (−78°C) solution of BuLi (1.7 M. in hexanes,
4.2.1. Tricarbonyl[(1,2,3,4,5-p)-1-(1%-methylethenyl)-
4-methoxycyclohexadienyl]iron(1+) hexafluoro-
phosphate(1−) 2a
t
four equivalents) in dry diethyl ether. The solution
was stirred for 1 h and then the temperature of the
reaction mixture was reduced to −100°C followed by
addition of 1 (or 10 for the formation of 11) as
slurry in dichloromethane via a wide bore cannula.
The mixture was allowed to warm to room tempera-
ture and was poured into a separating funnel charged
with water. Extraction with diethyl ether (three por-
tions), washing with water (three portions) followed
by drying (MgSO4) and removal of the solvent in
vacuo afforded brown oils in all cases. These were
partially purified by flash chromatography through a
3 cm plug of basic alumina in a 5 cm diameter glass
sinter funnel with diethyl ether as the eluant to afford
orange oils after removal of the solvent. The oils
were dissolved in acetic anhydride and cooled to 0°C
followed by dropwise addition of hexafluorophospho-
ric acid (40% in water). The mixtures were poured
into diethyl ether and the solid material was collected
by filtration. The salts were further purified by repre-
cipitation from acetone–ether several times to afford
the required cationic complexes as air stable yellow
solids.
Experimental procedures and work up were as de-
scribed in Section 4.2. Experimental details: 2-bromo-
propene (3.390 g, 28.0 mmol); 1 (4.670 g, 14.0 mmol).
Yield: 4.140 g, 68% based on 1. Obtained as a pale
1
yellow solid. H-NMR (acetone-d6, 270 MHz): l 6.84
(dd, 1H, J=2.6, 3.6 Hz, C3ꢀH), 5.72 (d, 1H, J=6.3
Hz, C2ꢀH), 5.62 (s, 1H, CH3CꢁCH2), 5.59 (s, 1H,
CH3CꢁCH2), 4.03 (m, 1H, C5ꢀH), 3.84 (s, 3H, OCH3),
3.41 (dd, 1H, J=6.6, 14.5 Hz, C6bꢀH), 2.20 (bd, 1H,
J=14.5 Hz, C6aꢀH), 1.75 (s, 3H, CH3CꢁCH2). IR
(cm−1):w (CO) 2108, 2067, 2038, 1635. EIMS; m/z: 290
([M]+, 1.5), 262 ([M+−CO], 23), 234 ([M+−2CO],
12), 206 ([M+−3CO], 48), 148 (100), 107 (54). Anal.
Found: C, 35.9; H, 2.9. Calc. for C13H13F6FeO4P: C,
35.9; H, 3.0%. The structure was also established by
X-ray analysis (Fig. 3, Tables 1 and 3).
4.2.2. (E)-Tricarbonyl[(1,2,3,4,5-p)-4-methoxy-1-
(2%-methylethenyl)cyclohexadienyl]iron(1+) hexafluoro-
phosphate(1−) 2b
Experimental procedures and work up were as de-
scribed in Section 4.2. Experimental details: (E/Z)-1-

96
R.D.A. Hudson et al. / Journal of Organometallic Chemistry 630 (2001) 88–103
4.2.4. Tricarbonyl[(1,2,3,4,5-p)-1-(ethenyl-1%-phenyl)-
4-methoxycyclohexadienyl]iron(1+) hexafluoro-
phosphate(1−) 2d
Experimental procedures and work up were as de-
scribed in Section 4.2. Experimental details: 1-bro-
mostyrene (1.990 g, 10.9 mmol); 1 (1.832 g, 5.50 mmol).
Yield: 1.428 g, 52% based on 1. Obtained as a pale
1
yellow solid. H-NMR (acetone-d6, 400 MHz): l 7.48
(m, 3H, Ph), 7.25 (m, 2H, Ph), 7.12 (dd, 1H, J=2.4,
6.0 Hz, C3ꢀH), 6.09 (s, 1H, PhCꢁCHH), 5.74 (s, 1H,
PhCꢁCHH), 5.65 (d, 1H, J=6.0 Hz, C2ꢀH), 4.44 (m,
1H, C5ꢀH), 3.99 (s, 3H, OCH3), 3.74 (dd, 1H, J=5.1,
15.2 Hz, C6bꢀH), 2.67 (bd, 1H, J=15.2 Hz, C6aꢀH).
IR (cm−1): w(CO): 2107, 2058, 2035. FABMS; m/z: 351
([M]+, 100), 323 ([M+−CO], 16), 295 ([M+−2CO],
2), 267 ([M+−3CO], 11), 211 (9), 103 (9). FABMS;
m/z: Mass calc. for [C18H15FeO4]+: 351.0320. Found:
351.0322. The structure was also established by X-ray
analysis (Fig. 4, Tables 1 and 3).
Scheme 4. Reagents and conditions: (1) 2-lithiopropene, THF, −
100°C; (2) HPF6, Ac2O, 0°C; (3) Me2CuLi, THF, −30°C; (4) HPF6,
Ac2O, 0°C; (5) CH(CO2Me)2Na, THF, 0°C; (6) Me3NO, PPh3,
acetone.
4.3. General procedure for the formation of
dicarbonyltriphenylphosphine deri6ati6es
bromopropene (0.660 g, 5.40 mmol); 1 (0.903 g, 2.70
mmol). Yield: 0.694 g, 59% based on 1. Obtained as a
pale yellow solid. H-NMR (acetone-d6, 270 MHz): l
1
Following the work of Birch and Kelley [24] the
neutral vinyl adducts (one equivalent) formed above as
intermediates were dissolved in acetone. Trimethy-
lamine-N-oxide (six equivalents) and triphenylphos-
phine (seven equivalents) were added and the mixtures
were heated at reflux for 24 h. After cooling, the
solutions were poured into a 1:9 mixture of diethyl
ether–petroleum ether (40/60). Most of the excess
triphenylphosphine was removed in this way as a fine
precipitate and the solutions were concentrated after
filtration to afford yellow solids. Purification by column
chromatography on silica gel with 1:30:69 triethy-
lamine–diethyl ether–petroleum ether (40/60) as the
eluant afforded the required compounds as yellow
solids.
7.23 (dd, 1H, J=2.6, 5.9 Hz, C3ꢀH), 6.72 (dq, 1H,
J=7.5, 15.5 Hz, HCꢁCHCH3), 5.98 (d, 1H, J=5.9 Hz,
C2ꢀH), 5.67 (d, 1H, J=15.5 Hz, HCꢁCHCH3), 4.30
(m, 1H, C5ꢀH), 3.99 (s, 3H, OCH3), 3.83 (dd, 1H,
J=6.53, 15.2 Hz, C6bꢀH), 2.76 (bd, 1H, J=14.5 Hz,
C6aꢀH), 1.91 (dd, 3H, J=1.7, 7.6 Hz, HCꢁCHCH3).
IR (cm−1): w(CO) 2106, 2043. EIMS; m/z: 290 ([M++
H], 0.4), 262 ([M++H−CO], 8), 233 ([M+−2CO],
0.3), 205 ([M+−3CO], 2), 178 (16), 121 (61). FABMS;
m/z: Mass calc. for [C13H14FeO4]+: 290.024. Found:
290.024.
4.2.3. (E)-Tricarbonyl[(1,2,3,4,5-p)-4-methoxy-1-
(1%,2%-dimethylethenyl)cyclohexadienyl]iron(1+)
hexafluorophosphate(1−) 2c
4.3.1. Dicarbonyl[2,3,4,5-p)-1,4-dimethoxy-1-
(1%-methylethenyl)cyclohexadiene]triphenylphos-
phineiron(0) 3a
Experimental procedures and work up were as de-
scribed in Section 4.2. Experimental details: (E/Z)-2-
bromobut-2-ene (0.737 g, 5.40 mmol); 1 (0.901 g, 2.70
mmol). Yield: 0.762 g, 62% based on 1. Obtained as a
Experimental procedures and work up were as de-
scribed in Section 4.3. Experimental details: 1 (1.290 g,
3.55 mmol). Yield: 0.767 g, 39% based on 1. Obtained
1
pale yellow solid. H-NMR (acetone-d6, 270 MHz): l
1
7.26 (dd, 1H, J=2.3, 5.9 Hz, C3ꢀH), 5.96 (m, 2H,
C2ꢀH and CH3CꢁCHCH3), 4.27 (m, 1H, C5ꢀH), 4.02
(s, 3H, OCH3), 3.62 (dd, 1H, J=7.3, 15.5 Hz, C6bꢀH),
2.76 (bd, 1H, J=15.5 Hz, C6aꢀH), 1.96 (s, 3H,
as a very pale yellow solid. H-NMR (CHCl3-d, 270
MHz):
l 7.3–7.7 (m, 15H, Ph), 6.74 (s, 1H,
CH3CꢁCHH), 6.65 (s, 1H, CH3CꢁCHH), 4.22 (m, 1H,
C3ꢀH), 3.38 (s, 3H, OCH3), 3.05 (m, 1H, C5ꢀH), 2.88
(s, 3H, OCH3), 1.94 (t, 2H, J=3.6 Hz, C6aꢀH,
C6bꢀH), 1.88 (m, 1H, C2ꢀH), 1.53 (s, 3H, CH3CꢁCH2).
