Sulfur-bridged phenoxide and naphthyloxide-based ligands for lanthanide chemistry and catalysis

Article abstract of DOI:10.1016/S0022-4596(02)00197-4

The development of some new lanthanide chemistry of aryloxide-based ligands is presented. The use of chelating, dianionic aryloxide ligand sets, such as sterically encumbered binolates, to allow a degree of geometrical control over the reaction chemistry of these large metal cations, is reviewed. We show how the development of potentially tridentate, dianionic, sulfur-bridged biphenolate and binaphtholate [OSO] ligands has allowed us to make new Ln(III) aryloxide complexes such as [Sm{1,1′-S(2-OC₆H₂Bu^t-3,-Me-5) ₂}(OC₆H₃Bu₂^t-2,6) (THF)] and [Sm{1,1′-S(2-OC₁₀H₄Bu₂ ^t-3,6)₂}(OC₆H₃Bu₂ ^t-2,6)(THF)]. Unusually, both symmetric and asymmetric derivatives of the [OSO] ligands may be prepared; reasons for this observation are suggested. Reactivity studies of these Sm(III) derivatives have shown them to be selective Lewis acid catalysts for the one-step monoacylation of 1,2-diols. Oxidation products of the sulfur-bridged binaphtholate ligand have been crystallographically characterized.

Full text of DOI:10.1016/S0022-4596(02)00197-4

Journal of Solid State Chemistry 171 (2003) 90–100
Sulfur-bridged phenoxide and naphthyloxide-based ligands
for lanthanide chemistry and catalysis
Louise S. Natrajan, Jonathan J. Hall, Alexander J. Blake, Claire Wilson,
and Polly L. Arnoldꢀ
School of Chemistry, University of Nottingham, University Park, Nottingham NG7 2RD, UK
Received 23 May 2002; received in revised form 15 August 2002; accepted 13 November 2002
Abstract
The development of some new lanthanide chemistry of aryloxide-based ligands is presented. The use of chelating, dianionic
aryloxide ligand sets, such as sterically encumbered binolates, to allow a degree of geometrical control over the reaction chemistry of
these large metal cations, is reviewed. We show how the development of potentially tridentate, dianionic, sulfur-bridged biphenolate
and binaphtholate [OSO] ligands has allowed us to make new Ln(III) aryloxide complexes such as [Sm{1,10-S(2-OC6H2But-
3,-Me-5)2}(OC6H3But2-2,6)(THF)] and [Sm{1,10-S(2-OC10H4Bu2t -3,6)2}(OC6H3Bu2t -2,6)(THF)]. Unusually, both symmetric and
asymmetric derivatives of the [OSO] ligands may be prepared; reasons for this observation are suggested. Reactivity studies
of these Sm(III) derivatives have shown them to be selective Lewis acid catalysts for the one-step monoacylation of 1,2-diols.
Oxidation products of the sulfur-bridged binaphtholate ligand have been crystallographically characterized.
r 2003 Elsevier Science (USA). All rights reserved.
Keywords: Samarium; Lanthanide; Aryloxide; Naphthyloxide; Sulfur hemilabile biphenolate catalysis; Desymmetrization; Acylation; Phenoxide
1. Introduction
reported since are Ln(OC6H3–But2)3 (Ln=Y and Ce) [4]
and Ln(OC6H3–R2)3 (R=Me, Ph, Pri, Ln=Ce, Sc, Sm,
and Yb) [5]. Unless specified otherwise, the arene
substituents in the complexes described in this paper
are in the 2, 6 or 2, 6, and 4 positions.
Following the synthesis of the first homoleptic
lanthanide aryloxide complexes 20 years ago, one of
the major themes in this area has been the development
of aryloxide-based ancillary ligands to support reactive
f-element centers. In recent decades, phenolates have
taken on a variety of roles as ligands for the stabilization
of f-element coordination complexes [1]. The Ln–OAr
bond is thermodynamically strong but unlike cyclopen-
tadienyl anions, aryloxide ligands do not impart a high
level of kinetic stability on a complexunless large ortho-
substituents such as tert-butyl groups are incorporated
[2]. The first isolated F-element aryloxide derivatives
were homoleptic, trivalent tris(aryloxide) complexes,
derived from phenolates incorporating bulky ortho-
substituents; the complexes Ln(OC6H2–But2Me)3
(Ln=Sc, Y 1, Fig. 1, La, Pr, Nd, Dy, Ho, Er, and Yb)
[3] are all hydrocarbon-soluble, air-sensitive, high-
melting, crystalline solids. Other homoleptic complexes
The complexes may be made via reactions such as the
direct treatment of activated metal (often with trace
Hg2+ added) with a phenol in liquid ammonia,
benzophenone or iso-propanol, group transfer from
pentafluorophenyl-lanthanide adducts formed in situ,
protonolysis of a coordinated amide or alkyl, or
metathesis of a trihalide with a Li, Na or K aryloxide
[6]. The last method has also been used to make mixed
halide–aryloxide complexes [7]. The solvent-free homo-
leptic adducts are often those most readily isolated due
to the use of sublimation as a purification technique.
Mononuclear complexes of aryloxides with smaller
ortho-arene substituents must be isolated as Lewis base
adducts [8], and in the absence of coordinated solvent
form O-bridged dimers and clusters [9]. Two interesting
structural forms have been identified which invoke
metal–arene interactions for additional electrostatic
stabilization of the electropositive metal center;
ꢀCorresponding author. Fax:+44-115-951-3563.
E-mail address: polly.arnold@nottingham.ac.uk (P.L. Arnold).
0022-4596/03/$ - see front matter r 2003 Elsevier Science (USA). All rights reserved.
doi:10.1016/S0022-4596(02)00197-4

L.S. Natrajan et al. / Journal of Solid State Chemistry 171 (2003) 90–100
91
Notably, the replacement of one Cpꢀ ligand in
SmCpꢀ2(THF)2 with a monodentate OArÀ ligand gives
a unique catalytic system that can polymerize styrene
and ethene separately, or into block styrene–ethene
copolymers [21]. Thermodynamic arguments suggest
that the incorporation of a hard aryloxide in an
organolanthanide alkene polymerization catalyst should
suppress b-elimination reactions; the complex
[YCpꢀ(OC6H3–Bu2t )(m-H)]2 is a single component cata-
lyst for the formation of isotactic poly(1-hexene) which
also cyclopolymerizes 1,5-hexadiene to poly(methylene-
1,3-cyclopentane) [22]. Derivatives of soluble silses-
quioxane-based polysilanolate ligands have also been
isolated, although no catalytic chemistry has yet been
reported: these include functionalized adducts of
f-elements coordinated to aryloxide in Sm(OC6H3But2-
2,6){(c-C5H9)7Si7O9(O)(OLi)(OSiMe2But)}2 [23], cyclo-
pentadienyl in [Li(THF)]SmCpꢀ2[Cy7Si8O12O]2 [24], and
chloride ligands in LaCl(THF)[Cy7Si7O11(OSiMe3)] [25].
