Trivalent lanthanide-alkene complexes: Crystallographic and NMR evidence for coordination of tethered alkenes in the solid state and solution

Article abstract of DOI:10.1016/j.jorganchem.2010.09.004

The reaction of 2,6-diallyl-4-methylphenol (H-DALP) and 2-allyl-4,6-dimethylphenol (H-MALP) with Ln[N(SiMe₃) ₂]₃ affords dimeric {Ln[DALP]₂} ₂[μ-DALP]₂ [Ln = La (3), Ce (4), Nd (5), Er (6),

Full text of DOI:10.1016/j.jorganchem.2010.09.004

Journal of Organometallic Chemistry 695 (2010) 2703e2712
Contents lists available at ScienceDirect
Journal of Organometallic Chemistry
journal homepage: www.elsevier.com/locate/jorganchem
Trivalent lanthanideealkene complexes: Crystallographic and NMR evidence for
coordination of tethered alkenes in the solid state and solution
,
b
c
David J. Berg a , Tosha Barclay , Xuening Fei
*
a Department of Chemistry, University of Victoria, P.O. Box 3065 STN CSC, Victoria, B.C., Canada V8W 3V6
b Lone Star College e North Harris, Houston, TX 77073-3499, USA
c Department of Material Science and Engineering, Tianjin Institute of Urban Construction, Tianjin 300384, PR China
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 7 July 2010
Received in revised form
23 August 2010
The reaction of 2,6-diallyl-4-methylphenol (H-DALP) and 2-allyl-4,6-dimethylphenol (H-MALP) with Ln
[N(SiMe3)2]3 affords dimeric {Ln[DALP]2}2[
m
-DALP]2 [Ln ¼ La (3), Ce (4), Nd (5), Er (6), Yb (7), Y (8)] and
{Ln[MALP]2}2[
m
-MALP]2 [Ln ¼ La (9), Sm (10), Y (11)] complexes, respectively. The solid state structures of
5, 6, 8, 10 and 11 reveal the presence of two weak Lnealkene interactions per lanthanide center involving
one o-allyl unit of each terminal aryloxide ligand. Variable temperature NMR studies reveal that the
dimeric structures remain intact in non-coordinating solvents with rapid bridgeeterminal aryloxide
exchange taking place above about 275e295 K for all DALP complexes except erbium. In the case of
yttrium complex 8, evidence for a second dynamic process involving alkene exchange is observed at
temperatures below about 220 K. 13C NMR chemical shift data also supports a weak Lnealkene inter-
action, although the magnitude of these shifts is small due to rapid averaging of free and bound alkene
resonances at higher temperatures. For all complexes, the addition of THF cleaves the dimeric structure
and disrupts the Lnealkene bond.
Accepted 1 September 2010
Available online 15 September 2010
This paper is dedicated to the memory of
Professor Herbert Schumann: a true pioneer
of f-element chemistry, an inspired teacher,
an esteemed mentor and above all, a great
gentleman.
Keywords:
Lanthanide
f-element
Alkene complex
NMR
Ó 2010 Elsevier B.V. All rights reserved.
Crystallography
X-ray
1. Introduction
latter studies placed the upper limit of
at ca. 40 kJ molꢀ1 indicating a weak lanthanideealkene interaction,
as might be expected in the absence of any -backdonation [13,14].
D
Gz for alkene dissociation
Lanthanideealkene complexes are assumed to be intermediates
in a number of catalytic processes mediated by these metals
including olefin [1e3] and diene [3e6] polymerization, hydroge-
nation [3], hydroamination [3, 7e10], hydrophosphination [1, 3]
and hydrosilation [1, 3]. Despite their importance, very few lan-
thanideealkene complexes have been structurally characterized
even though there is ample evidence for their existence from
previous NMR studies. Marks showed convincing paramagnetic
shift evidence for ethylene coordination to base-free (C5Me5)2Eu
[11], and Casey reported extensively on the dynamic NMR behavior
of a yttrium alkyl-tethered alkene system of general type (C5Me5)2Y
p
Given that the lanthanideealkene bond is expected to be weak
and labile, it is not surprising that all structurally characterized
lanthanideealkene complexes, with one exception, involve an
alkene that is tethered to another coordinating group. The one
exception is not a simple alkene at all but rather the Pt-coordinated
alkene complex of divalent ytterbium, (C5Me5)2Yb(m- :h
h2 2-C2H4)Pt
(PPh3)2, reported by Andersen and Burns over 20 years ago [21].
This was the first lanthanideealkene structure reported and
remains the only structurally characterized example without
alkene tethering. Evans et al. [22] and Schumann et al. [23,24]
reported the structures of several alkene-tethered cyclo-
pentadienyl complexes of divalent lanthanides (I, II, Chart 1) and
the closely related alkaline earths (III, Chart 1) that show alkene
coordination. In fact, other than the interesting alkylealkene
(
h1
:
h5-CH2CH2CR2CR0¼CH2) (R ¼ H, Me; R0 ¼ H, Me) [12e20]. The
* Corresponding author. Tel.: þ1 250 721 7161; fax: þ1 250 721 7147.
E-mail address: djberg@uvic.ca (D.J. Berg).
complex [Yb{h1
:h
2-C(SiMe3)2SiMe2CH¼CH2}{OEt2}]2(
m-I)2, (IV),
0022-328X/$ e see front matter Ó 2010 Elsevier B.V. All rights reserved.
doi:10.1016/j.jorganchem.2010.09.004

2704
D.J. Berg et al. / Journal of Organometallic Chemistry 695 (2010) 2703e2712
Chart 1.
these represented the extent of structurally characterized lantha-
nideealkene complexes prior to this work [25].
2.2. Ligand synthesis
In this contribution, we discuss the synthesis, solid state struc-
tural features and solution behavior of dimeric lanthanide aryl-
oxides bearing one or two allyl groups in the o-aryl positions.
Alkene coordination to the Ln3þ center from one allyl group of each
terminal aryloxide is observed crystallographically across the
lanthanide series. These complexes represent the first neutral,
trivalent lanthanideealkene complexes that have been structurally
characterized and NMR evidence suggests that the structures
obtained in the solid state are maintained in non-coordinating
solvents.
2.2.1. 2,6-diallyl-4-methylphenol, H-DALP (1)
4-Methylphenol (75.6 g, 0.70 mol) was dissolved in 250 mL
toluene in a 1 L round bottom flask and 250 mL water was added.
Excess NaOH (29 g), tricaprylmethylammonium chloride (12 g) and
allyl bromide (93.2 g, 1.1 equiv) were added and the two-phase
mixture was heated at 70 ꢁC overnight. After cooling, the organic
phase was separated and washed with a further 250 mL water,
followed by drying over anhydrous MgSO4. Filtration and removal
of the solvent from the filtrate under reduced pressure afforded
crude 4-methylphenyl allyl ether (94.3 g, 91%). The allyl ether was
then heated as a neat oil at 180 ꢁC for 24 h to afford the Claissen
product 2-allyl-4-methylphenol as a light yellow oil in 60% yield
(62.2 g) after column chromatography on silica gel (15% ether in
hexane). Repeating the above reactions with 2-allyl-4-methyl-
phenol (40 g, 0.27 mol) afforded 32.3 g (64%) of 2,6-diallyl-4-
methylphenol, H-DALP (1), after chromatography on silica gel
2. Experimental
2.1. General procedures
All manipulations were carried out under an argon atmosphere,
with the rigorous exclusion of oxygen and water, using standard
glovebox (Braun MB150-GII) or Schlenk techniques. Tetrahydro-
furan (THF), hexane and toluene were dried by distillation from
sodium benzophenone ketyl under argon immediately prior to use.
The appropriate lanthanide tris(silylamides), Ln[N(SiMe3)2]3, was
prepared according to a literature procedure [26].