IR (cm−1): w(CO) 1973, 1911, 1484, 1436, 1311, 1184,
1120. FABMS; m/z: very weak ions, 554 ([M]+), 526
([M+−CO]), 498 ([M+−2CO]). FABMS; m/z: Mass
calc. for [C30H28FeO3P]+: 523.1125. Found: 523.1125.
CH3CꢁCHCH3),
1.74
(d,
3H,
J=7.3
Hz,
CH3CꢁCHCH3). IR (cm−1): w(CO) 2104, 2053, 2038.
EIMS; m/z: 304 ([M++H], 0.2), 276 ([M++H−CO],
3), 248 ([M++H−2CO], 2), 220 ([M++H−3CO],
5), 219 ([M+−3CO], 1). CIMS; m/z: Mass calc. for
[C14H15FeO4]+: 304.0398. Found: 304.0398.

R.D.A. Hudson et al. / Journal of Organometallic Chemistry 630 (2001) 88–103
97
4.3.2. (E/Z)-Dicarbonyl[2,3,4,5-p)-1,4-dimethoxy-1-
(2%-methylethenyl)cyclohexadiene]triphenylphos-
phineiron(0) 3b
3.10 (bs, 4H, C5ꢀH and OCH3), 3.05 (s, 3H, OCH3),
2.34 (dt, 1H, J=3.6, 14.8 Hz, C6bꢀH), 2.16 (ddd, 1H,
J=2.3, 4.6, 14.8 Hz, C6aꢀH), 2.03 (dd, 1H, J=6.3, 6.3
Hz, C2ꢀH). IR(cm−1): w(CO) 1974, 1914, 1622, 1435,
1221, 1090. FABMS; m/z: 616 ([M]+, 22), 560 ([M+−
2CO], 3), 528 ([M+−2CO−OCH3−H], ([M+−
2CO−OCH3−H], 100), 349 (31), 318 (46), 298
([M+−2CO−PPh3], 3), 263 (16), 239 (11), 183 (20).
FABMS; m/z: Mass calc. for [C33H29FeOP]+, (M+−
2CO−OCH3−H): 528.1305. Found: 528.1276.
Experimental procedures and work up were as de-
scribed in Section 4.3. Experimental details: 1 (5.00 g,
14.0 mmol). Yield: 2.482 g, 32% based on 1. Obtained
1
as a very pale yellow solid. H-NMR (CHCl3-d, 270
MHz): l 7.3–7.7 (m, 15H, Ph), 5.30 (dq, 1H, J=7.3,
11.6 Hz, HCꢁCHCH3), 4.99 (dq, 1H, J=1.7, 11.6 Hz,
HCꢁCHCH3), 3.86 (m, 1H, C3ꢀH), 3.35 (s, 3H, OCH3),
3.09 (m, 4H, OCH3 and C5ꢀH), 2.17 (dt, 1H, J=3.6,
14.5 Hz, C6bꢀH), 2.07 (t, 1H, J=7.3 Hz, C2ꢀH), 1.89
(ddd, 1H, J=2.3, 4.6, 14.5 Hz, C6aꢀH), 1.62 (dt, 3H,
1.7, 7.3 Hz, HCꢁCHCH3). IR (cm−1): w(CO): 1973,
1911, 1482, 1434, 1219, 1091, 1073, 1028. EIMS; m/z:
264 ([M+−CO−PPh3], 2), 262 (PPh3, 100), 205 ([M+
−CO−PPh3−OCH3], 1), 183 (48), 181 ([M+−
2CO−PPh3+H], 2), 148 (73), 133 (18), 121 (35).
4.4. General procedure for the formation of salts 4
Compounds 3 (one equivalent) were dissolved in
acetic anhydride (10 ml) and cooled (°C). Excess hex-
afluorophosphoric acid was added dropwise and the
mixtures were stirred for 10 min. The solutions were
poured into dry diethyl ether (50 ml) and the yellow
precipitates were collected by filtration. These were
further purified by reprecipitation from acetone–diethyl
ether to afford the required compounds as orange
solids.
4.3.3. (E/Z)-Dicarbonyl[2,3,4,5-p)-1,4-dimethoxy-1-
(1%,2%-dimethylethenyl)cyclohexadiene]triphenylphos-
phineiron(0) 3c
Experimental procedures and work up were as de-
scribed in Section 4.3. Experimental details: 1 (5.00 g,
14.0 mmol). Yield: 1.164 g, 16% based on 1. Obtained
as a very pale yellow solid as a mixture of two insepara-
ble configurational isomers in the ratio 1:4, A:B; by
4.4.1. Dicarbonyl[(1,2,3,4,5-p)-1-(1%-methylethenyl)-4-
methoxycyclohexadienyl]triphenylphosphineiron(1+)
hexafluorophosphate(1−) 4a
Experimental procedures and work up were as de-
scribed in Section 4.4. Experimental details: 3a (0.615 g,
1.11 mmol). Yield: 0.475 g, 64%. Obtained as an orange
solid. 1H-NMR (acetone-d6, 270 MHz): l 7.60 (m, 15H,
Ph), 7.03 (bs, 1H, C3ꢀH), 5.72 (bs, 1H, C2ꢀH), 5.62 (s,
1H, CH3CꢁCHH), 5.55 (s, 1H, CH3CꢁCHH), 3.96 (bs,
1H, C5ꢀH), 2.80 (s, 3H, OCH3), 2.33 (d, 1H, J=14.4
Hz, C6bꢀH), 2.05 (m, 1H, C6aꢀH), 1.86 (s, 3H,
CH3CꢁCH2). IR (cm−1): w(CO) 2020, 1985, 1562, 1380,
1264, 1093. FABMS; m/z: 523 ([M]+, 100), 495 ([M+
−CO], 26), 467 ([M+−2CO], 15), 335 (10), 318 (12),
263 (12), 139 (11), 183 (13). FABMS; m/z: Mass calc.
for [C30H28FeO3P]+: 523.1123. Found: 523.1123.
1
NMR. H-NMR (CHCl3-d, 270 MHz). Configurational
isomer A: l 6.8–7.8 (m, 15H, Ph), 5.46 (1H, q, J=6.6
Hz, CH3CꢁCHCH3), 4.26 (ddd, 1H, J=2.3, 7.0, 7.3
Hz, C3ꢀH), 3.18 (m, 1H, C5ꢀH), 3.14 (s, 3H, OCH3),
2.98 (s, 3H, OCH3), 2.28 (dd, 1H, J=6.9, 7.0 Hz,
C2ꢀH), 2.00–2.20 (m, 2H, C6aꢀH, C6bꢀH), 1.52 (s,
3H, CH3CꢁCHCH3), 1.40 (d, 3H, J=6.6 Hz,
CH3CꢁCHCH3). Configurational isomer B: l 6.8–7.8
(m, 15H, Ph), 5.21 (1H, q, J=7.3 Hz, CH3CꢁCHCH3),
3.95 (ddd, 1H, J=2.3, 7.0, 7.3 Hz, C3ꢀH), 3.29 (m, 1H,
C5ꢀH), 3.15 (s, 3H, OCH3), 3.07 (s, 3H, OCH3), 2.28
(dd, 1H, J=6.9, 7.0 Hz, C2ꢀH), 2.00–2.20 (m, 2H,
C6aꢀH, C6bꢀH), 1.77 (d, 3H, J=7.3 Hz,
CH3CꢁCHCH3), 1.48 (s, 3H, CH3CꢁCHCH3). IR
(cm−1): w(CO) 1984, 1926, 1485, 1435, 1272, 1093,
1029, 744, 697. FABMS; m/z: 568 ([M]+, 16), 537
([M+−OCH3], 24), 509 ([M+−OCH3−CO], 10), 481
([M+−OCH3−2CO], 45), 480 (100), 349 (51), 318
(91), 263 (PPh3+H, 30), 239 (39), 183 (41).
