Dimeric samarium(III) alkoxides bearing N2O2 tetradentate Schiff bases and their utility for the catalytic epoxidation of trans-chalcone

Article abstract of DOI:10.1016/j.jorganchem.2004.11.011

Two anhydrous, dimeric samarium(III) complexes bearing a bulky μ-alkoxide (diphenylmethoxide) ligand and different "saturated" tetradentate Schiff bases {bis-5,5′-(1,3-propanediyldiimino)-2,2-dimethyl- 4-hexene-3-onato and bis-5,5′-(2,2-dimethyl-1,3-propanediyldiimino)-2,2- dimethyl-4-hexene-3-onato} were synthesized and fully characterized. The complexes differ only in alkyl substitution at the three-carbon amino linker between the ketoiminato halves, with one having a CH₂ group and the other a C(CH₃)₂ substituent along the free-ligand idealized mirror plane. Both metal-containing molecules were isolated in high-yields from direct alcoholysis of the corresponding 5-coordinate, mononuclear complexes and their catalytic activity for the epoxidation of 1,3-diphenyl propenone (trans-chalcone), an α,β-unsaturated ketone, investigated.

Full text of DOI:10.1016/j.jorganchem.2004.11.011

Journal of Organometallic Chemistry 690 (2005) 1011–1017
www.elsevier.com/locate/jorganchem
Dimeric samarium(III) alkoxides bearing N2O2 tetradentate
Schiff bases and their utility for the catalytic epoxidation
of trans-chalcone
Steven A. Schuetz, Elizabeth A. Bowman, Carter M. Silvernail,
*
Victor W. Day, John A. Belot
Department of Chemistry, University of Nebraska-Lincoln, 621 Hamilton Hall, Lincoln, NE 68588-0304, USA
Received 18 October 2004; accepted 8 November 2004
Available online 29 December 2004
Abstract
Two anhydrous, dimeric samarium(III) complexes bearing a bulky l-alkoxide (diphenylmethoxide) ligand and different ‘‘satu-
rated’’ tetradentate Schiff bases {bis-5,50-(1,3-propanediyldiimino)-2,2-dimethyl-4-hexene-3-onato and bis-5,50-(2,2-dimethyl-1,3-
propanediyldiimino)-2,2-dimethyl-4-hexene-3-onato} were synthesized and fully characterized. The complexes differ only in alkyl
substitution at the three-carbon amino linker between the ketoiminato halves, with one having a CH2 group and the other a
C(CH3)2 substituent along the free-ligand idealized mirror plane. Both metal-containing molecules were isolated in high-yields from
direct alcoholysis of the corresponding 5-coordinate, mononuclear complexes and their catalytic activity for the epoxidation of 1,3-
diphenyl propenone (trans-chalcone), an a,b-unsaturated ketone, investigated.
ꢀ 2004 Elsevier B.V. All rights reserved.
Keywords: Rare-earth; Alkoxide; Epoxidation; Catalyst; Schiff-base
1. Introduction
low-cost, rapid synthesis, and straightforward modifica-
tion at prochiral sites. Furthermore, the labile apical li-
gand in these 5-coordinate complexes undergoes facile
stoichiometric replacement without N2O2 ligand degra-
dation or displacement. More specifically, this apical li-
gand has been converted to both alkoxides and
phenoxides [1a], as part of a detailed study designed to
assess the effects of steric factors on final product nucle-
arity and/or reactivity.
Previously, rare-earth alkoxides have been used for
the preparation of oxide materials [2] and as active cat-
alysts for small molecule transformations, including the
ring-opening polymerization of lactides [3] and lactones
[4] and the epoxidation of a,b-unsaturated ketones [5].
They therefore represent potentially useful materials
for solving waste management problems through renew-
able resources (e.g., biobased feedstocks) and for alter-
native catalytic preparations of epoxides which are of
An area of considerable interest in contemporary lan-
thanide (Ln) chemistry is the identification of robust li-
gand frameworks which stabilize low metal
coordination numbers and support catalytic activity.
