An organoiron approach to thyroid hormone analogues

Article abstract of DOI:10.1016/j.jorganchem.2005.01.032

An approach to thyroid hormone analogues was proposed involving sequential substitution of cationic cyclopentadienyl(1,4-dichlorobenzene)iron(II) complexes with phenoxide/thiophenoxide and hydroxide/amine, followed by decomplexation. Although the selectivity for monosubstitution with phenolates and thiophenolates was poorer than previously observed, it was often possible to control the reaction with sterically less demanding phenolates of intermediate nucleophilicity. The subsequent introduction of a polar substituent into the monosubstituted product was successful with amine nucleophiles. A modified approach, based on the reverse order of substitution was also attempted. Whereas clean monosubstitution with hydroxide/hydroxide equivalents was unsuccessful, cyclopentadienyl(N-alkyl-1-chloro-4-aminobenzene)iron(II) complexes could be prepared in fair yields and further substituted with nucleophiles such as thiophenolates.

Full text of DOI:10.1016/j.jorganchem.2005.01.032

Journal of Organometallic Chemistry 690 (2005) 1784–1792
www.elsevier.com/locate/jorganchem
An organoiron approach to thyroid hormone analogues
a
Joulia Smirnova , Lars Engman , Carl-Magnus Andersson , Johan Malm
a,
b
b
*
a
Department of Chemistry, Organic Chemistry, Uppsala University, P.O. Box 599, S-751 24 Uppsala, Sweden
b
Chemistry Department, Karobio AB, Novum, S-141 57 Huddinge, Sweden
Received 27 March 2004; accepted 4 January 2005
Available online 8 March 2005
Abstract
An approach to thyroid hormone analogues was proposed involving sequential substitution of cationic cyclopentadienyl(1,4-
dichlorobenzene)iron(II) complexes with phenoxide/thiophenoxide and hydroxide/amine, followed by decomplexation. Although
the selectivity for monosubstitution with phenolates and thiophenolates was poorer than previously observed, it was often possible
to control the reaction with sterically less demanding phenolates of intermediate nucleophilicity. The subsequent introduction of a
polar substituent into the monosubstituted product was successful with amine nucleophiles. A modified approach, based on the
reverse order of substitution was also attempted. Whereas clean monosubstitution with hydroxide/hydroxide equivalents was unsuc-
cessful, cyclopentadienyl(N-alkyl-1-chloro-4-aminobenzene)iron(II) complexes could be prepared in fair yields and further substi-
tuted with nucleophiles such as thiophenolates.
ꢀ
2005 Elsevier B.V. All rights reserved.
Keywords: Thyroid hormones; (1,4-Dichlorobenzene)iron(II) complexes; Sequential substitution; Phenoxide/thiophenoxide; Hydroxide/amine
I
I
H2N
R1
R3
R4
1
. Introduction
COOH
HO
O
Y
X
Z
Selenoenzymes is a slowly but steadily growing family
X
Y
R2
of enzymes. To date, the structure and function of a doz-
en of them are known with some certainty [1]. Like other
selenoenzymes, the three varieties of the iodothyronine
deiodinase enzymes (ID-1, ID-2 and ID-3) contain a
selenocysteine residue at the active site. These enzymes
catalyze the interconversion of active and inactive forms
of thyroid hormones [2]. The follicular cells of the thy-
roid provide the only source of thyroxin (1a). However,
this is only thought to be a prohormone requiring outer
ring deiodination to provide the active hormone 3,5,3 -
triiodothyronine (1b). This reduction is brought about
by the ID-1 and ID-2 varieties of the enzyme. The ID-
3 variety is known to cause inner-ring deiodination of
thyroxin to give inactive 1c. Both compounds 1b and
1
a X=Y=I
2
X=O, S, Se
Y=OH, NH2
Z=CH2CH2COOR
R1-R4=H, Cl, Br, I, Me,
i-Pr, or t-Bu
1b X=H; Y=I
1
1
c X=I; Y=H
d X=Y=H
The thyroid hormone receptor (TR) functions as a li-
gand-dependent transcriptional regulator that controls
the expression of a specific set of genes involved in devel-
0
opment and homeostasis in response to 1b (T ) [3].
3
There are two known TR subtypes, a (TRa) and b
TRb), expressed from two different genes. In adults,
(
the TRb isoform is the most prevalent form in most tis-
sues, especially in the liver. The TRa isoform has been
found in highest concentration in skeletal muscle and
brain and is closely linked to cardiac function. At nor-
mal levels, thyroid hormones maintain body weight,
metabolic rate, body temperature, and mood, and
1
c can undergo further monodeiodination to provide 1d.
*
Corresponding author. Tel.: +46 18 471 3784; fax: +46 18 471 3818.
E-mail address: lars.engman@kemi.uu.se (L. Engman).
0
022-328X/$ - see front matter ꢀ 2005 Elsevier B.V. All rights reserved.
doi:10.1016/j.jorganchem.2005.01.032

J. Smirnova et al. / Journal of Organometallic Chemistry 690 (2005) 1784–1792
1785
influence serum lipid levels as well as basal hepatic glu-
R1
R3
R4
cose output. Thus, in hypothyroidism there is weight
gain, high levels of LDL cholesterol, and depression.
In hyperthyroidism, these hormones cause weight loss,
hypermetabolism, lowering of serum LDL cholesterol,
cardiac tachycardia, heart failure, muscle weakness,
bone loss in postmenopausal women, and anxiety. Thy-
roid hormones are currently used primarily as replace-
ment therapy for patients with hypothyroidism.
Therapy with 1b returns metabolic functions to normal.
However, replacement therapy, particularly in older
individuals, is limited by the deleterious effects of thy-
roid hormones. Some effects of thyroid hormones
should be therapeutically useful in non-thyroid disor-
ders provided that adverse effects could be minimized
or eliminated. Such disorders include atherosclerosis,
dyslipidemia, hypercholesterolemia, obesity, diabetes,
congestive heart failure, osteoporosis and various skin
disorders [4]. Development of tissue and/or subtype
selective thyroid hormone receptor agonists could lead
to novel therapies for these common disorders. For
avoiding cardiotoxicity, TR agonists that interact selec-
tively with the b-isoform of the thyroid hormone recep-
tor are especially attractive [5].
2
L
L
+
X
Z
R2
M
3
4
L = a good leaving group
Scheme 1.
5
zene)-g -(cyclopentadienyl)iron hexafluorophosphate
5) as a model compound.
(
2
. Results and discussion
The only practical method for preparation of cationic
arene iron(II) cyclopentadienyl complexes involves the
ligand exchange reaction with ferrocene discovered by
Nesmeyanov et al. [8]. However, product yields vary
dramatically depending on the arene substituents. With
haloaromatics, dehalogenation is frequently observed.