4.4.2. (E)-Dicarbonyl[(1,2,3,4,5-p)-1-(2%-methylethenyl)-
4-methoxycyclohexadienyl]triphenylphosphineiron(1+)
hexafluorophosphate(1−) 4b
Experimental procedures and work up were as de-
scribed in Section 4.4. Experimental details: 3b (1.300 g,
2.34 mmol). Yield: 1.189 g, 76%. Obtained as an orange
solid. 1H-NMR (acetone-d6, 270 MHz,): l 7.60 (m,
15H, Ph), 6.65 (bs, 1H, C3ꢀH), 6.34 (dq, 1H, J=6.9,
15.8 Hz, HCꢁCHCH3), 5.18 (bs, 1H, C2ꢀH), 3.98 (bs,
1H, C5ꢀH), 3.49 (d, 1H, J=15.8 Hz, HCꢁCHCH3),
2.93 (s, 3H, OCH3), 2.23 (dd, 1H, J=5.6, 14.4 Hz,
C6bꢀH), 1.70 (m, 1H, C6aꢀH), 1.77 (d, 3H, J=6.9 Hz,
HCꢁCHCH3). IR (cm−1): w(CO) 2032, 1984, 1495,
1464, 1434, 1377, 1239, 1044, 984, 835. Anal. Found: C,
54.1; H, 4.2. Calc. for C30H28F6FeO3P2: C, 53.9; H,
4.2%. The structure was also established by X-ray
analysis (Fig. 5, Tables 1 and 3).
4.3.4. Dicarbonyl[2,3,4,5-p)-1,4-dimethoxy-1-
(1%-phenylethenyl)cyclohexadiene]triphenylphos-
phineiron(0) 3d
Experimental procedures and work up were as de-
scribed in Section 4.3. Experimental details: 1 (1.00 g,
2.70 mmol). Yield: 0.965 g, 58% based on 1. Obtained
as a very pale yellow solid. 1H-NMR (270 MHz,
CHCl3-d): l 7.0–7.8 (m, 20H, Ph), 5.17 (s, 2H,
PhCꢁCH2), 4.32 (ddd, 1H, J=2.3, 6.3, 6.6 Hz, C3-H),

98
R.D.A. Hudson et al. / Journal of Organometallic Chemistry 630 (2001) 88–103
required compounds as yellow solids. Isolated yields are
quoted.
4.4.3. (E)-Dicarbonyl[(1,2,3,4,5-p)-1-(1%,2%-dimethyl-
ethenyl)-4-methoxycyclohexadienyl]triphenyl-
phosphineiron(1+) hexafluorophosphate(1−) 4c
Experimental procedures and work up were as de-
scribed in Section 4.4. Experimental details: 3c (1.100 g,
1.90 mmol). Yield: 1.088 g, 84%. Obtained as an orange
4.5.1. (E)-Tricarbonyl[methyl (2,3,4,5-p)-2%-
carbomethoxy-4-(4-methoxycyclohexadien-1-ylidene)-
pentanoate]iron(0) 5a
1
Experimental procedures and work up were as de-
scribed in Section 4.5. Experimental details: lithium
bis(trimethylsilyl)amide (1.5 ml, 1 M in THF, 1.5
mmol). Dimethyl malonate (0.184 g, 1.4 mmol); 2a
(0.300 g, 0.69 mmol). Yield: 0.061 g, 21% based on 2a.
Obtained as a yellow oil. 1H-NMR (CHCl3-d 270
MHz): l 5.11 (dd, 1H, J=2.6, 6.6 Hz, C3ꢀH), 3.71 (s,
3H, OCH3), 3.69 (s, 3H, OCH3), 3.65 (s, 3H, OCH3),
3.40–3.54 (m, 2H, COCHCO and C5ꢀH), 3.32 (d, 1H,
J=6.9 Hz, C2ꢀH), 2.55 (dd, 1H, J=8.3, 13.9 Hz,
CH3CꢁCHH), 2.36 (dd, 1H, J=7.3, 14.2 Hz,
CH3CꢁCHH), 2.53 (db, 1H, J=16.7 Hz, C6bꢀH), 2.29
(bd, 1H, J=16.7 Hz, C6aꢀH), 1.64 (t, 3H, 1.3 Hz,
CH3CꢁCH2). IR (cm−1): w(CO) 2039, 1962. EIMS;
m/z: 420 ([M]+, 3), 364 ([M+−2CO], 20), 337 (19),
336 ([M+−3CO], 100), 321 (13), 236 (28), 221 (17), 218
(19), 204 (57), 164 (16), 148 (97), 133 (31). EIMS; m/z:
Mass calc. for [C17H20FeO7]+: 392.058. Found:
392.058.
solid. H-NMR (acetone-d6, 270 MHz): l 7.3–7.7 (m,
15H, Ph), 7.00 (bs, 1H, C3ꢀH), 6.14 (q, 1H, J=6.8 Hz,
CH3CꢁCHCH3), 5.50 (bm, 1H, C2ꢀH), 3.34 (s, 3H,
OCH3), 3.30 (bm, 1H, C5ꢀH), 2.20 (bd, 1H, 14.5 Hz,
C6bꢀH), 1.75 (d, 3H, J=6.8 Hz, CH3CꢁCHCH3), 1.70
(bm, 1H, C6aꢀH), 1.62 (s, 3H, CH3CꢁCHCH3). IR
(cm−1): w(CO) 2037, 1993, 1489, 1435, 1422, 1094, 896,
846. FABMS; m/z: 537 ([M]+, 100), 509 ([M+−CO],
56), 481 ([M+−2CO], 19), 479 (41), 318 (46), 263 (37),
239 (29), 183 (37). FABMS; m/z: Mass calc. for
[C31H30FeO3P]+: 537.1282. Found: 537.1267.
4.4.4. Dicarbonyl[(1,2,3,4,5-p)-1-(1%-phenylethenyl)-4-
methoxycyclohexadienyl]triphenylphosphineiron(1+)
hexafluorophosphate(1−) 4d
Experimental procedures and work up were as de-
scribed in Section 4.4. Experimental details: 3d (0.974 g,
1.58 mmol). Yield: 0.923 g, 80%. Obtained as an orange
1
solid. H-NMR (acetone-d6, 400 MHz): l 7.2–7.6 (m,
20H, Ph), 6.98 (bs, 1H, C3ꢀH), 6.95 (s, 1H,
PhCꢁCHH), 5.46 (s, 1H, PhCꢁCHH), 5.28 (bs, 1H,
C2ꢀH), 3.64 (dd, 1H, J=5.6, 14.6 Hz, C6bꢀH), 3.59
(bs, 1H, C5ꢀH), 2.96 (s, 3H, OCH3), 2.45 (d, 1H,
J=14.6 Hz, C6aꢀH). IR (cm−1): w(CO) 2042, 2001,
1495, 1487, 1434, 1378, 1311, 1237, 1093. FABMS;
m/z: 585 ([M]+, 100), 557 ([M+−CO], 16), 529 ([M+
−2CO], 16), 335 (5), 318 (7), 263 (8), 183 (7). FABMS;
m/z: Mass calc. for [C35H30FeO3P]+: 585.1282. Found:
585.1261.
4.5.2. (E)-Tricarbonyl[methyl (2,3,4,5-p)-2%-
carbomethoxy-4-(4-methoxycyclohexadien-1-ylidene)-
3%-methylbutanoate]iron(0) 5b
Experimental procedures and work up were as de-
scribed in Section 4.5. Experimental details: lithium
bis(trimethylsilyl)amide (1.5 ml, 1 M in THF, 1.5
mmol). Dimethyl malonate (0.184 g, 1.4 mmol); 2b
(0.300 g, 0.69 mmol). Yield: 0.029 g, 10% based on 2b.