(hexane eluent). 1H NMR (C6D6)
d 6.73 (s, 2H, arylH), 5.90 (m, 2H,
CH]), 5.01 (d, 2H, trans- ]CH2), 4.96 (m, 2H, cis- ]CH2), 4.75 (s,
1H, OH), 3.23 (d, 4H, CH2), 2.12 (s, 3H, CH3); 13C{1H} NMR (C6D6)
d
150.8 (arylCOH), 137.0 (CH]), 129.6 (p-arylC), 129.5 (m-arylC),
125.7 (o-arylC), 116.0 (]CH2), 35.4 (CH2), 20.6 (CH3). High Res. MS
(EI) Found (Calcd): Mþ 188.1208 (188.1201).
All NMR spectra (1H, 360 MHz; 13C, 90 MHz) including variable
temperature experiments were recorded on a Bruker AMX-360 MHz
spectrometer and referenced to residual solvent resonances.
Deuterated solvents were dried over activated 4 Å molecular sieves
and spectra were recorded using 5 mm tubes fitted with a TeflonÔ
valve (Brunfeldt). NMR data for diamagnetic complexes are given at
ambient temperature unless otherwise specified; data for para-
magnetic complexes are given at selected temperatures. In the 13C
DEPT-135 spectra, resonance phases are indicated by þ or ꢀ in the
data below. Complete variable temperature NMR data for the para-
2.2.2. 2-Allyl-4,6-dimethylphenol, H-MALP (2)
2-allyl-4,6-dimethylphenol was prepared from 2,4-dimethyl-
phenol using a procedure analogous to that for 1 above. H-MALP (2)
was isolated as a yellow oil after chromatography on silica gel (15%
ether in hexane). Yield: 53% overall (4.20 g of 2 from 6.1 g starting
phenol). 1H NMR (C6D6)
d 6.77 (s, 1H, arylH), 6.70 (s, 1H, arylH), 5.93
(m, 1H, CH]), 5.02 (m, 2H, ]CH2), 4.60 (s, 1H, OH), 3.20 (d, 2H, CH2,
3JHH ¼ 6.6 Hz), 2.17 (s, 3H, CH3), 2.14 (s, 3H, CH3); 13C{1H} NMR
(C6D6)
d 150.3 (arylCOH), 137.0 (CH]), 129.9 (arylCH), 129.5 (p-
arylC), 128.5 (arylCH), 124.4, 124.0 (o-arylC), 116.2 (]CH2), 35.6
(CH2), 20.4 (CH3), 15.8 (CH3). High Res. MS (EI) Found (Calcd): Mþ
162.1043 (162.1045).
magnetic DALP complexes 4e7 including plots of d vs. 1/T, complete
listings of chemical shifts and peak widths at half height are given in
the Supplementary Material. Melting points were recorded using
a Büchi melting point apparatus and are not corrected. Elemental
analyses were performed by Canadian Microanalytical, Delta, B.C.
Mass spectra were recorded on a Kratos Concept H spectrometer
usingelectronimpact (70 eV)or FABmethods. Only metal complexes
3, 5 and 7 gave high mass peaks in the mass spectrum under any of
the conditions used.
2.3. Complex synthesis
2.3.1. {La[DALP]2}2[m-DALP]2 (3)
A solution of 1 (0.250 g,1.33 mmol) in 4 mL toluene was added to
a solution of La[N(SiMe3)2]3 (0.274 g, 0.442 mmol) in 16 mL toluene

D.J. Berg et al. / Journal of Organometallic Chemistry 695 (2010) 2703e2712
2705
in the glovebox. An immediate pale green color developed but the
solution soon faded to very pale yellow on stirring at room
temperature. After stirring overnight, the solvent was removed
under reduced pressure and the pasty solid residue was redissolved
in hot hexane, filtered though a Celite pad and the filtrate was
allowed to cool slowly to room temperature. The cream colored
flaky plates of 3 that deposited from solution were collected by
suction filtration and dried under vacuum. Yield: 0.23 g (74%). Mp.
Mp.130e134 ꢁC; MS (EI) m/z 1469 (Mþ isotopic cluster) amu; 1H NMR
(C7D8, 297 K)
d
36.07 (12H, CH], n1/2 ¼ 750 Hz), 25.14 (12H, CH],
n1/2 ¼ 500 Hz), 13.44 (12H, arylH, n1/2 ¼ 200 Hz), 3.48 (18H, CH3,
n1/2 ¼ 15 Hz), ꢀ12.46 (24H, CH2, n1/2 ¼ 540 Hz), one alkene H reso-
nance was not observed; 1H NMR (C7D8, 377 K)
d 22.95 (12H, CH],
n1/2 ¼ 42 Hz), 17.82 (12H, CH], n1/2 ¼ 56 Hz), 11.60 (12H, aryl H,
n1/2 ¼ 42 Hz), 2.78 (18H, CH3, n1/2 ¼ 5 Hz), ꢀ3.99 (24H, CH2, n1/
¼ 76 Hz), ꢀ19.78 (12H, CH], n1/2 ¼ 800 Hz); Anal. Calcd for
2
152e154 ꢁC; 1H NMR (C6D6)
CH]), 5.16 (m, 24H, ]CH2 overlapping), 3.50 (d, 24H, CH2), 2.20 (s,
18H, CH3); 13C{1H} NMR (C6D6)
140.7 (CH]),129.4 (m-arylC),117.0
(]CH2), 36.1 (CH2), 20.7 (CH3); the quaternaryaryl carbons were not
observed; 13C{1H} NMR (C6D6:C4D8O, 5:1)
160.0 (arylCO), 139.1
d
6.86 (s, 12H, arylH), 6.09 (m, 12H,
C78H90O6Yb2: C, 63.75; H, 6.17%; Found: C, 63.24; H, 6.18%.
d
2.3.6. {Y[DALP]2}2[m-DALP]2 (8)
Complex 8 was prepared using a procedure analogous to 3 from
Y[N(SiMe3)2]3 (0.252 g, 0.442 mmol) and 3 equiv of 1. Colorless
cubes of 8 were isolated from hexane solution. Recrystallization
from hot a hexane solution afforded crystals suitable for X-ray
diffraction. Yield: 0.275 g (95%). Mp. 163e164 ꢁC; 1H NMR (C6D6)
d
(CH]), 127.0 (m-arylC), 115.0 (]CH2), 35.5 (CH2), 20.8 (CH3); the
remaining quaternary aryl carbons were not observed; MS (EI) m/z
1214 ([La2(DALP)5þ] þ 1) amu; Anal. Calcd for C78H90O6La2: C, 66.85;
H, 6.47%; Found: C, 66.03; H, 6.63%. The elemental analysis on this
compound was repeated several times but this is the best result that
was obtained.
d
6.87 (s, 12H, arylH), 5.98 (m, 12H, CH]), 5.08 (m, 24H, ]CH2
overlapping), 3.50 (d, 24H, CH2), 2.15 (s, 18H, CH3); 1H NMR
(C6D6:C4H8O, 5:1) 6.85 (s, 12H, arylH), 5.97 (m, 12H, CH]), 4.92
(m, 24H, ]CH2 overlapping), 3.42 (d, 24H, CH2), 2.05 (s, 18H, CH3);
13C{1H} NMR (C6D6)
140.1 (very broad, CH]), 128.9 (m-arylC),
d
2.3.2. {Ce[DALP]2}2[m-DALP]2 (4)
d
Complex 4 was prepared using a procedure analogous to 3 from
Ce[N(SiMe3)2]3 (0.250 g, 0.403 mmol) and 3 equiv of 1. Deep yellow
plates of 4 were isolated from hexane solution. Yield: 0.155 g (55%).