In an extension of this work on monoanionic systems,
the exploration of rigid, dianionic biphenolate and
binaphtholate ligands as potential ancillary ligand sets
for stereochemical control in a-olefin polymerization has
now begun [26].
Fig. 1.
these are exemplified by the dinuclear complex
[Sm(OC6H3–Pri2)3]2 2, Fig. 1, and the mononuclear
Nd(OC6H3–Ph2)3 in which one of the ortho-arene
groups binds in an Z6-conformation to the metal [10].
Incorporation of Group 1 aryloxide precursors can
also give ‘ate’ complexes such as K[Sm(OC6H2–
But2Me)3(THF)] and K[Ln(OC6H3–Pr2i )4] (Ln=Er, La)
[11]. Both homoleptic and solvated aryloxides of the
divalent lanthanides are also readily prepared, including
the aryloxide bridged [Yb(OC6H2–But2Me)2]2 [12],
Yb(OC6H2–But2Me)2(THF)3 and Yb(OC6H2–Bu2t Me)2
(THF)2 [13], Sm(OC6H2–But2Me)2(THF)3 [14], Eu
Metal alkyl derivatives of ortho-trimethylsilyl and
triphenylsilyl-substituted BINOL (BINOL=[1,10]bi-
naphthalenyl-2.20-diol) have been synthesized via alkane
elimination from La(CH{SiMe3}2)3. The energy re-
quired to twist the two aryloxide planes past each other
in the biphenolate complex[La{CH(SiMe 3)2}{1,10-
(2-OC6H2But2-3,5)2}] is very low, so the two asymmetric
isomers of the complexinterconvert. However, in the
(OC6H3–But2)2(MeCN)4
and
Eu(OC6H2–But2Me)2
(Et2O)2 [13,15]. Divalent Tm, Nd, and Dy adducts of
this class of ligands have yet to be reported.
The tris(aryloxide) complexes are now commonly
used as halide-free starting materials for synthetic f-
element chemistry, and the tert-butyl substituted sol-
vent-free Ln(OAr)3 species are volatile and make
excellent precursors for ceramic materials [16]. The
complexSm(OC 6H2–But2Me)3 is a good Lewis acid
catalyst for the Tischenko reaction [17], whilst divalent
Sm(OC6H2–But2Me)2(THF)3 shows extremely high ac-
tivity as an initiator for the ring-opening polymerization
of both e-caprolactone and d-valerolactone, producing
polyesters with molecular weights as high as 600,000 and
narrow molecular weight distributions (o1.65).
Although the complexdoes not polymerize g-butyro-
lactone alone, it can copolymerize this monomer with
e-caprolactone [18].
Continued efforts to find alternative ancillary ligand
sets for Group 3 and f-element alkene polymerization
catalysts have yielded mixed alkyl/aryloxide complexes
such as YCpꢀ(CH{SiMe3}2)(OC6H3–But2) (Cpꢀ=
[Z–C5Me5]), YCpꢀ(OC6H3–Bu2t )2, and Yb(C5H4Me)
(OC6H3–But2)2(THF), SmCp2ꢀ(OC6H3–But2)2, [Li(THF)4]
[Lu(CH2SiMe3)2(OC6H3–But2)2], and NdCp2(OC6H3–
Ph2)(THF)2 [19]. The complexY(CH 2SiMe3)(OC6H3–
But2)2(THF)2 polymerizes both ethene and e-caprolac-
tone, although the related ‘ate’ complexes with incorpo-
rated LiCl – [Li(THF)3]2[Y(CH2SiMe3)2(OC6H3–But2)2
(THF)2]Cl and [Li(THF)4][Lu(CH2SiMe3)2(OC6H3–
But2)2] only polymerize e-caprolactone [20].
complex[La{CH(SiMe
3)2}{1,10-(2-OC10H5SiPh3-3)2}
(OEt2)] 3, Fig. 2, the fused rings render the C2-
symmetric conformation rigid. The complexundergoes
a clean CO insertion into the La–C bond, but no
catalysis studies have been reported on these systems
[27]. A lanthanum diiodide is readily made by treatment
of the monopotassium salt HKBINOL with LaI3,
affording LaI2{OC10H6–C10H6OH}(THF)2 4, Fig. 2
[28]. Other than these, complexes of BINOL are
generally formed as M3[Ln(BINOL)3], where M=Li,
Na, K. These very stable bimetallic complexes are
effective catalysts for asymmetric aldol reactions and the
enantioselective alkylation of aldehydes [29,30]. They
Fig. 2.

92
L.S. Natrajan et al. / Journal of Solid State Chemistry 171 (2003) 90–100
may be considered as a new class of bifunctional catalyst
containing both a Lewis acidic site and a Br^nsted base.
This is demonstrated by the use of the complexes as
promoters for the asymmetric nitroaldol reaction, the
hydrophosphonylation of aldehydes and imines, and
asymmetric Diels-Alder and Michael reactions [31].
Recently, we have become interested in developing
new alternative supporting ligands for f-element com-
plexes—in particular complexes that have open coordi-
nation sites and can participate in reactivity such as
small-molecule activation and catalysis. Aware of the
thermodynamic stability of the lanthanide–aryloxide
bond, but also the difficulty of retaining open, reactive
coordination sites, we began to investigate atom-bridged
biphenolates and binaphtholates which have been
expanded by the formal insertion of an atom X as
suitable dianionic supporting ligands for f-block metals.
For convenience, these are denoted as [OXO] ligands,
where the bracketed atoms are those capable of binding
to the metal, but the type of bond is not defined. In the
complexes described here the two O atoms derive from
the biphenolate structure, and the ligand carries a
dinegative charge. We wanted to know if the additional
donor atom, X, would increase the ease of synthesis of
lanthanide complexes of the ligand compared with the
binolate and binaphtholate systems, and if the donor
atom might then behave as a hemilabile functional
group, by providing steric protection to a mononuclear
complex, but not interfering with the subsequent
reactivity of the Lewis acidic center. The d-block
chemistry of bridged biphenolate ligands for X=C, N,
P, S, Se and Te (5), Fig. 3, has already been studied in
some depth [32]. The S-bridged biphenol is the most
widely used example of this type, originally as a ligand
in main group heterocycle chemistry [33], and more
recently as a supporting ligand for some interesting
d-block chemistry; Ti adducts catalyse alkene polymer-
ization while Cu adducts catalyse the selective aerial
oxidation of alcohols [34]. Related [OSO] ligands are
also prevalent in the patent literature, since nickel
complexes are effective antioxidants that are used to
protect polymers from UV radiation [35].