Obtained as yellow oil as a mixture of two inseparable
1
diastereoisomers in the ratio 2:3, A:B; by NMR. H-
NMR (CHCl3-d, 270 MHz). Diastereoisomer A: l 5.08
(dd, 1H, J=2.3, 6.6 Hz, C3ꢀH), 5.05 (d, 1H, J=11.0
Hz, CꢁCHCH), 3.69 (s, 3H, OCH3), 3.67 (s, 3H,
OCH3), 3.62 (s, 3H, OCH3), 3.49 (m, 1H, C5ꢀH), 3.19
(d, 1H, J=9.2 Hz, COCHCO), 3.08 (d, 1H, J=6.6
Hz, C2ꢀH), 2.85 (m, 1H, CHCHCH3), 2.60 (m, 1H,
C6bꢀH), 2.25 (m, 1H, C6aꢀH), 0.90 (d, 3H, J=6.3 Hz,
CHCHCH3). Diastereoisomer B: l 5.08 (dd, 1H, J=
2.3, 6.6 Hz, C3ꢀH), 5.05 (d, 1H, J=11.0 Hz,
CꢁCHCH), 3.69 (s, 3H, OCH3), 3.67 (s, 3H, OCH3),
3.62 (s, 3H, OCH3), 3.49 (m, 1H, C5ꢀH), 3.17 (d, 1H,
J=9.2 Hz, COCHCO), 3.08 (d, 1H, J=6.6 Hz,
C2ꢀH), 2.85 (m, 1H, CHCHCH3), 2.60 (m, 1H,
C6bꢀH), 2.25 (m, 1H, C6aꢀH), 0.90 (d, 3H, J=6.3 Hz,
CHCHCH3). IR (cm−1): w(CO) 2048, 1986, 1794, 1729,
1562, 1382. EIMS; m/z: 392 ([M+−CO], 1), 364 ([M+
−2CO], 1), 336 ([M+−3CO], 4), 204 (10), 148 (100),
133 (52), 84 (21), 49(19). EIMS; m/z: Mass calc. for
[C17H20FeO7]+: 392.058. Found: 392.058.
4.5. General procedure for the addition of dialkyl
malonates to 2 and 4
The dialkyl malonate (two equivalents) was dissolved
in dry THF (30 ml) and cooled to −78°C under
nitrogen. The appropriate base (lithium bis(trimethylsi-
lyl)amide or NaH) was added (2.1 equivalents) and the
mixture was stirred for 30 min. The salts 2 and 4 were
added as solids in one portion and the mixture was
stirred at −78°C for a further 30 min before being
allowed to warm to room temperature. The solution
was poured into a separating funnel charged with water
and extracted with diethyl ether (three portions). The
extracts were combined, washed (three portions of wa-
ter) and dried over MgSO4 followed by removal of the
solvent in vacuo. The residues were purified by column
chromatography on silica gel with 20:80; diethyl ether–
petroleum ether (40/60) as the eluant to afford the

R.D.A. Hudson et al. / Journal of Organometallic Chemistry 630 (2001) 88–103
99
4.5.3. (E)-Tricarbonyl[methyl (2,3,4,5-p)-2-
carbomethoxy-4-(4-methoxycyclohexadien-1-ylidene)-3-
methylpentanoate]iron(0) 5c
Experimental procedures and work up were as de-
scribed in Section 4.5. Experimental details: lithium
bis(trimethylsilyl)amide (1.5 ml, 1 M in THF, 1.5
mmol). Dimethyl malonate (0.184 g, 1.4 mmol); 2c
(0.300 g, 0.67 mmol). Yield: 0.022 g, 8% based on 2c.
Obtained as yellow oil as a mixture of two inseparable
bis(trimethylsilyl)amide (0.85 ml, 1 M in THF, 0.85
mmol). Dimethyl malonate (0.106 g, 0.8 mmol); 4a
(0.250 g, 0.37 mmol). Yield: 0.189 g, 78% based on 4a.
1H-NMR (CHCl3-d, 270 MHz): l 6.8–7.8 (m, 15H,
Ph), 3.98 (ddd, 1H, J=2.5, 4.6, 6.9 Hz, C3ꢀH), 3.71
(dd, 1H, J=6.6, 9.2 Hz, COCHCO), 3.51 (s, 3H,
OCH3), 3.40 (m, 1H, C5ꢀH), 3.29 (s, 3H, OCH3), 3.14
(s, 3H, OCH3), 2.90 (dd, 1H, J=9.2, 14.2 Hz,
CH3CCHHCH), 2.86 (bd, 1H, J=17.5 Hz, C6bꢀH),
2.64 (dd, 1H, J=6.9, 7.6 Hz, C2ꢀH), 2.56 (dd, 1H,
J=6.6, 14.2 Hz, CH3CCHHCH), 2.34 (dd, 1H, J=
5.6, 17.5 Hz, C6aꢀH), 1.58 (s, 3H, CH3CCH2CH). IR
(cm−1): w(CO) 1968, 1909. FABMS; m/z: 655 ([M++
H], 15), 654 ([M]+, 12), 625 ([M+−CO+H], 14), 598
([M+−2CO], 7), 449 (100), 349 (11), 336 ([M+−
2CO−PPh3], 17), 263 (15), 183 (13), 148 (8). FABMS;
m/z: Mass calc. for [C33H35FeO5P]+: 598.1572. Found:
598.1552.
1
diastereoisomers in the ratio 2:3, A:B; by NMR. H-
NMR (CHCl3-d, 270 MHz). Diastereoisomer A: l 5.15
(dd, 1H, J=2.6, 6.9 Hz, C3ꢀH), 3.75 (s, 3H, OCH3),
3.74 (s, 3H, OCH3), 3.67 (s, 3H, OCH3), 3.47 (dt, 1H,
J=2.3, 3.9 Hz, C5ꢀH), 3.40 (d, 1H, J=10.9 Hz,
COCHCO), 3.30 (d, 1H, J=6.9 Hz, C2ꢀH), 3.10 (dq,
1H, J=6.9, 11.2 Hz, CH3CCHCH3), 2.68 (m, 1H,
C6bꢀH), 2.42 (m, 1H, C6aꢀH), 1.58 (s, 3H,
CH3CCHCH3),
0.97
(d,
3H,
J=6.9
Hz,
CH3CCHCH3). Diastereoisomer B: l 5.11 (dd, 1H,
J=2.6, 6.9 Hz, C3ꢀH), 3.72 (s, 3H, OCH3), 3.66 (s, 3H,
OCH3), 3.65 (s, 3H, OCH3), 3.54 (dt, 1H, J=2.3, 3.9
Hz, C5ꢀH), 3.41 (d, 1H, J=11.2 Hz, COCHCO), 3.30
(d, 1H, J=6.9 Hz, C2ꢀH), 3.10 (dq, 1H, J=6.9, 11.2
Hz, CH3CCHCH3), 2.68 (m, 1H, C6bꢀH), 2.42 (m, 1H,
C6aꢀH), 1.58 (s, 3H, CH3CCHCH3), 0.89 (d, 3H, J=
6.9 Hz, CH3CCHCH3). IR (cm−1): w(CO) 2042, 1968,
1759, 1738, 1489, 1435, 1226, 1173, 1039. EIMS; m/z:
434 ([M]+, 0.3), 378 ([M+−2CO], 1), 350 ([M+−
3CO], 28), 218 (25), 162 (100), 147 (58). EIMS; m/z:
Mass calc. for [C19H22FeO8]+: 434.0664. Found:
434.0664.
4.5.6. (E)-Dicarbonyl[methyl(2,3,4,5-p)-2-
carbomethoxy-4-(4-methoxycyclohexadien-1-ylidene)-3-
methylbutanoate]triphenylphosphineiron(0) 5bPPh3
Experimental procedures and work up were as de-
scribed in Section 4.5. Experimental details: lithium
bis(trimethylsilyl)amide (1.0 ml, 1 M THF, 1.0 mmol).