116.5 (very broad, ]CH2), 36.7 (CH2), 20.8 (CH3); the quaternary
aryl carbons were not observed; 13C{1H} NMR (C6D6:C4H8O, 5:1)
d
158.0 (arylCO), 139.1 (CH]), 124.7 (m-arylC), 114.9 (]CH2), 35.2
(CH2), 20.9 (CH3); the remaining quaternary aryl carbons were not
observed; 13C DEPT-135 NMR (C7D8, 298 K)
139.8 (þ, very broad,
Mp. 120e122 ꢁC; 1H NMR (C7D8, 297 K)
d 9.22 (24H, CH2,
n1/2 ¼ 140 Hz), 2.66 (18H, CH3, n1/2 ¼ 400 Hz), 1.33 (12H, CH],
d
n1/2 ¼ 160 Hz), ꢀ3.29 (12H, CH], n1/2 ¼ 780 Hz), ꢀ7.19 (12H, CH],
CH]), 128.9 (þ, m-arylC), 116.7 (ꢀ, broad, ]CH2), 35.8 (ꢀ, CH2),
n1/2 ¼ 900 Hz), arylH not observed; 1H NMR (C7D8, 377 K)
d
6.28
20.7 (þ, CH3); 13C DEPT-135 NMR (C7D8, 203 K)
d
143.3 (þ,
(24H, CH2, n1/2 ¼ 60 Hz), 7.21 (12H, arylH, n1/2 ¼ 56 Hz), 2.44 (12H,
CH], n1/2 ¼ 48 Hz), 2.35 (18H, CH3, n1/2 ¼ 23 Hz), ꢀ0.65 (12H, CH],
n1/2 ¼ 38 Hz), ꢀ2.42 (12H, CH], n1/2 ¼ 40 Hz); Anal. Calcd for
C78H90O6Ce2: C, 66.74; H, 6.46%; Found: C, 66.34; H, 6.71%.
terminal-OAr CH]),136.4 (þ, bridge-OAr CH]),130.2 (þ, terminal-
OAr m-arylC), 128.8 (þ, bridge-OAr m-arylC), 119.9 (ꢀ, bridge-OAr
]CH2), 117.6 (ꢀ, terminal-OAr ]CH2), 37.6 (ꢀ, terminal-OAr CH2),
35.4 (ꢀ, bridge-OAr CH2), 27.0 (þ, bridge-OAr CH3), 21.9 (þ,
terminal-OAr CH3); Anal. Calcd for C78H90O6Y2: C, 71.99; H, 6.97%;
Found: C, 71.55; H, 6.92%.
2.3.3. {Nd[DALP]2}2[m-DALP]2 (5)
Complex 5 was prepared using a procedure analogous to 3 from
Nd[N(SiMe3)2]3 (0.277 g, 0.443 mmol) and 3 equiv of 1. Deep blue
prisms of 5, suitable for X-ray crystallography, were isolated from
hexane solution. Yield: 0.211 g (68%). Mp. 158e160 ꢁC; MS (EI) m/z
1226 ([Nd2(DALP)þ5 ] þ 1) amu showing the extensive isotope
2.3.7. {La[MALP]2}2[m-MALP]2 (9)
Complex 9 was prepared using a procedure analogous to 3 from
La[N(SiMe3)2]3 (0.085 g, 0.14 mmol) and 3 equiv of 2. Colorless
crystals of 9 were isolated from hexane solution. Yield: 0.058 g
pattern of Nd; 1H NMR (C7D8, 293 K)
d
16.76 (24H, CH2,
(67%). Mp. 144e147 ꢁC; 1H NMR (C6D6)
d 6.82 (s, 6H, arylH), 6.71 (s,
n1/2 ¼ 500 Hz), 3.7 (18H, CH3, n1/2 ¼1300 Hz), ꢀ3.46 (12H, CH],
6H, arylH), 5.89 (m, 6H, CH]), 5.02 (d, 6H, trans- ]CH2), 4.94 (m,
6H, cis- ]CH2), 3.31 (br m, 12H, CH2), 2.43 (s, 18H, CH3), 2.15 (s, 18H,
n1/2 ¼ 250 Hz), remaining alkene H and arylH not observed; 1H NMR
(C7D8, 373 K)
d
10.04 (24H, CH2, n1/2 ¼ 96 Hz), 7.83 (12H, arylH,
CH3); 13C{1H} NMR (C6D6)
d 143 (very broad, CH]), 130.5 (arylCH),
n1/2 ¼ 160 Hz), 2.77 (18H, CH3, n1/2 ¼ 64 Hz), 0.18 (12H, CH],
n1/2 ¼ 80 Hz), ꢀ3.39 (12H, CH], n1/2 ¼ 120 Hz), ꢀ6.46 (12H, CH],
n1/2 ¼ 145 Hz); Anal. Calcd for C78H90O6Nd2: C, 66.35; H, 6.42%;
Found: C, 65.84; H, 6.15%.
128.6 (arylCH), 116.5 (broad ]CH2), 36.4 (very broad, CH2), 20.6
(CH3), 17.9 (CH3); Anal. Calcd for C66H78O6La2: C, 63.67; H, 6.31%;
Found: C, 64.24; H, 6.78%. This is the best elemental analysis
obtained from several trials.
2.3.4. {Er[DALP]2}2[
m-DALP]2 (6)
2.3.8. {Sm[MALP]2}2[m-MALP]2 (10)
Complex 6 was prepared using a procedure analogous to 3 from Er
[N(SiMe3)2]3 (0.287 g, 0.443 mmol) and 3 equiv of 1. Well-formed,
bright pink cubes of 6, suitable for X-ray crystallography, were iso-
lated from hexane solution. Yield: 0.23 g (71%). Mp. 162e163 ꢁC; 1H
Complex 10 was prepared using a procedure analogous to 3
from Sm[N(SiMe3)2]3 (0.125 g, 0.200 mmol) and 3 equiv of 2. Small,
colorless crystals of 10 were isolated from hexane solution. Yield:
0.077 g (61%). Mp. 112e116 ꢁC; 1H NMR (C6D6)
d 6.45 (br s, 6H,
NMR (C7D8, 298 K)
d
189.4 (16H, n1/2 ¼ 3500 Hz), 117.3 (8H,
arylH), 6.35 (br s, 6H, arylH), 4.74 (br s, 12H, ]CH2), 4.23 (br s, 6H,
CH]), 3.27 (br s, 18H, CH3), 1.81 (br s, 18H, CH3); the allyl CH2
resonance was not observed but may lay under the 1.81 ppm
resonance based on integrated intensities; Anal. Calcd for
C66H78O6Sm2: C, 62.51; H, 6.20%; Found: C, 62.77; H, 6.40%.
n1/2 ¼ 1000 Hz), 108.6 (8H, n1/2 ¼ 2300 Hz), 73.3 (8H, n1/2 ¼ 850 Hz),
45.3 (4H, n1/2 ¼1200 Hz), 8.4 (4H, n1/2 ¼1000 Hz), ꢀ11.0 (4H,
n1/2 ¼ 1200 Hz), ꢀ28.0 (8H, n1/2 ¼ 320 Hz), ꢀ47.1 (6H, n1/2 ¼ 400 Hz),
ꢀ68.1 (12H, n1/2 ¼1200 Hz), ꢀ84.6 (8H,
n1/2 > 5000 Hz), one reso-
nance of relative integration 4H was not observed; Anal. Calcd for
C78H90O6Er2: C, 64.25; H, 6.22%; Found: C, 64.17; H, 6.22%.
2.3.9. {Y[MALP]2}2[m-MALP]2 (11)
Complex 11 was prepared using a procedure analogous to 3 from
Y[N(SiMe3)2]3 (0.115 g, 0.200 mmol) and 3 equiv of 2. Colorless
cubes of 11, suitable for X-ray crystallography, were isolated by
repeated crystallization from hexane solution. Yield: 0.038 g (33%).