Fig. 4.
methylalumoxane (MAO) cocatalyst. The results are in
good agreement with experimental observations of these
systems [36]. Of the three chalcogenide bridging atoms,
it follows from the order of electron-donating capacity
to the metal, SoSeoTe, that the ability of the bridging
atom to lower the barrier to ethene insertion follows the
order Te4Se4S. It follows that ligands with alkyl
bridges that do not interact with the metal center form
ineffective Group 4 alkene polymerization catalysts [37].
The most effective precatalysts are predicted to contain
unsaturated CC backbones between the two anionic
groups, and an [OOO]2À or an [SSS]2À set of donor
atoms, although neither of these ligands has been
synthesized to date [38]. Some studies of the PO, SO,
SO2, and even S2-bridged systems have now been
reported [39]. The macrocyclic analogue, tetrathia-
calixarene has recently been shown to be a selective
ligand for the UO22+ ion, but no lanthanide chemistry
has been reported [40].
Given the range of reactive metal complexes that have
been made from the sulfur-bridged biphenol 7, it is
surprising that no Group 3 or f-element derivatives of
this or any of the heteroatom-bridged ligands had been
reported until now. In the last few months, we have been
able to make some [OSO] derivatives of Ln(III) from
across the f-block by both protonolysis and metathesis
routes [41]; herein we focus on the Sm(III) chemistry of
7, 1,10-S(2-HOC6H2But-3-Me-5)2, denoted H2LS shown
in Fig. 4. We also describe how the use of new,
larger ligands based on a binaphtholate framework, 8
Density functional theory has been used to predict the
insertion barriers for ethene polymerization by cations
of the form [Ti[OSO]CH3]+, the active species that is
formed from Ti[OSO]Cl2 6, Fig. 3, in the presence of
in Fig. 4, 1,10-S(2-HOC10H4Bu2t -3,6, denoted H2LSN
gives rise to subtle variations in the chemistry of the
complexes.
,
2. Experimental details
All experimental procedures were carried out under
an atmosphere of dry dinitrogen or argon, using
standard Schlenk techniques (10À4 mbar) or in a glove
box(M-BRAUN or Saffron). NMR spectra were
recorded on a Bruker DPX 300 spectrometer, operating
Fig. 3.

L.S. Natrajan et al. / Journal of Solid State Chemistry 171 (2003) 90–100
93
frequency 300 MHz (1H), 72 MHz (13C), at 300 K.
Chemical shifts are reported in parts per million, and
referenced to residual proton resonances in d6-benzene,
and against external TMS. IR spectra were recorded on
an Avatar 360 FT–IR spectrometer as nujol mulls
between KBr discs, and mass spectra on a Kratos 320
spectrometer.
Solvents were freshly distilled from the appropriate
drying reagent under dinitrogen, and were thoroughly
degassed prior to use: diethylether and pentane from
sodium/benzophenone, toluene from sodium, n-hexane,
THF and benzene-d6, from potassium. All chemicals
were obtained from Aldrich Chemicals. The compounds
Sm(OAr)3, H2LS H2LSN, 11, 12, and 13 were synthesized
according to literature procedures [41,42].
X-ray data were collected using MoKa radiation
(l ¼ 0:71073 A) on a Bruker SMART1000 CCD area
detector diffractometer using o scans. Structure solution
and refinement was carried out using the SHELXTL
suite of programs [43]. Crystallographic data for the
structural analysis has been deposited with the
Cambridge Crystallographic Data Centre, CCDC
No. 186033 for compound 9 and No. 186034 for
compound 10.
18 mmol) and glacial acetic acid (3.06 ml). The reaction
mixture was heated to reflux temperature for 2 h, during
which time the solution turned from colorless to yellow.
Removal of volatiles under reduced pressure and
subsequent recrystallization from a minimum volume
of chloroform at À301C yielded analytically pure
colorless crystalline 3,6-di-tert-butyl-1,2-dicarboxylic
acid, 10, 27% yield, 0.15 g. Crystals of 10 suitable for
single-crystal X-ray diffraction were grown at room
temperature from a saturated solution of the compound
in chloroform.
3
1H NMR/CDCl3 d: 7.99 (d, 1H, 8-H JHH 8.18 Hz),
3
4
7.34 (dd, 1H, 7-H JHH 8.25 Hz, JHH 1.62 Hz), 7.38 (d,
1H, 5-H 4JHH 1.62 Hz) s 7.13 (s, 1H, 4-H), 1.28 (s, 18H,
3-tBu, 6-tBu), 1.36 (s, 9H, 3-tBu), 1.32 (s, 9H, 6-tBu).
ES-MS m=z 327 (24%, [M+Na]+) HR ES-MS m=z
287.1641 (calc. for [M–OH]+ 287.1647). Anal. Calcd.
for C18H24O4 C 71.03 H 7.95 S 0.00 found C 71.12 H
8.06 S 0.00.
2.3. Reaction of 13a with DME
In a Young’s PTFE tap equipped NMR tube, two
equivalents of DME (0.3 mL, 0.028 mmol) in d6-benzene
were added to a sunflower yellow d6-benzene solution
of 13a at 300 K (3.0 mg, 0.014 mmol). An immediate
color change to very pale yellow was observed due t0o
the formation of a complexcharacterized as [Sm{1,1 -
2.1. Attempted preparation of oxythiobis-3,6-di-tert-
butyl-2-naphthol, preparation of 9
To a solution of thiobis-3,6-di-tert-butyl-2-naphthol
(1.0 g, 1.85 mmol) in acetone was added H2O2 (1.0 ml,
9 mmol) and glacial acetic acid (0.51 ml). The reaction
mixture was heated to reflux for 2 h, during which time
the solution turned from colorless to orange. Removal
of volatiles under reduced pressure and subsequent
recrystallization from a minimum volume of n-hexane at
À301C yielded analytically pure needles of 9 in 22%
yield (0.11 g), m.p. 123–1251C. Crystals of 9 suitable for
single-crystal X-ray diffraction were grown at room
temperature from a saturated n-hexane solution of the
compound.
S(2-OC10H4But2-3,6)2}
(OC6H3But2-2,6)
(CH3OC2
H4OCH3), 14. The 1H NMR spectrum was recorded
at this point, and remained unchanged after a further
24 h.