Dimethyl malonate (0.125 g, 0.95 mmol); 4b (0.300 g,
0.45 mmol). Yield: 0.88 g, 64% based on 4b. Obtained
as a yellow solid as a mixture of two inseparable
1
diastereoisomers in the ratio 1:3, A:B; by NMR. H-
NMR (CHCl3-d, 270 MHz). Diastereoisomer A: l 6.8–
7.8 (m, 15H, Ph), 4.83 (d, 1H, J=9.9 Hz,
CꢁCHCHCH3), 3.71 (s, 3H, OCH3), 3.55 (m, 1H,
C3ꢀH), 3.34 (s, 3H, CO2CH3), 3.32 (s, 3H, CO2CH3),
3.25 (m, 1H, C5ꢀH), 3.24 (d, 1H, J=9.2 Hz,
COCHCO), 2.85 (ddq, 1H, J=6.9, 9.2, 9.9 Hz,
CH3CHCH), 2.69 (bd, 1H, J=17.5 Hz, C6bꢀH), 2.22
(m, 2H, C6bꢀH and C2ꢀH), 0.84 (d, 3H, J=6.9 Hz,
CH3CH). Diastereoisomer B: l 6.8–7.8 (m, 15H, Ph),
4.88 (d, 1H, J=9.9 Hz, CꢁCHCHCH3), 3.75 (s, 3H,
OCH3), 3.55 (m, 1H, C3ꢀH), 3.39 (s, 3H, CO2CH3),
3.33 (d, 1H, J=9.2 Hz, COCHCO), 3.26 (s, 3H,
CO2CH3), 3.25 (m, 1H, C5ꢀH), 2.90 (ddq, 1H, J=6.9,
9.2, 9.9 Hz, CH3CHCH), 2.69 (bd, 1H, J=17.5 Hz,
C6bꢀH), 2.22 (m, 2H, C6bꢀH and C2ꢀH), 0.75 (d, 3H,
J=6.9 Hz, CH3CH). IR (cm−1): w(CO) 1977, 1921.
FABMS; m/z: 655 ([M++H], 8), 654 ([M]+, 5), 598
([M+−2CO], 24), 449 (100), 349 (11), 336 ([M+−
2CO−PPh3], 26), 318 (14), 163 (130, 183 (15), 147 (14).
FABMS; m/z: Mass calc. for [C35H36FeO7P]+:
655.1548. Found: 655.1563.
4.5.4. (Z)-Tricarbonyl[methyl (2,3,4,5-p)-2-
carboethoxy-4-(4-methoxycyclohexadien-1-ylidene)-4-
phenylbutanoate]iron(0) 5d
Experimental procedures and work up were as de-
scribed in Section 4.5. Experimental details: lithium
bis(trimethylsilyl)amide (0.85 ml, 1 M in THF, 0.85
mmol). Diethyl malonate (0.128 g, 0.8 mmol); 2d (0.200
1
g, 0.40 mmol). Yield: 0.059 g, 29% based on 2d. H-
NMR (CHCl3-d, 400 MHz): l 7.35 (m, 3H, Ph), 7.10
(m, 2H, Ph), 5.01 (dd, 1H, J=2.4, 6.8 Hz, C3ꢀH), 4.06
(q, 4H, J=7.2 Hz, 2×CH3CH2O), 3.62 (s, 3H, OCH3),
3.55 (m, 1H, C5ꢀH), 3.13 (d, 1H, J=6.8 Hz, C2ꢀH),
3.10 (dd, 1H, J=7.3, 7.8 Hz, COCHCO), 2.84 (dd, 1H,
7.8, 14.4 Hz, PhCCHHCH), 2.73 (dd, 1H, 7.8, 14.4 Hz,
PhCCHHCH), 2.72 (d, 1H, J=17.8 Hz, C6bꢀH), 2.46
(dd, 1H, J=2.0, 17.8 Hz, C6aꢀH), 1.19 (t, 6H, J=7.2
Hz, 2×CH3CH2O). EIMS; m/z: Mass calc. for
[C25H26FeO8]+: 510.0977. Found: 510.0977.
4.5.5. (E)-Dicarbonyl[methyl(2,3,4,5-p)-2-
4.5.7. (Z)-Dicarbonyl[methyl(2,3,4,5-p)-2-
carbomethoxy-4-(4-methoxycyclohexadien-1-ylidene)-
pentanoate]triphenylphosphineiron(0) 5aPPh3
Experimental procedures and work up were as de-
scribed in Section 4.5. Experimental details: lithium
carbomethoxy-4-(4-methoxycyclohexadien-1-ylidene)-
4-phenylbutanoate]triphenylphosphineiron(0) 5dPPh3
Experimental procedures and work up were as de-
scribed in Section 4.5. Experimental details: lithium

100
R.D.A. Hudson et al. / Journal of Organometallic Chemistry 630 (2001) 88–103
bis(trimethylsilyl)amide (0.5 ml, 1 M in THF, 0.5
mmol). Diethyl malonate (0.080 g, 0.5 mmol); 4d (0.200
g, 0.27 mmol). Yield: 1.02 g, 51% based on 4d. Ob-
tained as a yellow solid. 1H-NMR (CHCl3-d, 270
MHz): l 7.0–7.4 (m, 20H, Ph), 4.16 (bm, 1H, C3ꢀH),
4.06 (dd, 1H, J=7.1, 8.5 Hz, COCHCO), 4.00 (q, 4H,
J=7.6 Hz, 2×CO2CH2CH3), 3.36 (s, 3H, OCH3), 3.28
(bs, 1H, C5ꢀH), 3.07 (dd, 1H, J=7.1, 8.3 Hz, C2ꢀH),
2.83 (dd, 1H, J=7.1, 14.4 Hz, PhCCHH), 2.76 (m, 1H,
C6bꢀH), 2.75 (dd, 1H, J=7.1, 14.4 Hz, PhCCHH),
2.43 (dd, 1H, J=5.4, 17.3 Hz, C6aꢀH), 1.17 (t, 3H,
J=7.6 Hz, CO2CH2CH3), 1.15 (t, 3H, J=7.6 Hz,
CO2CH2CH3). IR (cm−1): w(CO) 1970, 1912, 1727,
1594, 1483, 1430, 1262, 845, 757. FABMS; m/z: 744
([M]+, 8), 688 ([M+−2CO], 9), 687 ([M+−2CO−H],
20), 585 ([M+−malonate], 100), 557 (26), 529 (39), 318
(24), 263 (27), 183(26). FABMS; m/z: Mass calc. for
[C35H30FeO3P]+: 585.1220. Found: 585.1220.
(dd, 3H, J=1.7, 7.3 Hz, CHꢁCHCH3). IR (cm−1):
w(CO) 2054, 1986. EIMS; m/z: 420 ([M]+, 0.1), 392
([M+−CO], 0.1), 364 ([M+−2CO], 2), 336 ([M+−
3CO], 18), 204 (15), 148 (100), 133 (16). Anal. Found:
C, 51.6; H, 4.6. Calc. for C18H20FeO8: C, 51.5; H, 4.8%.
4.5.10. Tricarbonyl[methyl (2,3,4,5-p)-2-carbomethoxy-
2-(4-methoxy-1-(1%,2%-dimethylethenyl)cyclohexadien-
1-yl)ethanoate]iron(0) 6c
Experimental procedures and work up were as de-
scribed in Section 4.5. Experimental details: sodium
hydride (0.056 g, 60% dispersion in mineral oil, 1.4
mmol). Dimethyl malonate (0.185 g, 1.4 mmol); 2c
(0.300 g, 0.67 mmol). Yield: 0.116 g, 40% based on 2c.
Obtained as a yellow oil. 1H-NMR (CHCl3-d, 270
MHz): l 5.32 (q, 1H, J=7.6 Hz, CH3CꢁCHCH3), 5.08
(dd, 1H, J=1.4, 6.3 Hz, 3ꢀH), 3.93 (s, 1H, COCHCO),
3.74 (s, 3H, OCH3), 3.66 (s, 3H, OCH3), 3.62 (s, 3H,
OCH3), 3.26 m, 2H, C2ꢀH and C5ꢀH), 2.80 (m, 1H,
C6bꢀH), 2.27 (m, 1H, C6aꢀH), 1.91 (s, 3H,
4.5.8. Tricarbonyl[methyl (2,3,4,5-p)-2-carbomethoxy-
2-(4-methoxy-1-(1%-methylethenyl)cyclohexadien-
1-yl)ethanoate]iron(0) 6a
CH3CꢁCHCH3),
1.61
(d,
3H,
J=7.6
Hz,
CH3CꢁCHCH3). IR (cm−1): w(CO) 2043, 1980. EIMS;
m/z: 378 ([M+−2CO], 1), 350 ([M+−3CO], 7), 218
(10), 162 (100), 147 (65). Anal. Found: C, 52.7; H, 5.0.