2.3.5. {Yb[DALP]2}2[m-DALP]2 (7)
Complex 7 wasprepared usinga procedure analogous to 3 from Yb
[N(SiMe3)2]3 (0.290 g, 0.443 mmol) and 3 equiv of 1. Yelloweorange,
flaky plates of 7 deposited from hexane solution. Yield: 0.055 g (17%).
Mp. 167e170 ꢁC; 1H NMR (C6D6)
d 6.82 (s, 6H, arylH), 6.71 (s, 6H,

2706
D.J. Berg et al. / Journal of Organometallic Chemistry 695 (2010) 2703e2712
Table 1
Summary of crystallographic data.a
5
6
8
10
11
Formula
fw
Cryst syst
Space group
a (Å)
C
78H90O6Nd2
C78H90O6Er2
1458.0
Monoclinic
P21/n (No. 14)
13.178(2)
17.357(3)
16.233(3)
90
109.265(3)
90
3505.2(10)
2
C78H90O6Y2
1301.3
Monoclinic
P21/n (No. 14)
13.183(2)
17.327(2)
16.241(2)
90
109.159(3)
90
3504.3(8)
2
C66H78O6Sm2
1268.0
C66H78O6Y2
1145.1
Triclinic
P ꢀ 1 (No. 2)
10.654(2)
12.208(2)
12.803(2)
115.060(2)
96.102(3)
93.136(3)
1490.7(4)
1
1412.1
Monoclinic
P21/n (No. 14)
13.292(3)
17.272(4)
16.407(4)
90
109.006(3)
90
3561.6(13)
2
Triclinic
P ꢀ 1 (No. 2)
10.689(3)
12.237(4)
12.806(4)
114.441(5)
95.824(5)
93.432(5)
1507.5(8)
1
b (Å)
c (Å)
a
b
g
(deg)
(deg)
(deg)
V (Å3)
Z
rcalc (g cmꢀ3
)
1.32
1.49
1.38
2.43
1.23
1.70
1.40
1.98
1.28
1.99
m
(mmꢀ1
)
2
qmax (deg)
50
50
50
50
50
meas. refl.
unique refl.
18,483
6280
18,299
6184
18,486
6191
11,228
5289
11,132
5222
refl. I > 2
s(I)
4727
4241
3177
4953
4255
F000
1452
0.038, 0.100
0.054
1484
0.030, 0.071
0.059
1368
0.050, 0.110
0.119
646
0.023, 0.058
0.025
600
0.036, 0.092
0.049
R,b Rw
c
Rb (all data)
GOF
1.03
0.92
0.89
1.04
0.98
a
Collected using a Smart 1000 CCD system (graphite-monochromated Mo Ka l ¼ 0.71073 Å at 223 K).
b
c
R ¼ S(jFoj ꢀ jFcj)/SjFoj.
2
Rw ¼ [Sw(jFoj ꢀ jFcj )/Sw(jFoj)2]1/2
.
arylH), 5.89 (m, 6H, CH]), 5.01 (m, 6H, trans- ]CH2), 4.95 (m, 6H,
cis- ]CH2), 3.31 (broad m, 12H, CH2), 2.42 (s, 18H, CH3), 2.14 (s, 18H,
3. Results and discussion
3.1. Complex syntheses and solid state structures
CH3); 13C{1H} NMR (C6D6)
d 156.2 (broad, arylCO), 143.4 (broad,
CH]), 130.6 (arylCH), 128.7 (arylCH), 126.1 (broad, arylC), 116.5
(broad, ]CH2), 36.4 (CH2), 20.7 (CH3), 18.0 (CH3); Anal. Calcd for
C66H78O6Y2: C, 69.22; H, 6.87%; Found: C, 68.88; H, 6.73%.
Reaction of 3 equivalents of 2,6-diallyl-4-methylphenol (H-
DALP) or 2-allyl-4-methylphenol (H-MALP) with Ln[N(SiMe3)2]3
afforded the dimeric tris(phenoxide) complexes {Ln[DALP]2}2[
DALP]2 (Ln ¼ La (3), Ce (4), Nd (5), Er (6), Yb (7), Y (8)) or {Ln
[MALP]2}2[
(Scheme 1). In the case of the colorless complexes of yttrium and
lanthanum, initial mixing of the phenol with Ln[N(SiMe3)2]3
resulted in a green colored solution that faded to colorless after
m-
2.4. X-ray crystallographic studies
m
-MALP]2 (Ln ¼ La (9), Sm (10), Y (11)), respectively
Crystals suitable for X-ray diffraction were grown from a hot,
saturated hexane solution by slow cooling to room temperature in
the glovebox. Crystals were mounted on a glass fiber and sealed in
epoxy prior to analysis. Data was collected at 223(1) K using
a Bruker SMART 1000 instrument (Mo K
a
radiation,
l
¼ 0.71073 Å).
Data reduction and correction for Lorentzian polarization and
decay were performed using the SHELXTL software [27]. Absorp-
tion corrections were applied using SADABS [28]. The structure was
solved by direct methods and refined by least squares method on F2
using SHELXS-97 [29]. All non-hydrogen atoms were refined
anisotropically and hydrogen atoms refined isotropically using
a riding model. Disorder was observed in some of the non-coor-
dinating allyl groups in each structure. This was modeled using
major and minor fragments in approximately 80:20 ratio. Details of
the data collection and refinement are given in Table 1.
Scheme 1.
Fig. 1. ORTEP3 plot (30% probability) of {Nd[DALP]2}2[m-DALP]2 (5) [30].

D.J. Berg et al. / Journal of Organometallic Chemistry 695 (2010) 2703e2712
2707
Table 3
Bond angles around Ln (deg) for complexes 5, 6, 8, 10 and 11.a
5 (Nd)
6 (Er)
8 (Y)
10 (Sm)
11 (Y)
b
O
T1-Ln-OT2
135.77(3)
137.49(13) 136.51(12) 134.82(8)
135.51(9)
107.57(8)
110.57(8)
109.19(7)
105.60(7)
72.01(7)
OT1-Ln-OB
OT1-Ln-OB
OT2-Ln-OB
OT2-Ln-OB
103.64(11) 101.34(12) 112.36(11) 108.27(7)
111.00(11) 111.32(12) 101.68(11) 105.99(7)
110.10(11) 110.20(11) 104.89(11) 107.55(8)
106.87(11) 104.53(12) 110.63(11) 111.31(6)
0
0
0
OB-Ln-OB
70.19(10)
72.45(13)
72.44(12)
70.48(7)
Ct1-Ln-Ct2c 106.73
105.25
74.00
76.22
93.62
165.55
85.12
74.23
161.10
88.66
105.57
73.69
76.14
93.28
165.13
84.88
73.88
161.21
90.55
103.08
73.55
78.45
95.26
165.67
89.69
82.92
161.31
91.23
101.89
70.70
82.15
91.95
163.67
77.27
73.44
166.16
94.15
Ct1-Ln-OT1
Ct1-Ln-OT2
Ct1-Ln-OB
73.35
77.22
92.79
162.94
86.17
71.57
160.07
90.17
0
Ct1-Ln-OB
Ct2-Ln-OT1
Ct2-Ln-OT2
Ct2-Ln-OB
0
Ct2-Ln-OB
Ln-OB-Ln0
108.81(10) 107.55(13) 107.56(12) 109.56(12) 107.99(7)
a
Complete tables of bond angles including specific labels can be found in the
Supplementary Material.
b
OT1 and OT2 refer to the terminal aryloxide-O; OB and OB0 refer to the symmetry-
related bridging aryloxide-O pair.
c
Ct1 and Ct2 refer to the centroids (midpoints) of the two unique bound alkene
carbons.