Toward this goal, we previously synthesized heteroleptic
Ln3+ complexes which coordinate a ‘‘saturated’’ tetrad-
entate Schiff base (SB) at the four ‘‘basal’’ sites of a
square pyramid and a monodentate amido at the apical
position [1a–1c]. These complexes offer several potential
advantages over cyclopentadienyl systems; most nota-
bly, the stability associated with the Schiff-base Ln–O
and Ln–N bonds. They also offer inherent attributes of
*
Corresponding author. Tel.: +402 472 5666; fax: +402 472 9402.
E-mail address: jbelot2@unl.edu (J.A. Belot).
0022-328X/$ - see front matter ꢀ 2004 Elsevier B.V. All rights reserved.
doi:10.1016/j.jorganchem.2004.11.011

1012
S.A. Schuetz et al. / Journal of Organometallic Chemistry 690 (2005) 1011–1017
interest for the synthesis of pharmaceutical and natural
product intermediates.
tion of polar lactide monomers. Our reactions yielded
rigorously dimeric Sm3+complexes 3 and 4 that have
remarkably similar molecular structures even though
they contain propyl linkages with quite different steric
requirements, crystallize with totally different solid-state
packing arrangements and contain sterically different
solvent molecules of crystallization. The only molecular
difference for Sm3+ dimers 3 and 4 involves the substit-
uents on the central carbon of the propylene bridge be-
tween the two ketoiminato ‘‘halves’’ of the Schiff base.
This carbon atom [C(6) or C(26)] has two hydrogen
substitutents in 3 and has two methyl substitutents in
4. The final products were fully characterized and as-
sessed as active catalysts for the epoxidation of trans-
chalcone.
The catalytic asymmetric epoxidation of conjugated
enones (a,b-unsaturated) is well established for metal
catalysts such as Ti(OiPr)4 [6] and Mn(salen) complexes
(Jacobsenꢀs catalyst) [7] and poly(amino acids) [8]. It is
also successfully accomplished with rare-earth isoprop-
oxides with the addition of a bidentate chiral auxillary
ligand such as BINOL and its derivatives [9] yielding
product distributions with >90% ee purity. Further-
more, rare-earth BINOL systems offer many advantages
including mild reaction conditions that avoid the use of
halogenated solvents, reduced temperatures, and avoid-
ance of complex multi-step syntheses. However, the
impetus behind developing new epoxidation pathways
for conjugated enones that proceed with enantiospecific
conversion in good yields and minimized quantities of
catalyst offer inherent advantages, especially when con-
sidering large scale preparations. Thus, we sought to
combine the success of Jacobsenꢀs Salen systems with
the rare-earth alkoxides and build lanthanide complexes
of known nuclearity and conformational rigidity for the
epoxidation of conjugated enones, as a logical extension
of our prior work [1].
Although we have reported the synthesis and charac-
terization of numerous mononuclear and rigorously
dinuclear and dimeric Ln3+ complexes bearing ‘‘satu-
rated’’ Schiff bases [1,10], the subtle structural differ-
ences within these molecules and the significant amido
lability prompted evaluation of the replacement of the
apical, bulky bis(trimethylsilyl)amido group with 2,6-
bis(tert-butyl)-4-methylphenol [1a]. Unfortunately these
alkoxide species proved to be kinetically inert, thus we
sough to reduce alkoxide bulk to afford a more open me-
tal center for substrate approach. Herein, we present the
results of the stoichiometric alcoholysis results of the
Schiff-base complexes, [bis-5,50-(1,3-propanediyldi-
imino)-2,2-dimethyl-4-hexene-3-onato]samarium[bis(tri-
methylsilyl)amido], 1, and [bis-5,50-(2,2-dimethyl-1,3-
propanediyldiimino)-2,2-dimethyl-4-hexene-3-onato]-
samarium[bis(trimethylsilyl)amido], 2, with benzhydrol
(diphenylmethanol) (Scheme 1), a ligand previously em-
ployed by Tolman and Hillmyer [11] in the polymeriza-
2. Results and discussion
Although our current interest in 3 and 4 focuses on
their reactivity and potential as active catalysts for the
epoxidation of conjugated enones, we are interested in
understanding any structural basis for such behavior
as well as the apical ligand steric influence on nuclearity
[1a,10]. Differences in the reactivity or catalytic behavior
of 3 and 4 can presumably be attributed to the presence
of gem-dimethyl substituents on the propyl bridges in 4.