1H NMR/C6D6 dH 8.47 (s, 2H, 5-H), 8.23 (d, 2H,
8-H), 7.95 (s, 2H, 4-H), 7.36 (d, 2H, 7-H), 6.98 (d, 2H,
OAr-m), 7.73 (t, 1H, OAr-p), 1.12 (br s, 18H), 3.07, 1.35,
1.21 (s, 18H, 3-But, 6-But, OAr But) 0.80 (s, 4H, CH2),
0.43p (s, 6H, CH3), 1.57 (s, 18H).
2.4. Catalysis of monoacylation of meso-hydrobenzoin
3
1H NMR/CDCl3 dH: 7.96 (d, 1H, 8-H JHH 8.1 Hz),
3
4
7.41 (dd, 1H 7-H JHH 8.1 Hz, JHH 1.84 Hz), 7.23 (d,
Under an atmosphere of dry dinitrogen, a THF
solution of meso-1,2-diphenyl-1,2-ethanediol (meso-hy-
drobenzoin) (2.4 ml) was treated with 10 equivalents
of acetic anhydride and 10 mol% of a Sm(III) complex
at 251C. The extent of reaction was monitored over
24 h by regular extraction of aliquots for analysis by
TLC (1:1 diethylether:hexane). At the end of the
reaction period, a portion of ethyl acetate (25 ml) was
added to the sample, and the solution washed succes-
sively with saturated NaHCO3 (2 Â 20 ml) and brine
(20 ml). The organic solution was dried over MgSO4,
evaporated to dryness, and recrystallized from ethyl
acetate/hexane. The spectroscopic data for the products
were identical to that previously reported in the
literature [44].
4
1H, 5-H JHH 1.84 Hz) s 7.22 (s, 1H, 4-H), 1.36 (s, 9H,
3-tBu), 1.32 (s, 9H, 6-tBu). dC: 181.6, 179.7, 160.7, 148.1,
140.0, 135.9, 130.2, 128.7, 127.5, 127.1 (naphthol), 35.9,
35.5 (quaternary), 31.2, 29.6 (But Me). n (cmÀ1) 1693w,
1662s, 1590m, 1267m. ES-MS m=z 271 (100%,
[M+H]+) HR ES–MS m=z 271.1694 (calc. for
[M+H]+ 271.1698). Anal. Calcd for C18H22O2: C
79.96, H 8.20 Found: C 79.65 H 8.41.
2.2. Attempted preparation of dioxythiobis-3,6-di-tert-
butyl-2-naphthol, preparation of 10
To a solution of thiobis-3,6-di-tert-butyl-2-naphthol
(1.0 g, 1.85 mmol) in acetone was added H2O2 (2.0 ml,

94
L.S. Natrajan et al. / Journal of Solid State Chemistry 171 (2003) 90–100
3. Results and discussion
The synthesis of the [OSO] ligand 7, has been known
since the 1980s, and involves the ZnCl2-catalyzed
coupling of 2-tert-butyl-4-methylphenol and sulfur
dichloride. Under the same conditions, we have been
able to convert 3,6-di-tert-butyl-2-naphthol into 8,
although we find the reaction is an order of magnitude
faster. We considered that f-element derivatives of this
larger [OSO] ligand might be easier to isolate as
mononuclear complexes, and also that when bound to
emissive centers such as Eu(III) or Tb(III) it might act
as a light harnessing ‘antenna’ ligand, to generate
complexes with interesting optical properties.
Scheme 1. (i) 9 mmol H2O2, glacial acetic acid, acetone, reflux2 h. (ii)
18 mmol H2O2, glacial acetic acid, acetone, reflux2 h.
As one method of tuning the [OSO] ligand 8, we have
investigated oxidation of the sulfur bridge, 7 is selectively
oxidized to yield ligands with an SO and an SO2 bridge—
both of which should bind as [OOO] ligands with
reduced hemilability compared to an [OSO] ligand [45].
However, using the conditions identified for the mono-
and di-oxygenation of 7, 8 is instead converted into 1,2-
naphthoquinone 9 and a dicarboxylate 10, respectively,
Scheme 1. In the cyclic voltammetry experiment, no
electrochemical oxidation of 7 can be measured in the
solvent window but an irreversible, multiple electron
oxidation of 8 is observed at +0.78 V and a quasi-
reversible oxidation at +0.81 V (DEP 0.049 V) in THF.
The absence of clean oxidation chemistry of 8 has led
us to investigate the lanthanide coordination chemistry
of the simplest S-bridged, rather than the SQO or
SO2-bridged ligands, in the first instance.
We have found that the reaction of Sm(OC6H3But2-
2,6)3 with one equivalent of the biphenol 7 affords
the complex[Sm{1,1 0-S(2-OC6H2But-3-Me-5)2}(OAr)
(THF)] (OAr=OC6H3But2-2,6), 11 as bright yellow,
hexane-soluble, air-sensitive crystals, Scheme 2 [41]; this
was the first reported f-block complexof a sulfur-
biphenolate, or any anionic [OSO] type ligand. The
complexcan be converted into a dinuclear, THF-free
adduct 12 by heating in vacuo for a few hours. Similarly,
the reaction of Sm(OC6H3But2-2,6)3 with an equivalent
of binaphthol 8 affords sunflower yellow [Sm{1,10-S(2-
OC10H4But2-3,6)2}(OC6H3Bu2t -2,6)]2, 13a. The com-
plexes are both readily handled in an inert atmosphere,
thermally very stable, and soluble in common organic
solvents including pentane. Attempts to sublime 12 or
13 to purity (pyrextube, 10 À5 mbar) have so far been
unsuccessful. The most notable feature about both 12
and 13a is that the two ring systems of the sulfur-
bridged ligands are inequivalent in the NMR spectra
recorded in benzene solution, and in the solid state as
measured by single-crystal X-ray diffraction. One O
anion of the chelating ligand bridges two metal centers,
rendering the complexes C2-symmetric. This is the first
observation of this bridging mode of coordination for a
metal complexof the [OSO] ligand.
We have also investigated metathesis reactions to
generate Sm(III) derivatives of these [OSO] ligands,
since the reaction of Sm(OAr)3 with a lithium phenoxide
should eliminate the pentane-insoluble LiOAr as a
byproduct, a procedure already shown to be of use in
the reaction chemistry of the hexane-soluble lanthanide
tris(aryloxides) [46]. The dilithium salts of the ligands,
1,10-S(2-LiOC6H2But-3-Me-5)2 denoted Li2LS and 1,10-
S(2-LiOC10H4Bu2t -3,6)2, Li2LSN, may be prepared most
cleanly from the reaction of 7 or 8 with two equivalents
of LiN(SiMe3)2 in diethylether and LinBu in THF,
respectively [40]. Treatment of Sm(OC6H3But2-2,6)3 with
base-free Li2LS affords two equivalents of LiOC6H3But2-
2,6 as a byproduct, and 12, Scheme 2. Surprisingly, the
reaction of Sm(OC6H3But2-2,6)3 with THF-solvated
Li2LSN affords the C2n-symmetric dinuclear complex
13b, Eq. (1), rather than a THF-solvated complex
analogous to 11.