Therefore, under optimized conditions, Pearson and
Gelormini [9] could only obtain a 24% yield of 1,4-
dichlorobenzene complex 5. Selective displacement of
chlorides with phenolates from cyclopentadienyl(1,4-
dichlorobenzene)iron complexes is of crucial importance
for the strategy outlined in Scheme 2 [9,10]. As judged
by PearsonÕs study with compound 5 [9], monosubstitu-
tion occurred selectively when phenolates carrying para-
methoxy and para-carbomethoxy groups were added at
low temperature. In our experience, ortho-or ortho/para
substituted phenolates 6 often gave rise to inseparable
Compounds of potential use as thyroid hormone ago-
nists [6] are appropriately substituted diaryl ethers (2;
X = O), diaryl sulfides (2; X = S) and diaryl selenides
(
compounds 1. One or several of the iodines in the 3,3
2; X = Se) which show only a slight resemblance to
0
0
and 5,5 -positions may be replaced by hydrogen, chlo-
rine, bromine, or alkyls. The phenolic hydroxyl may
be replaced by an amine and the amino acid side-chain
may be substituted by some other polar group (e.g., a
carboxylic acid alkyl chain).
Due to the many useful properties conferred on the
aromatic ligand by complexation, transition metal arene
complexes have been extensively employed for the func-
tionalization of aromatic molecules. Introduction of
hydroxide, alkoxide, phenoxide, carboxylate, amine,
thiolate and various carbanion nucleophiles into cat-
ionic cyclopentadienyliron complexes by nucleophilic
displacement has been recently reviewed [7]. We thought
this chemistry could be useful for the preparation of a
variety of thyroxine analogues. A retrosynthetic analysis
of compounds 2 (Scheme 1) leads to a transition
metal-complexed aromatic 3, functionalized in the 1,4-
positions with a good leaving group, and a properly
substituted phenolate/thiophenolate/selenophenolate 4
as suitable starting materials. The transition metal
would not only facilitate aromatic nucleophilic substitu-
tion to give a diaryl chalcogenide, but could also assist
in the transformation of the remaining leaving group
R3
Cl
Cl
+
Na
O
Z
FeCp PF6
R4
a R =Et; R =Z=H
5
6
3
4
b R =t-Bu; R =Z=H
3
4
c R =Br;R =Z=H
3
4
d R3=R4=H; Z=CH2CH(NHCOOBut)COOMe
e R =R =H; Z=NH
3
4
2
f R =R =Me; Z=H
3
4
g R =R =i-Pr; Z=H
3
4
h R =R =t-Bu; Z=H
3
4
i R =R =t-Bu; Z=CH CH COOMe
3
4
2
2
R3
O
R3
R3
Cl
Z + Z
O
O
Z
PF6 FeCp R4
a R =Et; R =Z=H
b R =t-Bu; R =Z=H
c R =Br;R =Z=H
R4 PF6 FeCp R4
7
8
a R3=Et; R4=Z=H
b R =t-Bu; R =Z=H
3
4
3
4
3
4
c R =Br;R =Z=H
3
4
3 4
d R =R =H; Z=CH CH(NHCOOBut)COOMe d R =R =H; Z=CH CH(NHCOOBut)COOMe
3
4
2
3
4
2
e R3=R4=H; Z=NH2
f R =R =Me; Z=H
e R3=R4=H; Z=NH2
f R =R =Me; Z=H
3
4
3
4
L into a suitably protected OH or NH group before it
2
g R =R =i-Pr; Z=H
g R =R =i-Pr ;Z=H
3 4
3
4
h R =R =t-Bu; Z=H
h R =R =t-Bu; Z=H
3 4
i R =R =t-Bu; Z=CH CH COOMe
3 4 2 2
is finally removed from the molecule. In the following, we
describe an organoiron approach to thyroid hormone
analogues as outlined above, using g -(1,4-dichloroben-
3
4
iR3=R =t-Bu; Z=CH CH COOMe
4
2
2
6
Scheme 2.

1786
J. Smirnova et al. / Journal of Organometallic Chemistry 690 (2005) 1784–1792
mixtures of unreacted compound 5, the desired
monosubstitution (7) and disubstitution (8) products
ature overnight) was only low (19%). A series of
2,6-difluoro-, dichloro-, dibromo- and diiodosubstituted
phenolates 6j–6m were also tried in the substitution
reaction of complex 5. Surprisingly, none of these com-
pounds reacted under the usual reaction conditions
(THF, ambient temperature). This was also true at more
forcing conditions (THF, 60 ꢁC; DMF, 80 ꢁC) when
complex 5 started to decompose. The phenolates derived
from N-acetyl- and N-benzoyl-3,5-diiodo-L-tyrosine
ethyl ester were similarly unreactive. The low reactivity
of all these phenolates is probably due to a combination
of electronic and steric substituent effects. With small,
electron withdrawing o-substituents the former is pre-
dominating, whereas with larger less deactivating ones,
the latter effect is more important. The more nucleo-
philic sodium 2,6-dichlorothiophenolate was quite reac-
tive towards complex 5. Thus, with equimolar amounts
of reactants at ꢀ20 ꢁC, an inseparable 7/4/1-mixture of
mono-(9) and disubstitution (10a) products and unre-
acted starting complex was formed. With an excess of
the thiophenolate, disubstitution product 10a was iso-
lated in 38% yield. The even more nucleophilic thio-
phenolate derived from 2,6-dimethylthiophenol,
showed a similar reactivity. In this case the disubstitu-
tion product 10b could be isolated in pure form (50%
isolated yield). Out of curiosity, we also tried sodium
3,5-dibromo-4-methylphenolate in the displacement
reaction with complex 5. It turned out that pure mono-
substitution product 11 could be isolated in 49% yield at
ambient temperature when an equimolar amount was
used and that the disubstitution product 12 was isolated
in a modest 18% yield when an excess of the reagent was
employed.