These compounds also represent additional examples of
the ease with which alkoxide or phenoxide species can be
straightforwardly prepared from Ln(SB)(amido) com-
plexes without Schiff-base degradation or replacement
[1a].
The intimate solid-state structures of 3 and 4 are sim-
ilar. They both consist of discrete dimeric molecules
(Fig. 1(a) and (b)), each of which contains two Sm(SB)
moieties bridged by the oxygen anions of the diphenyl-
methoxide ligands. Each Sm3+ ion is therefore formally
6-coordinate and bonded to the two oxygens and two
nitrogens of a single dianionic SB ligand as well as the
l-oxygens of the two diphenylmethoxide anions. For
Sm(1) in both structures, the tetradentate Schiff base
occupies one square face of a trigonal prism and the
oxygens of the two bridging diphenylmethoxide ligands
O
HO
R
R
R
O
O
N
N
N
N
O
Sm
Sm
R
R
N
N
O
O
R
Sm N(TMS)2
O
C7H8,
reflux
O
1 R = H
2 R = CH3
3 R = H
4 R = CH3
Scheme 1. Synthesis of dimeric lanthanide alkoxide dimers, 3 and 4. HN(TMS)2 by products omitted for clarify.

S.A. Schuetz et al. / Journal of Organometallic Chemistry 690 (2005) 1011–1017
1013
3
4
˚
Fig. 1. X-ray crystal structure of compounds 3 and 4. Solvent molecules of crystallization omitted for clarity. Relevant bond lengths (A) with esds for
3: Sm1–O1 2.227(5); Sm1–O2 2.224(5); Sm1–O3 2.338(5); Sm1–O4 2.417(4); Sm1–N1 2.477(6); Sm1–N2 2.457(6); Sm2–O3 2.321(5); Sm2–O4
2.275(5); Sm2–O11 2.318(5); Sm2–O12 2.319(6); Sm2–N11 2.481(7); Sm2–N12 2.495(7); 4: Sm1–O1 2.256(4); Sm1–O2 2.246(5); Sm1–O3 2.369(4);
Sm1–O4 2.405(4); Sm1–N1 2.469(6); Sm1–N2 2.469(7); Sm2–O3 2.348(4); Sm2–O4 2.305(4); Sm2–O11 2.318(5); Sm2–O12 2.312(5); Sm2–N11
2.510(6); Sm2– N(12) 2.513(5).
Fig. 2. Identically orientated Sm3+ coordination polyhedra (a) and molecular structures without hydrogens (b) for compounds 3 and 4. The weak
Sm3+ꢀ ꢀ ꢀphenyl interaction is indicated by dashed lines.
complete the trigonal prismatic coordination polyhe-
dron (Fig. 2(a)). The 6-coordinate polyhedron for
Sm(2) in both structures is highly distorted and resem-
bles neither idealized octahedral nor trigonal prismatic
geometry. It almost appears to be a monocapped trigo-
nal prism with one vertex missing. Both Sm(2) ions have
a large vacant region on one side of their coordination
sphere and one of the alkoxide phenyl rings is folded
into the space above this void. The Sm(2)ꢀ ꢀ ꢀC(41) and
Sm(2)ꢀ ꢀ ꢀC(46) interactions (Fig. 2(b)) in 3 and 4 are
bonds in 3. Both trends can be explained by Sm(2) hav-
ing a highly distorted coordination polyhedron and by 4
containing a bulkier Schiff base ligand. The Sm(1)–OSB
˚
bond lengths average [13] 2.235(5,9) and 2.251(5,5) A
in 3 and 4, respectively; the Sm(2)–OSB bond lengths
˚
average 2.319(6,1) and 2.315(5,3) A, respectively. The
Sm(1)–NSB bond lengths average 2.467(6,10) and
˚
2.469(6,0) A in 3 and 4, respectively; the Sm(2)–NSB
˚
bond lengths average 2.488(7,7) and 2.512(6,1) A,
respectively. The Sm–O(3) bond lengths to the first
bridging diphenylmethoxide ligand average 2.330(5,9)
˚
3.12–3.30 A and are in good agreement with previously
˚
and 2.359(4,11) A in 3 and 4, respectively. The Sm–
O(4) bond lengths to the second bridging ligand average
˚
2.346(5,71) and 2.355(4,50) A in 3 and 4, respectively;
˚
Sm(1)–O(4) bond is significantly longer (>0.10 A) than
the Sm(2)–O(4) bond in both complexes. The remaining
observed Ln ꢀ ꢀ ꢀ arene p interactions [12].