Fig. 2. ORTEP3 plot (30% probability) of {Sm[MALP]2}2[m-MALP]2 (10) [30].
shown for 5 in Fig. 1 [31]. Similarly, the MALP complexes of Sm (10)
and Y (11) are isostructuraland crystallize in the triclinic space group
P ꢀ 1; a representative structure of this type is shown for 10 in Fig. 2.
Selected bond lengths and angles for all five complexes are given in
Tables 2 and 3, respectively. The solid state structures of the DALP
and MALP complexes are quite similar and are probably best
regarded as displaying a distorted octahedral geometry at the metal
center with the alkene centroids occupying positions trans to one
bridging aryloxide-O (Ct-Ln-OB-trans angle range: 160.07e166.16ꢁ,
Table 3). However, this geometry is obviously highly distorted as the
angle between the terminalaryloxide-O atoms deviates greatly from
180ꢁ (range: 134.82(8)e137.49(13)ꢁ) and the angle between the
terminal and bridging aryloxide-O is far wider than 90ꢁ (range:
101.34(12)e112.36(11)ꢁ). The distortion may be viewed as from
octahedral towards tetrahedral with the alkene ligands being forced
out of the inner coordination sphere. That being said, the distance
between the alkenyl carbons of one allyl group of each terminal
aryloxide and the metal center (LneCterm range: 2.928(5)e3.140(4)
Å; LneCint range: 3.123(3)e3.288(4) Å, Table 2) are far shorter than
the sum of the Ln and carbon van der Waals radii in all five structures
(van der Waals Ln/C distances range from 3.6 to 3.7 Å for Y to La,
assuming rVDW for C is 1.5 Å) [32]. The carbon atoms of the other allyl
group on the terminal aryloxide of the DALP complexes (5, 6 and 8)
are not within the sum of van der Waals radii and the same is true of
all allyl group carbons on the bridging aryloxides in both the DALP
and MALP structures. The lanthanide 2,6-diisopropylphenoxide ate
about 15 min, possibly suggesting the formation of transient DALP
or MALP phenoxyl radicals [30]. The complexes were recrystallized
from hot hexane to afford crystals suitable for X-ray crystallography
in most cases. However, crystals of 4 (Ce) and 7 (Yb) were of lower
quality. This may indicate the presence of some impurities as these
two complexes also gave consistently lower melting points than the
rest of the series.
The DALP complexes of Nd (5), Er (6) and Y (8) are isostructural in
the monoclinic space group P21/n; a representative structure is
Table 2
Selected bond lengths and bond length differences (Å) for complexes 5, 6, 8, 10 and
11.a
5 (Nd)
6 (Er)
8 (Y)
10 (Sm)
11 (Y)
3.090(3)
3.140(4)
b
LneCterm
3.122(6) 3.122(6) 3.128(5) 3.120(4)
3.024(5) 2.928(5) 2.970(5) 3.099(3)
Bound alkene
c
LneCint
3.185(6) 3.186(6) 3.187(5) 3.149(4)
3.256(5) 3.126(5) 3.153(5) 3.252(3)
3.123(3)
3.288(4)
Bound alkene
Difference
ꢀ0.063
ꢀ0.232
ꢀ0.060
ꢀ0.198
ꢀ0.059
ꢀ0.183
ꢀ0.029
ꢀ0.153
ꢀ0.033
ꢀ0.148
D
¼ (LneCterm
)
ꢀ (LneCint
)
Coordinated C]C
1.311(2) 1.314(7) 1.323(2) 1.279(5)
1.331(7) 1.314(7) 1.310(6) 1.294(5)
1.296(5)
1.297(5)
Free C]C
Terminal-OAr
1.274(2) 1.275(4) 1.263(2) 1.267(7)
1.294(2) 1.272(2) 1.289(2)
1.272(7)
complexes, [La(OAr)2(THF)(
m
-OAr)2]ꢀ[M(THF)n]þ (M ¼ Li, n ¼ 1;
M ¼ Na, n ¼ 2) and [Ln(OAr)4]ꢀMþ (Ln ¼ Nd, M ¼ K; Ln ¼ La, M ¼ Cs),
are interesting in this regard because the saturated isopropyl group
is similar to an allyl group yet in this case, the distance from the
lanthanide ion to all three isopropyl carbons lies well outside the
sum of the van der Waals radii (Ln/C distance > 3.65 Å in all cases
and usually >4 Å) [33,34]. Interestingly, these ate complexes do
Free C]C
Bridging OAr
1.300(2) 1.295(2) 1.293(2) 1.271(10) 1.287(11)
1.262(8) 1.255(8) 1.256(2)
Ln-OAr
Terminal-OAr
2.147(3) 2.065(3) 2.064(3) 2.121(2)
2.160(3) 2.076(3) 2.075(3) 2.135(2)
2.066(2)
2.074(2)
Ln-
Ln-
m
m
-OAr
-O0Ar
2.345(3) 2.244(3) 2.273(3) 2.319(2)
2.371(3) 2.256(3) 2.254(3) 2.319(2)
2.260(2)
2.248(2)
feature p-interactions but these are between the arene rings and the
alkali metal cations not the lanthanide ions.
Bridge asymmetry 0.026
j(Ln-OAr)
0.012
0.019
0.0
0.012
The interaction between the allyl alkene carbons and the metal
is distinctly asymmetric with the terminal ]CH2 group approach-
ing the metal center more closely than the internal CH] group by,
on average, 0.12 Å (range: 0.029e0.232 Å, Table 2). It is interesting
to note that in all five structures the coordinated alkene C]C
ꢀ (Ln-O0Ar)j
a
Complete tables of bond lengths including specific labels can be found in the
Supplementary Material.
b
Cterm refers to the terminal ]CH2 carbon of the allyl chain.
Cint refers to the internal CH] carbon of the allyl chain.
c

2708
D.J. Berg et al. / Journal of Organometallic Chemistry 695 (2010) 2703e2712
distance is slightly longer than the non-coordinated alkene units,
whether part of the bridging or terminal aryloxide ligand, in the
same molecule (Table 2). The difference in coordinated and free
alkene bond lengths is much more pronounced in the case of the
DALP complexes compared with those containing the MALP ligand.
It is not obvious why this should be the case and the fact that there
is some disorder in the free allyl groups of some structures suggests
that these differences should be viewed with caution.
The aryloxide bonding is largely unexceptional. As expected, the
LneO bonds for the terminal aryloxide are about 0.2 Å shorter than
the bridging aryloxide LneO bonds (Table 2). The aryloxides bridge
the two lanthanide centers in a symmetrical fashion (asymmetry
range: 0.00e0.026 Å). Both the bridging and terminal LneO bond
distances are somewhat shorter than the distances normally
observed in 6-coordinate lanthanide aryloxide dimers. For example,
when all distances are corrected to an equivalent 6-coordinate Y3þ
radius of 0.90 Å [35], the terminal and bridging LneO distances in
the DALP and MALP complexes fall in the ranges 2.063e2.086 and
2.248e2.288 Å, respectively, while the literature complexes span
the ranges 2.103e2.138 and 2.272e2.426 Å for bridging and
terminal LneO distances [36e40]. The only complexes with
comparably short distances are [Yb(ONaph)2(THF)(NCCH3)]2[m-
ONaph]2
C6H3(CH2Ph)2)2]2[
the former complex, the shorter LneO distance is explained by the
(ONaph ¼ 1-naphthoxide)
and
[Ln(O-2,6-
m
-O-2,6-C6H3(CH2Ph)2]2 (Ln ¼ La, Yb) [41,42]. In
smaller CH3CN ligand. However, the latter pair of complexes possess
secondary
oxide benyzl groups and the lanthanide centers (
case of La and both
3 and 2 in the case of Yb) [42]. As observed in
this work, these interactions are weak and the bonds long,
p
-arene interactions between one of the bridging aryl-
h
6-arene in the
Fig. 3. Observed and simulated 1H NMR (360 MHz) spectra for the m-aryl proton
region of {Y[DALP]2}2[ -DALP]2 (8) between 333 and 255 K in d14-methylcyclohexane.