(
Scheme 2). Thus, when an equimolar amount of
sodium 2-ethylphenolate (6a) in dry THF was allowed
to react with a suspension of complex 5 in dry THF at
ꢀ
35 ꢁC for 3 h, a 2.5/1-mixture of compounds 7a and
was formed according to H NMR analysis. Raising
1
5
the temperature to ꢀ20 ꢁC caused an increase in the con-
version of starting material and a 37% yield of com-
pound 7a was isolated after chromatographic
purification on alumina with acetone elution. Stirring
of the reactants at 0 ꢁC for 5 h returned an inseparable
8
/1-mixture of compounds 7a and 8a. By allowing a
threefold excess of compound 6a to react overnight at
ambient temperature with complex 5, pure disubstitu-
tion product 8a was isolated in a modest 22% yield after
crystallization from ethyl ether. Sodium 2-tert-butyl-
phenoxide (6b) and 2-bromophenoxide (6c) behaved
similarly. Thus, at ꢀ20 ꢁC, an inseparable 1/8-mixture
of mono- and disubstitution products 7b and 8b was
formed in ca. 20% yield, whereas, at ambient tempera-
ture, pure compound 7c was isolated in 15% yield after
chromatographic purification. Again, using an excess of
the respective phenolates at ambient temperature, disub-
stituted products 8b (17%) and 8c (86%) could be iso-
lated. At ꢀ20 ꢁC, para-substituted phenolate 6d,
derived from N-(tert-butyloxycarbonyl)-L-tyrosine,
afforded a 1/1-mixture of mono-(7d) and disubstitution
(
8d) products. However, only compound 8d could be
isolated in pure form in a modest 13% yield. Sodium
4
6
-aminophenolate (6e) was less reactive than compound
d as a nucleophile. Crude product 7e obtained at ambi-
ent temperature was always contaminated by unreacted
complex 5. After filtration through alumina and crystal-
lization from ethyl ether, pure material was isolated in
1
7% yield. Several o,o-disubstituted phenolates were
R3
R4
Cl
Cl
also studied in the displacement reaction. Sodium 2,6-
dimethylphenolate (6f), 2,6-di-i-propylphenolate (6g),
and 2,6-di-tert- butylphenolate (6h) all afforded mono-
substitution products (compounds 7f, 7g, 7h) in fair iso-
lated yields (64%, 46% and 44%, respectively). However,
the optimal reaction temperature had to be increased in
the series (ꢀ20, 0 ꢁC and ambient temperature, respec-
tively). Obviously, this is to compensate for the decrease
in nucleophilicity of the phenolate due to steric hin-
drance at the ortho-positions. For the similar reason,
the sterically encumbered phenolate 6h in excess failed
to produce a disubstitution product with complex 5.
Phenolates 6f and 6g, though, produced complexed tria-
ryl diethers 8f and 8g in 46% and 54% yields, respec-
tively, under these conditions. Phenolate 6i, similarly
hindered as compound 6h but carrying an additional
Na
Z
O
Cl
S
PF6
FeCp
6
j R3=R4=F; Z=H
k R3=R4=Cl; Z=H
l R3=R4=Br; Z=H
m R3=R4=I; Z=Me
9
R
R
Br
Br
S
S
Cl
O
Me
Me
R
PF6 FeCp
10a R=Cl
R
PF6 FeCp
11
b R=Me
Br
Br
Br
R
S
R
2
-carbomethoxyethyl group in the 4-position, also failed
Me
O
O
Me
O
to give a disubstitution product. Still, the isolated yield
of monosubstitution product 7i under the best condi-
tions (equimolar amounts of reactants, ambient temper-
Br
PF6 FeCp
PF6 FeCp Me
1
2
13a R=Cl
b R=Me

J. Smirnova et al. / Journal of Organometallic Chemistry 690 (2005) 1784–1792
1787
We conclude from this part of the study that selec-
tivity for monosubstitution of complex 5 with pheno-
lates is more difficult to control than previously
thought. Phenolates where the nucleophilicity is low-
ered due to steric or electronic reasons fail to react
in the displacement reaction even at elevated tempera-
ture whereas highly nucleophilic phenolates tend to
undergo disubstitution even at low temperature. How-
ever, by the proper choice of stoichiometry and reac-
tion temperature, it is often possible to effect
selective mono- substitution with phenolates of inter-
mediate reactivity.
As a further step in the synthetic direction towards
thyroid hormone analogues (Scheme 1), we proposed
substitution of the remaining leaving group in the
arene iron(II) cyclopentadienyl complex. Chloroarene
complex 7f was used as a model compound for these
types of transformation. Substitution with sodium
uct of disubstitution were always contaminating the
desired monosubstitution product in the crude product
mixtures obtained with alcoholates derived from 2-
trimethylsilylethanol and 4-methoxybenzyl alcohol.
Only when complex 5 was added to a threefold excess
of sodium 4-methoxybenzylalcoholate in tetrahydrofu-
ran was it possible to isolate a pure product from the
reaction mixture. However, this was the undesired
disubstitution product 15 which was isolated in 48%
yield. Surprisingly, stirring of complex 5 at ambient
temperature with a twofold excess of ethylenediamine
in THF afforded a monosubstitution product 16 in
high yield (89%). Substitution of the remaining chlo-
ride using sodium 2,6-dimethylthiophenolate afforded
complex 17 in a 35% yield. However, attempts to
use less nucleophilic phenolates in this substitution
reaction were met with failure.
2
,6-dichlorothiophenolate and 2,6-dimethylthiopheno-
late occurred readily to afford compounds 13a and
3b, respectively, in 77% and 53% yields. However,
Me
H2N(CH2)2NH
Cl H2N(CH2)2NH
S
1
introduction of hydroxyl proved more difficult. Direct
substitution with hydroxide as described by Helling
and Hendrickson [11] was found to cause decomposi-
tion of complex 7f. Potassium 2-trimethylsilylethoxide
has been used in iron chemistry for converting com-
plexed halobenzenes to phenols via tetra-n-butylammo-
nium fluoride deprotection [12]. However, both the
sodium and potassium alcoholates were unreactive
towards complex 7. Salts of 4-methoxybenzyl alcohol-
ate [13] also turned out to react sluggishly with com-
plex 7 in THF or DMF and we were unable to free
the substitution product from unreacted starting mate-
rial by Al O -chromatography.
Amines turned out to be more reactive [14,15]. Thus,
stirring of complex 7f for 5 h in THF with a twofold
excess of ethylenediamine and potassium carbonate
afforded substitution product 14a in 94% yield. Chloro-
benzene complex 7h, when submitted to similar reaction
conditions, afforded complex 14b in 89% yield.
PF6 FeCp
PF6 FeCp Me
16
17
This part of the work can be summarized as fol-
lows: Whereas cationic cyclopentadienyl(1-chloro-4-
aryloxybenzene)iron(II) complexes 7 allow substitution
of the remaining chloride with good nucleophiles such
as thiophenolates and amines, they are generally not
reactive enough towards hydroxide or alcoholates that
are convertible to OH. A slightly modified approach
to thyroid hormone analogues, based on the reversal
of the order of substitution of dichlorobenzene com-
plex 5 with hydroxyl/amine and phenolate/thiopheno-
late did not turn out to be generally applicable.