In both complexes, the Sm–O bond lengths about
Sm(2) generally tend to be slightly longer than the cor-
responding Sm–O bond lengths for Sm(1) and those
for 4 tend to be slightly longer than the corresponding

1014
S.A. Schuetz et al. / Journal of Organometallic Chemistry 690 (2005) 1011–1017
Scheme 2. Epoxidation reaction (above) and 1H NMR of epoxidation product using 3 (below) at t = 24 h.
bond lengths and angles in both compounds are
unexceptional.
The 1H NMR (Scheme 2) demonstrates that ꢁ50% con-
version of trans-chalcone to 2,3-epoxy-1,3-diphenyl
propanone occurs at ambient temperature over a 24 h.
period. These results are in good agreement with those
reported using La(OiPr)3/(R)- BINOL/Ph3PO as the cat-
alytically active species [14]. Initial attempts at the cata-
lytic epoxidation of trans-chalcone under identical
conditions without activated molecular sieves failed,
perhaps through the formation of tetranuclear clusters
[1b]. This can be attributed to the presence of spurious
water, most likely from the commercial source of TBHP,
which results in inactivity by accelerated oligomeriza-
tion and/or degradation of the rare-earth catalyst. These
results demonstrate the epoxidation of trans-chalcone
using 3 proceeds under mild conditions without elevated
temperatures.
The only noticeable difference between the solid-state
structures of 3 and 4 involves the 6-membered –Sm(1)–
N(1)–C(5)–C(6)–C(7)–N(2)– diamine chelate ring which
adopts a flattened boat conformation in 3 and a flat-
tened chair conformation in 4. The boat conformation
˚
in 3 permits a 2.63 A agostic Smꢀ ꢀ ꢀH interaction that
is not possible in 4 where the methylene hydrogens on
the central carbon of the propyl bridge have been re-
placed by methyl groups. This same behavior has been
previously observed for diamine chelate rings in a vari-
ety of mononuclear Ln(SB)(L) complexes [1a]. The sec-
ond diamine chelate ring in both molecules is twisted
into a conformation midway between the boat and
chair. This twisted chelate ring conformation produces
˚
two ꢁ3.09 A Sm(2)ꢀ ꢀ ꢀH contacts in each molecule; they
involve one hydrogen on each of the terminal methylene
groups [C(25) and C(27)] of the propyl bridge.
3. Conclusions
The close similarity of the structures for 3 and 4 is
indicated by the fact that the two metals, six oxygens
and four nitrogens of each dimer superimpose with a
Presented in this manuscript are the syntheses and
characterization of two dimeric Schiff-base lanthanide
complexes whose molecular structures are formed by
bridging diphenylmethoxides. The crystallographically
observed open coordination sites about the metal cen-
ters prompted us to explore their utility as epoxidation
catalysts toward the electron-rich, a,b-unsaturated ke-
tone, trans-chalcone with TBHP as the oxidant. The
epoxidation is successful when molecular sieves are pres-
ent, but not observed in their absence.
˚
rms deviation of 0.10 A When carbon atom C(6) and
tert-butyl methyl carbons are excluded, the remaining
63 non-hydrogen atoms common to both dimers super-
˚
impose with a rms deviation of 0.23 A. This close simi-
larity was unanticipated since it did not appear that the
methylene hydrogens on the central carbons of the pro-
pyl bridges in 3 could be replaced by methyl groups
without substantially distorting the structure. The mole-
cule easily accommodated one gem-dimethyl substituent
by changing the conformation of one diamine chelate
ring. It accommodated the second gem-dimethyl substi-
tuent through small displacements of nearby groups to-
ward the exterior of the molecule as previously observed
in dinuclear Ln2(SB)3 species [1d].
4. Materials and methods
4.1. X-ray crystallographic characterization of
compounds 3 and 4
Preliminary data indicates that 3 is an active catalyst
for the asymmetric epoxidation of a,b-unsaturated ke-
tones in THF using a 185:1 trans-chalcone:catalyst ratio.