Calc. for C19H20FeO8: C, 52.5; H, 5.1%.
Experimental procedures and work up were as de-
scribed in Section 4.5. Experimental details: sodium
hydride (0.056 g, 60% dispersion in mineral oil, 1.4
mmol). Dimethyl malonate (0.185 g, 1.4 mmol); 2a
(0.300 g, 0.69 mmol). Yield: 0.084 g, 30% based on 2a.
Obtained as a yellow oil. 1H-NMR (C6H6-d6, 270
MHz): l 5.00 (s, 1H, CH3CꢁCHH), 4.92 (s, 1H,
CH3CꢁCHH), 4.68 (dd, 1H, J=2.3, 6.9 Hz, C3ꢀH),
3.72 (s, 1H, COCHCO), 3.27 (s, 3H, OCH3), 3.26 (s,
3H, OCH3), 3.1 (m, 2H, C2ꢀH and C5ꢀH), 2.99 (s, 3H,
OCH3), 2.59 (dd, 1H, J=2.6, 15.5 Hz, C6bꢀH), 1.87
(dd, 1H, J=3.3, 15.2 Hz, C6aꢀH), 1.60 (s, 3H,
CH3CꢁCH2). IR (cm−1): w(CO) 2039, 1962. 13C-NMR
(C6H6-d6, 67 MHz): l 210.7, 168.3, 167.9, 147.1, 140.3,
114.2, 65.0, 60.7, 53.0, 52.3, 52.1, 52.0, 50.0, 37.2, 19.4.
The structure was also established by X-ray analysis
(Fig. 6, Tables 1 and 3).
4.5.11. Tricarbonyl[ethyl (2,3,4,5-p)-2-carboethoxy-2-
(4-methoxy-1-(1%-phenylethenyl)cyclohexadien-
1-yl)ethanoate]iron(0) 6d
Experimental procedures and work up were as de-
scribed in Section 4.5. Experimental details: sodium
hydride (0.056 g, 60% dispersion in mineral oil, 1.4
mmol). Diethyl malonate (0.224 g, 1.4 mmol); 2d (0.300
g, 0.67 mmol). Yield: 0.116 g, 40% based on 2d. Ob-
1
tained as a yellow oil. H-NMR (CHCl3-d, 400 MHz):
l 7.35 (3H, m, Ph), 7.10 (m, 2H, Ph), 5.40 (s, 1H,
PhCꢁCHH), 5.15 (s, 1H, PhCꢁCHH), 5.00 (dd, 1H,
J=2.4, 6.8 Hz, C3ꢀH), 4.13 (q, 4H, J=7.2 Hz, 2×
CH3CH2O), 3.76 (s, 1H, COCHCO), 3.61 (s, 3H,
OCH3), 3.65 (m, 1H, C5ꢀH), 3.36 (d, 1H, J=6.8 Hz,
C2ꢀH), 2.83 (bd, 1H, J=15.4 Hz, C6bꢀH), 2.32 (dd,
1H, J=3.18, 15.4 Hz, C6aꢀH), 1.10 (q, 6H, J=7.20
Hz, 2×CH3CH2O). IR (cm−1): w(CO) 2042, 1971.
EIMS; m/z: Mass calc. for [C25H26FeO8]+: 510.0977.
Found: 510.0977.
4.5.9. Tricarbonyl[methyl (2,3,4,5-p)-2-carbomethoxy-
2-(4-methoxy-1-(2%-methylethenyl)cyclohexadien-
1-yl)ethanoate]iron(0) 6b
Experimental procedures and work up were as de-
scribed in Section 4.5. Experimental details: sodium
hydride (0.120 g, 60% dispersion in mineral oil, 3.0
mmol). Dimethyl malonate (0.396 g, 3.0 mmol); 2b
(0.660 g, 1.52 mmol). Yield: 0.287 g, 45% based on 2b.
Obtained as a yellow oil. 1H-NMR (CHCl3-d, 270
MHz): l 5.15 (dq, 1H, J=1.7, 15.3 Hz, CHꢁCHCH3),
5.38 (dq, 1H, J=6.9, 15.3 Hz, CHꢁCHCH3), 5.02 (dd,
1H, J=2.3, 6.6 Hz, C3ꢀH), 3.80 (s, 1H, COCHCO),
3.73 (s, 3H, OCH3), 3.67 (s, 3H, OCH3), 3.62 (s, 3H,
OCH3), 3.27 (dt, 1H, J=2.6, 3.3 Hz, C5ꢀH), 3.12 (d,
1H, J=6.6 Hz, C2ꢀH), 2.66 (dd, 1H, J=2.6, 14.8 Hz,
C6bꢀH), 2.21 (dd, 1H, J=3.3, 14.8 Hz, C6aꢀH), 1.63
4.6. General procedure for the addition of
organocuprates to 2a
The appropriate organolithium reagent (four equiva-
lents) was stirred in THF (10 ml) at −30°C and copper
bromide dimethyl sulfide complex (two equivalents) was
added. The mixture was stirred for a further 1 h until
all the copper salt had been consumed. 2a (one equiva-
lent) was added as a solid in one portion and the
mixture was stirred for a further 2 h at −30°C fol-

R.D.A. Hudson et al. / Journal of Organometallic Chemistry 630 (2001) 88–103
101
4.6.3. (1E)-Tricarbonyl[(2,3,4,5-p)-4-methoxy-1-
lowed by the addition of saturated aqueous ammonium
chloride solution. The mixture was extracted with di-
ethyl ether (3×50 ml) and the combined extracts were
washed with water (2×50 ml) and brine (1×50 ml).
The solution was dried over magnesium sulfate and the
solvent removed in vacuo. The residue was purified by
column chromatography on silica gel to afford the
required compounds as yellow oils.
(1-methyl-2-phenyl)ethylidene)cyclohexadiene]iron(0) 8c
Experimental procedures and work up were as de-
scribed in Section 4.6. Experimental details: diphenyl-
cuprate (4.60 mmol); 2a (0.500 g, 1.15 mmol). Yield:
0.282 g, 67% based on 2a. Obtained as a yellow oil.
1H-NMR (CHCl3-d, 270 MHz): l 7.15 (m, 5H, ArH),
5.15 (dd, 1H, J=2.6, 6.9 Hz, C3ꢀH), 3.67 (s, 3H,
OCH3), 3.55 (m, 1H, C5ꢀH), 3.46 (d, 1H, J=6.6 Hz,
C2ꢀH), 3.27 (d, 1H, J=14.8 Hz, CH2Ph), 3.05 (d, 1H,
J=14.8 Hz, CH2Ph), 2.60 (bd, 1H, J=17.2 Hz,
C6bꢀH), 2.40 (bd, 1H, J=17.2 Hz, C6aꢀH), 1.60 (s,
3H, CH3C). 1H-NMR (CHCl3-d, 270 MHz, n.O.e.):
irradiation at l 1.60 (s, 3H, CH3C). Observed n.O.e.: l
3.46 (4.5%, d, 1H, J=6.6 Hz, C2ꢀH). IR (cm−1):
w(CO) 2040, 1965. Anal. Found: C, 62.6; H, 4.8. Calc.
for C19H18O4Fe: C, 62.3; H, 4.9%. EIMS; m/z: 338
([M+−CO], 1), 310 ([M+−2CO], 1), 282 ([M+−
3CO], 8), 224 (13), 204 (13), 148 (36), 135 (100).
4.6.1. (1E)-Tricarbonyl[(2,3,4,5-p)-4-methoxy-1-
(1-methylpropylidene)cyclohexadiene]iron(0) 8a
Experimental procedures and work up were as de-
scribed in Section 4.6. Experimental details: dimethyl-
cuprate (2.0 mmol); 2a (0.250 g, 0.57 mmol). Yield:
0.135 g, 78% based on 2a. Obtained as a yellow oil.