Thus, cationic cyclopentadienyl(1-chloro-4-alkoxyben-
zene)iron(II) complexes where the alkoxy group was
convertible to OH were inaccessible and cyclopenta-
dienyl(N-alkyl-1-chloro-4-aminobenzene)iron(II) com-
plexes were only reactive towards good nucleophiles
such as thiophenolates.
2
3
R
H2N(CH2)2NH
PF6
O
4-MeOC6H4CH2O
OCH2C6H4-4-OMe
Despite some limitations in terms of product yields
and generality, we feel the synthetic methodology out-
lined in this paper is of interest for the preparation of
certain members of the structurally diverse family of
potentially useful thyroid hormone agonists 2.
R
PF6 FeCp
FeCp
1
4 a R=Me
15
b R=t-Bu
In view of the facile substitution of compound 7f
with thiophenolates, a slightly modified route to thy-
roid hormone analogues was also considered, involv-
ing monosubstitution of complex 5 with hydroxyl/
amine before substitution of the remaining chloride
with a phenolate/thiophenolate. However, despite vari-
ations in solvent, temperature and stoichiometry of the
reactants, dichlorobenzene complex 5 and/or the prod-
3. Experimental
1
13
H NMR (200 MHz) and C NMR (50.28 MHz)
spectra were recorded in acetone-d . Melting points are
6
uncorrected. Elemental analyses were performed by Ana-
lytical Laboratories, Lindlar, Germany. All reactions

1
788
J. Smirnova et al. / Journal of Organometallic Chemistry 690 (2005) 1784–1792
6
5
were performed under a nitrogen atmosphere. Ether and
tetrahydrofuran were freshly distilled under nitrogen
from Na/benzophenone. Except for the addition
of water, g -(1,4-dichlorobenzene)-g -cyclopentadienyl-
iron(II) hexafluorophosphate (1) was prepared according
to the literature [9]. Methyl 3-(2,6-tert-butyl-4-hydroxy-
phenyl)propionate was prepared according to a known
procedure [16,17].
3.3. g -[1-Chloro-4-(2-tert-butylphenoxy)benzene]-g -
cyclopentadienyliron(II) hexafluorophosphate (7b)
6
5
The substitution reaction was run at ꢀ20 ꢁC for 2.5 h.
This afforded a 8/1 mixture of mono- and disubstitution
products, corresponding to a 21% yield of the title com-
1
pound. H NMR d: 7.60–7.10 (4H, m), 6.95 (2H, d,
J = 9.0 Hz), 6.62 (2H, d, J = 9.0 Hz), 5.45 (5H, s), 1.28
(9H, s).
6
5
3
.1. Typical procedure for preparation of monosubstitution
6
3.4. g -[1-Chloro-4-(2-bromophenoxy)benzene]-g -
cyclopentadienyliron(II) hexafluorophosphate (7c)
products 7 and 11. g -[1-chloro-4-(2,6-dimethylphenoxy)-
benzene]-g -cyclopentadienyliron(II) hexafluorophosphate
5
(
7f)
The substitution reaction was carried out at room
temperature for 16 h. The crude product was dissolved
in the minimum amount of acetone and passed through
a short column of Al O . Removal of the solvent affor-
2,6-Dimethylphenol (0.21 mmol, 25.7 mg) was stir-
red with sodium hydride (0.4 mmol, 16 mg of 60% dis-
persion in mineral oil) in THF (2 mL) for 15–20 min.
This solution was then added dropwise to a stirred
suspension of the complex 5 (0.2 mmol, 82.6 mg) in
THF (4 mL) cooled to ꢀ78 ꢁC. The mixture was
warmed to ꢀ20 ꢁC and stirring continued. The reac-
tion was then quenched by addition of a few drops
of water and the mixture was passed through a pad
of Celite, which was subsequently washed with
CH Cl . The solvent was removed in vacuo and the
2
3
ded the pure product. Yield: 15%, m.p. 199–202 ꢁC (d).
1
2
3
H NMR d: 7.85 (1H, dd, J = 8.0 Hz, J = 2.0 Hz),
7.63–7.30 (3H, m), 6.88 (2H, d, J = 7.0 Hz), 6.56 (2H,
1
3
d, J = 7.0 Hz), 5.43 (5H, s). C NMR d: 149.3, 133.5,
131.2, 129.2, 127.5, 121.6, 113.6, 103.4, 86.0, 78.9,
75.6. Anal. Calc. for C H BrClF FeOP: C, 37.16; H,
1
7
13
6
2.38. Found: C, 37.00; H, 2.51.
6
5
2
2
3.5. g -[1-Chloro-4-(4-aminophenoxy)benzene]-g -
cyclopentadienyliron(II) hexafluorophosphate (7e)
residue redissolved in CH Cl (25 mL). This solution
2
2
was washed with NH PF (1.5 mL of a saturated solu-
4
6
tion further diluted with 3.5 mL of H O), then 0.05 M
2
The substitution reaction was run at room tempera-
ture for 16 h and after quenching the mixture was fil-
tered through neutral Al O . Yield: 17%, m.p. 178–179
NaOH (2 · 5 mL) and finally with water (10 mL). The
organic layer was dried over MgSO and the solvent
was removed in vacuo. The oily residue was triturated
with dry ether to give the title compound, 64 mg
4
2
3
1
ꢁC (d). H NMR d: 7.05 (2H, d, J = 9.0 Hz), 6.85–6.75
13
(4H, m), 6.39 (2H, d, J = 9.0 Hz), 5.56 (5H, s).
C
1
(
s), 6.87 (2H, d, J = 7.0 Hz), 6.45 (2H, d, J = 7.0 Hz),
64%), m.p. 185–186 ꢁC (d). H NMR d: 7.21 (3H,
NMR d: 146.6, 142.1, 133.6,120.5, 114.4, 102.6, 85.9,
78.4, 74.2. Anal. Calc. for C H ClF FeNOP: C,
42.05; H, 3.11. Found: C, 41.79; H, 3.30.
1
7
15
6
5.41 (5H, s), 2.98 (6H, s). For some reason, the aro-
matic protons in the uncomplexed ring of this and
many other (7g, 7h, 8f, 8g, 13a, 13b, 14a, 14b) com-
6
3.6. g -[1-Chloro-4-(2,6-di-iso-propylphenoxy)benzene]-
1
3
5
plexes appear as singlets. C NMR d: 149.5, 133.3,
28.9, 135.8, 130.0, 86.0, 78.6, 74.6, 15.0. Anal. Calc.
for C H ClF FeOP: C, 45.44; H, 3.64. Found: C,
g -cyclopentadienyliron(II) hexafluorophosphate (7g)
1
1
9
18
6
The substitution reaction was run at 0 ꢁC for 3 h.
1
4
5.97; H,3.86.
The following compounds were similarly prepared.