Crystallographic data for the X-ray structural analy-
sis of solvated crystals of 3 and 4 are summarized in

S.A. Schuetz et al. / Journal of Organometallic Chemistry 690 (2005) 1011–1017
1015
Table 1
Crystallographic data for 3 and 4a
Compound No.
3
4
Formula
Fw
Sm2C67.5H94N4O6
1358.17
Sm2C37.5H51N2O3
728.15
Crystal system, Space groupb
Triclinic, P1
Triclinic, P1
ꢀ
ꢀ
˚
a (A)
12.611(2)
14.584(2)
13.027(7)
13.520(7)
˚
b (A)
˚
c (A)
18.441(2)
87.973(3)
22.782(1)
96.589(9)
a (ꢁ)
b (ꢁ)
c (ꢁ)
82.343(3)
87.324(4)
104.536(10)
109.184(9)
3582(3)
3
˚
V (A )
3356.4(8)
2, 1.344
1.782
Z, qcalc (Mg/m3)
4, 1.350
1.675
l (Mo Ka) (mmꢂ1
)
Crystal size (mm3)
0.08 · 0.07 · 0.04
100% (26.00)
Empirical
0.6014/0.5384
2.312/ꢂ1.528
173(2)
0.56 · 0.08 · 0.06
98.6% (25.41)
Empirical
0.8060/0.6668
1.681/ꢂ1.563
193(2)
Data completeness (max hMo Ka
)
Absorption correction
Max/min transmission
3
Largest diff. peak/hole (e/A )
˚
Temperature (K)
No. of reflns (collected/unique)
40948/13195
0.0730 (9150)
0.0953 (13195)
1.240
26538/13039
0.0513 (8403)
0.1339 (13039)
0.970
c
R1 (I > 2rI)
wR2 (unique reflections)
d
Goodness of fit indicatore
a
G.M. Sheldrick, SHELXTL v.5.1, Bruker-AXS, Madison, WI, 1999.
b
c
N.F.M. Henry, K. Lonsdale (Eds.), International Tables for X-ray Crystallography, Symmetry Groups, vol. 1, Kyntoch Press, England, 1969.
R1 = |Fo| ꢂ |Fci/ |Fo|.
P
P
P
P
d
e
R2 = [ w(|Fo| ꢂ |Fc|)2/ w(|Fo|2)]1/2
.
P
Goodness of fit [ w(|Fo| ꢂ |Fc|)2/(Nobs ꢂ Nparam)]1/2
.
Table 1. Lattice constants and final orientation matrices
for solvated single crystals of each compound were
determined using peak centers for 4799(3) or 922(4)
reflections. Complete spheres (2474 30-s frames for 3)
or hemispheres (1868 15-s frames for 4) of diffraction
data were collected for both compounds using graph-
ite-monochromator Mo Ka radiation on a Bruker
SMART (4) or APEX (3) CCD diffractometer. X-rays
for both studies were provided by a fine-focus (3) or nor-
mal-focus (4) sealed X-ray tube operated at 50 kV and
35 or 40 mA. Frame data were collected using the Bru-
ker SMART software package and final lattice con-
stants were determined with the Bruker SAINT
software package. The Bruker SHELXTL-NT software
package was used to solve and refine both structures.
‘‘Direct methods’’ techniques were used to solve each
structure and the resulting structural parameters were
refined with F2 data to convergence using counter-
weighted full-matrix least-squares techniques.
The final structural model for heptane-solvated crys-
tals of 3 incorporated anisotropic thermal parameters
for all non-hydrogen atoms and isotropic thermal
parameters for all hydrogen atoms of the metal com-
plex. The heptane solvent molecule of crystallization is
disordered about the crystallographic inversion center
at (1/2, 0,1/2) in the unit cell and each of its carbon
atoms was included in the structural model as an isotro-
pic atom with an occupancy factor of 0.50. A continu-
ous chain of only 6 carbon atoms could be located for
this molecule from difference Fouriers and chemically
reasonable metrical parameters could not be achieved
unless all of the solvent C–C bond lengths and Caꢀ ꢀ ꢀCc
separations (for the tetrahedral Ca–Cc–Cc angles) were
restrained to values which were 1.000 and 1.633 times
a common distance. This distance was included in the
least-squares refinement cycles as a common free vari-
˚
able which refined to a final value of 1.47(2) A. All
methyl groups for metal complex 3 were included in
the structural model as rigid rotors (assuming idealized
sp3 hybridization of the carbon atom and a C–H bond
˚
length of 0.98 A) allowed to rotate about their C–C
bonds in least squares refinement cycles. All other
hydrogens were incorporated in the structural model
as fixed idealized (assuming idealized sp3 hybridization
of the carbon atom carbon and C–H bond lengths of
˚
0.95–1.00 A) atoms ‘‘riding’’ on their respective carbon
atoms. The isotropic thermal parameter for each of
these idealized hydrogens was fixed at a value that was
1.2 (non-methyl) or 1.5 (methyl) times the equivalent
carbon atom isotropic thermal parameter to which it is
bonded. Hydrogen atoms were not included for the sol-
vent molecule of crystallization.