h
h
m
p
resulting in less steric crowding at the lanthanide ion and a closer
approach of the aryloxide ligands.
range from 2.94 to 3.14 Å, more than 0.2 Å longer. Similarly, the Sr2þ
and Ba2þ complexes (IIIbec) predict comparable distances (range:
2.71e2.76 Å) [24]. However, the Ca2þ complex (IIIa) predicts
somewhat longer distances of 2.85e2.95 Å and the Mg2þ complex
shows no alkene coordination in the solid state at all. These results
suggest that steric pressure plays a role and it seems likely that the
longer LneC distances observed in the MALP and DALP complexes is
a reflection of weaker alkene bonding due to increased steric
crowding for the smaller trivalent lanthanide ions with a larger
There are very few structurally characterized lanthanide-
ealkene complexes with which to compare the complexes reported
here. The majority of these involve divalent lanthanide complexes
containing Cp ligands with pendant alkene groups (I and II) [22,23],
although there is one example of an alkyl complex with a coordi-
nated tethered alkene (IV, Chart 1) [25]. A series of alkaline earth
alkene complexes (III) that are closely related to the divalent
lanthanide derivatives II have also been reported (Chart 1) [24]. The
bridging ethylene derivative (C5Me5)2Yb(m-C2H4)Pt(PPh3)2 was
aryloxide ligand. It is noteworthy that the Lnearene p-interactions
reported several years ago but since backbonding from Pt clearly
dominates the structural parameters, this complex is not compa-
rable to those reported here [21]. There are also two reports of
lanthanide 1,4-diphenylbutadiene complexes but in both cases, the
diene ligand is clearly reduced and cannot be compared with the
neutral alkene complexes discussed here [43,44]. The only trivalent
lanthanideealkene complex that has been structurally character-
ized is the [(C5Me4SiMe2CH2CH¼CH2)2Sm]þ cation of Id (Chart 1)
[22]. All of these complexes, with the exception of IV, also show
shorter terminal vs. internal LneC distances by ca. 0.2 Å on average
compared with an average value of 0.12 Å here. In contrast, the
alkylealkene complex IV shows a significantly closer internal vs.
terminal YbeC distance
(
D
¼ (YbeCterm) ꢀ (YbeCint) ¼ þ0.30 Å).
Given the very different type of tethering group, it seems likely that
whether the internal or terminal carbon approaches the metal most
closely is a function of the nature of the tether rather than any
intrinsic preference of the metal.
The terminal LneC distances in the DALP and MALP complexes
described here differ considerably from those found in complexes I,
II and IV. When the metal ionic radii are corrected to that of six-
coordinate Y3þ according to Shannon [35], the divalent complexes I
and II, and the trivalent cationic complex Id, predict LneCterm
distances in the range 2.73e2.82 Å [22,23]. In fact, the corrected
distances in the aryloxide tethered alkene complexes discussed here
Fig. 4. Eyring plot for the bridgeeterminal aryloxide exchange process of {Y
[DALP]2}2[ -DALP]2 (8) in d14-methylcyclohexane.
m

D.J. Berg et al. / Journal of Organometallic Chemistry 695 (2010) 2703e2712
2709
the observed and simulated spectra for this resonance between 333
and 255 K [45]. An Eyring plot for this resonance, shown in Fig. 4,
yiezlded
D
Hz ¼ 40 ꢂ 1 kJ molꢀ1
,
D
Sz ¼ ꢀ59 ꢂ 4 J molꢀ1 Kꢀ1 and
D
G298 ¼ 58 ꢂ 3 kJ molꢀ1. The Eyring plot for the same process and
resonance of 3, shown in Fig. 5, gave
D
Hz ¼ 37.8 ꢂ 0.6 kJ molꢀ1
,
D
Sz ¼ ꢀ51 ꢂ3 J molꢀ1 Kꢀ1 and
D
Gz298 ¼ 53 ꢂ 2 kJ molꢀ1. The nega-
tive entropy term for bridgeeterminal aryloxide exchange is
noteworthy since it argues against cleavage of the bridge during
exchange. One possible explanation for the negative entropy term
is that exchange occurs by way of a crowded, triple aryloxide bridge
like that shown in Eq (2). This would no doubt involve alkene
dissociation (see evidence for this below) which would make
a positive entropy contribution, but the crowding of such an
intermediate would likely prevent aryloxide rotation and the
increased rigidity would be expected to result in an overall negative
entropy term. The marginally lower barrier to bridgeeterminal
aryloxide exchange for the La complex 3 vs. Y complex 8 could also
be taken as evidence for the process shown in Eq (2) given the
considerably larger size of La3þ
.
Fig. 5. Eyring plot for the bridgeeterminal aryloxide exchange process of {La
[DALP]2}2[ -DALP]2 (3) in d14-methylcyclohexane.
m
observed in [Ln(O-2,6-C6H3(CH2Ph)2)2]2[m-O-2,6-C6H3(CH2Ph)2]2
(Ln ¼ La, Yb) give similar Ln-
p
-arene C distances after correction for
differences in radii (estimated range 2.9e3.2 Å) [42].
The median bond C]C bond length in complexes IeIV is 1.32 Å,
very similar to the 1.31 Å found in the present work, but in both the
literature structures and those reported here, the coordinated C]C
bond length covers avery wide range (1.267(11)e1.361(2) Å [22e25];
Table 2). This variability makes it difficult to state with confidence
whether the small increase in average C]C bond length for the
coordinated vs. non-coordinated alkenes is real but it is clear that any
changes in bond length that do occur on coordination are minor.
3.2.2. Evidence for weak alkene binding
On further cooling below 255 K, the 1H NMR of 8 in d14-meth-
ylclohexane again begins to broaden and the terminal DALP reso-
nances eventually decoalesce just before the solubility limit of the
complex at 188 K. While there is significant broadening of all reso-
nances at this temperature due to viscosity effects, the larger m-aryl
proton resonance observed at 255 K splits into two resonances of
equal intensity by 188 K. This is consistent with the solid state
structure and indicates that at this temperature alkene exchange is
slow. The observed and simulated spectra for this process and the
corresponding Eyring plot are shown in Figs. 6 and 7, respectively.
3.2. Solution behavior of the diamagnetic complexes
In view of the alkene coordination observed in the solid state,
we sought evidence for alkene bonding in solution by NMR spec-
troscopy. At room temperature, the diamagnetic complexes 3 and 8
show a single set of resonances for the DALP ligands in deuterated
benzene, toluene or methylcyclohexane. This suggests either that
the complexes exist as monomers or are undergoing rapid
bridgeeterminal aryloxide exchange. On cooling below room
temperature in either d8-toluene or d14-methylcyclohexane, the 1H
NMR resonances of either 3 or 8 broaden and eventually decoalesce
into two sets of DALP resonances in 2:1 ratio, clearly indicating that
these complexes are dimeric in non-coordinating solvents at room
temperature. No concentration effects are observed over at least
a five-fold concentration range, consistent with an intact dimer.
However, the addition of a small amount of d8-THF (5:1 d8-tolu-
ene:d8-THF) results in sharp spectra with a single set of DALP
resonances down to at least 200 K, clearly indicating that THF
cleaves the dimeric structure into a monomer (Eq (1)).