ð1Þ

L.S. Natrajan et al. / Journal of Solid State Chemistry 171 (2003) 90–100
95
Scheme 2. Synthesis of Sm LS derivatives from a homoleptic Sm(III) aryloxide complex.
Scheme 3. Synthetic routes to LS and LSN adducts of Sm(III).
The lack of any THF coordination by the binaphthol
derivatives 13 provides the first observed difference in
the metal chemistry of the two ligands. But a more
interesting observation is that for Sm(III) we can
selectively make the C2n-symmetric analogue of the
binaphtholate complexby a different procedure, so
either a symmetric or asymmetric coordination of the
ligand to Sm can be readily accessed. This chemistry is
summarized in Scheme 3.
Fig. 5.
When all three potentially ligating atoms of the LS
and LSN ligands are coordinated to a large metal cation
the OSO functional groups are opened out, and the two
ring systems are pushed together, and must either fold
down against each other, reducing the CSC angle, or
alternatively twist in opposite directions to reduce
unfavorable steric interactions in the complexes (respec-
tively (a) and (b) in Fig. 5). All the structurally
characterized d-block complexes reported to date show
a facially capping coordination environment as seen in
11 and 13b, except for one Cu(II) derivative in which the
tridentate ligand is close to a planar geometry about the
pseudo-D4h metal; a preferred geometry for Cu(II) [32].
A single-crystal X-ray diffraction study of 13a shows
the twisted ligand conformation, Fig. 6. The two O
atoms of the Sm2O2 core derive from naphthyloxides of
two separate ligands, producing a tightly bound core
with a short Sm–Sm distance of 3.6084(5) A [47]. The
Sm–S distance is 3.117(1) A. This Sm–S distance is
shorter than the maximum distance anticipated for a
non-contact interaction between the two ions, implying
that there is a significant Sm–S interaction in the solid

96
L.S. Natrajan et al. / Journal of Solid State Chemistry 171 (2003) 90–100
state [48]. A wide range of Sm–S distances have been
measured in which the S donor is part of a covalently
bound anionic ligand such as in tris(2,4,6-tert-butylphe-
nylthiolato)Sm(III), ave. Sm–S 2.644 A [49], or part of a
delocalized [–C(QS)(–S)]À ligand such as in bis(hydro-
tris(3,5-dimethylpyrazolyl)borato)(diethyldithiocarbama-
to)Sm(III), ave. Sm–S 2.88A [50]. Two other structurally
characterized lanthanide complexes, which exhibit
similarly close LnyS interactions with a neutral sulfur
donor atom of a thioether have been reported recently.
La–S distances in the solid state of between 3.0635(4)
and 3.1263(4) [51]. These are longer than in the analogous
U(III) adduct, indicating a weaker, but still significant
interaction with the soft sulfur donors. The other is
catena-sodium(bis(m5-thioglycolato)Nd(III)) [52], in
which the central thioether of a dicarboxylate ligand
forms a close (3.45 A) contact with the neodymium cation
resulting in a puckering of the nine-membered metalla-
cyclic ring into two five-membered rings—a nearly
identical conformation to that seen for 13.
A complexof the facially capping thia-crown, La(9S3)I
(CH3CN)2 (9S3=1,4,7-trithiacyclononane), displays
The range of conformations of the phenolate and
naphtholate ligands is shown diagrammatically in
Fig. 7. Thus, it seems reasonable that the large
samarium ion should accommodate the twisted con-
formation more readily than the symmetrical fac
conformation. Since crude 13a is heated to moderate
(501C) temperatures to remove eliminated phenol,
perhaps the absence of heat in the synthesis of 13b
allows the sterically more crowded C2n symmetric
product to be isolated. We have seen no evidence for
the interconversion of 13a and 13b in solution.
3
The ‘wraparound’ coordination that places the two
aromatic planes perpendicular to each other, affords the
complexes an intrinsic chirality, in a similar manner to
BINOL complexes, but neither predetermined nor
necessarily fixed at ambient temperatures. A variable
temperature study of a d8-toluene solution of the C2-
symmetric 13a has been undertaken (see supplementary
information). It shows that upon warming the solution
above room temperature, in addition to the tempera-
ture-dependent paramagnetic shifts of the ligand 1H
resonances, the ortho-aryloxide resonances of the tert-
butyl groups begin to coalesce. The coalescence
temperature cannot be reached in standard NMR
spectroscopic solvents, being in excess of 380 K, but
the minimum free energy associated with this process is
calculated as 69.7 kJ molÀ1. Importantly, over the range
of temperatures studied it is clear that the dinuclear
complexis neither dissociating nor undergoing any
dynamic process that would destroy the asymmetry of
the complex. The dynamic processes we can envisage
Fig. 6. ORTEP drawing of the molecular structure of 13a (50%
probability) with Me groups omitted. Inset, a Pluton view of the M2O2
core structure from above. Selected distances (A) and angles (deg):
Sm1–O5 2.138(3), Sm1–O2 2.157(3), Sm1–O1 2.392(3), Sm1-O1a
2.296(3), Sm1-S4 3.1172(11), Sm1–Sm1a 3.6084(5), O2–Sm1–O1
123.98(11), O2–Sm1–O1a 95.61(11).
Fig. 7. Possible geometries for d- and f-block LS and LSN derivatives rionic[Cu(II)] 0.87, [Ti(IV)] 0.745, [Sm(III)] 1.098 A [52].

L.S. Natrajan et al. / Journal of Solid State Chemistry 171 (2003) 90–100
97
Table 1
Product distributions from the use of selected lanthanide complexes as
promoters for the monoacylation of meso-hydrobenzoin
include most probably the inversion of the [OSO] set at
the metal via labilization of the sulfur, the bridging of
alternative oxygen atoms from the ligand set, or the
direct dissociation of the two halves of the dimer.
The reaction with DME in either toluene or benzene
in a Young’s tap NMR tube affords a pale yellow
adduct [Sm{1,10-S(2-OC10H4But2-3,6)2}(OC6H3But2-2,6)
(CH3OC2H4OCH3)] 14, in which at ambient tempera-
Ratio
ComplexReaction time (h)
Diol
Monoacyl
Diacyl
No catalyst
SmCl3 (anhyd.)
SmCl3(THF)3
Sm(OAr)3
12
24
24
24
24
24
48
24
84
11
0
14
86
40
34
50
66
50
0
3
60
0
1
ture the H chemical shifts of two naphthyl groups and
66
50
34
50
the tert-butyl groups of the monodentate aryloxide are
identical on the NMR timescale, although the DME
resonances are broadened, Eq. (2).