The final structural model for toluene-solvated crys-
tals of 4 incorporated anisotropic thermal parameters
for all non-hydrogen atoms and isotropic thermal
parameters for all hydrogen atoms. One of the t-butyl
groups is disordered with two preferred (30% and 70%
occupancy) orientations about the C(21)–C(22) bond.

1016
S.A. Schuetz et al. / Journal of Organometallic Chemistry 690 (2005) 1011–1017
All methyl groups were included in the structural model
as rigid rotors (assuming idealized sp3 hybridization of
Found: C, 59.36; H, 7.02 (the presence of heptane was
confirmed by X-ray diffraction).
˚
the carbon atom and a C–H bond length of 0.98 A) al-
lowed to rotate about their C–C bonds in least squares
refinement cycles. All other hydrogens were incorpo-
rated in the structural model as fixed idealized (assuming
idealized sp3 hybridization of the carbon atom and C–H
4.4. Bis-5,50-(2,2-dimethyl-1,3-propanediyldiimino)-2,2-
dimethyl- 4-hexene-3-one benzhydrol samarium(III)
dimer (4)
˚
bond lengths of 0.95–1.00 A) atoms ‘‘riding’’ on their
In a 100 mL Schlenk flask, 0.228 g (0.346 mmol) of 2
was dissolved in 30 mL C7H8. While stirring, 0.0640 g
(0.347 mmol) of benzhydrol dissolved in 30 mL C7H8
was added in one portion and the reaction heated at re-
flux overnight. Upon cooling to ambient temperature,
the light yellow solution was concentrated in vacuo
(ꢁ10 mL) and pale yellow crystals resulted from the
mother liquor after 2 weeks at ꢂ18 ꢁC. Yield: 90.7%
(0.210 g); m.p. 150.5–152 ꢁC (dec.); 1H NMR (d,
C6D6, all peaks paramagnetically broadened): ꢂ14.0,
ꢂ6.45(b), ꢂ2.85 (b), ꢂ1.92 (b), ꢂ0.80, 0.725(b), 0.89
(m), 1.049 (b), 1.24 (b), 1.28 (b), 1.91 (b), 2.11 (b),
2.32 (b), 2.68 (m, b), 3.32 (b), 6.89 (b), 9.18 (b); 13C
NMR (d, C6D6): 14.7, 17.2 (b), 19.4 (b), 20.4, 20.6,
23.4 (b), 26.6, 27.5, 28.7, 29.0, 29.8, 30.2 (b), 30.6 (b),
30.6, 32.6, 32.7, 36.8, 37.4 (b), 43.8 (b), 52.7 (b), 58.7,
98.7, 102.2, 126.3 (b), 172.8, 176.2, 191.9. Anal. Calcd.
for Sm2N4O6C68H94Æ1/2C7H8: C, 60.22; H, 7.13. Found
C, 60.39; H, 6.85 (the presence of toluene was confirmed
by X-ray diffraction).
respective carbon atoms. The isotropic thermal parame-
ter for each of these idealized hydrogens was fixed at a
value that was 1.2 (non-methyl) or 1.5 (methyl) times
the equivalent carbon atom isotropic thermal parameter
to which it is bonded.