Yield: 46%, m.p. 188–189 ꢁC (d). H NMR d: 7.36
(3H, s), 6.90 (2H, d, J = 7.0 Hz), 6.49 (2H, d, J = 7.0
Hz), 5.40 (5H, s), 3.01 (2H, m, J = 7.0 Hz), 1.18 (12H,
1
3
d, J = 7.0 Hz). C NMR d: 147.2, 139.5, 134.3, 126.8,
124.3, 103.1, 86.1, 78.6, 74.5, 26.2, 21.6. Anal. Calc.
for C H ClF FeOP: C, 49.80; H, 4.72. Found: C,
6
5
3
cyclopentadienyliron(II) hexafluorophosphate (7a)
.2. g -[1-Chloro-4-(2-ethylphenoxy)benzene]-g -
2
3
26
6
4
9.61, H, 4.72.
The substitution reaction was run at ꢀ15 to ꢀ20 ꢁC
for 3 h. The crude product was dissolved in the mini-
mum amount of acetone and passed through a short col-
umn of Al O . Evaporation of the solvent gave pure
6
3.7. g -[1-Chloro-4-(2,6-di-tert-butylphenoxy)benzene]-
5
g -cyclopentadienyliron(II) hexafluorophosphate (7h)
2
3
1
product. Yield: 37%, m.p. 141–142 ꢁC (d). H NMR:
.56–7.25 (4H, m), 6.85 (2H, d, J = 9.0 Hz), 6.50 (2H,
d, J = 9.0 Hz), 5.41 (5H, s), 2.57 (2H, q, J = 7.6 Hz),
.09 (3H, t, J = 7.6 Hz).
The substitution reaction was run at room tempera-
ture for 16 h using a 1:1.5-ratio of 5 and 2,6-di-tert-
butylphenol (4). Yield: 44%, m.p. 246–247 ꢁC (d). H
NMR d: 7.72 (2H, s), 7.07 (2H, d, J = 9.0Hz), 6.96
7
1
1

J. Smirnova et al. / Journal of Organometallic Chemistry 690 (2005) 1784–1792
1789
(
2H, d, J = 9.0 Hz), 6.72 (1H, s), 5.21 (5H, s), 1.52 (18H,
1
15.1. Anal. Calc. for C H F FeO P: C, 55.50; H,
27 27 6 2
3
s). C NMR d: 155.5, 137.4, 123.8, 123.8, 105.0, 104.5,
7.4, 84.5, 78.8, 33.5. Anal. Calc. for C H ClF FeOP:
C, 51.53; H, 5.19. Found: C, 51.51; H, 5.33.
4.66. Found: C, 55.31; H, 4.65.
The following compounds were similarly prepared.
8
2
5
30
6
6
5
3
.11. g -[1,4-Bis(2-ethylphenoxy)benzene]-g -
6
3
.8. g -[1-Chloro-4-[2,6-di-tert-butyl-4-(2-carbometh-
cyclopentadienyliron(II) hexafluorophosphate (8a)
5
oxyethyl)phenoxy]benzene]-g -cyclopentadienyliron(II)
hexafluorophosphate (7i)
The reaction was quenched with water. Yield: 22%,
1
m.p. 203–204 ꢁC (d). H NMR: 7.52–7.20 (8H, m),
6.37 (4H, s), 5.37 (5H, s), 2.61 (4H, q, J = 7.6 Hz),
The substitution reaction was run at room temper-
1
13
ature for 16 h. Yield: 18%, m.p. 87–88 ꢁC (d).
NMR d: 6.96 (2H, s), 6.89 (2H, d, J = 7.0 Hz), 6.73
H
1.19 (6H, t, J = 7.6 Hz). C NMR d: 150.8, 134.5,
130.1, 129.5, 127.0, 125.6, 119.1, 77.1, 74.1, 21.5, 12.7.
(
(
(
2H, d, J = 7.0 Hz), 5.37 (5H, s), 3.62 (3H, s), 2.61
2H, t, J = 7.6 Hz), 2.25 (2H, d, J = 7.6 Hz), 1.31
18H, s). C NMR d: 184.4, 171.7, 148.5, 140.0,
Anal. Calc. for C H F FeO P: C, 55.50; H, 4.66.
6
2
7
27
2
Found: C, 55.29; H, 4.68.
1
3
1
3
14.8, 109.6, 106.0, 87.2, 84.5, 78.8, 78.2, 50.1, 43.9,
4.2, 32.5.
6
5
3
.12. g -[1,4-Bis(2-tert-butylphenoxy)benzene]-g -
cyclopentadienyliron(II) hexafluorophosphate (8b)
6
3
.9. g -[1-Chloro-4-(3,5-dibromo-4-methylphenoxy)-
1
Yield: 17%, m.p. 219–221 ꢁC (d). H NMR d: 7.51
2H, dd, J = 7.0 Hz, J = 2 Hz), 7.26 (4H, dt, J = 7.0
5
benzene]-g -cyclopentadienyliron(II)
hexafluorophosphate (11)
2
3
2
(
Hz, J = 2 Hz), 7.06 (2H, dd, J = 7.0 Hz, J = 2 Hz),
.60 (4H, s), 5.46 (5H, s), 1.42 (18H, s). C NMR d:
53.5, 139.1, 129.3, 127.1, 126.8, 124.3, 117.9, 77.4,
6.8, 33.4. Anal. Calc. for C H F FeO P: C, 58.14;
3
2
3
1
3
6
1
7
The substitution reaction was run at 0 ꢁC for 3 h.
1
Yield: 49%, m.p. 189–191 ꢁC (d). H NMR d: 7.68
3
1
35
6
2
(
2H, s), 7.90 (2H, d, J = 6.7 Hz), 6.72 (2H, d, J = 6.7
H, 5.51. Found: C, 57.92; H, 5.46.
1
3
Hz), 5.44 (5H, s), 2.58 (3H, s). C NMR d: 153.2,
34.4, 130.8, 124.3, 123.0, 103.6, 86.1, 78.9, 76.7, 21.3.
Anal. Calc. for C H Br ClF FeOP: C, 33.66; H,
1
6
5
3
.13. g -[1,4-Bis(2-bromophenoxy)benzene]-g -
cyclopentadienyliron(II) hexafluorophosphate (8c)
1
8
14
2
6
2
.20. Found: C, 33.40; H, 2.31.