1H-NMR (CHCl3-d, 270 MHz): l 5.05 (dd, 1H, J=2.6,
6.6 Hz, C3ꢀH), 3.58 (s, 3H, OCH3), 3.46 (m, 1H,
C5ꢀH), 3.36 (d, 1H, J=6.9 Hz, C2ꢀH), 2.36 (db, 1H,
J=14.4 Hz, C6bꢀH), 2.17 (bd, 1H, 14.4 Hz, C6aꢀH),
1.78 (dq, 1H, J=7.6, 13.5 Hz, CH2CH3), 1.59 (t, 3H,
J=1.3 Hz, CH3CCH2), 0.81 (t, 3H, J=7.6 Hz,
4.7. Formation of cations 9
1
CH2CH3). H-NMR (CHCl3-d, 270 MHz, n.O.e.): irra-
4.7.1. Tricarbonyl[(1,2,3,4,5-p)-4-methoxy-1-
diation at l 1.59 (t, 3H, J=1.3 Hz, CH3CCH2). Ob-
served n.O.e.: l 3.36 (3.5%, d, 1H, J=6.9 Hz, C2ꢀH).
IR (cm−1): w(CO) 2043, 1978, 1968. EIMS; m/z: 304
([M]+, 31), 276 ([M+−CO], 49), 248 ([M+−2CO],
25), 220 (95), 218 (100), 216 (43), 178 (44). EIMS; m/z:
Mass calc. for [C14H16O4Fe]+: 304.0398. Found:
304.0398.
(1-methylpropyl)cyclohexdienyl]iron(1+) hexafluoro-
phosphate(1−) 9a
Experimental procedures and work up were as de-
scribed in Section 4.2. Experimental details: 8a (0.120 g,
0.39 mmol). Yield: 0.127 g, 72%. Recovered as a single
diastereoisomer as a pale yellow solid. 1H-NMR
(CH3CN-d3, 270 MHz): l 6.80 (dd, 1H, J=2.3, 5.9 Hz,
C3ꢀH), 5.56 (d, 1H, J=5.9 Hz, C2ꢀH), 3.90 (d, 1H,
J=5.9 Hz, C5ꢀH), 3.78 (s, 3H, OCH3), 3.03 (dd, 1H,
J=6.3, 15.5 Hz, C6bꢀH), 2.25 (bd, 1H, J=15.5 Hz,
C6aꢀH), 1.89 (sextet, 1H, J=6.9 Hz, CH3CHCH2),
1.48 (m, 2H, CHCH2CH3), 1.18 (d, 3H, J=6.9 Hz,
CH3CH), 0.87 (t, 3H, J=7.3 Hz, CH2CH3). 13C-NMR
(CH3CN-d3, 67 MHz): l 150.7 (C4), 100.2 (C1), 94.9
(C2), 72.5 (C3), 58.0 (OCH3), 43.2 (C5), 42.6
(CH3CHCH2), 29.8 (C6), 27.1 (CHCH2CH3), 19.06
(CH3CH), 11.9 (CHCH2CH3). IR (cm−1): w(CO) 2102,
2057. Anal. Found: C, 37.5; H, 3.7. Calc. for
C14H17O4FePF6: C, 37.4; H, 3.8%. EIMS; m/z: 304
([M+−H], 3), 277 ([M+−CO], 4), 276 ([M+−CO−
H], 23), 249 ([M+−2CO], 2), 248 ([M+−2CO−H],
14), 221 ([M+−3CO], 10), 220 ([M+−3CO−H], 59),
218 (40), 135 (100).
4.6.2. (1E)-Tricarbonyl[(2,3,4,5-p)-4-methoxy-1-
(1-methylhexylidene)cyclohexadiene]iron(0) 8b
Experimental procedures and work up were as de-
scribed in Section 4.6. Experimental details: dimethyl-
cuprate (2.0 mmol); 2a (0.230 g, 0.53 mmol). Yield:
0.128 g, 70% based on 2a. Obtained as a yellow oil.
1H-NMR (CHCl3-d, 270 MHz): l 5.10 (dd, 1H, J=2.4,
6.6 Hz, C3ꢀH), 3.64 (s, 3H, OCH3), 3.49 (m, 1H,
C5ꢀH), 3.43 (d, 1H, J=6.6 Hz, C2ꢀH), 2.35 (m, 1H,
C6bꢀH), 1.80 (m, 1H, C6aꢀH), 1.65 (t, 3H, J=1.3 Hz,
1
CH3CCH2), 1.2 (m, 11H, CH2CH2CH2CH2CH3). H-
NMR (CHCl3-d, 270 MHz, n.O.e.): irradiation at l
1.65 (t, 3H, J=1.3 Hz, CH3CCH2). Observed n.O.e.: l
3.42 (4.0%, d, 1H, J=6.6 Hz, C2ꢀH). 13C-NMR
(CHCl3-d, 67 MHz): l 211.3 (MCO), 139.4 (C4), 129.0
(C1), 125.6 (CH3CCH2), 65.6 (C3), 54.4 (OCH3), 52.6
(C2), 52.4 (C5), 35.0 (C6), 31.8 (CH3CCH2), 29.4
(CH3CCH2CH2), 27.1 (CH3CCH2CH2CH2), 22.6
(CH3CCH2CH2CH2CH2), 17.3 (CH3CCH2), 13.9
(CH2CH3). IR (cm−1): w(CO) 2041, 1970. Anal.
Found: C, 59.1; H, 6.4. Calc. for C17H22O4Fe: C, 59.0;
H, 6.4%. EIMS; m/z: 346 ([M]+, 2), 318 ([M+−CO],
26), 290 ([M+−2CO], 9), 262 ([M+−3CO], 36), 190
(30), 135 (100).
4.7.2. Tricarbonyl[(1,2,3,4,5-p)-4-methoxy-1-
(1-methylhexyl)cyclohexdienyl]iron(1+) hexafluoro-
phosphate(1−) 9b
Experimental procedures and work up were as de-
scribed in Section 4.2. Experimental details: 8b (0.121 g,
0.35 mmol). Yield: 0.100 g, 66%. Recovered as a single
1
diastereoisomer as a pale yellow solid. H-NMR (ace-
tone-d6, 270 MHz): l 6.80 (br, 1H, C3ꢀH), 5.56 (br,
1H, C2ꢀH), 3.90 (br, 1H, C5ꢀH), 3.79 (s, 3H, OCH3),

102
R.D.A. Hudson et al. / Journal of Organometallic Chemistry 630 (2001) 88–103
3.01 (br, 1H, C6bꢀH), 2.50 (m, 1H, CH3CHCH2), 2.28
(br, 1H, C6aꢀH), 0.9–1.4 (m, 14H, CH3CHCH2-
CH2CH2CH2CH3). 13C-NMR (acetone-d6, 67 MHz): l
213.0 (CO), 148.2 (C4), 98.6 (C1), 92.2 (C2), 69.9 (C3),
55.5 (OCH3), 38.4 (C5), 38.2 (CH3CHCH2), 31.5 (C6),
29.5 (CH3CHCH2), 27.5 (CHCH2CH2), 24.7 (CHCH2-
CH2CH2), 20.4 (CH3CH), 16.9 (CHCH2CH2CH2CH2),
11.5 (CH2CH3). IR (cm−1): w(CO) 2107, 2057. EIMS;
m/z: 347 ([M++H)], 346 ([M)+], 318 ([M++H−
CO], 4), 290 ([M+−2CO)], 262 ([M+−3CO)], 135
(100). EIMS; m/z: Mass calc. for [C17H23O4Fe]+:
347.095. Found: 347.095.
6. Supplementary material
Crystallographic data for the structural analysis have
been deposited with the Cambridge Crystallographic
Data Centre, CCDC nos. 154839 for compound 2a,
154841 for compound 2d, 154840 for compound 4b and
155022 for compound 6a. Copies of this information
may be obtained free of charge from The Director,
CCDC, 12 Union Road, Cambridge CB2 1EZ, UK
(Fax: +44-1223-336033; e-mail: deposit@ccdc.cam.
ac.uk or www:http://www.ccdc.cam.ac.uk).
Acknowledgements
4.7.3. Tricarbonyl[(1,2,3,4,5-p)-1-(1-benzylethyl)-4-
methoxycyclohexdienyl]iron(1+) hexafluoro-
This work was supported by EPSRC and Parke–
Davis (financial support to R.D.A.H.). We thank the
EPSRC Mass Spectrometry Centre (Swansea) for FAB
and high-resolution mass spectrometric measurements.