0
12
13a
0
0
monoacylation is complete, and if the reaction is
allowed to proceed beyond 24 h at room temperature,
the catalyst is still active.
ð2Þ
It appears that catalysts that are very soluble and very
coordinatively unsaturated give fast but unselective
acylation. These data are best explained by a reaction
mechanism in which both the diol and the acylating
agent coordinate to the Lewis acid metal before the OH
is acylated. Our initial thoughts on the mechanism were
that either the coordination of a diol is favored over
coordination of the monoacylated product molecule
(since both are THF-soluble and present in the mixture),
which would disfavor the second acylation, or a
situation exists where only an uncoordinated, but
adjacent hydroxyl group is nucleophilic enough to
attack the Ln-bound acylating agent, so if only one
hydroxyl group is present the substrate is effectively
deactivated upon complexation. If there is only one
mode of catalysis, then our observations that the
soluble, coordinatively unsaturated complexes give
overacylation, and that the sterically encumbered
complexes give only monoacylation products, both give
weight to the former mechanism. A solution of 12 in
dichloromethane is completely decomposed after stir-
ring at room temperature for 24 h, preventing compar-
ison with some of the LnCl3-catalyzed reactions that
have been studied previously.
Further reactivity studies have been undertaken to
identify possible intermediates in this catalyzed reaction,
and to help demonstrate how the acylating agent and
monoacylated product fail to react further at the
sterically encumbered catalysts, by treatment of para-
magnetic 13a with a racemic mono-propionate ester,
Eq. (4). In this propionate adduct 15 the asymmetric
naphtholate framework remains virtually unperturbed,
and we find that two equivalents of either the R- or S-
form of the ester bind to any one dimer. The substrate
is sigma-bound to the metal as an alkoxide, so is
undoubtedly not a real intermediate in the catalysis, but
this result shows that there is a suitable metal-accessible
cavity in the asymmetric dinuclear unit.
We have studied the ability of the [OSO] complexes to
catalyse the monoacylation, or desymmetrization, of 1,2
diols using meso-hydrobenzoin as a model substrate,
Eq. (3). It has recently been discovered that lanthanide
trihalides are more efficient promoters for this one-step
reaction than traditional synthetic routes [53], which
also are often multi-step or require specific substrate-
dependent conditions.
ð3Þ
A range of Sm(III) complexes and the simple
tris(aryloxide), Sm(OAr)3, were tested as promoters for
this reaction, since nothing is currently known of the
mechanism. Under an atmosphere of dry dinitrogen, a
THF solution of a suitable test-substrate diol, meso-
hydrobenzoin, was treated with an excess (10 equiva-
lents) of acetic anhydride and a Sm(III) complexat
10 mol% at 25 1C. The mixture was magnetically stirred,
and the extent of the reaction followed by TLC. The
1
product distributions determined by H NMR spectro-
scopy after 24 h (or 48 h) are gathered in Table 1.
Anhydrous SmCl3 is clearly an effective catalyst for
this reaction, affording only a little overacylation of the
diol, but is not particularly soluble in THF. The soluble
trichloride is an unselective catalyst, whilst the soluble,
but sterically encumbered Sm(OAr)3 is a poor, slow
catalyst, although still selective for the monoacylation.
The former two compounds do not allow the reaction to
be studied by NMR spectroscopy. It is clear that
although the aryloxide-derived complexes are slow
catalysts for the acylation reaction, the selectivity for

98
L.S. Natrajan et al. / Journal of Solid State Chemistry 171 (2003) 90–100
ð4Þ
Work is now in progress to identify whether the
[3] The melting points of the complexes are in the range 150 to
2211C, P.B. Hitchcock, M.F. Lappert, A. Singh, Chem. Commun.
(1983) 1499;
optically pure analogues of 13a are effective promoters
for the stereoselective acylation of this meso-diol. We are
also working to identify whether the ligand 8 binds
solely in a bridging sense for the larger f-elements
including Ce(III) and in a symmetrical fac conformation
for Yb(III).
C.L. Zhang, Y. Yao, Y. Luo, Q. Shen, J. Sun, Polyhedron 19
(2000) 2243.
[4] P.B. Hitchcock, M.F. Lappert, R.G. Smith, Inorg. Chim. Acta
139 (1987) 183;
H.A. Stecher, A. Sen, Inorg. Chem. 27 (1988) 1130.
[5] G.B. Deacon, P.E. Fanwick, A. Gitlits, I.P. Rothwell, B.W.
Skelton, A.H. White, Eur. J. Inorg. Chem. (2001) 1505;
Y.M. Yao, Q. Shen, Y. Zhang, M.Q. Xue, J. Sun, Polyhedron
20 (2001) 3201.
8. Conclusions
[6] J. van den Hende, P.B. Hitchcock, S.A. Holmes, M.F. Lappert,
W.-P. Leung, T.C.W. Mak, S. Prashar, J. Chem. Soc. Dalton
Trans. (1995) 1427;
The sulfur-bridged dianionic [OSO] ligands derived
from 2,20-thiobis(6-tert-butyl-4-methylphen-2-ol) and thio-
binaphthol
2,20-thiobis(3,6-di-tert-butylnaphth-2-ol),
R.C. Mehrotra, A. Singh, U.M. Tripathi, Chem. Rev. 91 (1991)
1287;
form air-sensitive but thermally stable complexes of
Sm(III) via a number of synthetically straightforward
routes. The steric demands of the ligands on these large
metals allow both fac and twisted conformations of the
ligands to be isolated. In the latter, which is unprece-
dented in d-block chemistry to date, the ligands bridge
to form C2-symmetric dinuclear complexes. The di-
nuclear systems can be broken up into mononuclear
Lewis base adducts with DME, but are not interconver-
tible on the NMR timescale up to 380 K in the absence
of coordinating solvents. It is still unclear in these
systems whether the S atom is acting as a hemilabile
donor atom.
P.B. Hitchcock, M.F. Lappert, A. Singh, Chem. Commun. (1983)
1499;
G.B. Deacon, C.M. Forsyth, S. Nickel, J. Organomet. Chem. 647
(2002) 50;
G.B. Deacon, G. Meyer, D. Stellfeldt, Eur. J. Inorg. Chem. (2000)
1061.
[7] Y.-M. Yao, Q. Shen, Y. Zhang, M.-Q. Xue, J. Sun, Polyhedron
20 (2001) 3201.
[8] W. Chen, G. Yu, X. Bao, C. Bao, Huaxue Xuebao 43 (1985) 79.