The reaction was quenched with water. Yield: 86%,
1
3
.10. Typical procedure for preparation of disubstitution
6
m.p. 193–194 ꢁC (d). H NMR d: 7.86 (2H, dd,
J = 8.0 Hz, J = 2 Hz), 7.61–7.30 (6H, m), 6.48 (4H,
products 8, 10, 12. g -[1,4-Bis(2,6-dimethyl-
phenoxy)benzene]-g -cyclopentadienyliron(II)
hexafluorophosphate (8f)
2
3
5
1
3
s), 5.43 (5H, s). C NMR d: 149.8, 133.5, 129.4,
29.1, 127.2, 121.3, 113.5, 77.7, 74.3. Anal. Calc. for
C H Br F Fe- O P: C, 40.27; H, 2.50. Found: C,
1
2
3
17
2
6
2
2
,6-Dimethylphenol (0.6 mmol, 73.3 mg) was stirred
3
9.97; H, 2.64.
with sodium hydride (1.2 mmol, 48 mg of 60% disper-
sion in mineral oil) in THF (2 mL) for 15–20 min.
The resulting solution of the phenolate was then
added to a stirred suspension of the complex 5 (0.2
mmol, 82.6 mg) at room temperature. The reaction
was stirred for 12–16 h and then quenched by addi-
tion of a few drops of 10% aqueous HCl to dissolve
precipitate formed. The resulting mixture was filtered
through Celite and the pad washed with CH Cl .
6
3.14. g -[1,4-Bis(4-[N-(tert-butoxycarbonyl)-L-2-
5
amino-2-carbomethoxyethyl]phenoxy)benzene]-g -
cyclopentadienyliron(II) hexafluorophosphate (8d)
The typical procedure for preparation of monosubsti-
tution products was followed, but the substitution reac-
tion was run at ꢀ20 ꢁC for 5 h.
2
2
The solvent was removed in vacuo and the residue
redissolved in CH Cl (25 mL). This solution was then
washed with a dilute solution of NH PF (1.5 mL of a
The title compound was isolated by passing the
solution of the 1/1-mixture of mono- and disubstitu-
tion products in the minimum amount of acetone
through a short column of Al O . Yield: 17 mg
2
2
4
6
saturated solution further diluted with 3.5 mL of
H O), 0.05 M NaOH (2 · 5 mL) and water (10 mL).
2
3
1
(13%), m.p. 163–165 ꢁC (d). H NMR d: 7.46 (4H,
d, J = 8.2 Hz), 7.28 (4H, d, J = 8.2 Hz), 6.36 (4H,
s), 6.27 (2H, broad d, J = 6.7 Hz), 5.36 (5H, s),
2
Drying over MgSO and removal of the solvent in
4
vacuo afforded a semi-solid residue, which was tritu-
rated with dry ether to give the title compound, 63
4.42 (2H, broad m), 3.71 (6H, s), 3.26–2.95 (4H, m)
13
1
mg (54%), m.p. 253–254 ꢁC (d). H NMR d: 7.18
1.35 (18H, s).
135.0, 130.5, 130.0, 119.3, 77.6, 77.1, 74.1, 54.1,
C NMR d: 171.1, 154.3, 151.6,
1
3
(
6H, s), 6.31 (4H, s), 5.37 (5H, s), 2.19 (12H, s).
C
NMR d: 150.3, 131.2, 128.8, 126.1, 125.6, 77.1, 73.2,
50.4, 35.8, 26.6. Anal. Calc. for C H F FeN O P:
2
4
1
49
6
10

1790
J. Smirnova et al. / Journal of Organometallic Chemistry 690 (2005) 1784–1792
C, 52.92; H, 5.31; N, 3.01. Found: C, 52.79; H, 5.31;
N,3.01.
tion was added via syringe to a solution of complex 7f
(0.16 mmol, 84.7 mg) in THF (4 mL). The reaction
was stirred for 3 h, quenched by addition of a few drops
of water and the mixture filtered through a Celite pad.
The pad was washed well with CH Cl and the solvent
6
5
3
.15. g -[1,4-Bis(2,6-di-iso-propylphenoxy)benzene]-g -
cyclopentadienyliron(II) hexafluorophosphate (8g)
2
2
removed in vacuo. The residue was redissolved in
CH Cl (25 mL) and the resulting solution was washed
with 1/4 saturated aqueous NH PF (5 mL), 0.05 M
The reaction was quenched with water. The crude
product was dissolved in the minimum amount of ace-
tone and passed through a short column of Al O .
2
2
4
6
NaOH (2 · 5 mL) and H O (10 mL). The organic layer
2
3
2
Evaporation of the solvent gave the pure product. Yield:
1
was separated, dried over MgSO and the solvent was
removed in vacuo. The semi-solid was triturated with
ether to give the title compound, 79 mg (77%), m.p.
4
4
6%, m.p. 106–109 ꢁC (d). H NMR d: 7.33 (6H, s), 6.35
4H, s), 5.34 (5H, s), 3.05 (4H, m, J = 7.0 Hz), 1.18
(
(
1
3
1
12H, d, J = 7.0 Hz). C NMR d: 147.8, 139.5, 132.4,
190–192 ꢁC. H NMR d: 7.78–7.62 (3H, m), 7.19 (3H,
1
3
1
26.6, 124.2, 77.0, 73.2, 26.2, 21.7.
s), 6.31–6.46 (4H, m), 5.35 (5H, s), 2.07 (6H, s).
C
NMR d: 149.8, 140.0, 132.8, 132.5, 128.9, 128.8, 126.5,
125.7, 103.4, 82.7, 78.3, 74.3, 15.0. Anal. Calc. for
C H Cl F FePOS: C, 46.83; H, 3.30; S, 5.00. Found:
6
5
3
.16. g -[1,4-Bis(2,6-dimethylphenylthio)benzene]-g -
cyclopentadienyliron(II) hexafluorophosphate (10a)
2
5
21
2 6
C, 47.01; H, 3.41; S, 4.84.
The crude product was dissolved in the minimum
amount of acetone and passed through a short column
of neutral Al O . Evaporation of the solvent and tritur-
6
3.20. g -[1-(2,6-Dimethylphenylthio)-4-(2,6-
5
dimethylphenoxy)benzene]-g -cyclopentdienyliron(II)
hexafluorophosphate (13b)
2
3
ation of the residue with ether gave the title compound
1
in 38% yield, m.p. 188–189 ꢁC (d). H NMR d: 7.27–
7
.40 (6H, m), 6.26 (4H, s), 5.24 (5H, s), 2.43 (12H, s).
C NMR d: 142.3, 130.0, 128.2, 126.6, 107.8, 82.9,
8.8, 20.1.
Except that 2,6-dimethylbenzenethiol was employed,
the procedure for compound 13a was followed. The
crude product was dissolved in the minimum amount
of acetone and passed through a short column of neutral
Al O . Removal of the solvent in vacuo afforded the
pure title compound in 53% yield, m.p. 204–205 ꢁC
d). H NMR d: 7.40–7.31 (3H, m), 7.18 (3H, s), 6.35–
1
3
7
6
5
3
cyclopentadienyliron(II) hexafluorophosphate (10b)
.17. g -[1,4-Bis(2,6-dichlorophenylthio)benzene]-g -
2
3
1
(
The reaction was quenched with water. The crude
product was purified as described above. Yield: 50%,
6.26 (4H, m), 5.32 (5H, s), 2.46 (6H, s), 2.16 (6H, s).