Values of
D
D
Hz ¼ 50 ꢂ 4 kJ molꢀ1
,
D
Sz ¼ þ46 ꢂ 15 J molꢀ1 Kꢀ1 and
Gz298 ¼ 36 ꢂ 15 kJ molꢀ1 were obtained from Fig. 7. Although these
activation parameters are prone to considerable error, the entropy
term obtained for this process is so substantially positive that we
believe it is significant. If so, it indicates a dissociative alkene
exchange process and the barrier observed would represent the sum
of the barrier to alkene dissociation and aryloxide rotation.
It might be expected that alkene bonding would be detectable by
chemical shift differences between alkene proton or carbon reso-
nances in the complex relative to free alkene by NMR spectroscopy.
3.2.1. Dynamic bridgeeterminal aryloxide exchange
Bridgeeterminal aryloxide exchange was examined in more
detail for 8 in d14-methylcyclohexane. The dynamic behavior of the
m-aryl proton resonance was the most easily followed. Fig. 3 shows
Fig. 6. Observed and simulated 1H NMR (360 MHz) spectra for the m-aryl proton
region of {Y[DALP]2}2[m-DALP]2 (8) between 255 and 188 K in d14-methylcyclohexane.

2710
D.J. Berg et al. / Journal of Organometallic Chemistry 695 (2010) 2703e2712
Fig. 9. 1H NMR Chemical shift
d vs. 1/T for {Nd[DALP]2}2[m-DALP]2 (5) in 5:1 d8-tolu-
ene:d8-THF (360 MHz).
Fig. 7. Eyring plot for the terminal aryloxide alkene exchange process of {Y
[DALP]2}2[m-DALP]2 (8) in d14-methylcyclohexane.
to alkene coordinationwith the terminal carbon shifting upfield and
the internal carbon downfield byseveral ppm. Forexample, the Cp2Y
Casey et al., for example, have shown that the 1H NMR chemical shift
difference, Dd, between the terminal ]CH2 protons in Cp2Y(h1 h2
(
h1
:
h2-CH2CH2CR2CR0¼CH2) (R ¼ H, Me; R0 ¼ H, Me) complexes
:
-
CH2CH2CR2CR0¼CH2) (R ¼ H, Me; R0 ¼ H, Me) is between about 0.6
and 1.5 ppm in the coordinated alkene complex but <0.1 ppm for
either the free alkene equivalent or the complex in the presence of
d8-THF (where alkene displacement by THF is assumed) [12e17].
However, in the diamagnetic complexes 3, 8, 9 and 11 at room
temperature in d8-toluene, Dd is <0.1 ppm. Similarly, other lantha-
nideealkene complexes I, II and the alkaline earth complexes III also
show Dd < 0.1 ppm in non-coordinating solvents [22e24]. The
alkene carbon 13C NMR chemical shifts, however, are more sensitive
reported by Casey et al, show an increase in the coordinated alkene
internaleterminal 13C chemical shift difference of, on average, ca.
17 ppm (DDdc range: 13.8e25.0 ppm) relative to the free ligand
[12e16]. The divalent Yb complexes I and II show DDdc values of 9.6
and 18.1 ppm, respectively, while the related alkaline earth
complexes III have an average DDdc value of 9.7 ppm (range:
8.9e10.9 ppm) [22e24]. At room temperature, the diamagnetic
DALP complexes of La (3) and Y (8) show smaller DDdc values of 2.7
and 2.5 ppm, respectively, while this difference is somewhat larger
Fig. 8. 1H NMR Chemical shift
d vs. 1/T for {Nd[DALP]2}2[m-DALP]2 (5) in d8-toluene (360 MHz).

D.J. Berg et al. / Journal of Organometallic Chemistry 695 (2010) 2703e2712
2711
In contrast to 4 and 5, Er2(DALP)6 (6) shows evidence for up to
12 distinct resonances throughout the temperature range studied
(253e357 K) indicating that bridgeeterminal aryloxide exchange
remains slow on the NMR timescale. This is a reflection of the large
pseudo-contact shifts observed for Er3þ as evident from the total
chemical shift range of >250 ppm at 298 K. In the presence of d8-
THF, the spectra again simplify to only 6 resonances indicating
formation of a monomeric THF adduct, Er(DALP)3(THF)n (Fig. 10).
Interestingly, the resonances of this THF adduct show distinct
curvature between 293 and 353 K. This may be due to an equilib-
rium between species with different numbers of coordinated THF
molecules, Eq (3).
4. Conclusions
The ortho-allyl substituted aryloxide ligands DALP and MALP
form dimeric complexes with trivalent lanthanides and yttrium
that contain weak Lnealkene bonds in the solid state with allyl
groups on the terminal aryloxides. The Lnealkene structural
parameters indicate, in common with the small number of reported
lanthanide and alkaline earth alkene complexes, that the terminal
alkene carbon approaches the metal more closely than the internal
alkene carbon. However, bond length comparisons clearly indicate
that the LneC distances observed in our work are significantly
longer than those reported in IeIV after correction for differences
in ionic radii.
The dimeric complexes undergo bridgeeterminal aryloxide
exchange at room temperature in solution for all complexes except
that of Er. In the case of the diamagnetic complex Y2(DALP)6,
a second dynamic process consistent with alkene exchange was
observed at very low temperature. This fact, coupled with the
observation of a significant increase in the chemical shift difference
between the terminal and internal alkene carbons relative to free
alkene for the terminal but not bridging aryloxide ligands, strongly
suggests that the Lnealkene interaction persists in non-coordi-
nating solvent such as toluene or methylcyclohexane. In contrast,
the complex is completely cleaved to a monomer in the presence of
THF and shows no evidence for alkene coordination in this solvent
as might be expected.
Fig. 10. 1H NMR Chemical shift
ene:d8-THF (360 MHz).
d vs. 1/T for {Er[DALP]2}2[m-DALP]2 (6) in 5:1 d8-tolu-
in the corresponding MALP complexes (9, 5.7; 11, 6.0 ppm). The 13
C
NMR spectrum of 8 recorded at 203 K shows two distinct sets of
resonances in 2:1 ratio for the terminal and bridging aryloxides,
respectively. At 203 K, the alkenyl carbon DDdc value for the terminal
aryloxide increases by 4.7 ppm while that for the bridging aryloxide
decreases by ꢀ4.6 ppm. At this temperature, alkene exchange is still
rapid in the 13C NMR spectrum, so the observed DDdc value is an
average of that for free (by definition DDdc ¼ 0) and coordinated
alkene. In order to properlycompare this parameter tothat observed
in other alkene complexes mentioned above, it therefore must be
doubled to 9.4 ppm, a value that is very similar to that observed for I
and III [22,24], but smaller than those observed in the tethered
alkylealkene complexes [12e16].
3.3. Paramagnetic complexes: variable temperature 1H NMR
The paramagneticDALPcomplexes4 (Ce), 5(Nd), 6 (Er) and7 (Yb)
were studied by variable temperature 1H NMR in both d8-toluene
and 5:1 d8-toluene:d8-THF between about 210 and 380 K. Plots of
d
vs. 1/T in d8-toluene and 5:1 d8-toluene:d8-THF are shown for
Nd2(DALP)6 (5) in Figs. 8 and 9, respectively; complete tables of NMR
chemical shifts including line widths at half height and vs.1/T plots
Acknowledgment
d
are given for 4, 6 and 7 in the Supplementary Material. In the case of
5, decoalescence for bridgeeterminal aryloxide exchange occurs
between ca. 275 and 295 K, depending on the separation of the
resonances involved (Fig. 8). There is no evidence for decoalescence
due to the alkene exchange process at lower temperatures but the
extremely broad line widths, in some cases >2000 Hz, below about
250 K make it impossible to discern the onset of this second process.
Addition of d8-THF simplifies the spectra to 6 resonances and
completely alters the variable temperature behavior (Fig. 9). The
The support of the Natural Sciences and Engineering Research
Council of Canada is gratefully acknowledged.