4.2. Materials and methods
All reactions were carried out under an atmosphere
of dry dintrogen using standard Schlenk and glovebox
techniques. Compounds 1 and 2 were synthesized by a
previous published procedure [1a,1c]. Diphenylmetha-
nol (benzhydrol) (Aldrich) was doubly sublimed on a
vacuum line fitted with an oil diffusion pump operating
at 10ꢂ5–10ꢂ6 Torr. Heptane (C7H16) and tetrahydrofu-
ran (THF) were dried and distilled from CaH2 and
Na/benzophenone ketal, respectively. Mallinckrodt,
˚
grade 514GT, 4 A molecular sieves (VWR) and trans-
chalcone (Lancaster) were predried for 24 h (120 ꢁC,
10ꢂ4–10ꢂ5 Torr and 50 ꢁC, 10ꢂ4–10ꢂ5 Torr, respec-
tively). Tert-butyl hydroperoxide (5.0–6.0 M, Aldrich)
in decane was used as received. H and 13C NMR were
recorded on a Bruker DRX Avance 500 MHz (1H fre-
quency) or DRX Avance 400 MHz (1H frequency). Ele-
mental analyses were performed at Midwest Microlabs
(Indianapolis, IN). Melting points were collected on a
modified Mel-Temp II apparatus providing digital ther-
mocouple readouts and are uncorrected.
4.5. Epoxidation of trans-chalcone
1
In a 100 mL 2-neck round-bottom flask is placed
0.0191 g (0.140 mmol) of 3 and ꢁ1.00 g of predried 4
˚
A molecular sieves. The flask is fitted with a rubber sep-
tum and a Merlic solid addition funnel containing 0.548
g (2.60 mmol) of predried 1,3-diphenyl-2-propenone.
The flask is removed from the glovebox, interfaced to
an N2-containing Schlenk line and 10mL of dry THF
is introduced via syringe. While stirring, the contents
of the Merlic solid addition funnel are added and the
reaction allowed to stir for 1 h. at ambient temperature.
Following this, 1.00 mL of 5–6 M TBHP in decane was
added in one portion via syringe and the reaction stirred
for an additional 24 h. TLC analysis followed by flash
chromatography of the reaction mixture using 90:10
hexane:ethyl acetate mobile phase confirmed the pres-
ence of 2,3-epoxy-1,3-diphenyl propanone, which was
4.3. Bis-5,50-(1,3-propanediyldiimino)-2,2-dimethyl-4-
hexene-3- one benzhydrol samarium(III) dimer (3)
In a 100 mL Schlenk flask, 0.336 g (0.532 mmol) of 1
was dissolved in 30 mL C7H8. While stirring, 0.0980 g
(0.532 mmol) of benzhydrol dissolved in 30 mL C7H8
was added in one portion and the reaction heated at re-
flux overnight. Upon cooling to ambient temperature,
the pale yellow solution was concentrated in vacuo
and the solids recrystallized from C7H16. Pale yellow
crystals resulted after 1 week at ꢂ18 ꢁC. Yield: 85.6%
1
subsequently calculated via H NMR integration to be
>50% conversion.
1
(0.298 g); m.p. 150–151 ꢁC (dec.); H NMR (d, C6D6,
all peaks paramagnetically broadened): ꢂ25.34, ꢂ6.35
(b), ꢂ4.02 (b), ꢂ1.62 (b), 0.87, 1.13 (b), 1.27 (b), 1.78
(b), 3.72(b), 4.66 (b), 6.86 (b), 14.31 (b), 17.22 (b); 13C
NMR (d, C6D6): 14.3, 18.8, 23.0, 29.4 (b), 30.0, 32.2,
36.9, 44.5, 101.9 (b), 123.7 (b), 176.1 (b); Anal. Calcd.
for Sm2N4O6C64H86Æ1/2C7H16: C, 59.69; H, 6.98.
Acknowledgements
This work was supported by the Nebraska Research
Initiative and by the donors of the American Chemical
Society Petroleum Research Fund. The authors also
thank Dr. Joanna Clark and Ms. Sara Basiaga of the

S.A. Schuetz et al. / Journal of Organometallic Chemistry 690 (2005) 1011–1017
1017
´ ˆ
[4] (a) E. Martin, P. Dubois, R. Jerome, Macromolecules 36 (2003)
7094;
(b) E. Martin, P. Dubois, R. Jerome, Macromolecules 33 (2000)
University of Nebraska–Lincoln for their crystallo-
graphic and paramagnetic NMR assistance, respectively.