1
3
C NMR d: 149.8, 142.2, 132.4, 129.9, 128.8 (2 signals),
128.3, 127.1, 125.6, 106.3, 82.3, 78.0, 74.2, 20.1, 15.0.
1
m.p. 189–191 ꢁC (d). H NMR d: 7.78–7.63 (6H, m),
1
3
6
.43 (4H, s), 5.32 (5H, s). C NMR d: 140.2, 132.7,
29.0, 125.7, 105.6, 83.1, 79.5. Anal. Calc. for
Anal. Calc. for C H F FePOS: C, 54.01; H, 4.53; S,
7
2
27 6
1
C H Cl F FePS : C, 39.57; H, 2.17; S, 9.19. Found:
C, 39.37; H, 2.37; S, 8.99.
5.34. Found: C, 53.83; H, 4.66; S, 5.15.
2
3
15
4
6
2
6
3.21. g -[1-(2,6-Dimethylphenoxy)-4-(N-(2-aminoethyl)-
5
mino)benzene]-g -cyclopentadienyliron(II) hexafluoro-
phosphate (14a)
6
3
.18. g -[1,4-Bis(3,5-dibromo-4-methylphenoxy)-
5
benzene]-g -cyclopentadienyliron(II) hexafluoro-
phosphate (12)
Except for the use of complex 7f instead of 5 and a
reaction time of 5 h, the procedure was as described
for compound 16. Yield: 94%, m.p. 159–162 ꢁC (d).
The reaction was quenched with water. Yield: 18%,
1
1
m.p. 229–231 ꢁC. NMR H d: 7.06 (4H, s), 6.64 (4H,
NMR H d: 7.18 (3H, several peaks), 6.01 (2H, d,
1
3
s), 5.43 (5 H. s), 2.56 (6H, s). NMR C, d: 151.9,
J = 7.2 Hz), 5.88 (2H, d, J = 7.2 Hz), 5.09 (5H, s),
3.54–3.41 (4H, m), 2.21 (6H, s).
1
for C H Br F FeO P: C, 34.44; H, 2.20. Found: C,
33.9, 129.0, 124.2, 122.6, 77.7, 75.8, 21.3. Anal. Calc.
2
5
19
4
6
2
6
3
4.64; H, 2.32.
3.22. g -[1-(2,6-Di-tert-butylphenoxy)-4-(N-(2-
5
aminoethyl)-amino)benzene]-g -cyclopentadienyliron-
(II) hexafluorophosphate (14b)
6
3
.19. g -[1-(2,6-Dichlorophenylthio)-4-(2,6-
5
dimethylphenoxy)benzene]-g -cyclopentadienyliron(II)
hexafluorophosphate (13a)
Except for the use of complex 7h instead of 7f and a
reaction time of 6.5 h, the procedure was as described
for compound 16. Yield: 89%, m.p. 188–190 ꢁC (d).
2
,6-Dichlorobenzenethiol (0.32 mmol, 57.3 mg) was
1
stirred with NaH (0.6 mmol, 24 mg of 60% dispersion
in mineral oil) in THF (2 mL) for 15–20 min. This solu-
NMR H d: 7.65 (2H, s), 6.58 (2H, d J = 7.0 Hz), 6.54
(1H, s), 6.06 (2H, d, J = 7.0 Hz), 4.90 (5H, s), 3.67–3.51

J. Smirnova et al. / Journal of Organometallic Chemistry 690 (2005) 1784–1792
1791
1
3
(
4H, m), 1.52 (18H, s). C NMR d: 154.3, 137.1, 125.4,
a Celite pad which was washed well with CH Cl . The
2 2
1
24.7, 123.4, 97.5, 82.6, 75.3, 66.5, 48.7, 42.9, 33.5.
combined filtrate was diluted with CH Cl to 25 mL
2 2
and washed with 1/4 saturated aqueous NH PF (5
4
6
6
5
3
.23. g -[1,4-Bis(4-methoxybenzyloxy)benzene]-g -
cyclopentadienyliron(II) hexafluorophosphate (15)
mL), 0.05 M NaOH (2 · 5 mL) and H O (10 mL), dried
2
over MgSO and the solvents were removed in vacuo.
The brownish oily residue was triturated with dry ether
and pentane to give the title compound as a yellow-or-
4
p-Methoxybenzyl alcohol (0.6 mmol, 82.9 mg) was
stirred with sodium hydride (1.2 mmol, 48 mg of a
1
ange powder, 38 mg (35%), m.p. 87–89 ꢁC. NMR H
6
2
0% dispersion in mineral oil) in THF (2 mL) for 15–
0 min. The complex 5 (0.2 mmol, 82.6 mg) was added
d: 7.37–7.26 (3H, m), 6.04–5.90 (4H, m), 5.04 (5H, s),
3.55 (4H, broad s), 2.48 (6H, s).
as a solid and the mixture allowed to stir for 16 h. The
reaction was quenched by addition of a few drops of
water and the mixture filtered through a Celite which
was washed well with CH Cl . The solvent was removed
Acknowledgements
2
2
in vacuo and the residue redissolved in CH Cl (25 mL).
2
2
We thank Carl-Gustaf Saluste for performing initial
experiments and J. Peter Storm for many helpful sugges-
tions. Financial support by the Royal Swedish Academy
of Sciences, the Swedish Research Council and Karobio
AB is gratefully acknowledged.
This solution was washed with 1/4 saturated aqueous
NH PF (5 mL), 0.05 M NaOH (2 · 5 mL), H O (10
4
6
2
mL) and dried over MgSO . Removal of the solvent in
4
vacuo gave a crystalline product, which was washed sev-
eral times with ether and dried in vacuo. Yield: 33 mg
1
(
48%), m.p. 168–169 ꢁC (d). H NMR d: 7.49 (4H, d,
J = 8.8 Hz), 7.01 (4H, d, J = 8.8 Hz), 6.37 (4H, s), 5.29
1
3
References
(
1
4H, s), 5.10 (5H, s), 3.83 (6H, s). C NMR d: 159.3,
29.9, 129.2, 125.9, 113.0, 75.8, 71.3, 70.4, 53.8. Anal.