Appendix A. Supplementary material
Part 1: Tables of atomic coordinates, bond distances and angles
and anisotropic thermal parameters for 5, 6, 8, 10 and 11.
Part 2: Variable temperature 1H NMR data for paramagnetic
complexes 4e7 in d8-toluene and d8-toluene:d8-THF (5:1).
Supplementary data associated with this article can be found, in
the online version, at doi:10.1016/j.jorganchem.2010.09.004.
resonances in Fig. 9 follow fairly linear d vs. 1/T behavior consistent
with formation of a monomeric Nd(DALP)3(THF)n complex at all
temperatures. The behavior of Ce complex 4 is very similar to 5 with
approximately the same decoalescence temperature for bridge-
eterminal aryloxide exchange. The variable temperature 1H NMR
behavior of Yb2(DALP)6 (7) was only examined at room temperature
and above (297e377 K) but a maximum of 6 resonances was
observed indicating that this complex too is in the fast bridge-
eterminal aryloxide exchange region in this temperature range.
References
[1] F.T. Edelmann, Chem. Soc. Rev. 38 (2009) 2253e2268.
[2] B. Corneils, W.A. Hermann (Eds.), Applied Homogeneous Catalysis with
Organometallic Compounds, Wiley-VCH, Weinheim, Germany, 1996.
[3] F.T. Edelmann, Coord. Chem. Rev. 253 (2009) 343e409.
[4] W. Chen, F. Wang, Sci. China, Ser. B: Chem. 52 (2009) 1520e1543.

2712
D.J. Berg et al. / Journal of Organometallic Chemistry 695 (2010) 2703e2712
[5] L. Porri, G. Ricci, A. Giarusso, N. Shubin, Z. Lu, Recent developments in
lanthanide catalysts for 1,3-diene polymerization. in: P. Arjunan, J.E. McGrath,
T.L. Hanlon (Eds.), Olefin Polymerization: Emerging Frontiers, ACS Symposium
Series, vol. 749, 1999, pp. 15e30.
[25] C. Eaborn, P.B. Hitchcock, K. Izod, Z.R. Lu, J.D. Smith, Organometallics 15 (1996)
4783e4790.
[26] D.C. Bradley, J.S. Ghotra, F.A. Hart, J. Chem. Soc., Dalton Trans. (1973)
1021e1023.
[6] W.J. Evans, D.G. Giarikos, J.W. Ziller, Organometallics 20 (2001) 5751e5758.
[7] H.F. Yuen, T.J. Marks, Organometallics27 (2008) 155e158 and references therein.
[8] G. Zi, L. Xiang, H. Song, Organometallics 27 (2008) 1242e1246.
[9] D.V. Vitanova, F. Hampel, K.C. Hultzsch, J. Organomet. Chem. 69 (2007)
4690e4701.
[10] S. Hong, T.J. Marks, Acc. Chem. Res. 37 (2004) 673e686.
[11] S.P. Nolan, T.J. Marks, J. Am. Chem. Soc. 111 (1989) 8538e8542.
[12] C.P. Casey, S.L. Hallenback, D.W. Pollock, C.L. Landis, J. Am. Chem. Soc. 117
(1995) 9770e9771.
[27] G.M. Sheldrick, SHELXTL: v6.10, Structure Determination Software Suite
[Computer Program]. Bruker AXS, Madison, WI, 2001.
[28] G.M. Sheldrick, SADABS: v2.01, Empirical Absorption Correction Program
[Computer Program]. Bruker AXS, Madison, WI, 2001.
[29] G.M. Sheldrick, SHELXS97: Structure Solution Software [Computer Program].
Bruker AXS, Madison, WI, 1997.
[30] I.S. Orlov, A.A. Moiseeva, K.P. Butin, M.P. Egorov, Russ. Chem. Bull., Int. Ed. 51
(2002) 1961e1962.
[31] L.J. Ferrugia, J. Appl. Cryst. 30 (1997) 565e566.
[13] C.P. Casey, S.L. Hallenback, J.M. Wright, C.L. Landis, J. Am. Chem. Soc. 119
(1997) 9680e9690.
[14] C.P. Casey, M.A. Fagan, S.L. Hallenback, Organometallics 17 (1998) 287e289.
[15] C.P. Casey, J.J. Fisher, Inorg. Chim. Acta 270 (1998) 5e7.
[16] C.P. Casey, J. Fisher Klein, M.A. Fagan, J. Am. Chem. Soc. 122 (2000)
4320e4330.
[17] C.P. Casey, T.Y. Lee, J.A. Tunge, D.W. Carpenetti II, J. Am. Chem. Soc. 123 (2001)
10762e10763.
[18] C.P. Casey, J.A. Tunge, T.Y. Lee, D.W. Carpenetti II, Organometallics 21 (2002)
389e396.
[32] S.-Z. Hu, Z.-H. Zhou, B.E. Robertson, Z. Kristallogr. 224 (2009) 375e383.
[33] D.L. Clark, J.G. Watkin, J.C. Huffman, Inorg. Chem. 31 (1992) 1554e1556.
[34] D.L. Clark, R.V. Hollis, B.L. Scott, J.G. Watkin, Inorg. Chem. 35 (1996) 667e674.
[35] R.D. Shannon, Acta Cryst. A 32 (1976) 751e767.
[36] T.J. Boyle, L.J. Tribby, S.D. Bungey, Eur. J. Inorg. Chem. (2006) 4553e4563.
[37] G.B. Deacon, C.M. Forsyth, R. Harika, P.C. Junk, J.W. Ziller, W.J. Evans, J. Mater.
Chem. 14 (2004) 3144e3149.
[38] W.J. Evans, J.M. Olofson, J.W. Ziller, Inorg. Chem. 28 (1989) 4308e4309.
[39] G.R. Giesbrecht, J.C. Gordon, D.L. Clark, B.L. Scott, J.G. Watkin, K.J. Young, Inorg.
Chem. 41 (2002) 6372e6379.
[19] C.P. Casey, J.A. Tonge, M.A. Fagan, J. Organomet. Chem. 663 (2002) 91e97.
[20] C.P. Casey, J.A. Tunge, T.Y. Lee, M.A. Fagan, J. Am. Chem. Soc. 125 (2003)
2641e2651.
[40] W.J. Evans, M.A. Ansari, S.I. Khan, Organometallics 14 (1995) 558e560.
[41] J. Carretas, J. Branco, J. Marcalo, P. Isolani, A. Domingos, A.P. de Matos, J. Alloys
Compd. 323 (2001) 169e172.
[21] C.J. Burns, R.A. Andersen, J. Am. Chem. Soc. 109 (1987) 915e917.
[22] W.J. Evans, J.M. Perotti, J.C. Brady, J.W. Ziller, J. Am. Chem. Soc. 125 (2003)
5204e5212.
[23] H. Schumann, A. Heim, S. Schutte, D. Lentz, Z. Anorg. Allg. Chem. 632 (2006)
1939e1942.
[42] M.L. Cole, G.B. Deacon, P.C. Junk, K.M. Proctor, J.L. Scott, C.R. Strauss, Eur. J.
Inorg. Chem. (2005) 4138e4144.
[43] K. Mashima, H. Sugiyama, A. Nakamura, J. Chem. Soc., Chem. Commun. (1994)
1581e1582.
[44] N.S. Emelyanova, A.A. Trifonov, L.N. Zakharov, A.F. Shestakov, Y.T. Struchkov,
M.N. Bochkarev, J. Organomet. Chem. 540 (1997) 1e6.
[45] H.J. Reich, WINDNMR, Vers. 7.1.13 [Computer Program], Madison, WI, 2008.
[24] H. Schumann, S. Schutte, H.J. Kroth, D. Lentz, Angew. Chem. Int. Ed. Engl. 43
(2004) 6208e6211.