´ ˆ
1530;
(c) M. Yamashita, Y. Takemoto, E. Ihara, H. Yasuda, Macro-
molecules 29 (1996) 1798.
Appendix A. Supplementary data
[5] (a) J. Inanaga, H. Furuno, T. Hayano, Chem. Rev. 102 (2002)
2211;
ORTEP plots of compounds 3 and 4 are available and
may be obtained using any current masthead page. Crys-
tallographic files, in CIF format, are available from the
Cambridge Crystallographic Data Center, 12 Union
Road, Cambridge, CB2 1EZ, UK (fax: +44-1223-
336033; email: deposit@ccdc.cam.ac.uk) under assigned
CCDCnumbersof250074(3)and250073(4). Supplemen-
tary data associated with this article can be found, in the
online version at doi:10.1016/j.jorganchem.2004.11.011.
(b) M. Shibasaki, N. Yoshikawa, Chem. Rev. 102 (2002)
2187.
[6] R.A. Johnson, K.B. Sharpless, in: B.M. Trost, I.Fleming
(Eds.),Comprehensive Organic Synthesis, vol. 7, New York,
1991, p. 389.
[7] (a) C.E. Song, S.-G. Lee, Chem. Rev. 102 (2002) 3495;
(b) T. Katsuki, Coord. Chem. Rev. 140 (1995) 189.
[8] (a) C. Lauret, M. Roberts, Aldrichim. Acta 35 (2002) 47;
(b) M.J. Porter, J. Skidmore, Chem. Commun. (2000) 1215.
[9] (a) R. Chen, C. Qian, J.G. deVries, Tetrahedron 57 (2001)
9837;
(b) T. Nemoto, T. Ohshima, K. Yamaguchi, M. Shibasaki, J.
Am. Chem. Soc. 123 (2001) 2725;
References
K. Daikai, M. Kamaura, J. Inanaga, Tetrahedron Lett. 39
(1998) 7321.
[10] S.A. Schuetz, V.W. Day, A.L. Rheingold, J.A. Belot, J. Chem.
Soc., Dalton Trans. (2003) 4303.
[1] (a) S.A. Schuetz, C.M. Silvernail, C.D. Incarvito, A.L. Rheingold,
J.L. Clark, V.W. Day, J.A. Belot, Inorg. Chem. 43 (2004) 6203;
(b) S.A. Schuetz, V.W. Day, J.L. Clark, J.A. Belot, Inorg. Chem.
Commun. 5 (2002) 706;
[11] B.J. OꢀKeefe, L.E. Breyfogle, M.A. Hillmyer, W.B. Tolman, J.
Am. Chem. Soc. 124 (2002) 4384.
[12] G.B. Deacon, G.M. Forsyth, P.C. Junk, B.W. Skelton, A.H.
White, Chem. Eur. J. 5 (1999) 1452.
(c) S.A. Schuetz, V.W. Day, R.D. Sommer, A.L. Rheingold, J.A.
Belot, Inorg. Chem. 40 (2001) 5292;
[13] Throughout this manuscript, a standard formalism of crystal-
lography-derived estimated standard deviations is used to
provide a more concise, yet comprehensive, synopsis of metrical
parameters. Hence, the first number in parenthesis following an
(d) S.A. Schuetz, M.A. Erdmann, V.W. Day, J.L. Clark, J.A.
Belot, Inorg. Chim. Acta. 357 (2004) 4045.
[2] L.G. Hubert-Pfalzgrapf, New J. Chem. (1995) 727.
[3] (a) B.J. OꢀKeefe, M.A. Hillmyer, W.B. Tolman, J. Chem. Soc.,
Dalton Trans. 15 (2001) 2215;
average value of
a bond length is the root-mean-squared
estimated standard deviation of an individual datum. The
second number is the average deviation from the average value,
respectively..
(b) M. Yamashita, Y. Takemoto, E. Ihara, H. Yasuda, Macro-
molecules 33 (2000) 3920;
(c) B.M. Chamberlain, Y. Sun, J.R. Hagadorn, E.W. Hemmesch,
V.G. Young, M. Pink, M.A. Hillmyer, W.B. Tolman, Macro-
molecules 32 (1999) 2400.
[14] K. Daikai, T. Hayano, R. Kino, H. Furuno, T. Kagawa, J.
Inanaga, Chirality 15 (2003) 83.