Calc for C H F FeO P: C, 52.62; H, 4.42. Found: C,
[
[
[
1] R.A. Sunde, in: B. OÕDell, R.A. Sunde (Eds.), Handbook of
Nutritionally Essential Mineral Elements, Marcel Dekker, New
York, 1997 (Chapter 18);
2
7
27
6
4
5
2.40, H, 4.40.
D. Behne, A. Kyriakopoulos, Ann. Rev. Nutr. 21 (2001) 453;
M. Birringer, S. Pilawa, L. Floh e´ , Nat. Prod. Rep. 19 (2002)
6
3
.24. g -[1-Chloro-4-(N-(2-aminoethyl)-amino)-
6
93.
5
benzene]-g -cyclopentadienyliron(II)
hexafluorophosphate (16)
2] J.R. Arthur, G.J. Beckett, in: R.F. Burk (Ed.), Selenium in
Biology and Human Health, Springer-Verlag, New York, 1994, p.
9
3;
J.L. Leonard, J. K o¨ hrle, in: L.E. Braverman, R.D. Utiger (Eds.),
The Thyroid, Lippincott-Williams & Wilkins, Philadelphia, 2000,
pp. 136–173.
Complex 5 (0.2 mmol, 82.6 mg) and ethylenediamine
0.4 mmol, 24.0 mg) were stirred with K CO (0.5 mmol,
2 3
(
6
9.1 mg) in dry THF (4 mL) for 14 h at room tempera-
3] P.D. Leeson, A.H. Underwood, in: J.C. Emmet (Ed.), Compre-
hensive Medicinal Chemistry, Pergamon Press, Oxford, 1991, pp.
1145–1173;
ture. The inorganic precipitate was filtered off and
washed with CH Cl . The solvents were removed
2
2
N. Yokoyama, G.N. Walker, A.J. Main, J.L. Stanton, M.M.
Morrissey, C. Boehm, A. Engle, A.D. Neubert, J.M. Was-
vary, Z.F. Stephan, R.E. Steele, J. Med. Chem. 38 (1995)
in vacuo and the oily residue was triturated with dry
ether to yield the title compound as an orange precipi-
tate, 78 mg (89%), m.p. 108–109 ꢁC. NMR H d: 6.52
1
6
4] J. Malm, Exp. Opin. Ther. Pat. 14 (2004) 1169.
95.
(
J = 7.0 Hz), 5.12 (5H, s), 3.64–3.44 (4H, m). NMR
2H, d, J = 7.0 Hz), 6.27 (1H, broad s), 6.00 (2H, d,
1
[
3
C
[5] L. Ye, Y.-L. Li, K. Mellstr o¨ m, C. Mellin, L.-G. Bladh, K.
Koehler, N. Garg, A.M.G. Collazo, C. Litten, B. Husman, K.
Persson, J. Ljunggren, G. Grover, P.G. Sleph, R. George, J.
Malm, J. Med. Chem. 46 (2003) 1580;
d: 125.3, 99.8, 85.0, 76.8, 65.6, 48.6, 43.0. Anal. Calc.
for C H ClF FeN P: C, 35.77; H, 3.69; N, 6.42.
Found: C, 35.95; H, 3.78; N, 6.29.
1
3
16
6
2
G.J. Grover, K. Mellstr o¨ m, L. Ye, J. Malm, Y.-L. Li, L.-G.
Bladh, P.G. Sleph, M.A. Smith, R. George, B. Vennstr o¨ m, K.
Mookthiar, R. Horvath, J. Speelman, D. Egan, J.D. Baxter,
Proc. Natl. Acad. Sci. USA 100 (2003) 10067.
6
3
.25. g -[1-(2,6-Dimethylphenylthio)-4-(N-(2-
5
aminoethyl)-amino)benzene]-g -cyclopentadienyliron-
II) hexafluorophosphate (17)
[
6] D.E. Ryono, J.G. Grover, K. Mellstr o¨ m, in: BurgerÕs Medicinal
(
Chemistry and Drug Discovery, sixth ed., Wiley, New York, 2003
(
Chapter 12).
2
,6-Dimethylthiophenol (0.4 mmol, 55.3 mg) was stir-
[
7] For accounts of the use of organoiron compounds in synthesis see:
M.F. Semmelhack, in: B.M. Trost, I. Fleming, M.F. Semmelhack
red with sodium hydride (0.8 mmol, 32 mg of 60% dis-
persion in mineral oil) in THF (3 mL) for 15–20 min.
A solution of complex 16 (0.2 mmol, 87.3 mg) in THF
(Eds.), Comprehensive Organic Synthesis, vol. 4, Pergamon Press,
Oxford, 1991 (Chapter 9.1);
J.A. Pearson, Iron Compounds in Organic Synthesis, Academic
Press, London, 1994;
(
4 mL) was added via syringe and stirring continued
for 3 h. The reaction was quenched by addition of a
few drops of water and the mixture was filtered through
M.J. Morris, in: J.A. Labinger, M.J. Winter, E.W. Abel, F.G.A.
Stone, G. Wilkinson (Eds.), Comprehensive Organometallic

1792
J. Smirnova et al. / Journal of Organometallic Chemistry 690 (2005) 1784–1792
Chemistry II, vol. 5, Pergamon Press, Oxford (Chapter 8);
M.S. Holden, S.R. Petrich, Tetrahedron Lett. 40 (1999)
285;
[11] J.F. Helling, W.A. Hendrickson, J. Organomet. Chem. 168 (1979)
87.
4
[12] M.S. Holden, A.D. Brosius, M.A. Hilfiker, E.J. Humbert,
Tetrahedron Lett. 41 (2000) 6275.
D. Astruc (Ed.), Modern Arene Chemistry: Concepts,
Synthesis and Applications, Wiley-VCH, Weinheim, 2002
[13] J.P. Storm, Arene iron complexes in organic synthesis, Disserta-
tion, Lund University, 2000, p. 40.
(Chapters 11 and 12).
[
[
8] A.N. Nesmeyanov, N.A. VolÕkenau, I.N. Bolesova, Tetrahedron
Lett. (1963) 1725.
9] A.J. Pearson, A.M. Gelormini, J. Org. Chem. 59 (1994)
[14] C.C. Lee, A.S. Abd-El-Aziz, R.L. Chowdhury, U.S. Gill, A.
Pi o´ rko, J. Organomet. Chem. 315 (1986) 79.
[15] F. Moulines, M. Kalam-Alami, V. Martinez, D. Astruc, J.
Organomet. Chem. 643–644 (2002) 125.
4
561.
10] A.S. Abd-El-Aziz, C.R. de Denus, J. Chem. Soc., Perkin Trans. 1
1993) 293.
[
[16] L.-C. Yu, Himont Inc., Eur. Pat. Appl., 467063A2.
[17] E.J. Evain, K. Raman, Himont Inc., US Pat. 5,264,612.
(