[9] D.M. Barnhart, D.L. Clark, J.C. Gordon, J.C. Huffman,
R.L. Vincent, J.G. Watkin, B.D. Zwick, Inorg. Chem. 33 (1994)
3487.
[10] W.J. Evans, M.A. Greci, J.W. Ziller, Inorg. Chem. 39 (2000)
3213;
W.J. Evans, M.A. Ansari, S.I. Khan, Organometallics 14 (1995)
558.
[11] D.L. Clark, J.C. Gordon, J.C. Huffman, R.L. Vincent-Hollis,
J.G. Watkin, B.D. Zwick, Inorg. Chem. 33 (1994) 5903;
G.B. Deacon, T. Feng, B.W. Skelton, A.H. White, Aust. J. Chem.
48 (1995) 741.
Acknowledgments
The authors are grateful to Dr. E.S. Davies for help
with the electrochemistry, and to the EPSRC and the
Royal Society for funding.
Supplementary material available: Variable tempera-
ture NMR spectra of 13a, full X-ray crystallographic
data and atomic coordinates for 9 and 10, cyclic
voltammograms of 7 and 8 (16 pages).
[12] W.J. Evans, R. Anwander, M.A. Ansari, J.W. Ziller, Inorg.
Chem. 34 (1995) 5;
D.L. Clark, J.C. Gordon, J.C. Huffman, R.L. Vincent-Hollis,
J.G. Watkin, B.D. Zwick, Inorg. Chem. 33 (1994) 5903;
D.L. Clark, G.B. Deacon, T. Feng, R.V. Hollis, B.L. Scott,
B.W. Skelton, J.G. Watkin, A.H. White, Chem. Commun. (1996)
1729.
[13] J.R. van den Hende, P.B. Hitchcock, M.F. Lappert, Chem.
Commun. (1994) 1413.
[14] G.B. Deacon, T. Feng, P. MacKinnon, R.H. Newnham,
S. Nickel, B.W. Skelton, A.H. White, Aust. J. Chem. 46
(1993) 387; G.B. Deacon, P.B. Hitchcock, S.A. Holmes,
M.F. Lappert, P. MacKinnon, R.H. Newnham, Chem. Commun.
(1989) 935.
References
[1] F.T. Edelmann, Angew. Chem. Int. Ed. 34 (1995) 2466.
[2] W.J. Evans, New. J. Chem. 19 (1995) 525.

L.S. Natrajan et al. / Journal of Solid State Chemistry 171 (2003) 90–100
99
[15] W.J. Evans, R. Anwander, M.A. Ansari, J.W. Ziller, Inorg.
Chem. 34 (1995) 5;
R.M.L. Mercado, A. Chandrasekaran, R.O. Day, R.R. Holmes,
Organometallics 18 (1999) 906.
W.J. Evans, M.A. Greci, J.W. Ziller, J. Chem. Soc. Dalton Trans.
(1997) 3035.
[34] J. Okuda, E. Masoud, Makromol. Chem. Phys. 199 (1998) 543;
T. Miyatake, K. Mizunuma, Y. Seki, M. Kakugo,
Makromol. Chem. Rapid Commun. 10 (1989) 349; Y. Yoshio,
K. Masahiro, M. Koozi. M. Tatsuya US Patent 5043408
(1991); M. Hess, K. Wieghardt, P. Chaudhuri, US Patent
6153779 (2000).
[16] Y.K. Gunko, F.T. Edelmann, Comm. Inorg. Chem. 19 (1997) 153.
[17] M.R. Burgstein, H. Berberich, P.W. Roesky, Chem. Eur. J. 7
(2001) 3078.
[18] M. Nishiura, Z.M. Hou, T. Koizumi, T. Imamoto, Y. Wakatsuki,
Macromolecules 32 (1999) 8245.
[35] D.J. Carlsson, T. Suprunchuk, D.M. Wiles, US Patent 3871901
(1975).
[19] C.J. Schaverien, J.H.G. Frijns, H.J. Heeres, J.R. van den Hende,
J.H. Teuben, A.L. Spek, Chem. Commun. (1991) 642;
R.J. Butcher, D.L. Clark, J.C. Gordon, J.G. Watkin,
J. Organomet. Chem. 577 (1999) 228;
[36] R.D. Froese, D.G. Musaev, K. Morokuma, Organometallics 18
(1999) 373.
[37] S. Fokken, T.P. Spaniol, J. Okuda, F.G. Sernetz, R. Mulhaupt,
Organometallics 16 (1997) 4240;
Y. Yao, Q. Shen, J. Sun, F. Xue, Acta Crystallogr. C 54
(1998) 625;
Y. Nakayama, K. Watanabe, Y. Nakayama, N. Ueyama, A.
Nakamura, A. Harada, J. Okuda, Organometallics 19 (2000)
2498.
Z. Hou, Y. Zhang, T. Yoshimura, Y. Wakatsuki, Organometallics
16 (1997) 2963;
[38] A. van der Linden, C.J. Schaverien, N. Meijboom, C. Gantner,
A.G. Orpen, J. Am. Chem. Soc. 117 (1995) 3008.
[39] J. Okuda, S. Fokken, T. Kleinhenn, T.P. Spaniol, Eur. J. Inorg.
Chem. (2000) 1321;
W.J. Evans, R.N.R. Broomhall-Dillard, J.W. Ziller, J. Organo-
met. Chem. 569 (1998) 89;
X.-G. Zhou, Z.-Z. Wu, Z.-S. Jin, J. Organomet. Chem. 431 (1992)
289; G.B. Deacon, S. Nickel, E.R.T. Tiekink, J. Organomet.
Chem. 409 (1991) C1;
J. Okuda, S. Fokken, H.C. Kang, W. Massa, Polyhedron 17
(1998) 943; F. Amor, S. Fokken, T. Kleinhenn, T.P. Spaniol,
J. Okuda, J. Organomet. Chem. 621 (2001) 3;
W.T. Klooster, L. Brammer, C.J. Schaverien, P.H.M. Budzelaar,
J. Am. Chem. Soc. 121 (1999) 1381.
2,20-PPh(4,6-But-C6H2OH)
and
2,20-PPh(=O)(4,6-But-
[20] W.J. Evans, R.N.R. Broomhall-Dillard, J.W. Ziller, J. Organo-
met. Chem. 569 (1998) 89.
C H OH), R. Siefert, T. Weyhermuller, P. Chaudhuri, J. Chem.
¨
6
2
Soc. Dalton Trans. 24 (2000) 4656.
[21] Z.M. Hou, S. Kaita, Y. Wakatsuki, Pure Appl. Chem. 73 (2001)
291;
[40] Z. Asfari, A. Bilyk, J.W.C. Dunlop, A.K. Hall, J.M. Harrowfield,
M.W. Hosseini, B.W. Skelton, A.H. White, Angew. Chem. Int.