Microanalyses and low-resolution mass spectrometric
measurements were made by Mr I. Saunders, Univer-
sity of East Anglia, Norwich.
phosphate(1−) 9c
Experimental procedures and work up were as de-
scribed in Section 4.2. Experimental details: 8c (0.105 g,
0.29 mmol). Yield: 0.093 g, 65%. Recovered as a single
diastereoisomer as a pale yellow solid. 1H-NMR
(CH3CN-d3, 270 MHz): l 7.23 (m, 5H, ArH), 7.09 (dd,
1H, J=2.3, 5.9 Hz, C3ꢀH), 5.83 (d, 1H, J=5.9 Hz,
C2ꢀH), 4.31 (d, 1H, J=6.3 Hz, C5ꢀH), 3.96 (s, 3H,
OCH3), 3.36 (dd, 1H, J=6.3, 15.2 Hz, C6bꢀH), 2.65
(d, 2H, J=7.1 Hz, CHCH2Ph), 2.51 (bd, 1H, J=15.2
Hz, C6aꢀH), 2.43 (m, 1H, CH3CHCH2), 1.27 (d, 3H,
J=6.6 Hz, CH3CH). 13C-NMR (CH3CN-d3, 67 MHz):
l 150.8 (C4), 139.4, 129.8, 129.4, 127.5 (ArC), 98.3
(C1), 95.5 (C2), 72.4 (C3), 58.0 (OCH3), 43.4 (C5), 43.1
(CH3CHCH2), 40.4 (CH3CHCH2), 29.6 (C6), 19.3
(CH3CH). IR (cm−1): w(CO) 2107, 2057. EIMS; m/z:
339 ([M++H−CO)], 310 ([M++H−2CO)], 282
([M++H−3CO)], 135 (100). CIMS m/z: Mass calc.
for [C18H20O3Fe]+: 339.068. Found: 339.068.
References
[1] (a) R.D.A. Hudson, B.M. Foxman, M. Rosenblum,
Organometallics 18 (1999) 4098. (b) R.D.A. Hudson, B.M. Fox-
man, M. Rosenblum, Organometallics 19 (2000) 469.
[2] E.W. Abel, F.G.A. Stone, G. Wilkinson (Eds.), Comprehensive
Organometallic Chemistry 11, vols. 5–9, Pergamon Press/El-
sevier Science, Oxford, 1995 (for applications in synthesis, see
vol. 12).
[3] (a) M. Franck-Neumann, L. Meisch-Gross, C. Gateau, Tetrahe-
dron Lett. 40 (1999) 2829. (b) T.J. Benvegnu, L.J. Toupet, R.L.
Gre´e, Tetrahedron 52 (1996) 11 811. (c) B. Dasgupta, W.A.
Donaldson, Tetrahedron: Asymmetry 9 (1998) 3781.
[4] (a) B.M.R. Bandara, A.J. Birch, L.F. Kelly, J. Org. Chem. 49
(1984) 2496. (b) A.J. Birch, L.F. Kelly, D.V. Weerasuria, J. Org.
Chem. 53 (1988) 278. (c) A.J. Pearson, A. Alimardanov, A.A.
Pinkerton, D.M. Fouchard, K. Kirschbaum, Tetrahedron Lett.
39 (1998) 5919. (d) H.-J. Kno¨lker, E. Baum, T. Hopfmann,
Tetrahedron 55 (1999) 10 391. (e) H.-J. Kno¨lker, W. Fro¨hner,
Tetrahedron Lett. 40 (1999) 6915.
5. X-ray data collection
5.1. Crystallography
Crystallographic data for compounds 2a, 2d, 4b and
6a are summarised in Table 3.
Data collections were conducted on a CAD4 auto-
matic four-circle diffractometer at ambient temperature
using Mo–Ka radiation (u=0.71069 A) for 2a, 2d and
[5] G.R. Stephenson (Ed.), Advanced Asymmetric Synthesis,
Blackie/Chapman & Hall, London, 1996, pp. 319–320.
[6] G.R. Stephenson, H. Finch, D.A. Owen, S. Swanson, Tetrahe-
dron 49 (1993) 5649.
[7] E.J. Sandoe, G.R. Stephenson, S. Swanson, Tetrahedron Lett. 37
(1996) 6283.
,
4b and on a Siemens P4 for 6a. Full-matrix anisotropic
refinement was implemented in the final least-squares
cycles throughout. Four of the fluorines in the PF6
anion were seen to be disorder in the ratio 3:2 with
their primed counterparts in compound 2a. Hydrogens
were included at calculated positions as appropriate,
with the exception of H4, H6 and H7 in 2a, H211,
H231, H241, H281 and H291 in 4b, and H4, H6 and
H7 in 2d. These hydrogens were readily located in their
respective penultimate difference Fourier maps and
freely refined.
[8] S.T. Astley, G.R. Stephenson, Synlett (1992) 507.
[9] (a) G.R. Stephenson, I.M. Palotai, W.J. Ross, D.E. Tupper, J.
Organomet. Chem. 364 (1991) C11. (b) G.R. Stephenson, I.M.
Palotai, W.J. Ross, D.E. Tupper, Synlett (1991) 586. (c) M.
Tinkl, PhD Thesis, University of East Anglia, Norwich, 1995.
[10] D.A. Owen, G.R. Stephenson, H. Finch, S. Swanson, Tetrahe-
dron Lett. 30 (1998) 2607.
[11] N. Genco, D. Marten, S. Raghu, M. Rosenblum, J. Am. Chem.
Soc. 98 (1976) 848. For further examples of Rosenblum’s
cationic Fp complexes with alkenes and their regio- and stereo-
chemical reactions with nucleophiles see W. Zhen, K.-H Chu, M.
Rosenblum, J. Org. Chem. 62 (1997) 3344 and references cited
therein.

R.D.A. Hudson et al. / Journal of Organometallic Chemistry 630 (2001) 88–103
103
[12] P. Lennon, A.M. Rosan, M. Rosenblum, J. Am. Chem. Soc. 99
(1977) 8426.
[13] N. Genco, D. Marten, S. Raghu, M. Rosenblum, J. Am. Chem.
Soc. 98 (1976) 848.
[18] (a) G.R. Stephenson, M. Voyle, S. Williams, Tetrahedron Lett.
31 (1990) 3979. (b) Preliminary communication: R.D.A. Hudson,
S.A. Osborne, E. Roberts, D.G. Smyth, G.R.Stephenson, Tetra-
hedron Lett. 37 (1996) 9009. (c) R.D.A. Hudson, S.A. Osborne,
G.R.Stephenson, Tetrahedron 53 (1997) 4095.
[19] (a) A.J. Pearson, P.R. Raithby, J. Chem. Soc. Dalton Trans.
(1981) 884. (b) J.A.D. Jeffreys, C. Metters, J. Chem. Soc. Dalton
Trans. (1977) 729.
[20] A.J. Birch, P.E. Cross, J. Lewis, D.A. White, S.B. Wild, J.
Chem. Soc. A. (1968) 332.
[21] C.E. Anson, R.D.A. Hudson, D.G. Smyth, G.R. Stephenson,
Appl. Organomet. Chem. 15 (2001) 16.
[14] P. McArdle, J. Appl. Crystallogr. 28 (1995) 65.
[15] (a) M.F. Semmelhack, H.T. Hall Jr., R. Farina, M. Yoshifuji, G.
Clark, T. Bargar, K. Hirotsu, J. Clardy, J. Am. Chem. Soc. 101
(1979) 3535. (b) N.A. Bailey, E.H. Blunt, G. Fairhurst, C. White,
J. Chem. Soc. Dalton Trans. (1980) 829. (c) M. Bottrill, M.
Green, E. O’Brien, L.E. Smart, P. Woodward, J. Chem. Soc.
Dalton Trans. (1980) 292. (d) B. Etermadi, D.S. Moss, R.A.
Palmer, J. Inorg. Nucl. Chem. 43 (1981) 1997. (e) P. Janse Van
Nuuren, R.J. Fletterich, J. Meinwald, R.E. Hughes, J. Am.
Chem. Soc. 93 (1971) 4394.
[22] Preliminary communication: C.E. Anson, R.D.A. Hudson, S.A.
Osborne, G.R. Stephenson, Tetrahedron Lett. 39 (1998)
7603.
[16] J. March, Advanced Organic Chemistry, 4th ed., Wiley, New
York, 1992 (pp. 19–21).
[23] D. Seebach, H. Neuman, Terahedron Lett. 52 (1976) 4839.
[24] A.J. Birch, L.F. Kelly, J. Organomet. Chem. 286 (1985) C5.
[17] O. Eisenstein, W.M. Butler, A.J. Pearson, Organometallics 3
(1984) 1150.
.