Ed. 40 (2001) 721.
Z.M. Hou, Y. Wakatsuki, J. Alloys Compds. 303 (2000) 75.
[22] C.J. Schaverien, J. Mol. Catal. 90 (1994) 177.
[23] P.L. Arnold, A.J. Blake, S.N. Hall, B.D. Ward, C. Wilson,
J. Chem. Soc. Dalton Trans. (2001) 488.
[41] P.L. Arnold, L.S. Natrajan, J.J. Hall, C. Wilson, J. Organomet.
Chem. 647 (2002) 205.
[42] S.D. Pastor, J.D. Spivack, L.P. Steinhuebel, J. Heterocyclic.
Chem. 21 (1984) 1285.
[24] V. Lorenz, A. Fischer, F.T. Edelmann, J. Organomet. Chem. 647
(2002) 245.
[43] Bruker, SHELXTL 5.10, Bruker AXS Inc., Madison, Wisconsin,
USA, 1997;
[25] J. Annand, H.C. Aspinall, J. Chem. Soc. Dalton Trans. (2000)
1867.
G.M. Sheldrick, Acta Crystallogr.
A 46 (1990) 467; G.M.
[26] A. van der Linden, C.J. Schaverien, N. Meijboom, C. Gantner,
A.G. Orpen, J. Am. Chem. Soc. 117 (1995) 3008.
[27] C.J. Schaverien, N. Meijboom, A.G. Orpen, Chem. Commun.
(1992) 124.
Sheldrick, SHELXL97, University of Goettingen, Germany,
1997.
[44] P.C. Zhu, J. Lin, C.U. Pittman, J. Org. Chem. 60 (1995)
5729.
[28] N. Giuseppone, J. Collin, A. Domingos, I. Santos, J. Organomet.
Chem. 590 (1999) 248.
[45] J. Okuda, S. Fokken, H.-C. Kang, W. Massa, Polyhedron 17
(1998) 943;
[29] N. Yoshikawa, J.M.A. Yamada, J. Das, H. Sasai, M. Shibasaki,
J. Am. Chem. Soc. 121 (1999) 4168.
C.R. Cornman, K.M. Geiser-Bush, J.W. Kamp, Inorg. Chem. 38
(1999) 4304;
[30] H.C. Aspinall, J.L.M. Dwyer, N. Greeves, A. Steiner, Organo-
metallics 18 (1999) 1366;
A. Chandrasekarau, R.O. Day, R.R. Holmes, Organometallics 15
(1996) 3182.
H.C. Aspinall, J.F. Bickley, J.L.M. Dwyer, N. Greeves, R.V.
Kelly, A. Steiner, Organometallics 19 (2000) 5416.
[31] M. Shibasaki, H. Sasai, T. Arai, Angew. Chem. Int. Ed. 36 (1997)
1237;
[46] P.B. Hitchcock, M.F. Lappert, W.-P. Leung, L. Diansheng, T.
Shun, Chem. Commun. (1993) 1386; P.B. Hitchcock, M.F.
Lappert, R.G. Smith, R.A. Bartlett, P.P. Power, Chem. Commun.
(1988) 1007.
N. Yoshikawa, J.M.A. Yamada, J. Das, H. Sasai, M. Shibasaki,
J. Am. Chem. Soc. 121 (1999) 4168.
[47] D.R. Click, B.L. Scott, J.G. Watkins, J. Chem. Cryst. 29 (1999)
921;
[32] P. Chaudhuri, M. Hess, T. Weyhermuller, K. Wieghardt, Angew.
Chem. Int. Ed. 38 (1999) 1095;
S. Daniele, L.G. Hubert-Pfalzgraf, J.-C. Daran, S. Halut,
Polyhedron 13 (1994) 927.
P. Chaudhuri, M. Hess, K. Hildenbrand, E. Bill, T.
Weyhermuller, K. Wieghardt, Inorg. Chem. 38 (1999) 2781;
O.S. Filipenko, S.M. Aldoshin, E.P. Ivakhnenko, V.A. Valiulin,
V.I. Minkin, Dokl. Akad. Nauk. 370 (2000) 345;
D.K. Petrikevich, V.A. Timoshchuk, O.I. Shadyro, O.T.
Andreeva, V.I. Votyokov, V.E. Zhelobkovich, Khim.-Farm. Zh.
29 (1995) 32; K. Ohkata, T. Yano, T. Kuwaki, K. Akiba, Chem.
Lett. (1990) 1721.
[48] This maximum distance is defined as 0.15 A longer than a single
Sm–S bond, i.e. 3.148 A (using the longest of the range of Sm–S
distances), A. Bondi, J. Phys. Chem. 68 (1964) 441.
[49] B. Cetinskaya, P.B. Hitchcock, M.F. Lappert, R.G. Smith, Chem.
Commun. (1992) 932.
[50] I. Lopes, A.C. Hillier, S.Y. Liu, A. Domingos, J. Ascenso, A.
Galvao, A. Sella, N. Marques, Inorg. Chem. 40 (2001) 1116.
[51] L. Karmazin, M. Mazzanti, J. Pecaut, Chem. Commun. (2002)
654.
[33] T.K. Prakasha, R.O. Day, R.R. Holmes, J. Am. Chem. Soc. 115
(1993) 2690;
T.K. Prakasha, S. Srinivasan, A. Chandrasekaran, R.O. Day,
R.R. Holmes, J. Am. Chem. Soc. 117 (1995) 10003;
[52] C.J. Kepert, B.W. Skelton, A.H. White, Aust. J. Chem. 52 (1999)
617.

100
L.S. Natrajan et al. / Journal of Solid State Chemistry 171 (2003) 90–100
[53] J.H. Babler, M.J. Coghlan, Tetrahedron Lett. 20 (1979) 1971;
T. Nishiguchi, H.J. Taya, J. Am. Chem. Soc. 111 (1989) 9102;
M. Kinugasa, T. Harada, A. Oku, Tetrahedron Lett. 39 (1998)
4529;
Takeda, A. Sakamoto, T. Tanaka, C. Iwata, Chem. Commun.
(1994) 1345;
M. Oikawa, A. Wada, F. Okazaki, S. Kusumoto, J. Org. Chem.
61 (1996) 4469;
H. Fujioka, Y. Nagatomi, N. Kotoku, H. Kitagawa, Y. Kita,
Tetrahedron 56 (2000) 10141; N. Maezaki, M. Soejima, M.
P.C. Zhu, J. Lin, C.U. Pittman Jr., J. Org. Chem. 60 (1995)
5729.